The first round is complete, and the top two logos are #59, with 289 votes, and #1 (the "tiles" logo) with 253 votes. These logos will now proceed to the second round, which will last until January 31st. --Yair rand00:22, 1 January 2010 (UTC)[reply]
i do not want to come across as contumelious but please consider (re)casting your vote for the tile logo in m:Wiktionary/logo/refresh/voting as--besides using English--the book logo has a clear directionality of horizontal left-to-right, starkly contrasting with Arabic and Chinese, two of the six official UN languages. As such, the tile logo is the only translingual choice left and it was also elected in m:Wiktionary/logo/archive-vote-4. Beyond this, several Wiktionaries, especially Dutch, have commented that this vote is being forced on the rest of us by the English Wiktionary. A tile logo outcome would bring back harmony between the Wiktionaries in a way that book logo would not because many of us boycotting the vote see the tile logo as the only legitimate choice.
a) tile logo is ugly as hell b) it has a big red Latin letter W in the middle. Why not Cyrillic, Arabic or Armenian? According to you it too must be considered Anglo-centric. c) no one is going to use a microscope to zoom in and try read stuff on the book logo, to see if its English, left, right or whatever. d) you should not canvass votes for your preferred variant: it's unethical. --Vahagn Petrosyan06:15, 2 January 2010 (UTC)[reply]
Well, that big red Latin letter W in the middle is actually a variable. It displays differently depending on the language of the wikitionary. See for example ko, lt, and zh. Armenian Wiktionary will display a big red Վ. --Stephen06:35, 2 January 2010 (UTC)[reply]
Hmm, indeed. On the other hand, the English wording "Wiktionary, the free dictionary" of the book logo too will be variable by language. I guess, we should drop the Anglo-centric argument and choose by design, each one according to his taste. --Vahagn Petrosyan07:12, 2 January 2010 (UTC)[reply]
How is it unethical, Vahagn Petrosyan, when well-informed users like yourself are completely unaware that Chinese texts bear no gaps on the left-hand or right-hand sides but, rather, on the bottom? Without any magnification it is abundantly clear that the text on the book logo has gaps on the right. It would have to be at least reflected horizontally for Arabic but it simply cannot work with Chinese. Besides, why are we ignoring the original vote and pushing our views on to the rest of the Wiktionaries in the first place? Warmest Regards, :)--thecurranSpeak your mindmy past06:59, 2 January 2010 (UTC)[reply]
Re the second part: "Following [the second round], each language Wiktionary will hold their own vote on whether to accept the winning logo. In the event that less than 60% of the Wiktionaries approve of the logo, none of the Wiktionaries will use the logo." So, no one is pushing anyone. The Dutch can reject the book logo in their local vote. Re Chinese and Arabic: you must surely understand that for languages writing in the wrong direction the logo can be slightly redesigned in minutes. And yes, I said wrong direction :D --Vahagn Petrosyan07:12, 2 January 2010 (UTC)[reply]
If it can be done so easily, please do so within 24 hours for each of the six official UN languages. That gives you four hours apiece, which should be plenty if it takes minutes and since English is already done. While U are at it, can U or anyone else please explain why we are throwing away the original widely-accepted vote?
Of course I did only a horizontal flip, nothing else. With a little more time and professionalism the text on the page can be made look like Arabic or Chinese. And even if we leave it as it is, it's still more multi-national than the tile-logo: why are there only 9 tiles? Why are Cyrillic, Latin, Arabic, Hebrew, Greek, Chinese, Korean, that other hieroglyph and that stupid face more important than other scripts? I want Armenian, Georgian, Syriac, Cuneiform, Avestan, Pahlavi and Runes there. But I don't write to others agitating for my favorite logo: you should stop too. This is my main point. --Vahagn Petrosyan08:43, 2 January 2010 (UTC)[reply]
Regarding the proposed book logo, the left page will have to be redone or at least smeared a bit. Prompted by Yair rand, I did a little research and found that the proposed logo includes a barely readable page from the Macmillan Dictionary for Children, which is definitely under copyright. So there will be a bit of work ahead of us if the book logo wins, but any anglocentrism will hopefully be gone by the time Wiktionary deploys a new logo. – Minh Nguyễn (talk, contribs) 03:58, 18 January 2010 (UTC)[reply]
That "stupid face" is the Japanese Katakana character for si, シ, and that "other hieroglyph" is Devanagari for śa, श, as is used by several languages in South Asia. The central character is meant to be Armenian on the Armenian Wiktionary, etc. This is clearly stated on m:Wiktionary/logo/archive-vote-4. Until there is notice of such caveats for the book logo, it is most decidedly not translingual and it is important that Wiktionary admins know this. i have already stopped alerting them on their talk pages but i see no reason not to continue. i would sincerely like at least one person to explain why we have restarted the old vote. Warmest Regards, :)--thecurranSpeak your mindmy past09:17, 2 January 2010 (UTC)[reply]
The whole point of the logo vote is to find out which logo people prefer, not which one they were forced to vote for. Stop canvassing, please. L☺g☺maniac☃15:21, 2 January 2010 (UTC)[reply]
I have no objection to "canvassing" of this sort, since the editor has no personal gain to obtain by raising the issue. On the other hand, I don't particularly care which logo we use. bd2412T03:34, 3 January 2010 (UTC)[reply]
I don't see anywhere that says "YOU MUST VOTE FOR A NEW LOGO OR ELSE" or anything suchly demanding. I'm sorry that the other Wiktionaries feel the vote has been forced upon them. L☺g☺maniac☃00:57, 3 January 2010 (UTC)[reply]
The problem is that we already had a vote and it was accepted. Then some upstarts from en WT decided to start it all over again. It seems like a careless, monumental waste of resources. They did not even phrase it as a re-vote. They completely omitted references to the original. If we permit this, what is there to prevent another group from doing the same thing to this vote in another three years? Warmest Regards, :)--thecurranSpeak your mindmy past01:23, 3 January 2010 (UTC)[reply]
What do you think? This vote was started while two logos are being used about equally, one that basically nobody likes, and the other that there is significant opposition to. Under these circumstances, with 71 supporters for starting a new vote, a new vote was begun. If the situation was that there was a logo used almost universally among Wiktionaries, and there was no significant support for starting a new vote, of course the vote would not go through. Does that answer your question? --Yair rand01:43, 3 January 2010 (UTC)[reply]
I'd like to point out that the canvassing done by User:Thecurran was in fact most annoying overall, and has if anything prompted me to vote in the other direction. Even if I had had any respect for the tile logo to begin with, which I didn't, because my main opinion of it the first time was that it made the entire project look like a dubious, childish game. --Neskayacontribs – talk?19:39, 3 January 2010 (UTC)[reply]
The proposed change has no effect on inclusion of entries.
I have created the vote to last only two weeks, for the vote's being merely cosmetic and not touching the substance of CFI in any way.
I realize there was no discussion before the vote, but went for it anyway, estimating the change should be wholly uncontroversial. Let's see whether I was right in this.
The vote starts on 8 January 2010, and the wording may be still adjusted. However, let's aim at good enough rather than perfect. --Dan Polansky15:45, 1 January 2010 (UTC)[reply]
The vote starts on 8 January 2009 and lasts one month.
The vote is based on an ongoing discussion that has largely proceeded.
While there is at least one person who disagrees with the proposal, I deem the proposal rather incontroversial, as it keeps the terms "male" and "female" in place in the names of the appendixes. Who does not like the use of "male" and "female", insisting that we should deal with grammatical gender rather than object or referent gender, should still, I estimate, accept that the result of the proposed renaming is no worse than the current situation. --Dan Polansky22:49, 1 January 2010 (UTC)[reply]
If there is still time, can I propose altering this vote to run along the lines of
Please vote on which pair of terms you would prefer to describe given names throughout wiktionary.
I'd rather leave the vote phrased the way it is, being specifically about the names of appendixes and not about the names of categories, and being about approval rather than preference. I admit that it is implied that the names of categories are also at stake, but that is not what the vote is about. It is an approval vote, meaning that a voter who gives support in the vote gives an approval to the renaming without saying that he actually prefers the renaming to the alternative with masculine and feminine; the voter thus says that the proposed renaming is good enough, is acceptable. Put differently, it implements the idea that a person who finds the renaming acceptable (can live with it) while he finds masculine and feminine the best possible options approves the renaming anyway, so that we avoid the block or deadlock resulting from the use of 70-75% majority voting scheme to non-constitutional changes in policy, in which a plain majority-50%, or a stable majority-60% would be more appropriate. The renaming option that I propose for approval is the one that has so far gained a plain majority of supporters. The voter can approve in the "I can live with it" way by abstaing, that is, by avoiding the oppose option. --Dan Polansky11:29, 6 January 2010 (UTC)[reply]
In this case, as my vote is merely a preferential one, it is not trying to change existing policy (of which we have none on this matter), merely clarify opinion. I see no reason why even the preference that wins by one vote would not be acceptable, so long as that is made clear from the offset (this is all complete bikeshedding anyway). The vote is currently tangential, there is no opposition to renaming the appendices, there is only disagreement as to the preferred title. I think it sets bad precedence if we are to vote on every time we rename a group of appendices, and would prefer that the underlying issue is cleaned up instead of ignored. That said, I could just open the other vote on a similar topic which seems to be more what I want, and then we could let this vote run to completion (and presumably pass, but if it fails, then what?). Conrad.Irwin11:43, 6 January 2010 (UTC)[reply]
I admit that it is not really a policy vote but an executive vote; it concerns the executive branch of Wiktionary government rather than the legislative branch. But the choice of the name of the appendixes has generated a lot of discussion and disagreement, and is controversial enough that no one up to now dared or bothered to rename the appendixes to the name that was so far informally preferred by the majority.
The preferences of the voters should be clear from the vote even with the current phrasing, as each voter will (a) enthusiastically support the vote by voting "support", or (b) vote "support" with a comment to the effect of "I prefer 'masculine' but I can live with 'male'", (c) vote "oppose", (d) vote "abstain" with a comment to the effect of "I prefer 'masculine' but I can live with 'male'", or (e) vote "abstain" without a comment.
It seems that a vote is needed because an informal consensus has not been achieved in the previous discussions. Votes do stand in a contradistiction to an informal consensus, and there is not way around this fact, not even by raising the bar to 70-75% majority voting scheme. The idea that every decision concerning a wiki should be made by consensus is flawed I think. --Dan Polansky12:35, 6 January 2010 (UTC)[reply]
Wiktionary as a mathematical lexicon
Well, maybe there was already a discussion here about "Wiktionary as a lexicon", in general; in print, it is common to distinguish between those two.
Though, I have not found even a single source in the net that translates well between concepts, terms and phrases of mathematical languages in different spoken languages; trying to read a math article in Deutsch, for example, I find myself going to Wikipedia (DE) again and again to see what a term means in English (same with Hebrew).
Wiktionary can solve this, if we allow entries for mathematical terms and phrases. I'll give here a few examples:
degree. This example is more or less already done well in Wiktionary: degree can be translated (arguably) to דרגה in Hebrew and to Grad or to Valenz in German, when in Graph Theory context; it can be translated to מעלה in Hebrew (and still to Grad in German, but not to Valenz) when in Polynomials context.
More ambiguous are the terms regarding different types of convergence or continuity. Here are a few examples:
a ring in common language is translated to טבעת in Hebrew, but to חוג in an algebraic context; while "חוג" itself is translated back to English as class or group in everyday context. group is translated in everyday context to קבוצה, which in mathematical context is translated back to a set; but in mathematical context a group would be translated to חבורה, which in everyday context will probably be translated to a gang, or a clique, which also has a completely different mathematical meaning (which will be translated in that context to קליקה in Hebrew, and so on). A class in a mathematical context would be מחלקה in Hebrew, which will be translated in everyday context back to department or division (which - again - means something completely different).
Even general fields of interest have really different names in different languages, that cannot always be literally translated; a few examples:
set theory is תורת הקבוצות which is quite similar ("the theory of sets"), but in German it is Mengenlehre, which means more or less "the doctrine of sets", but a "set" here is Menge, which means "quantity" in a general context.
One might also find a lot of theorems which have different names in different languages, leaving no way of literal translation. For example:
And, at last, there are some common expressions and phrases used in math, which are not literally translated from language to another. I can think of a few examples in Hebrew and in English, but in some cases I don't even know the proper way of saying this in German. Examples:
to prove by contradiction, which is sometimes called Reductio ad Absurdum, is להוכיח בשלילה. It gets complicated when you try to translate conjugations of this (for example, "נניח בשלילה" - something like "let us negatively assume", which is not a correct usage)
and so on.
The list can be really wrong, but the important thing here is my general question: do you think that Wiktionary should do that job? If the answer is yes, I am ready to make the effort, and start a "Wiktioary Project" dealing with this. If the answer is no, I believe I should establish a new platform that will handle this, since I find it crucial.
Yes, Wiktionary functions as a lexicon for all areas of knowledge in all languages. Fell free to add missing mathematical senses of words (marked with appropriate context labels) and a properly glossed translation. It would be the best to focus on the normal entries in the main namespace rather than on a specialized appendix (although the approach of a giant comparison table has its own merits) because that's where most of the people will look it up. Starting up a wikiproject might be an overkill since this project is seriously undermanned, but that might even work given that there are several mathematically inclined regulars around here. --Ivan Štambuk16:13, 2 January 2010 (UTC)[reply]
It is my pleasure :) You can help by letting people you know know about this project, and where it is located at the moment. In the meanwhile, I'll work slowly on it. Peleg19:45, 6 January 2010 (UTC)[reply]
Hidden categories
Is there any sort of logic to which categories are hidden or not? For example, requests for deletion. IMO categories that are related to maintenance and are not lexical or topical should be hidden. For example, we are very inconsistent on [[Category:<langname> words needing attention]] if they are hidden or not. Mglovesfun (talk) 08:49, 5 January 2010 (UTC)[reply]
Support the proposal categories that are related to maintenance and are not lexical or topical should be hidden, including [[Category:<langname> words needing attention]]. --Daniel.11:16, 5 January 2010 (UTC)[reply]
Of course, only regular editors will see hidden categories (and only some regular editors). So the idea i think is to hide any category that we don't want readers (who are not usually editors) to see in an entry. The "Requests for" categories (except deletion, verification, autoformat, and perhaps a coupla others) I think should be visible, so that even such readers can see that something should be added, and maybe they'll add it. No harm in it, and it can certainly help the entries, especially (but not only) for languages in which no regular editors are proficient. Some of the other categories perhaps should be hidden, though, including "Languagename words needing attention" and "Translations to be checked (Languagename)". Just by way of comparison, enWP hides maintenance categories except for its stub categories and its "uncategorized" categories.—msh210℠17:01, 5 January 2010 (UTC)[reply]
From what I see on current categorization scheme, number is a numerical symbol or group of such symbols (for instance, the contents of [[Category:Khmer numbers]] and [[Category:Japanese numbers]]), numeral is a numerical word, when not considered of other part of speech (such as [[Category:English numerals]] which doesn't contain the adjective second). Finally, ordinal number and cardinal number are topical categories that specify types of numerical words. However, this scheme is not exactly suitable for Portuguese, because "cardinal" and "ordinal" are distinct grammatical classes in this language — that is, both ordinal numerals and cardinal numerals have a set of different characteristics from each other and from other parts of speech; these characteristics include position in sentences, presence or absence of a distinct feminine form and pluralization. --Daniel.16:19, 7 January 2010 (UTC)[reply]
Number is polysemic and refers to (a) the number of pieces in a group (or elements of a set) such as the number of * in "*****", (b) a numeral such as "345", "two hundred" or "fifth". Numeral is polysemic too, and includes "345", "two hundred", "1st" and "fifth". The decadic numeral "15", the numeral "fifteen" and the binary numeral "1111" all denote the same number in the sense (a). A numeral is a syntactic object, while a number is a semantic object, a meaning of a numeral.
Not only Portuguese but also English cardinal numerals and ordinal numerals are grammatically or syntactically distinct: two (cardinal), second (ordinal); three, third; four, fourth; etc.
The term "ordinal number" is ambiguous and refers to (i) ordinal numeral such as "fifth", and (ii) the set-theoretic or order-theoretic concept that includes such individuals as omega, omega + 1, and epsilon zero.
This acount is still simplified in that it omits further meanings of "number" and "numeral".
In any case, the term "ordinal number" means in one of its senses the same as the term "ordinal numeral", but for this purpose, "ordinal numeral" is less ambiguous. However, "ordinal number" is possibly more commonly used than "ordinal numeral" in English grammar.
Addressing several comments above: The Category:Portuguese numerals (and the like for other languages) is intended to be the POS category for all the numerals. The problem with categorizing Cardinals and Ordinals as parts of speech is that, in many languages, not all the cardinal numbers function as numerals grammatically. In English, for example, hundred is a cardinal number, but it is not a numeral; it is a noun. Similar situations occur in Spanish and Latin, and presumably in other related languages. In most modern romance languages, the ordinal numbers are functional adjectives, not numerals, despite being numeric and relating to a cardinal. So, the compromise we currently use is to have the Category:pt:Cardinal numbers and Category:pt:Ordinal numbers, so that we can list all those words in one place regardless of the part of speech it actually is. These categories also allow us to include nonfinite cardinals, such as aleph-null within a subcategory of mathematics, even though the word does not function grammatically as a numeral. To summarize, I see no advantage to the proposed change, but see several logistical headaches, such as having no category for a language that includes all the cardinal numbers because they would be sorted by grammatical function instead of by topic. --EncycloPetey03:45, 8 January 2010 (UTC)[reply]
Seeking examples of types of flawed definitions
I am interested in accumulating examples of types of flaws in definitions for purposes of Wiktionary:Definitions, Wiktionary:Improving definitions, or some associated page(s). Examples that come to mind are: unsatisfactory technical definitions, unnecessary hyperspecialization, and "An X is when..." definitions (of "X").
The associated talk pages or the body of the Wiktionary pages would be good places for such things.
These Wiktionary pages are intended to provide something more comprehensive than Help:Writing definitions. They would allow that page to be more explicitly simplified for the benefit of new users. Ideas about and contributions to these pages are desired. At this stage, additions are preferred to deletions. Restructuring and massive revision are likely if the effort does not peter out. DCDuringTALK * Holiday Greetings! 16:07, 5 January 2010 (UTC)[reply]
Much needed, but spreading this info out across a bunch of different pages is not a recipe for success IMO. Given our limited editorial resources, could we concentrate this at WT:STYLE until/unless the volume of data becomes unmanageable? -- Visviva17:18, 5 January 2010 (UTC)[reply]
I didn't realize we had such a thing. It looks good. I have been reading Landau and realized that all dictionaries have a style manual of some considerable length. I doubt that we could get a copy of a modern one, but perhaps someone has some headings that would be a useful partial model. I will attempt to not duplicate or be inconsistent with WT:STYLE.
I was intending "Improving" to be narrow in focus. It is to provide the rationale and structure for checklist-type tasks that were less format-oriented than many of our current cleanup lists and more aimed at laying the groundwork for revising definitions. I got started writing it and realized that I was not close to ready. That is when I turned to the more general "Definitions".
Help:Writing definitions needs to be simplified so that it would be actually read by a newbie will waiting for his/her block to expire. Simplification and updating might be good for all of our Help. We need more contributors who can help us update and expand Wiktionary, including English definitions. DCDuringTALK * Holiday Greetings! 22:31, 5 January 2010 (UTC)[reply]
In some cases we do that (using a single synonym) to avoid recopying the first definition, esp. when one term is more common than the others. Mglovesfun (talk) 11:16, 13 January 2010 (UTC)[reply]
Single-synonym definitions sometimes seem appropriate for words that are obsolete. But if the defining synonym is polysemous, it would seem necessary to say which sense applies, in which case a full definition seems necessary. Multiple (>3) synonym definitions are especially suspect to me, especially when the defining words are polysemous. It may be possible for a native speaker to locate the common sense among the synonyms, but I thought we were trying to serve other types of users too. DCDuringTALK12:21, 13 January 2010 (UTC)[reply]
Proposed Wikisaurus style changes
Wikisaurus doesn't really put up a good appearance, not having a small logo like most Wikiprojects on WP, or even a real main page like other Wikimedia sub-projects, Wikijunior and Wikimedia cookbook. Also, the Wikisaurus doesn't have much of a style in it's entries. Thus, I propose the following:
Using the bookglobe logo designed by commons:User:Ephemeronium as the Wikisaurus logo, to be used in Template:Wikisaurus-link and external interwiki links to Wikisaurus. (To the best of my knowledge, no other Wiktionary has a logo for Wikisaurus, so this doesn't have to be a huge multiple-Wiktionary issue.)
Changing the templates {{ws header}} and {{ws}} to the designs shown, again, on this page
If there is consensus for these changes, what are the chances this could be done without the usual mess of bureaucracy (read: votes) which wastes a huge amount of time? --Yair rand19:25, 5 January 2010 (UTC)[reply]
Separating a portal page from a project page for Wikisaurus could be okay, but I really do not see a need for doing so. I don't believe creating a flashy portal page will help us (or help me :p) build Wikisaurus. The project page "Wiktionary:Wikisaurus" contains the essential information for starting contributing to Wikisaurus, all in one page.
I disagree with the proposed style: User:Yair_rand/WS-main. While I admit that it looks superficially nice, it sacrifices the function of Wiktionary:Wikisaurus to flashiness. Instead of organizing information by sections, it uses a tabular layout, which makes it harder to find information, for me anyway. It distracts the attention by excessive flashiness.
Wikisaurus entries: I disagree with making the changes to appearance of Wikisaurus entries that are shown in User:Yair_rand/WS-main, including the putting of a small logo at every bullet, which I deem wholly non-functional and unduly flashy.
Logo: I still think that Wikisaurus does not need any logo, being a namespace and subproject of Wiktionary rather than a standalone project. The problems of building Wikisaurus are not tied to a logo or appearance; someone has to do the real work of finding the best headwords and term and concept structures, and idetifying the semantic relations between words—the one that does not consist of setting up flashy CSS-styled boxes.
Discussions: I disagree that discussions about Wikisaurus should be channelled to a talk page rather than to Beer Parlour, based on my experience with trying to find out information in
Wiktionary:Thesaurus considerations -- starting in 2002 and 2003, getting more traffic in 2004, with most discussion ended by the end of 2006
Wiktionary:Wikisaurus/Improvements 1 -- created in February 2005, and stopped immediately; a surge of activity appeared in July 2008
Wiktionary:Wikisaurus/Improvements 2
Searching in Beer Parlour for Wikisaurus-related discussions is fairly easy, using the keyword "Wikisaurus" in the namespace "Wiktionary:".
Wikisaurus-link: I oppose the use of {{Wikisaurus-link}}; I encourage to link to Wikisaurus through "Synonyms" section, as (a) Wikisaurus is not a standalone project like Wikipedia or Commons, and (b) "Synonyms" section is the one where people come looking for synonyms. --Dan Polansky10:55, 6 January 2010 (UTC)[reply]
I was not in any way advocating the use of {{Wikisaurus-link}} (though I do use it myself sometimes). The main page design on User:Yair rand/WS-main is, well, not really supposed to be useful for finding information itself, the current Wiktionary:Wikisaurus is far better for that, which is why it's linked to from the my design. What a main page is supposed to do is really show the reader a nice design, give a bit of a "visual identity", show a short outline and a couple of useful links, and showcase either a "featured" piece of content or some new content so that the reader can tell what it's supposed to look like, and show a bit of what people can do to help. As for the bullet, maybe that was a bad idea, but I still like having the header be more than just a block of text and a search box. --Yair rand18:15, 6 January 2010 (UTC)[reply]
More Integration with Wiktionary
I think that Wikisaurus should be more integrated with Wiktionary, not less. In my opinion, the best scheme would be to make the regular Wiktionary entries act as a index to a "keyless" Wikisuarus entry, like in Roget's International Thesaurus by Harper-Collins. (RIT uses numbers for their thesaurus entries but we can hide those numbers within the software.)
For example, the regular Wiktionary entry for immigrant would have a Wikisaurus index section like:
WIKISAURUS FOR Immigrant
migrant
incomer
citizen
settler
newcomer
Clicking on "migrant" would bring up this Wikisaurus list:
Noun
migrant
migrator
trekker
immigrant
wetback (informal)
etc.
See Also
Traveler
Clicking on "immigrant" would bring you back to the regular Wiktionary entry for immigrant. Clicking on "trekker" would bring you to the Wiktionary entry for trekker, which would have a WIKISAURUS FOR Trekker index (which, in turn, would include the entry for "migrant", among possibly others) and so forth.
Clicking on "incomer" in the original WIKISAURUS FOR Immigrant entry would bring up:
Noun
incomer
entrant
visitor
immigrant
etc.
Clicking on "Traveler" in See Also would bring up a super-index of:
Nouns
traveler
wanderer
vagabond
nomad
migrant
etc
See Also
Places
Clicking on any of these would bring up the appropriate lowest-level index. Clicking on "Places" in See Also would bring up a super-super-index of:
Space
Location
Displacement
Traveler
etc.
Software (probably a template) would have to be developed so that, when a Wikisausus index is added to or removed from a word entry in Wiktionary, the software would automatically change the Wikisausus page. We would need editor software to change the order of the Wikisaurus word entries. The first entry would be the index tag (in the first case above, "migrant") of the reguler Wiktionary page. We would also have to edit the "See Also" entries and the usage notes (like "informal").
This is just an idea and I apologize if it's been proposed and slammed before. It is a big change but it does seem to be a method for including virtually all words in Wiktionary in one or multiple Wiksaurus entries. I put it into a separate sub-section because I think it relates to the Proposed Wikisaurus style changes but it's a somewhat different line of thought. --RoyGoldsmith14:44, 9 January 2010 (UTC)[reply]
It doesn't, per se. But after you reach the Thesaurus (by whatever means), the contents of the thesaurus would be (I hope) more organized. Rather than having an alphabetic list of terms (some of which overlap -- see copulate, sexual activity and sexual intercource, for example) you would be presented with a number of highly-abstact concepts: The Body, The Senses, Feelings, Places, etc.
Click on, say, The Body, and you would be presented with another list of more-refined conceptual terms: Birth, The Human Body, Hair, Clothing, Nutrition, etc. Click on Birth and you would get: NOUNS birth; VERBS be born, give birth; ABJECTIVES born. Click on (say) give birth and you would be presented with the actual words that relate to the concept of giving birth: give birth, bear, have young, have a baby, bear a child, whelp, litter, labor, etc. Click on one of these and you would be redirected to the actual Wiktionary entry. This entry would then have a reverse link to the lowest level of Wikisaurus.
Thus there would now be two main entry points for Wikisaursus. The first is from the top down, through the current way or your portal or something else entirely. The second would be from the bottom up, through the entry in Wiktionary (via the new "WIKISAURUS FOR xxx" section or something akin to it) up through a list of highly related terms (give birth, have young, etc.) and from there up to even more abstract concepts: for example, Birth in general and from there to The Body.
That's all I'm saying. You gave a new method for accessing Wikisaurus from the top; I'm giving a new method of accessing Wikisaurus from the bottom. (Of course, my method changes the entire structure of Wikisaurus and is therefore much harder to implement.) If you would prefer to have my proposal treated totally separate from yours, simply change the header "=== More Integration with Wiktionary ===" to "== More Integration with Wiktionary ==" (two equal signs on either side instead of three) at the top of this subsection and, magically, it will be transformed into two, independent proposals. --RoyGoldsmith15:08, 10 January 2010 (UTC)[reply]
Preposition forms and prepositional phrases.
A number of languages have what are called "inflected prepositions". The details vary from language to language, but the general idea is that certain preposition+pronoun combinations are expressed by inflecting the preposition to incorporate the pronoun. There are a few minor examples of this in Romance languages (such as Spanish conmigo(“with me”), which historically results from roughly con + mí + con, but nowadays amounts to a special form of the preposition con that incorporates the first-person singular personal pronoun mí), but in some languages, such as certain Celtic and Semitic languages, it's very pervasive: for example, all Hebrew prepositions are inflected in this way, though some prepositions are defective (meaning they can't take certain objects) and/or suppletive (meaning that not all of their forms have descended from the same ancestral preposition). Incidentally, Hebrew doesn't even have stand-alone object forms for the animate personal pronouns, since they're always expressed via an inflected form of whatever they're the object of.
From one perspective, these are forms of prepositions; but from another perspective, they are complete prepositional phrases (since the preposition's object is built-in), so they don't behave quite like English's transitive prepositions. Header-wise, this makes for a bit of a conundrum. As I see it, we have a few options:
===Preposition===. This is what I've been using.
===Adjective=== and ===Adverb=== (with identical or nearly-identical information under each). This accords with how we've been handling prepositional phrases in English.
===Idiom=== or ===Phrase===. Either of these would be misleading, since these are one-word forms; even in cases where the pronoun was a separate word in some ancestor language, it's not one now.
===Contraction===. A bit better, but still misleading, since conmigo is longer than *con mí, and since in Hebrew it's not clear what they're contractions of (since the personal pronouns simply don't have standalone object forms).
===Preposition form=== (currently not sanctioned by ELE). This is the most explicit and the most precise, but I'm not sure how much clearer it actually is for an English-speaker, even one who speaks an affected language. And we avoid headers like "Noun form" and "Verb form", so this would introduce some inconsistency.
===Prepositional phrase=== (currently not sanctioned by ELE). This is still a bit misleading, what with the word "phrase" in there, but overall, I think it would work pretty well. It could also be used for other kinds of prepositional phrases, which would be a bit clearer and more accurate than our current ===Adjective===/===Adverb=== approach (especially for languages where true adjectives inflect for number and such). Admittedly, there's still a bit of inconsistency, in that we avoid "Noun phrase" and so on, but I think it's clear that a noun phrase is noun-like in a way that a prepositional phrase is not preposition-like, so hopefully that's O.K.
All in all, I think my preferred option is to start allowing ===Prepositional phrase===, and to use it for object-including preposition forms as well as for regular prepositional phrases; but what do y'all think?
I certainly see the advantage of Prepositional phrase as a PoS-type header for English: eliminating low-value duplication of senses in Adjective and Adverb sections at no additional (possibly even less) user confusion. Category:English prepositional phrases would make it relatively easy to rapidly transition to such a header. The simple nature of English prepositional phrases would make it easy to find omissions, too. DCDuringTALK * Holiday Greetings! 22:48, 5 January 2010 (UTC)[reply]
I, too, like the idea of "Prepositional phrase" for English. I'll keep out of the conversation w.r.t. Hebrew though, at least for now: having thought about it and discussed it (elsewhere), I'm really not sure what's best. Ruakh, if you have the time and inclination, how do Even Shoshan and his counterparts list לי?—msh210℠17:24, 6 January 2010 (UTC)[reply]
Aside from Wiktionaries and Wiktionary mirrors, I know of no dictionary, online or in print, in any language or between any pair of languages, that has entries for individual word-forms. Online dictionaries will sometimes just redirect you to the main entry for the lexeme, and print ones will sometimes do the print equivalent for forms that are extremely irregular or appear very far away in alphabetical order (like, between the entries for sommer and sommet, you might find "sommesseeêtre"), but so far as I know, our informative soft redirects are without precedent. So I think we're on our own for this. —RuakhTALK20:54, 6 January 2010 (UTC)[reply]
At least "prepositional phrase" is a term that seems to be used in English fairly consistently in virtually every vintage and level of grammar book. We should try to anticipate problems and make sure that we have supporting material (glossary definition, entry, category text, Appendix) of good quality to support the idea. We might discover problems as we try to prepare such material. One small problem I can anticipate would be with the "postpositions" (eg, "notwithstanding", "apart", aside") if they head any CFI-meeting phrases. It may also lead us to face the question of whether to analyze some terms (eg, with regard to) as "compound prepositions" (Quirk et al, en.wikt) vs. elements of layered prepositional phrases (perCGEL) and also such odd cases as to do with. DCDuringTALK * Holiday Greetings! 22:11, 6 January 2010 (UTC)[reply]
Postposition is an ELE-approved POS header, so I don't see any problem with having Postpositional phrase if we have Prepositional phrase. The only issue would be where to put such an (English) phrase, and we can, I think, cross that bridge when we come to it.—msh210℠17:52, 7 January 2010 (UTC)[reply]
We don't really want the Postposition header in English, whatever value it has for other languages. Few normal people notice or are bothered by "preposition" being a misnomer etymologically when applied to the few terms that are atypical in this regard. Stranded prepositions are always postpositive anyway, without the name being what upsets most of those who complain about such things. In any event, the position of a preposition relative to the rest of the phrase it is in is not the important grammatical feature of a preposition, nor is it unique to prepositions ("The" is prepositive.).
My point was only that, given that we don't want the Postposition header, any includable prepositional phrase that uses one of these oxymoronic normally-postpositioned prepositions may look funny and cause comment. DCDuringTALK19:03, 7 January 2010 (UTC)[reply]
O.K., strike my earlier comment. I totally should have checked before writing it. I'm sorry. I was totally wrong. It turns out that Even-Shoshan does include inflected forms for some of the really basic prepositions. In general, it doesn't give any definition at all for these, or only a definition like "רְאֵה מִן" ("see from"), but for לִי(li, “to me”) it actually goes all out for some reason, defining it as "אֵלַי, לְעַצְמִי וְכוּ׳" ("toward me, to myself &c."). For most of the preposition forms I looked at, it gives the POS as מ״י(M.Y., “prep.”); the only one for which it gives a different POS is לוֹ(lo), for which it gives מ״ג(M.G., “pron.”). I'm inclined to regard that inconsistency as a mistake, though. —RuakhTALK02:05, 7 January 2010 (UTC)[reply]
So, I don't understand why you object to using ===Contraction=== and then labelling the function on the inflection line or in the usage notes. The problem with ===Prepositional phrase=== is that it isn't a full description of the term, nor is it the functional part of speech. Prepositional phrases are prepositions plus an object, so a label that emphasizes only the prepoisiton part is misleading and focusses on what is usually the minor particle in the combination. Also, prepositional phrases tend to function as adverbs or adjectives, and not a prepositions, so the label doesn't describe the word's function either. For the aforementioned Romance languages, I prefer ===Contraction=== for these situations, since that at least avoids giving the impression that the term somehow is a preposition. That doesn't mean this would necessarily be ideal for Hebrew or for other languages, but I haven't seen a convincing reason for ===Prepositional phrase===, and that was a header we tried to deprecate before. --EncycloPetey03:35, 8 January 2010 (UTC)[reply]
I thought I explained why I object to using "contraction": it's not accurate. Spanish conmigo is not exactly a "contraction" of *con mí, because it's actually longer than said; and Hebrew לי is not a "contraction" of anything, any more than English babies is.
"Prepositional phrase" focuses on the preposition part because that's the part that determines the grammar, no? I don't understand your statement that "prepositional phrases tend to function as adverbs or adjectives, and not a prepositions, so the label doesn't describe the word's function either". I agree that "preposition" doesn't describe the function of a prepositional phrase, but surely "prepositional phrase" does?
Re: "For the aforementioned Romance languages, I prefer ===Contraction=== for these situations, since that at least avoids giving the impression that the term somehow is a preposition": I don't understand. Are you saying that the header "prepositional phrase" gives the impression that the term somehow is a preposition?
BTW, I should say that while I'm not a big fan of the ===Contraction=== header for terms like conmigo, I do at least consider it to be an O.K. approach for those. It's not ideal, but it's not the end of the world. —RuakhTALK04:33, 8 January 2010 (UTC)[reply]
If I understand correctly, then the Hebrew term you're concerned with is not actually a prepositional phrase, since it's written as a single word. Phrases are inherently multi-word, so a label of "X phrase" would be inaccurate on that count alone. And no, "prepositional phrase" does not describe the function because prepositional phrases can have many different functions. Most often, they function as an adjective or adverb, but they can appear in other functions as well, such as an interjection or even as the subject of a sentence. --EncycloPetey17:22, 9 January 2010 (UTC)[reply]
Re: "Phrases are inherently multi-word, so a label of 'X phrase' would be inaccurate on that count alone": Yes, as I said.
Re: "And no, 'prepositional phrase' does not describe the function because prepositional phrases can have many different functions. Most often, they function as an adjective or adverb, but they can appear in other functions as well, such as an interjection or even as the subject of a sentence": Yes. That's why it's problematic to try to cover them in "Adjective" or "Adverb" sections: they're not really adjectives or adverbs, and their range of functions is not the same as that of adjectives or adverbs. Hence my suggestion that we give them their own POS header.
So, you want to use a POS tag that describes neither the structure nor the function? I don't understand how that could be a good idea. --EncycloPetey23:56, 9 January 2010 (UTC)[reply]
If by "neither the structure nor the function" you mean "both the structure and the function", then yes. ;-)
It may not describe the structure perfectly, but it describes it better than anything else besides "preposition form", which you don't seem to be advocating. And it does describe the function perfectly. You seem to be contradicting yourself, simultaneously arguing (1) that prepositional phrases have their own set of functions, not exactly the same as those of adjectives and adverbs and (2) that "prepositional phrase" doesn't describe their functions.
You are misunderstanding, so let me try to explain it this way: Why not use the POS "Word", which would "perfectly" describe the function? The answer is that word is too broad a description encompassing too many possible functions. Likewise, a "prepositional phrase" has many possible functions, so that such a label does nothing to distinguish which possible functions an entry might have. This is why we use Adverb, Adjective, etc. as POS headers for prepositional phrases, since such phrases as a group could have any of quite a number of functions, but individually they have one or two of those functions only. So when I said that the header you are proposing describes neither the structure nor the function, I meant exactly that, and was not using the "opposite-speak" you have credited me with. --EncycloPetey05:03, 10 January 2010 (UTC)[reply]
Thanks; if that's true, then it makes sense as a reason. But do you have any evidence for your statement that any given prepositional phrase is only used in a few ways? That's not obvious to me. (In Wikipedia terms, [citation needed]. :-) —RuakhTALK05:11, 10 January 2010 (UTC)[reply]
Perhaps I should clarify that those which are used in more than one way have a different meaning when so used, so there would be separate senses functioning in different ways (as diffeent POSes) and with different meanings. Do you have evidence that individual prepositional phrases can regularly be used in both an adjectival and adverbial sense? I have yet to see more than a tiny handful that can. --EncycloPetey07:42, 10 January 2010 (UTC)[reply]
The use of Adjective and Adverb headers, jointly or separately, for prepositional phrases is not perfectly accurate and requires duplication of essentially the same meanings.
The duplication of content argument has been used as part of the justification for excluding PoS sections for attributive use of nouns and common noun use of proper nouns. I would extend it in English to exclude some similar duplication in the case of -ing forms in English.
In English the accuracy issue arises from prepositional phrases not normally meeting tests for the adjective and adverb PoSs. For example, they normally do not accept modification by "too" and "very", sometimes accept modification by "right" and "straight", and cannot serve as the complement of "become".
One departure from the pattern is instructive. A prepositional phrase modified by "very" or "too" prefers to be distinguished by its stress pattern or orthographically by quotes or hyphens. "It was very in-the-moment". "He was speaking in-the-moment." "It was a very in-the-moment speech." DCDuringTALK12:00, 10 January 2010 (UTC)[reply]
International Wikisaurus
I'm not really sure about the state of Wikisaurus, but I'm sensing that it's not in any way interconnected across language versions... which is too bad, because synonyms have to be in the same language as the given term, and therefore the words should be exactly the same way under all the wikis. There has to be some way to take advantage of that -- otherwise, you end up doing the same work for the gazillion language version there are, or some people are missing out.
By the way, I'd really like other language synonyms in the wikisaurus... — This unsigned comment was added by MirekDve (talk • contribs).
Wikisaurus is not multilingual, meaning that the English language Wikisaurus only included entries for English words. The reason for this is, as you pointed out, we would end up doing the same work for the gazillion language version there are, with no benefits, because anyone could easily look up the word in the Wikisaurus of that language. --Yair rand02:17, 7 January 2010 (UTC)[reply]
I assume that the subject of this thread is "Multilingual Wikisaurus", meaning whether Wikisaurus in English Wiktionary should contain non-English entries.
It has not yet been decided whether Wikisaurus should be multilingual. I support its being multilingual, but have so far not pushed the idea very far, as there has been some opposition, and I am trying to build up the English section of Wikisaurus first.
To be most useful, each Wikisaurus entry should include comments (in English) when needed, e.g. about how senses are slightly different between words. Also remember that a thesaurus is not only about synonyms, but should include all words readers are likely to need when reading the entry, all words which come to mind when addressing the subject (e.g. kennel in the entry for dog, look at Roget's thesaurus for other such examples). Therefore, the entry should be organized into several sections (with section titles in English). This means that a good thesaurus cannot be common to all wiktionaries, and that there is no reason to limit Wikisaurus to the English language. But the titles should always be in English: Wikisaurus entries are not about a specific word (whatever its language), but about words of some language relating to something. Good titles could be something like Wikisaurus:dog (English) or Wikisaurus:dog (Dutch). Lmaltier22:47, 9 January 2010 (UTC)[reply]
Wikisaurus entries stand for senses AKA meanings rather than terms; right.
Non-English entry titles in Wikisaurus can be designed using various approaches. The current approach is (a) that the headwords should be in the given language: "Wikisaurus:příbuzný". One alternative proposed was (b) that the language should be indicated using its code before the term: "Wikisaurus:cs:příbuzný". If the codes are disliked, (c) the disambiguation can proceed in natural language: "Wikisaurus:příbuzný (Czech)". And there is also the option of (d) using English headwords: "Wikisaurus:relative (Czech)".
I tend to favor the approach (a), given the current state of discussion. There are level-2 headings for "English" in Wikisaurus entries already in place for the cases where more languages occupy one headword. The approach (d) would cause some complications in linking between Wikisaurus entries using {{ws}}: in a list of hyponyms, the template links a hyponym not only to mainspace but also to Wikisaurus. So if there is the hyponym {{ws|ďítě}} of "příbuzný", the template is automatically looking for a Wikisaurus page entitled "dítě" rather than "child (Czech)". It would be necessary to tell explicitly to the template where to look for the Wikisaurus entry, like {{ws|dítě|child|ws=child (Czech)"}} or {{ws|dítě|child|ws=child|lang=cs}}. Alternatively, it could be assumed that the gloss (the second parameter of ws) matches exactly an entry headword, so it would suffice to enter {{ws|dítě|child|lang=cs}}, but this assumption would be made only if the language is not English. In any case, (a) can be tried immediately without further technical adjustments. --Dan Polansky10:37, 10 January 2010 (UTC)[reply]
Hi there all. I would like to ask the community what they think about me requesting for {{io}} to be delinked due to the size/amount of entries that have been made in Ido so far. I believe that it is starting to become more used on here, and that having it linked isn't necessary any longer because of the popularity of the language. Ido is very similar to Esperanto, and Esperanto is not linked, so I think that on that basis alone, {{io}} should be delinked. Thanks, Razorflame16:49, 7 January 2010 (UTC)[reply]
Usually, headers aren't supposed to be wikilinked, although that might not hold true here. Anyways, I'd like it to not be wikilinked, as that is what I am most used to. There probably isn't any disadvantage with doing it, and there doesn't seem to be any advantages to not doing it, so we might as well do it :) Razorflame01:22, 8 January 2010 (UTC)[reply]
Then don't link the headers. However, please do not remove the link from {{io}}, as that template is primarily linked there for Tbot in dealing with Translations sections. We decided long ago which languages should and should not be linked within the Translations sections, and our ISO templates have been set up accordingly. If people are using the ISO templates to insert headers, then that is fine, but that's not their primary function. Frankly, the commonness of a language on Wiktionary is not a consideration in deciding which language templates are linked. It's based on recognition of the language name by the general population and whether the name of the language sufficiently resembles the name of the country where the language originated or is principally spoken. Ido is a language I would never have heard of if it weren't for working on Wiktionary. --EncycloPetey03:28, 8 January 2010 (UTC)[reply]
This page is for matters of "policy" that span some class of entries. We have at least two places for similar matters relating to individual entries. If you were asking an informational question you could have taken this to WT:TR, inserting {{rft}} in the entry. If the question was really a request for action, you could take it to WT:RFC (cleanup), using {{rfc}}. HTH. DCDuringTALK18:58, 8 January 2010 (UTC)[reply]
I was considering that this was in adherence with guidelines, and if that was so, the guidelines needed to be changed since it was confusing. I guess I didn't know how to check whether it was complying (i.e. with which guideline). __meco19:21, 8 January 2010 (UTC)[reply]
Bing
Have you tried Bing, the new Microsoft search engine (http://www.bing.com)? It's really incredible. With default settings (for France), I tried a few French and English common words. Here are results I get for the Wiktionnaire:
Thanks for the info. This must have something to do with the country... But I would have expected similar results in other countries (with the local language dictionary instead of fr.wiktionary, of course). Lmaltier16:38, 9 January 2010 (UTC)[reply]
Formatting note 1 states “The year is always in bold face, the title of the work in italics or, for shorter works, quotation marks.” (my emboldenment). Really? I never see the use of quotation marks for this purpose, and it would be more consistent to allow only italics, per the de facto standard (note that the rest of the page uses only italics). Or is this meant to apply to works published as a part of a compilation (e.g., an essay in a multi-authored book or an article in a journal)? Could we get some clarification or revision of this point, please? †﴾(u):Raifʻhār(t):Doremítzwr﴿19:39, 9 January 2010 (UTC)[reply]
My understanding is that this clause, if you will, exists because that is the standard form used for citing poetry in essays, as shown by the MLA format (outlined here and in many other websites found in a typical Google search). Cdhaptomos19:43, 9 January 2010 (UTC)[reply]
That talks about quoting poetry in an essay, not citing it in a dictionary; lexicographical practice is likely to be very different. Does anyone know the OED’s practice in citing poetry? Or can someone offer a rationale for departing from our usual standard? I don’t see why we should have different formatting standards for quotations from poetry from those that hold for those from prose. †﴾(u):Raifʻhār(t):Doremítzwr﴿20:41, 9 January 2010 (UTC)[reply]
The is the norm for citing just about anywhere in English. Shorter works such as scientific articles, short poems, and newspaper articles should have their titles in quotes. Longer works, such as epic poems, journals, novels, and other books should have their titles italicized. This is the norm here, as well as on Wikipedia, because it's standard in most style and publication guides. --EncycloPetey21:17, 9 January 2010 (UTC)[reply]
This was totally unknown to me. Could someone please update the policy page to state this fact more explicitly? Cdhaptomos, do you care to do the honours? †﴾(u):Raifʻhār(t):Doremítzwr﴿22:41, 9 January 2010 (UTC)[reply]
I'm not sure this is accurate. Shorter works do usually appear in quotes, but that's because they are part of larger works, which appear in italics. "Article", Newspaper. "Scholarly article", Journal. "Poem", Anthology. Length doesn't factor into directly AFAICT. Consider for example Frankfurter's On Bullshit. It is considerably shorter than many journal articles, yet its title is always italicized, because it is published as a freestanding book.
I would tend to limit quotes-only to special cases where a very small work appears in isolation, such as broadside ballads, or poems that are known to have been written down long before they appeared in print. Usenet posts would also qualify, the de-facto-standard "Post title", Usenet format notwithstanding. -- Visviva03:34, 10 January 2010 (UTC)[reply]
This is my understanding also. Except that when citing poems, it's fairly common to do so in isolation, since they were usually first published in a magazine or elsewhere. Ƿidsiþ06:21, 10 January 2010 (UTC)[reply]
Most shorter poems are able to be recognized as separate works, regardless of publication in an anthology or whatever. Fair enough, you could put "Poem", as published in Anthology, but at the end of the day, a poem is a poem. Cdhaptomos18:56, 10 January 2010 (UTC)[reply]
My belief was pretty much as Visviva explains. I regard the citation information we give as the means of verifying the existence of the quoted extract, ideally in the place it first occurred; if the work in question does not occur independently, but rather only as part of a larger work, then that justifies the use of quotation marks, whereas italics should be used otherwise. Presumably, no one publishes individual poëms or scholarly articles, hence they always appear as parts of larger works only, hence they are always cited using quotation marks, not italics; the fact that the stubby On Bullshit and the epic Iliad are cited with italics, and not quotation marks, supports this. In no way would I recommend that we depart from established citational practice, only that we work out properly what the rational distinction being made is. †﴾(u):Raifʻhār(t):Doremítzwr﴿00:43, 11 January 2010 (UTC)[reply]
I've been doing some investigating, but unfortunately do not have access to a copy of the book from which I first learned this convention. The books I have been able to examine either give no complete set of rules (the Chicago Manual of Style is frustrating in this respect), or do tend to support the idea of quotes used for portions of a larger work (this is noted in Peyton Hurt's book Bibliography and Footnotes: A Style Manual for Students). --EncycloPetey01:13, 11 January 2010 (UTC)[reply]
I don;t expect to obtain a copy, in part because it was a school text whose appearance I would recognize, but whose title and author were of no interest to me at the time. So, I do not know the bibliographic information needed to track it down quickly. --EncycloPetey05:24, 17 January 2010 (UTC)[reply]
OK. Well, in the absence of the availability that authority which states the contrary, I believe we should amend our policy page to prescribe formatting in line with what Peyton Hurt states. †﴾(u):Raifʻhār(t):Doremítzwr﴿02:58, 18 January 2010 (UTC)[reply]
Titles for rhymes pages
Are there any norms for the titles of rhymes pages? For example Rhymes: -e (French) could be enormous, isn't it better to use that as an index page for Rhymes: -be and the other consonants we can put ahead of it? What about more syllables than just one? Mglovesfun (talk) 19:31, 10 January 2010 (UTC)[reply]
The rhymes are all named according to the IPA symbols beginning from the vowel (or diphthong) in the stressed syllable through to the end of the word. So, if there are a lot of French words ending in /e/, and with the stress on that final syllable, then yes there could be an enormous page as a result. You could conceiveably choose to arrange the contents of the page differently for French (in English we group according to the number of syllables), but I wouldn't recommend using a different page naming scheme. --EncycloPetey01:10, 11 January 2010 (UTC)[reply]
Technically, Rhymes in French are not stress-based: all meaningful words (i.e. minus prepositions, pronouns...) in -/e/ are considered to rhyme! The French tradition is different and "length"-based. A rhyme based on only a final vowel is "poor" based on a final vowel and the preceding sound, or a vowel+consonant coda is "sufficient", and one with three or more phonemes is "rich". Some school of poetry have at time held that poor rhymes were not valid (so that marée and allée are not considered to rhyme)! Circeus14:01, 11 January 2010 (UTC)[reply]
You're right; I forgot about the absence of stress in French. Perhaps the entries could use /e/ as a navigational page then, and French rhymes could be grouped according to sequence beginning from the first vowel in the word. Thus, the /e/ listing page would itself include only French words with a single terminal vowel sound. --EncycloPetey05:22, 17 January 2010 (UTC)[reply]
RFD and RFV request
I'm trying to archive these pages, it would greatly facilitate the process if when listing multiple terms for one discussion you used the seperate level three headings (even if all discussion happens only under one of them) instead of the comma-separated level two heading format. Additionally, when closing such discussions (striking the heading, and removing the {{rfv}} or {{rfd}} from the page), if you say one of the following "magic" words it will be picked up so that I don't have to manually intervene to tell it which template to use.
Common variations on the above are also matched, providing they are emboldened ('''RFVFailed''', '''rfd passed''' etc.) . This is really a request for things to "stay as they are" as the code for this detection was written based on what normally happens on those pages - certainly if it would be a hassle to do, don't bother, I can do it for you. Conrad.Irwin19:15, 11 January 2010 (UTC)[reply]
Reverse index
Would it make sense to create a reverse index by language? A reverse index (I don't know the proper English expression) is similar to the current index except it's sorted by a reverse order of the letters, starting with the final letter. This index may also serve as a rhyme dictionary for some of the languages. --Panda1023:02, 11 January 2010 (UTC)[reply]
For some languages, this is highly useful. Latin and Ancient Greek make use of such lists often, particularly by those working with inscriptions or old manuscripts where the beginning of a word has been damaged beyond recognition. --EncycloPetey05:19, 17 January 2010 (UTC)[reply]
Is "idiom" still a valid header?
Noting our to do list, if ===Idiom=== is no longer a valid header for English, why should it be for other languages? Idiom is not a part of speech in the same way that verb and noun are. Shouldn't we keep the categories <langname> idioms but remove the header? Mglovesfun (talk) 17:22, 12 January 2010 (UTC)[reply]
Each language could come to its own conclusion until such time as there is a clear benefit from PoS standardization. Is there any clear benefit from amending WT:ELE to remove it? I would like to hear from those active in other languages why it was useful so as to know whether it would be something to discourage or deprecate or whether it should be explicitly "legislated" by each language.
Several other permitted PoS headers are also not "parts of speech" (Proverb, Phrase, the 3 abbreviations headers, Contraction, Symbol, Letter, Number). Note that they are approximately equal in number to the number that are actual parts of speech. DCDuringTALK18:16, 12 January 2010 (UTC)[reply]
Oh I agree, I think "phrase" and "contraction" are only to be used when nothing else fits. Personally I'd always advocate replacing "Idiom" with "Phrase" at worst. Mglovesfun (talk) 10:26, 13 January 2010 (UTC)[reply]
From the standpoint of Portuguese language, I say the "Idiom" header is not necessary at all. The current practice of labelling English idioms as other parts of speech or as phrases also works well for Portuguese. --Daniel.04:46, 17 January 2010 (UTC)[reply]
Likewise for Latin. The are a few entries currently categorized as Latin idioms, but I don't see "Idiom" as a necessary or desirable PoS header in Latin. --EncycloPetey05:17, 17 January 2010 (UTC)[reply]
Earlier in the year (December 2009), I posted a request for Darkicebot to be unblocked. I was told that I should do the few test edits for my bot through my main account, which is what I have done. I have made all the forms of the verb frapar and batar using my bot code through my main account, and I was wondering if maybe my bot could run for the bot flag again now? Thanks, Razorflame08:19, 14 January 2010 (UTC)[reply]
I'd imagine you can run for bot flag as often as you like, the question is if there's no chance of you getting it, why bother? It will only hurt your feelings. Mglovesfun (talk) 12:59, 14 January 2010 (UTC)[reply]
Why do you say that there isn't a chance of me getting it if I've proven that I know the language? I updated the Ido verb conjugation charts to include all of the possible form-ofs for verbs, and I've written the program to be exactly like the one I used for Esperanto, so I honestly don't see what the big deal is? Razorflame13:02, 14 January 2010 (UTC)[reply]
At this point in time, it's obvious that the community doesn't have the trust in you to grant the bot flag, Razorflame. I would suggest a longer period of time doing manual edits and other things that would convince people that a bot flag would actually be used rather than misused. --Neskayacontribs - talk?09:30, 15 January 2010 (UTC)[reply]
By manual edits, you mean using the bot program through my main account, like I have been doing to prove that it wouldn't be misused? Razorflame09:32, 15 January 2010 (UTC)[reply]
It's easy, just make some high quality edits over a long period of time (Note that "lots" of edits doesn't help, "time" is the key and "quality" is important). Conrad.Irwin13:10, 15 January 2010 (UTC)[reply]
Italian plurals
After going about my business, editing Ido and Esperanto, I've noticed quite a few Italian plural entries for both Adjectives and nouns that do not include the |p in them. My question is this: Why aren't they included, and shouldn't we include them? All opinions are welcome :). Cheers, Razorflame12:50, 14 January 2010 (UTC)[reply]
It doesn't matter a great deal as the definition says […] Plural form of […] Note that on the French Wiktionary a decision was made not to put {{m|p}} (etc.) as it duplicates the definition line. Mglovesfun (talk) 12:58, 14 January 2010 (UTC)[reply]
All of our Spanish entries already automatically add the p, so I don't see why Italian shouldn't if we already use it for Spanish, yeah? Razorflame13:00, 14 January 2010 (UTC)[reply]
Or we could remove it from Spanish, or we could leave it alone. The fact that we do things one way for one language should not necessarily be a guide to how other languages are done. --EncycloPetey05:14, 17 January 2010 (UTC)[reply]
Yes, of course we can just leave it alone. I just thought that maybe people might think that it should be the same as the other languages, but I guess not. Cheers, Razorflame23:32, 21 January 2010 (UTC)[reply]
Pre-English section stuff
When most people search Wiktionary for a word, they're probably just looking for a bit of information on the English word. Right now, we have quite a few things that are before that section:
The "Table of Contents" box. Nobody uses this, and it gets new users really confused. People arrive at a page, and all they see is a huge block of text on the left, sometimes a WP link, nothing to help them find the information that they're looking for. I know that users can click "hide" to shrink it, but that's not clearly visible, and no one bothers to click it anyway. I think that either the Table of Contents box should be collapsed by default, or it should be removed entirely. (I don't know technically how either of these things could be done, but there's probably a way.)
The ==Translingual== header. Generally, nobody's looking for this. I really think that this should be placed below the ==English== section, before the other languages. (English > Translingual > Aari > Aasax, etc.)
Interwiki boxes, images, and other random pieces that are really language-specific and should be placed in one of the sections. I seem to recall someone proposing a bot to place these correctly, but I don't remember what came of that.
Should these be fixed, the information on the English word will be visible as soon as the page loads, right near the top, which is generally where people would expect it to be. --Yair rand22:41, 14 January 2010 (UTC)[reply]
I don't know why you think nobody uses the TOC, and certainly not why you say it's "nothing to help them find the information that they're looking for": a TOC is precisely something to help them find the info they seek. (Perhaps we should do an analysis of the feedback clicks, figure out what the best and worst pages had in common; for example, does a longer/short/nonexistent TOC affect, statistically significantly, the feedback?) I also don't know why you say nobody's looking for Translingual. The translingual definitions are used in English, so people interested only in English should be interested in them, too, as should those interested in any single language, which is why it's best first. (Also, imagine if we put Translingual in its alphabetical order at "T" among the languages (which is not what you proposed, Yair, but is how I recently saw a newbie "fix" a page): everyone looking for "Germany" at [[de]] would add it to the English section, not thinking to look further down the page.) As for the (e.g.) Spanish-specific boxes in the English section, yes, those need fixing.—msh210℠23:12, 14 January 2010 (UTC)[reply]
I agree that left-hand side ToC is a terrible waste of precious screen real estate. Have you tried right hand side ToC? We have had an effort to eliminate the barriers to making rhs ToC the default.
The material on the right-hand side generally doesn't seem too bad and sometimes in useful. Sometimes a picture constitutes an ostensive definition, superior to words. OTOH, I usually put all sister project boxes under the See also header, except for a link to a WP disambiguation page, especially for a Proper noun. DCDuringTALK01:06, 15 January 2010 (UTC)[reply]
Bequw has a smaller pedia box so that even the top-appearing WP dab links that I like could be less intrusive. I had dispensed with lhs ToC as soon as rhs became available. It seems to be about the best we can do with wiki-type software. At first it required Javascript. Hippietrail made it easier to select when he put it in WT:PREFS. It would be possible to make it the default for all users. DCDuringTALK01:42, 15 January 2010 (UTC)[reply]
Right hand ToCs would be good, leaving the ToC still there for those who use it, while saving the necessary screen space. --Yair rand04:58, 15 January 2010 (UTC)[reply]
Re #3 (a bot to move stuff above the first L2 into the first L2), I brought it up and people thought it was fine. I just haven't gotten around to writing code to do it (ideally it would be part of AF). If someone's got time, it certainly would be a good cleanup. --Bequw → ¢ • τ04:28, 15 January 2010 (UTC)[reply]
An idea: an extension which, depending on the country of origin the user is coming from (or locally, depending or their language settings in browser/OS, if this can be detected), L2s are arranged so that the relevant section comes first (above English if necessary). --Ivan Štambuk07:03, 15 January 2010 (UTC)[reply]
I disagree with the baseless (and incorrect) assertions #1 and #2. As a result, I can't agree to ideas that stem from such assertions. --EncycloPetey05:03, 17 January 2010 (UTC)[reply]
{{selfref}} also goes in the header. Character info is probably on top because a lot of the other sections also deal with the letter. I happen to think it should go in Translingual, but that's just my opinion. --Yair rand18:22, 18 January 2010 (UTC)[reply]
Two reasons: 1) Not all symbols are Translingual (some are only used in one language). 2) Encoding information is completely different than linguistic information (it's "a-lingual" not "pan-lingual"). --Bequw → ¢ • τ20:02, 18 January 2010 (UTC)[reply]
Opio was desysoped a short while ago, by the steward Kylu, for unblocking himself after he was blocked by Prince Kassad for a comment on his talk page. I do not see this as necessary. This was not an emergency, there was no need for the desysoping. If there actually is support for desysoping Opi for this (which I doubt), a vote can be held, and if the majority support desysoping, one of the bureaucrats can implement it. I suggest that Opi be quickly given his admin tools back, and we can forget the whole incident. --Yair rand06:21, 15 January 2010 (UTC)[reply]
The fact of the matter is that unblocking oneself after another administrator blocks you, in most cases, would get you desysopped because it is abusing the tools. Opiaterein knowingly unblocked himself, fully knowing that the block had not expired yet. A more responsible administrator would have stopped editing and gone and done something else for the duration of the block. It is his unblocking himself that prompted the desysop because that is an egregious breach in the policy for administrative tools. Therefore, I fully support the desysop and believe it should be kept. Razorflame06:24, 15 January 2010 (UTC)[reply]
In that case, why don't we resysop Opi for the moment, you can start a vote for his desysoping, and we can get back to where we were before the "emergency" desysoping. --Yair rand06:31, 15 January 2010 (UTC)[reply]
I would be fine with him being resysopped AFTER his 24 hour block that he originally had was set to expire. Therefore, I wouldn't care if he was resysopped AFTER 5PM tonight. The whole idea was not to have him desysopped permanently, but desysopped for the duration of the block so that he couldn't unblock himself. Knowing him, I knew that he would do so, so that is why I believed the extra precaution needed to be taken. Razorflame06:35, 15 January 2010 (UTC)[reply]
More reasonable administrator would, if it was necessary to block anybody, block both of you for a short period. You caused Opiaterein's overreaction by unnecessary pestering. Blocking established contributors, especially admins, that can otherwise do productive work, is completely pointless. It is not the "abuse of tools" to lift a pointless block. Blocks should be used only against vandals and proven trolls, in situations when problems cannot be talked out, i.e. solved by means of a discussion. If you're looking for some kind of emotional satisfaction by Opiaterein enduring that 24 hour "punishment", you seriously need to grow up. --Ivan Štambuk06:46, 15 January 2010 (UTC)[reply]
Did I ever say that I was satisfied with the desysop? No. I got no emotional satisfaction with him getting either blocked, nor desysopped. I believe that it was the correct and appropriate progression of events. That is all that I am going to say. Razorflame06:49, 15 January 2010 (UTC)[reply]
Razorflame your expectations of the wiki world are somewhat at odds with our procedures and ways of thinking. You have now tried to get 3? maybe more administrators desysopped and/or banned because of what you think about their behaviour. Please note that this implies you are horrendously outnumbered in your opinions, and desist from causing yet more trouble. You are hanging on here by a tenuous thread anyway, and persistently causing disruption is not the way to gain the support and trust of the community. Conrad.Irwin09:43, 15 January 2010 (UTC)[reply]
The emergency desysop was the correct course of action here. Opiaterein not only unblocked himself (which in and of itself was a violation of the blocking policy), but then performed other logged actions on his own account which raised some suspicion as the nature of the account itself. It could have been easily interpreted as a compromised account (which was my first take on the situation), and if not that a rouge administrator who, in the best interests of the project, needed to be dealt with quickly. Personally, I am of the opinion that a formal discussion should take place locally but that Opiaterein should remain desysopped until consensus is reached. Tiptoetytalk06:39, 15 January 2010 (UTC)[reply]
Compromised account? o_O Oh c'mon. This was a gross abuse of steward privileges and you're just trying to cover up for your mate. You are not supposed to intervene into local communities at all, if they're large enough. There was absolutely no need for that absurd "emergency desysop". There was no discussion on it anywhere, and obviously we're dealing with some kind of premeditated off-wiki plot. Your friend Razorflame betrays himself above "The whole idea was not to have him desysopped permanently, ..." what idea? Whose idea? --Ivan Štambuk06:53, 15 January 2010 (UTC)[reply]
Ivan, please keep this as civil as possible. The desysoping was not abuse, it was most likely a simple mistake (stewards generally have to act quickly with desysops to prevent damage in actual cases of abuse), there is no "plot" to get Opi desysoped, and nobody has acted incorrectly here. It is very clear that Opi's account has not been compromised, he's not a rouge admin, and neither Razorflame nor Kylu have been "plotting". This whole thing was a series of mistakes. --Yair rand07:00, 15 January 2010 (UTC)[reply]
Mistake 1) Razorflame pesters Opiaterein despite a very very large number of requests by Opiaterein that imply he does not want to talk to Razorflame.
Mistake 2) Opiaterein is unnecessarily rude in reply.
Mistake 3) Prince Kassad blocks Opiaterein, without warning.
Mistake 4) Kylu desysops Opiaterein.
Perhaps we should undo all of these mistakes and get on with building a dictionary, something at which Opiaterein excels. I don't think it will be necessary to have a vote to sysop Opiaterein given that he has already passed one; we can start discussion how to procede once these mistakes have been rectified. Conrad.Irwin09:43, 15 January 2010 (UTC)[reply]
Opiaterein pisses me off sometimes, however he is a good Wiktionary administrator and I'd strongly support an immediate re-sysopsing. FWIW Opiaterein and Razorflame have been exchanging pleasantries for a while on the IRC channel. Mglovesfun (talk) 09:50, 15 January 2010 (UTC)[reply]
Mistake 1) can never be undone: Opi has become sort of a father figure to Razorflame and he will keep pestering Opi seeking his approval no matter what. I say we indef-block RF, return admin powers to Opi and let those stewards know their interference was uninvited. --Vahagn Petrosyan11:36, 15 January 2010 (UTC)[reply]
Actually, that is where you are wrong. I view Opiaterein as a competitor. Furthermore, I've already decided that it isn't worth it to talk to Opiaterein any longer. Thirdly, Opiaterein can have the sysop tools back because his 24 hour block would've expired by now. Razorflame11:41, 15 January 2010 (UTC)[reply]
Can anyone explain why Razorflame (RF) can revert edits and move pages and suppress the redirect despite not being an admin? Btw yes, no offense RF but when we played on the Internet Scrabble Club together you did pester me quite a bit. As I said to you in private, you need to calm down and stop taking every negative comment as a personal attack, which they are not. Mglovesfun (talk) 11:53, 15 January 2010 (UTC)[reply]
I am a global rollbacker Mglovesfun, so, as with all global rollbackers, I can move pages and surpress redirects, and I can rollback here. I don't believe that I've misused the rollback tool here, though. Razorflame11:58, 15 January 2010 (UTC)[reply]
We currently have two separate topical category prefixes for Mandarin Chinese: zh and cmn. I think we should consolidate. Currently zh is much more common, though cmn is probably more specific/accurate. As A-Cai says:
Actually, a case could be made for either one. The zh code is the most well known, and refers generically to "Chinese", which most often simply means standard Mandarin. That has been the default position on Wiktionary to date. On the other hand, cmn refers specifically to Mandarin, which is the language header that we use. I'm honestly not sure which is better. Perhaps another beer parlor discussion is in order? -- A-cai 22:34, 13 January 2010 (UTC)
A brilliant example of a category that could and/or should be both topical and non-topical. The category for emoticons should be called [[Category:Emoticons]], and the topical category related to English words for emoticons (like smiley, lolcat and so on) should also be called [[Category:Emoticons]]. Had 'my' motion to add en: to topical categories passed, which it sort of did, this wouldn't be a problem. FWIW I thought a policy change voted required a 60% approval to pass, and the en: one failed with about 63%. Is this a mistake, or have I read something that wasn't accurate (alternatively, I might just be totally wrong). Mglovesfun (talk) 09:49, 18 January 2010 (UTC)[reply]
The % needed is at the discretion of the closer, but for controversial issues, erring on the side of a failed vote lends a bit of weight to "current consensus". I don't think there is an actual number set in stone, and I also think this is a good thing (I can only recall one instance of a possibly controversial issue being closed in an "unfair" manner, though there is a lot more debate about what to do when there has been little or no participation in the vote).
For what it's worth, having a topcial category and a non-topical category with the same name would be even more confusing if they don't happen to be the same Category; and besides, aren't most emoticons more or less translingual? Conrad.Irwin13:11, 18 January 2010 (UTC)[reply]
Even though there is an informal range of acceptable passing percentages (2/3 - 3/4), why do you think the passing % needed for a specific vote should be up to the closer rather than determined ahead of time? --Bequw → ¢ • τ01:23, 19 January 2010 (UTC)[reply]
Simply counting votes ignores the discussion that regularly occurs on such pages; while I can't give exact examples, the few times I have closed a vote, I have read the discussion just prior to doing so. Conrad.Irwin09:35, 19 January 2010 (UTC)[reply]
Emoticons are usually Translingual, not English, so "en" would have caused more problems. Emoticons are also used as a kind of punctuation, so they are not topical but functional. As it is, this is probably better handled in an appendix than a category anyway. --EncycloPetey16:22, 18 January 2010 (UTC)[reply]
No, because it isn't an emoticon. Category:Countries does not include country, nation, or state; it includes only items that are countries. Your issue is with the distinction between inclusion based on strict hyponymy versus loose semantic relationships, not on the issue of topical versus grammatical categories. --EncycloPetey19:21, 18 January 2010 (UTC)[reply]
Vote on deleting Template:Wikisaurus-link
I am about to create a vote on deleting {{Wikisaurus-link}}.
The vote is probably needed rather than mere RFDO, as the following paragraph will need to be removed from Wiktionary:Entry layout explained:
{{Wikisaurus-link|money}} The link to a Wikisaurus entry can now be made by using the Template {{Wikisaurus-link|headword}}. This throws up a banner with a link to the Wikisaurus entry. For example, {{Wikisaurus-link|money}}
The paragraph was added by Richardb to ELE without a vote in this edit on 6 May 2006.
Justification of the deletion: The template looks like a link to a sister project such as Wikisource or Commons, but Wikisaurus is not a sister project; Wikisaurus is a part of Wiktionary. I think it better to link to Wikisaurus from "Synonyms" sections using "See also Wikisaurus:entry", where people naturally come to look for synonyms.
Please provide any feedback such as support or opposition, although I understand that this subject is of no interest to most fellow Wiktionarians. --Dan Polansky09:06, 19 January 2010 (UTC)[reply]
Idea: a different side-box format that could be used for indicating that there is information about a word in another section of Wiktionary, for example, Wikisaurus or appendixes. Also, a sidebar box could be made above "in other projects" and "in other languages" for "in other sections of Wiktionary". --Yair rand22:15, 19 January 2010 (UTC)[reply]
Many of the activities are down to User:Connel MacKenzie - who doesn't come here very often these days. His name probably needs to be removed and replaced by who?
Now what the heck... When I finish clicking these links until something useful comes I'll forget what I was originally checking...
How hard is it to add a box to the right or something on the inference page that would look something like this
Inference is
That which is inferred [box: reasoned with conclusion]
Or just add the whole English entry for infer in the right side box...— This unsigned comment was added by 79.175.67.156 (talk) at 17:39, 21 January 2010.
Yes, but for example inferred is defined like I said, and there are countless other examples (original poster)
Ah, but we have three definitions of infer, q.v. Do you want to have all three listed at [[inferred]]? That, too, can get unwieldy. And what happens when someone adds another sense to [[infer]]? (We're missing one, after all: "{{nonstandard|proscribed}} To imply: to have as a necessary consequence".) He'll need then to add it also to [inferred]], [[inferring]], and [[infers]], which is more work. And if, because of laziness or not realizing he should, he doesn't, a casual reader may wonder why inferred isn't given the fourth sense: is it only a past tense of the other three? This will confuse people. It makes more sense, especially for polysemic words, to put either no gloss or a very general one that applies to all sense (if such exists) on the inflected forms.—msh210℠18:12, 22 January 2010 (UTC)[reply]
I agree with the original complaint. We should try much harder to avoid defining words primarily using forms of the same word or using words that are closely related etymologically. I recently found the verb "usher" defined as "to act as an usher". Very uninformative. --EncycloPetey18:56, 30 January 2010 (UTC)[reply]
Latvian
SIL just completed their review of ISO 639-3 code changes. One change that may affect our work here is with Latvian. They changed {{lav}} "Latvian" from being an individual language into being a macrolanguage composed of "Standard Latvian" {{lvs}} and Latgalian {{ltg}}. We can probably react in two ways:
Treat the subdivisions ["Standard Latvian" (which we'd probably just call "Latvian") and "Latgalian"] as individual languages. Recode everything that was lav to lvs.
Treat the macrolanguage as the individual language (and treat Latgalian as a dialect). Continue using lav and merge the Latgalian content (at present only 2 translations) into general Latvian content.
Wiktionary has a big problem right now, which is could seriously affect our ability to function on the long term. To start with, the blocking policy is rarely followed, causing us to lose a lot of potential editors. New users are blocked for tests on sight, users making small mistakes are blocked without warning, and newbies that are viewed as making general annoyances of themselves are often quickly blocked. Explanations are rarely given, and when they are, they are generally unfriendly and unwelcoming. Another problem is that incivility is rampant on Wiktionary, and it often leads to further incivility, which, besides for being completely unproductive and being a huge waste of time, gives a horrible impression to new users, which is very likely to cause them to leave the project.
Worse yet, both of these problems are self-causing. Admins block because there isn't enough time to try and give warnings or explanations, which is because our editor:vandal ratio is horrible, which is because new users are quickly scared away from Wiktionary or blocked. Incivility causes people to respond with more incivility, which causes the general atmosphere to make everyone incivil, which causes the "standard" of civility to drop, which makes it far more likely that someone will start the whole cycle over again.
Which brings me to my main point: Wiktionary does not have a friendly atmosphere. We block people too easily, we aren't as civil as we should be, and so we don't get new editors as much as we should. We need to pull our act together, raise the standards of civility, and stop blocking people. Overall, the project is not easy to work with. If we could all work together to fix these problems, the rate of new editors would rise, and more content would be added. These problems have been - to my view, at least - made a bit more clear with the recent leave of User:Logomaniac, who could really be an example for all of us (the amazing friendliness bit, not the leaving Wiktionary). Civility breeds more civility, which makes the whole project a lot easier to work with. Less blocking means more editors, which makes it easier to get things done. We need to fix this up, or the situation will continue to worsen. There is no simple solution to this; I think we should all try and reread through Wiktionary:Civility and Wiktionary:Blocking policy (which actually might need a cleanup IMO) and follow them as much as we can. What we have now is that Wiktionary is unfortunately one of the least friendly, welcoming, and civil projects in Wikimedia. We should try to improve in these areas, pull our act together, become one of the most friendly, welcoming, and civil projects around, and get more contributors with less arguments, so we can all work together to build the dictionary. --Yair Rand06:34, 22 January 2010 (UTC)[reply]
Yes, I agree. It would be interesting to compare blocking statistics between projects. They would probably show that blocking too much is very counterproductive, not only for above reasons (most vandals like to be blocked). But, in some cases, blocking is necessary, of course. Lmaltier07:02, 22 January 2010 (UTC)[reply]
I agree as well. Though I'm a bit less bothered by the hair-trigger blocking than by the curtness in replying to newbies when they complain about blocks, rollbacks, and/or page deletions. A very small number of admins do a very large amount of patrolling, so it's probably inevitable that they will frequently either take wrong actions, or take right actions without enough explanation. (Heck, the rollback feature doesn't even give the opportunity to enter an explanation.) But when a user comments back angrily, I think the response should always begin with an apology. (For example: "Sorry; I appreciate your contribution, but that's not a 'real word' by Wiktionary standards."; or "Sorry; I understand why you felt that your edit was an improvement, but it went against our formatting rules. We've put a lot of thought into these rules, and while they don't always work perfectly, it's better for all entries to follow them than for entries to be formatted inconsistently.") —RuakhTALK04:44, 25 January 2010 (UTC)[reply]
At the moment it is on the judgement of the patroller(s) who gets blocked and who doesn't, by far the easiest way for you to fix this is to be the one who patrols - see also User:Robert Ullmann/Rat Patrol if it is too tedious to do manually.
On the topic of policies, I agree that WT:BLOCK and WT:CIVIL should be rewritten (WT:CIVIL in particular, but WT:BLOCK is showing its age). I would like to try a "new" style of policy page, where the letter of the law is brief, unambiguous, to the point, and requires a vote to change; however accopanying this there should be much more lengthy explanation of how the community currently interprets this, the explanation would not require a vote to tweak and improve, but would require discussion to change considerably - this would, I hope, avoid some of the problems we see with WT:CFI. Any kind of rewrite requires a chunk of effort from a dedicated contributor or two; people like me saying "it should be done" doesn't achieve anything. Conrad.Irwin14:06, 22 January 2010 (UTC)[reply]
I, too, find that we need to change our blocking policy. Currently, blocking here comes way too easily and can be for reasons that aren't really great enough reasons for them to exist. Currently, one can be blocked for stupidity, but I ask, "What defines an edit that can be considered stupidty?" It is clearly subjective when it needs to be objective. In addition to this, I agree with Conrad.Irwin that we are probably losing lots of potential great editors because of how easy blocking comes. What I think we need to do is redefine the blocking policy to include a warning system unless the user is commiting clear-cut vandalism. It could be a kind of two strikes and your out thing where the user in question is warned once and then if they continue on with what they are doing, then they are blocked. Basically, we need to tighten our reins on this because I feel like this wiki has a negative atmosphere from just working here, so I definitely see a need to update both the civility and blocking policies. Razorflame16:17, 22 January 2010 (UTC)[reply]
So we have a number of points to consider, such as whether warnings are needed in every case, whether to block blatant vandalism without warning, whether to block for "stupidity", whether/how to enforce civility (the Removing uncivil comments section of WT:CIVIL seems rather controversial, but overall having something like that might be a good idea). I think that we might benefit from having Wiktionary:Blocking policy/editable and Wiktionary:Civility/editable, but they would probably be best started from scratch, rather than starting them based on the current policy pages. --Yair rand18:37, 22 January 2010 (UTC)[reply]
Well, first of all, I believe that there are certain instances in which case a block without warning is OK: Clear-cut vandalism. Other forms could be stuff like people being stupid on purpose. A warning for most other things are good, and I think that a single warning is sufficient enough to deal with this, then block. Hopefully, this will allow us to get more active editors. Razorflame18:44, 22 January 2010 (UTC)[reply]
Well, what is clear-cut vandalism? Is an instance where a new user who is so amazed that they can actually edit the dictionary puts random strings of letters across the page clear-cut vandalism? Is a case where a user sees a formatting mistake elsewhere and duplicates the same formatting mistake on dozens of other pages clear-cut vandalism? Is a user who, upon seeing foreign language entries in the English Wiktionary, gets confused and starts deleting the foreign language sections, committing clear-cut vandalism? When a user finds what looks like, to his view, a graffiti board, and starts posting swear words all over the place, completely unaware that this is a serious dictionary, is that clear-cut vandalism? My answer would be that none of these instances are vandalism, and in general we have no way at all to tell what the circumstances may be, so blocking without warning or explanation wouldn't be the correct response to any case. --Yair rand19:02, 22 January 2010 (UTC)[reply]
Clear cut vandalism includes adding "ahhahahaha your gay and sos your mom she has a peenus LOOLLLLZZZZ" to a page; maybe blanking lots of pages in quick succession (sure 1 or at most 2 might be mistakes); logging in with usernames like "Yair rand Smells"; . I have tried to draft a "new" style policy at WT:Blocking policy/new, please expand and elaborate the Interpretation section so that it makes sense to everyone; and, of course, feel free to re-word or completely change the statement of policy itself. Maybe we should start a vote for this in a week's time. Conrad.Irwin19:35, 22 January 2010 (UTC)[reply]
[1], q.v., is a description of research underway on "improving socialization of newcomers in [English Wikipedia] WikiProjects" (w:wp:WikiProject). The postdoc apparently in charge of recruiting subjects for the study e-mailed me on the 6th inst. to see if I'd be interested to join the study as a subject (as I'm still listed as a member of a small enWP wikiproject). I replied, in part, about enwikt:
Although it scarcely has WikiProjects, you may be interested in it, as it has relatively few active editors, and newcomers often have a hard time fitting in: perhaps you can treat English Wiktionary for your purposes as an English Wikipedia WikiProject (which it is not). Regular editors of English Wiktionary -- or some of them -- recognize the problem of newcomers' fitting in, and seek ways to improve the Web site in this regard, so may be interested in your research. (I can try to help you recruit other Wiktionarians, as we style ourselves, to your research project.)
She replied (two weeks ago today) that she was interested and that she'd get back to me "sometime next week". I'm awaiting her further reply, but will keep you posted. Anyone interested, please post ~~~ at user talk:Msh210/CMU.—msh210℠19:07, 22 January 2010 (UTC)[reply]
Far worse than a plain comment, IMO. They are generic cookie-cutter things, not even a friendly personal message with individual meaning to it, and strike me as clutter. Maybe I'm a Scrooge, but "wikilove" seems to tend towards doing less work and more backslapping. Remember Esperanza [2]? Equinox◑16:52, 23 January 2010 (UTC)[reply]
I think having barnstars might be good, especially if they were often given to newbies. I'd say that a new-ish user who is given a barnstar for doing a good chunk of work is significantly more likely to stay than otherwise, but that's just my opinion. --Yair rand01:21, 24 January 2010 (UTC)[reply]
I have to confess to siding with Equinox on this one. While the idea is not a bad one, the problem is that there is little way to ensure than the correct people are rewarded; it gets to the point where having the reward is more important than having earned the reward. Perhaps if we were to give someone a notification when they get whitelisted, it would be a note of encouragement - though maybe that is too late? It also then makes being whitelisted more important, which could cause people trying to game the system. (There have of course been cases where we've had to unwhitelist someone, but they are rare enough it probably doesn't matter). Maybe I'm too cynical... (By the way, did you see WT:Blocking policy/new?) Conrad.Irwin02:53, 24 January 2010 (UTC)[reply]
Well, you can (and are encouraged to) put a message in the barnstar. Make them as personal as you want. @Conrad I see little problem in having the "incorrect" people receive barnstars. More importantly, we won't know if that's a large problem until we try. See {{The Original Barnstar}}. Use it if you'd like. --Bequw → ¢ • τ23:52, 24 January 2010 (UTC)[reply]
The barnstar template has been deleted by Leftmostcat. As for the new blocking policy, I think it needs expanding. Also, I am very strongly opposed to any type of warning-less blocks. Besides for the very likely possibility of making a mistake, it is probably much more likely for the actual vandals to come back if they've been blocked for one edit than if their one edit was quickly reverted, with them being given a warning. As Lmaltier pointed out, most vandals probably enjoy being blocked. (To the best of my knowledge, Wikipedia doesn't even block until the user has been given four separate warnings, but I don't think we need to go as far as that.) --Yair rand04:02, 25 January 2010 (UTC)[reply]
I rarely edit the French Wiktionary because of the awful atmosphere, I'm always amazed how much friendlier we are here with 50 regular editors compared to 10 regular editors on fr, where it's been tribal warfare since about June. I try to revert bad edits without blocking unless they are patent examples of vandalism ("so-and-so sucks balls" or something like that). I'm probably guilty of not taking enough time to explain things to newbies too. Plus Wikipedia gets all the headlines, and hence the contributors too. Mglovesfun (talk) 04:12, 25 January 2010 (UTC)[reply]
This all really boils down to one thing: either we, the kinder, gentler admins, start helping SemperBlotto in his patrolling, or he continues to chop off newbies’ heads. --Vahagn Petrosyan11:22, 25 January 2010 (UTC)[reply]
I appreciate what Semper is saying; I find him the harshest of all of us for blocking, deleting and reverting, however at least he's doing it. I'm gonna review our blocking policy. As pointed out on that page, we shouldn't block user for tests unless they aer blatant vandalism. Most of the time it's better to revert and move on, which saves time for patrolling more edits. Mglovesfun (talk) 13:19, 25 January 2010 (UTC)[reply]
My thanks and cheers to SemperBlotto for patrolling. An anonymous newbie truly interested in improving the Wiktionary project should not have a mental breakdown when he gets reverted by an admin; he can find another entry to work on. --Dan Polansky14:35, 25 January 2010 (UTC)[reply]
Communication issues
A related problem seems to be that we are unable to get across to people exactly why they are not doing things "right". This is partly because we don't explain things, or when we do don't explain them well; and partly because, even if we do, people don't read what has been written to them, or fail to understand it, or plain disagree with it. I am not sure what to do about this, perhaps more templated messages like {{asdfg}} but more specific or better written, or perhaps a few help pages on the topic of "why was my edit undone". It would also be good to have a clear rationale behind CFI, (I tried to write one at Wiktionary:Descriptivism, but given the overwhelming lack of interest in that page, it would seem that someone needs to do it more betterly). Clarifying when and where information from other dictionaries can and can't be used would also be a useful, it's a common source of confusion (pun intended). The problem with building help pages is we then need to link people to them, I believe we could modify the revert text to contain a link to "Help:Why was my edit undone?", which might be a good half-way step as oppposed to leaving impersonal messages on a hundred talk pages. Conrad.Irwin13:26, 25 January 2010 (UTC)[reply]
This arises also at OTRS. I have access to the info-en@wiktionary.org email address and I frequently see emails from users who have been blocked for a single misstep. At the very least it would be good to see a guide to appealing blocks, but ultimately, blocking people for test edits or adding nonsense entries is a very good way of driving away potentially good users. Stifle11:06, 2 March 2010 (UTC)[reply]
Unilateral vs. consensus
"Unilateral" is a horrible sounding word, yet consensus is very slow and often there is none. Being a wiki administrator is political, politics is slow. If you've noticed me deleting a lot of unused categories and templates recently, it's because nominating them all individually would be time consuming and possibly just a bit silly. In this case, if I delete something that should be kept and use, restore it or ask for it to be restored. I think there's a good inter-sysops policy of not openly reverting other admins when they make a bad edit, and we all do that sometimes. It's better to quietly go back to the last good version in the page history.
I hope Semper who uses a similar style to me feels roughly the same. I've restored some stuff that he's deleted when I figured I could quietly clean it up. Mglovesfun (talk) 19:39, 25 January 2010 (UTC)[reply]
Abuse of administrative power
Admins: After posting extensive rationale on the discussion page, I made a good faith update to the definition for the word "authorization".
User "SemperBlotto" immediately reverted the definition without any rationale given as to why the change was incorrect or how it might be improved. While it is good that volunteers monitor Wiktionary and correct instances of vandalism, it is not good when they abuse power to destroy other's input for no reason.
Background: I am a computer security professional and have some knowledge of the subject matter in question. My change cited as a reference the Internet Security Glossary (IETF rfc:4949), a widely-respected reference in the field. Although I sincerely believe that my change corrects errors and improves the quality of the entry, I could accept a consensus within the Wiktionary community that it is not an improvement. What is difficult to accept, however, is that a single user would choose to revert the change with no discussion at all.
I respectfully request that other administrators review the proposed change and it's discussion page, and take whatever action is deemed apropriate, either reversing SemperBlotto's reversion or providing a rationale for why the proposed change is "disruptive" (in his words) and should not be accepted. Banning a user for making a reasonable change is both immature and a good way to drive off legitimate content contributions. And Wikimedia is not well-served by those who arbitrarily abuse the administrative power entrusted to them. Parcheesy219:13, 25 January 2010 (UTC)[reply]
Don't know about "disruptive" etc., but bear in mind that a term being declared in an ISO document doesn't automatically make it a dictionary word. It has to be used by others with that same exact sense. Equinox◑21:16, 25 January 2010 (UTC)[reply]
You removed all mention of "countable" and "uncountable". ← Mglovesfun had already commented on the talk-page, informing you that this was wrong. Rather than reply to Mglovesfun and try to understand why it was wrong, you simply went ahead with the change.
You split the entry into two senses, “(abstract noun)” and “(concrete noun)”, each with two subsenses. We don't generally use subsenses at Wiktionary, except for words with very many senses, where there's really no other way to keep them organized. You did not propose the sense/subsense split on the talk-page, but you did mention abstract vs. concrete there, and Mglovesfun told you that we don't use them. Again, rather than reply to him, you simply went ahead with the change.
You formatted these subsenses incorrectly.
You added a few subsenses. This part was fine, except that you included references in such a way that they weren't instantly recognizable as references, and looked like some external links that you considered relevant.
You removed a sense as "inappropriate", together with all of its translations. On the talk-page, you argued that this sense is just a special case of a more general sense; and you may be right. Unfortunately, the sense doesn't have any example sentences or quotations, so it's hard to see what its author was getting at. In a case like this, an RFV or RFD discussion is probably in order; I'm sorry that no one told you that before you went ahead with it.
Note that, after SemperBlotto reverted your changes, he did add the computing sense.
All told, I think that his reverts were justified, but that the block was unfortunate. I wouldn't characterize it as "abuse of administrative power", though: I understand why you were annoyed at his initial revert, but re-reverting was not the correct next step. If we look at it from his standpoint, since he now had to re-re-revert (since your edits were, in fact, wrong), it made sense to prevent you from simply re-re-re-reverting.
I apologize for not following the Wiktionary process. I am familiar with the Wikipedia process of "Be Bold" (make the edits yourself) instead of asking someone else to make them, however I have now submitted an RFC asking someone else to make the changes I made on 20 January: http://en.wiktionary.org/w/index.php?title=authorization&oldid=8272382.
Mglovesfun does not communicate his ideas very clearly. On 11 January he said "We use countable/uncountable. Concrete/abstract is an entirely different idea." This statement is neither relevant nor probative:
if he meant "We use countable/uncountable on Wiktionary" then of course there are many words for which countable/uncountable is an appropriate categorization. That is irrelevant to a discussion of whether it is appropriate for this particular word.
if he meant "We use countable/uncountable for this specific word (authorization), then that was obvious simply from looking at the page. But stating that something is what it is has no probative value if the question is to decide whether or not it should be changed.
He should have said "We should use countable/uncountable here because ..." and give some reason that can be supported or refuted. I have given an example of why it is incorrect on the discussion page (a bank can perform a specific, countable number of acts of authorization per day), and thus sense 1 is not uncountable.
I gave Mglovesfun a specific example of a case (oxidation) where countable/uncountable did apply on 14 January, in order to contrast it to this case where it does not apply. There is little point in categorizing the senses when they are all the same category: sense 1 is countable, sense 2 is countable, and sense 3 (which is a subsense of sense 2 and should be deleted) is also countable. If one wants to categorize the senses at all, then concrete/abstract is the meaningful and relevant difference between sense 1 and sense 2. I am not arguing that the senses should be categorized, I am simply saying that the existing categorization (countable/uncountable) is wrong and if a categorization is desired, then concrete/abstract is a suitable replacement.
Six days later on 20 January, after seeing no further discussion from Mglovesfun, SemperBlotto, or anyone else, I assumed that the rationale for change was sufficient, so I made the change. Again, I apologize for failing to understand that Wiktionary uses a different process (RFC) than Wikipedia (Be Bold). I have now submitted the RFC, and you can now decide whether it is convincing. If it is, I suggest http://en.wiktionary.org/w/index.php?title=authorization&oldid=8272382, which cites rfc:4949 as a referece rather than showing a 4th "computer science" sense. The word should have only 2 senses; the two rfc:4949 definitions are just restatements/reinforcements of those two, not additional senses. The reference section can be completely omitted if you prefer.
And a few Chinese secret agents steal the secret nuclear launch codes of the USA and proceed to get to the nuclear missile launch site, planning to fire nuclear missiles to important European cities including Berlin, Paris and Rome. When they get to the launch site and try to fire up some nuclear missiles, a US secret agent pops up and says: "Hello. Firing nuclear missiles is against the international law and may kill hundreds of millions of people. Please read the international law for more information. Thanks!" The Chinese agents, unsurprisingly, just smile a little and continue firing nuclear missiles. The US secret agent continues: "Please stop. If you continue to fire nuclear missiles, you will be shot." Madrid, Budapest and Vilnius have already been hit by nuclear missiles and are completely devastated, with the Chinese continuing to fire nuclear missiles. The US secret agent speaks again: "Stop. If you continue to fire nuclear missiles, you will be shot." And again, the Chinese fire up more and more nuclear missiles. Oslo, Helsinki and Athens now exist only in the history books. Now the secret agent shouts: "This is your last warning. The next time you fire up a nuclear missile, you will be shot!". The Chinese fire up another nuclear missile, after which the US secret agent does shoot them. However, by now 30 European capital cities are destroyed, almost 80,000,000 people dead and 20,000 sq mi irradiated. This could have been prevented if the US secret agent had shot the Chinese secret agents at the very beginning, however the US community decided that everyone gets warned before he is shot. -- Prince Kassad17:46, 26 January 2010 (UTC)[reply]
PK is right, the only way for the world to properly function is to carefully follow the doctrine of preemptive strike: exercise power first, ask questions later. Preferably explaining it in terms of parable involving cartoonish-villain depiction of the only non-imperialist superpower left on the planet. --Ivan Štambuk18:07, 26 January 2010 (UTC)[reply]
Great example PK :). Does WT:Blocking policy/new meet with people's expectations, unless someone gives me feedback on it soon I will just start the vote anyway. (I'm interested both in improvements to the text and content of that particular page and the idea of separating the "policy" from the intepretation) Conrad.Irwin19:40, 26 January 2010 (UTC)[reply]
Actually, I don't think it was a good analogy for the simple fact that his example is so extreme. It might have been good if we were talking about WonderFool and Kassad's cities were the mainpage and the vote page etc.
I suppose I'm the only person who is bothered by all this bureaucratic structuring. Have none of you dealt with Wikipedia recently? — [ R·I·C ] opiaterein — 15:46, 27 January 2010 (UTC)[reply]
Reverting isn't an administrative power, but blocking is.
Another parable: Bambi the deer steps out of the forest into a residential neighborhood. Helicopter hunter spots him and launches a heat-seeking grenade, leaving nothing but flecks of meat and a yellow stain on the grass. Moral: since the deer might have crossed the road and damaged a car, and probably was planning to eat the flower buds off of Grandma's azaleas, it is essential to strike preemptively with overwhelming force in order to protect the community.Parcheesy200:03, 27 January 2010 (UTC)[reply]
You might want to kick off a separate Beer Parlor topic or create the discussion page for Blocking Policy, but since there aren't any ...
Having a policy in the first place is a great idea. Almost any set of guidelines, written down, is preferable to having individuals making their own ad hoc decisions without guidance, or having "secret" rules passed along in private discussion.
Your initial draft recommends proportional response, including warnings prior to blocking and block duration based on severity of offense, which are important features of any policy.
Beyond that, it is not appropriate for outsiders to recommend specifics. In order to be effective, any policy must have buy-in from those expected to follow it. The best way to get buy-in is for those closest to the problem to collaborate on the solution. Various types of misbehavior by users and by administrators must be addressed, but how they are addresed should be up to the administrators themselves.
I don't. Generally speaking, people just look for an orange band at the top of the page, they don't read the message in the band. That being said, I like this new color more, but the message is pointless (for me). Mglovesfun (talk) 18:02, 26 January 2010 (UTC)[reply]
That message is controlled by MediaWiki:Youhavenewmessages. There's some code in MediaWiki:Common.js to hide the second sentence from users once they're auto-confirmed, but said code is commented out. Plus, this change wasn't discussed anywhere so far as I can tell. (?) So I've deleted MediaWiki:Youhavenewmessages for now, thereby restoring the old, default text; we can re-add the new sentence once we have it in a form people are O.K. with. —RuakhTALK18:21, 26 January 2010 (UTC)[reply]
That was an attempt by me, in light of #Pulling_our_act_together, to make sure newbies read warnings before they're blocked. It told them to check their messages before editing (I debated using "saving" but decided it was less clear), not just that they have such. It was, as Ruakh noted, meant to be seen only by newbies, but I can't get that code to work. I know this is a GP issue, but if anyone can get said code to work (and thinks this is wise, and there's no uproar here against it), can you please help?—msh210℠18:36, 26 January 2010 (UTC)[reply]
I haven't tested or anything, but the only problem I see by looking at your code is that it tries to call document.getElementById before the page has loaded. This could be addressed by placing it in an event handler (e.g., by using the addOnloadHook utility function), or by removing the call to document.getElementById (seeing as the code isn't very expensive anyway). —RuakhTALK19:33, 26 January 2010 (UTC)[reply]
I think the idea is a good one, though I also found the message a little patronizing; that said I'm unable to come up with a better phrase. "You have new messages. IGNORE IT AT YOUR PERIL. Mwhahahahahahaaaaa" doesn't seem to quite cut the mustard, but is kind of what we want to say. "You have a new message, right click to open it in a new tab now!" - cunningly disguising the "please read it" as a neat feature. Until we can actually be sure the message will contain advice, we can't really put "You have new messages, these will help you become a better person". Any better ideas? I think something along the following lines would work for your javascript (though I've used a classname not an ID - maybe we want other things like this too?): Conrad.Irwin19:05, 26 January 2010 (UTC)[reply]
msh210: "That was an attempt by me, in light of #Pulling_our_act_together, to make sure newbies read warnings before they're blocked." That to me is unnecessary babying. If they want to ignore their talk page, then they frankly deserve it. As far as "Pulling our act together", I think it's too late for that. Recently there has been far too much of a move toward political bullshit and bureaucracy, away from encouraging more thorough verification of new entries. Coughrazorflameaddingtelugu. — [ R·I·C ] opiaterein — 19:31, 26 January 2010 (UTC)[reply]
I'm only concerned that newbies — really, anonIPs mostly (or maybe even only) — may well not know they have a talkpage, and view "You have new messages" as a banner ad attempting to disguise itself as a new-e-mail notification, which is what it looks like.—msh210℠19:43, 26 January 2010 (UTC)[reply]
If we want to have more assured initial communication with contributing newbies, why pussyfoot around? Why not fix it so unregistered users and newly registered users (first day/week/month; first 10/100/1000 edits) cannot scroll away from it? Why not prevent them from editing until they look at such a message? DCDuringTALK20:50, 26 January 2010 (UTC)[reply]
One message is sufficient. "Congratulations! You received your first message on Wiktionary. It's probably a warm welcome with useful guidelines, so read it." --Daniel.18:02, 29 January 2010 (UTC)[reply]
Interesting. Somehow intuitively this seems to me to differ from the other glossaries, which are typically context-specific.
On second look, doesn't the "/A" mean simply that another click is required to see something of substance. Why is that a good idea?
In any event, I think the thrust was to get the word English in the title so that English could feel like a stranger in its own home under the color of some kind of "fairness". Since we don't want to make things too easy for native-language only newbies and want to remind them that they should consider going elsewhere if they want a monolingual dictionary, the even more cumbersome Appendix:Glossary of English idioms - entries beginning with A would be appropriate.
DCDuring, I don't think I would be able to answer precisely why it is a good idea to require an additional click to see substantial information, because this idea is yours, not mine; furthermore, even you seem to reject it and I'm not really interested in looking for places where a first link to simply reach the second one would be desirable. I'd like to know if you have other ideas for lexical appendice names which might solve your concern of English superiority in English Wiktionary. The name "Appendix:Glossary of English idioms - entries beginning with A" would not be easily memorable or typeable, then it excludes most people, monolingual newbies or not, so it is a good example of how much information may not be feasible to be part of such titles (particularly, I still think that simply Appendix:English idioms/A is better, since the unresolved "Titles of glossaries" discussion from July 2009 did not provide arguments on the contrary as a whole - Dan Polansky's comments on practical differences between lists and glossaries were never confronted against my how-to-merge-them-into-a-better-topical-appendix argument as ways to reach a consensus). For your convenience, I can remember the naming scheme of Appendix:Idioms, Appendix:Idioms:Spanish and Appendix:Idioms:Italian, which was used in earlier days of Wiktionary; although, it is rather user-hostile due to the unusual colon. --Daniel.15:29, 30 January 2010 (UTC)[reply]
I would see the renaming by standardizing on either of the two options:
(a) Slash as a separator:
"Wiktionary:Requested entries:Spanish" to "Wiktionary:Requested entries/Spanish"
"Category:Translations to be checked (Spanish)" to "Category:Translations to be checked/Spanish"
(b) Dash surrounded by spaces as a separator:
"Wiktionary:Requested entries:Spanish" to "Wiktionary:Requested entries - Spanish"
"Category:Translations to be checked (Spanish)" to "Category:Translations to be checked - Spanish"
The colon's technical meaning in Mediawiki is that of namespace separator; any further nesting should better be done using "/", which is a standard way of marking up subpages in Mediawiki. The bracketed name seems clumsy to me.
That said, I think that the current colon polysemically used also as a nesting character is good enough, and I prefer it to the bracketed names. --Dan Polansky21:46, 31 January 2010 (UTC)[reply]
It is a list of terms with definitions. I admit that the terms are not selected per domain, so it satisfies only in part the definition "A list of terms in a particular domain of knowledge with the definitions for those terms". --Dan Polansky19:41, 9 February 2010 (UTC)[reply]
Does spade meet our criteria for inclusion
No really, terms have to be attested and idiomatic. Spade is attested, but not idiomatic. Can I RFD it then? I see nothing in CFI saying that single words like spade, dog, chicken etc. are exempt from being idiomatic. And people wander why nobody ever reads CFI. Mglovesfun (talk) 20:47, 26 January 2010 (UTC)[reply]
As I said recently to you on RFD, polcies must be applied with common sense. There are known problems with CFI, hence CFI/editable. Instead of moaning, fix problems; it's much more satisfying in the long run. Maybe we could even try for a new policy along the lines of WT:Blocking policy/new with a few carefully chosen criteria in the box and a lengthy explanation underneath (though I would imagine this would only deal with a specific subsection of CFI, trying to include the entire kerbang would take too much time). Mayhaps you want to have a go at writing it? Conrad.Irwin20:52, 26 January 2010 (UTC)[reply]
Well, unless you can somehow easily derive the meaning of "spade" from its separate components, ("sp"+"ade"?) its definitely "idiomatic" by CFI's rather odd definition... --Yair rand22:20, 26 January 2010 (UTC)[reply]
Yes, this is the general principle. However, the problem is inconsistency with this first sentence: other sections seem to explain that words must meet some conditions to be included (e.g. very recent words seem to be excluded, even when everybody knows that they are actual words of the language). CFI should explain instead what we call word and what we accept in addiiton to words (characters, proverbs...), when a word is considered to be used in a particular language, and how all this must be verifiable. When something is a word (without any doubt) in some language (used in this language without any doubt), CFI should explain that its other sections are not to be used, that they should be used only if there is a doubt. Lmaltier06:51, 27 January 2010 (UTC)[reply]
We could make the following edit to CFI:
"Each one-word term is idiomatic; An expression a multi-word term such as "black hole" is “idiomatic” if its full meaning cannot be easilysafely derived from the meaning of its separate componentsconstituent words."
Would that solve the problem that you percieve in CFI?
Sure enough, this stipulates "word" as being space-free.
The current definition of idiomacity is from this edit in May 2005, meaning it did its job well enough for more than four years.
The current CFI formulation of "idiomacity" is good enough: by the current definition, "spade" is idiomatic, as its meaning cannot be derived from the meanings of its components. --Dan Polansky09:36, 29 January 2010 (UTC)[reply]
Please don't, we don't need them at all. The only categories for proto-languages are PoS and they use full language name. There are also language family codes (e.g. Germanic:gem, Slavic:sla, Indo-European:ine etc.) that can be utilized in rare instances where there exists need for e.g. template covering several individual languages, and specifically relating to a certain proto-language (usually the proto-language of the family in question; cf. recently created Category:Proto-Indo-European reference templates). --Ivan Štambuk04:10, 28 January 2010 (UTC)[reply]
Please take note of this Latin entry which I just created. I’ve attested the phrase’s use not only in Latin, but also in Dutch, English, French, and German. The problem with this sort of term is that their entries tend to contain a hell of a lot of unnecessary duplication; please take a look at the entry to see whether what I’ve done is any good and to make any improvements as you think best. †﴾(u):Raifʻhār(t):Doremítzwr﴿00:02, 29 January 2010 (UTC)[reply]
As requæsted on the Requæsts for cleanup page, I shall comment mine edit of the article: I added a German quotation from one of Nietzsche’s masterpieces and rectified the lemma of the verb impendo, inf. impendere. The other verb, impendeo, with which it shares the same infinitive, has another meaning and is from the 2nd conjugation group (impendo, devote, is from the 3rd). The uſerhight Bogormconverſation09:21, 29 January 2010 (UTC)[reply]
Thanks for those corrections and additions, Bogorm. However, according to my dictionary, the ō in impendō(“I devote”) ought indeed to remain; conversely, according to the same dictionary, the present active infinite form of that verb is impendere, not impendēre. Besides that, have you any comments to pass on the layout? †﴾(u):Raifʻhār(t):Doremítzwr﴿03:29, 30 January 2010 (UTC)[reply]
Examples in other languages seem to be mentions, not uses. Or am I wrong? If a language really uses such a phrase, it's worth adding a section for this language (especially for the pronunciation in this language). Lmaltier13:06, 30 January 2010 (UTC)[reply]
Read w:Use–mention distinction for a clearer understanding of when a term is used and when it is mentioned; whilst italics are often used to mark a term that is merely being mentioned, they also have other purposes (in this case, marking a foreign phrase). The problem with adding language sections for every language which uses this Latin phrase is that we get a lot of unnecessary repetition, whereas the useful information could be far more concisely presented (as I envisage, in a way similar to the one I’ve used for the entry in its present state). Pronunciatory transcriptions are definitely what make adding information for other languages useful — besides the Latin [ˈwiːtã imˈpendere ˈweːroː] and the English [ˈviːtæm ɪmˈpɛndɪɹeɪ ˈviːɹəʊ] (or whatever), there will undoubtedly be semi-naturalised pronunciations in pretty much every other language, dependent on the particular phonemes &c. thereof; whether they’re presented in a Latin Pronunciation section or a Translingual Pronunciation section makes virtually no difference there. Presenting this information in a Translingual (rather than a Latin) language section has the disadvantage of obscuring the fact that this phrase is undeniably Latin, and is treated as such in its grammar and is indicated by certain typographical details like italicisation; important Latin information like the display of macra (to mark vowel length) cannot, without inconsistency, be shown in a Translingual section. Having the quotations in the entry itself is very valuable, and allows comparison of the phrase’s usage in diverse languages; consigning them to the Citations: diminishes this. †﴾(u):Raifʻhār(t):Doremítzwr﴿14:41, 30 January 2010 (UTC)[reply]
It's not translingual, and I still consider that, if something is really used in a language, there should be a section in the page for this language. This allows for pronunciation, but also usage notes, anagrams, etc. (and citations). This may lead to a few additional sections (or many, e.g. in Paris), but principles should be followed. Examples which are not in Latin should not be in the Latin section. Keep the Wiktionary principles simple! Lmaltier15:31, 30 January 2010 (UTC)[reply]
What about the argument that this phrase is so unnaturalised that it cannot be considered to have entered the languages in question, and that therefore there ought not to be an entry for it in those languages? I doubt there will exist any anagrams of this phrase in any language. And what about usage notes? Surely they go as far as “This is a foreign phrase, so it’s treated as such, retaining its Latin grammar and marked as foreign with italics.” — that applies to its treatment in all cases in the four languages cited thus far. Let’s not get too strict with principles — if a principle results in an absurdity, it is a sign that the principle needs to be revised. Can you demonstrate a practical downside to this merged presentation? †﴾(u):Raifʻhār(t):Doremítzwr﴿18:01, 30 January 2010 (UTC)[reply]
This breaks the mold of what we consider translingual entries because when it appears outside of Latin it's a quotation from a particular source and author, rather than a general term in the language. I agree that we ought to consider that in desciding how to handle "translingual" entries. --EncycloPetey18:38, 30 January 2010 (UTC)[reply]
I don't propose to create many sections for this case: at least the 1st and the 3rd citations for English and the French citation are only mentions, not uses (the other ones, I don't know). But the limit between naturalized words and unnaturalized ones is subjective, it's a question of personal opinion. This is why applying very simple principles is so important (otherwise, there are endless discussions). Lmaltier19:06, 30 January 2010 (UTC)[reply]
There are many more where they came from (google books:"vitam impendere vero"), if you want to take a look. Anyway, whether or not every language section is CFI-satisfying in this particular case has no bearing on the general question of how we should deal with cases like this. The exact point of sufficient naturalisation may be difficult to pin down in ambiguous cases, which is exactly why I chose this phrase, which, I should think, no one would be willing to say is naturalised. †﴾(u):Raifʻhār(t):Doremítzwr﴿19:31, 30 January 2010 (UTC)[reply]
I thought it would be good to start this page (when I get home) to deal with some of the confusion over Anglo-Norman and Old French. I was down at Leeds University Library and scholars seem to agree the Anglo-Norman is a dialect of Old French. However I feel compelled to follow ISO 639 on this one (I always do) so Category:Anglo-Norman Old French should be deleted and replaced with an explicit Anglo-Norman entry in each case. All Anglo-Norman entries should have an Old French entry as well. Does anyone object to this? Mglovesfun (talk) 13:50, 30 January 2010 (UTC)[reply]
But there may be words used only in Anglo-Norman. If Anglo-Norman and Old French are considered as separate languages, there is no reason to include an Old French section for words used only in Anglo-Norman. Lmaltier15:35, 30 January 2010 (UTC)[reply]
Why not just include all Anglo-Norman categories in Old French and make sure that all mentions of Anglo-Norman (other than in etymologies) also clearly indicate that Anglo-Norman is a dialect of Old French. Very little is thereby lost -- except the labor and duplication. Are we yet in a position to gainsay Webster 1913 or whatever other authorities we have relied on to mark these individually as Anglo-Norman (presumably reflecting some Norse, Old Dutch, and maritime influence) and not Old French? or vice-versa? DCDuringTALK16:18, 30 January 2010 (UTC)[reply]
Four different people with four different opinions. You see why I brought this up, then? I dunno Lmaltier, pichon is only used in Picardy to mean fish, so should we therefore exclude it from [[Category:French nouns]]? I wonder how the Arabic or Chinese languages would compare to this? For example, we don't accept ==Chinese== as a header because it has to be Madarin, Min Nan et al. Mglovesfun (talk) 20:52, 30 January 2010 (UTC)[reply]
No, regional French words are French words, of course. Don't forget the If Anglo-Norman and Old French are considered as separate languages... We must be consistent, that's all. But I think the best solution is to follow ISO. Good reasons not to do so, whatever they are, are disputable, because ISO also has good reasons to take its decisions. Lmaltier21:41, 30 January 2010 (UTC)[reply]
The assumption (not mine but from reading in Leeds University library is that Anglo-Norman is a dialect. Ergo, anything written in Anglo-Norman must also be Old French. I agree with Lmaltier, but I don't mind reduplication, I mean would you want to remove Galician and Asturian sections from words that already have Spanish? No. I think it's the same sort of thing (language vs. dialect). FWIW I've been looking for an address to write to the ISO 639 committee about it, but I can't yet find one. Mglovesfun (talk) 13:18, 31 January 2010 (UTC)[reply]
If a word is used in a language (as we define the language, or as ISO defines it), it gets a section for this language. If it's not used in this language (as we define the language, or as ISO defines it), it does not get a section, I don't want to say anything else. If we consider Anglo-Norman and Old French as separate languages, then, if a word is used only in Anglo-Norman, it does not need a section for Old French. You assume that Old French includes Anglo-Norman. If we assume this, you are right, but that's not what ISO assumes, it seems to consider it as a separate language (close to Old French). Lmaltier13:31, 31 January 2010 (UTC)[reply]
Right-floating declension templates
{{sv-adj-reg}} is currently a right-floating template. The idea is to have a nice compact template that can appear alongside the rest of the entry. Is there a particular reason not to do this, i.e. for templates like {{nl-decl-adj}}? --CodeCat14:38, 30 January 2010 (UTC)[reply]
I don't remember the particular reasons, but the prevalent common practice in English Wiktionary is to avoid right-floating declension and conjugation templates, although there are some deviations from it. I for one prefer non-right-floating ones, and I think this should be unified across English Wiktionary for various languages rather than some languages having it one way while other languages having it another way. --Dan Polansky15:16, 30 January 2010 (UTC)[reply]
I don't know, I think this looks quite nice. All the essential usage information is on the left, and additional declension information is on the right, out of the way. --CodeCat15:27, 30 January 2010 (UTC)[reply]
In your browser maybe, if you open that page in a small screen, for example on a mobile, you will notice that everying gets misplaced - particularly the definitions end up after the template. (You can simulate this by simply resizing your browser to a few hundred pixels wide - many mobile divice screens are narrower than that template itself) Conrad.Irwin15:34, 30 January 2010 (UTC)[reply]
Hmm, I hadn't considered that. So what would the best way to place this template? In a section such as ===Declension=== by itself? Collapsible perhaps? --CodeCat15:39, 30 January 2010 (UTC)[reply]
The Swedish ones are the only ones that still float right and there's been a slow but steady cleanup process to turn them into left-floating tables. Aside from layout issues it's also semantically cleaner if all inflection tables are in their own section rather than some in, some out. For just showing a few key inflections we use the inflection-line templates. --Bequw → τ16:14, 30 January 2010 (UTC)[reply]
Why do we let the needs of small-screen mobile devices govern our layout decisions? Our entries are mostly much too long to be useful on them. Meeting their needs would seem to require customization or other changes at the level of MediaWiki software. DCDuringTALK16:36, 30 January 2010 (UTC)[reply]
If that were the only reason, then I'd agree with you. However, that is not the only reason this is a problem. Data structuring is also at issue. When the declension might be in the inflection line or might be in the inflection/declension section or might be somewhere else, then it's a problem. It's also a problem to have certain huge declension tables floating right and displacing content for following language sections out of their section as a result. --EncycloPetey18:35, 30 January 2010 (UTC)[reply]
14 hopefully, maybe #4 (or just a small #1) - can we find any generic senses to divide up bridge? "something that connects" and "something that resembles a bridge over a river" maybe? Conrad.Irwin01:11, 1 February 2010 (UTC)[reply]
An option is to make them Wikisaurus pages (which makes sense, as they actually are thesaurus pages), with one of these syntaxes. I prefer the second one (round brackets), the current one is not user-friendly at all. Lmaltier07:45, 31 January 2010 (UTC)[reply]
I certainly prefer the ISO-code prefix over "in language name", since that latter option severely limits potential appendix names. I'm not personally fond of parenthetical language identification for these appendices, but again, it would be far better than "in language name". --EncycloPetey07:41, 31 January 2010 (UTC)[reply]
Flood flag proposal
Hello there all. I would like to propose that we add the flood flag here on the English Wiktionary. Basically, if you know that you are going to make a lot of edits fairly quickly, then an administrator can give another user or themselves the flood flag, and the edits that they make will be marked as bot edits, while not actually being a bot. It is just a way to help ease the load on the RC window for the other users of the English Wiktionary. Basically, if you know that you are going to make a fair amount of edits in a short period of time, you can ask one of the administrators to give you the flood flag to help prevent those edits from clogging up the Recent Changes. Comments, suggestions? Razorflame00:20, 31 January 2010 (UTC)[reply]
I don't think putting the bot flag on non-bot edits is ever going to make sense, since it's blatantly untrue. Wouldn't it be better to have a new "high-speed mode" flag if we were going to do something like this? (The current "minor edit" checkbox is probably enough, isn't it?) Equinox◑00:25, 31 January 2010 (UTC)[reply]
I see the logic of that. Why reinvent the wheel, if the point is just to keep mass-edits off the recent edits screen? bd2412T02:26, 31 January 2010 (UTC)[reply]
I think this is a reasonably sane proposal, lots of our admins run through clean-up lists, making a few hundred boring edits an hour; whether people would actually use this functionality is up to them, so it might end up just being complicated for no reason. Conrad.Irwin11:14, 31 January 2010 (UTC)[reply]
What would be useful would be to separate the edits that are the assisted addition/balancing of translations from other edits. I find it hard to see through to actual content changes on my watchlist currently. I can't think how to do this though. Thryduulf14:56, 31 January 2010 (UTC)[reply]
If we were to install the AbuseFilter extension, we could get it to mark edits that are just adding translations with a "tag" and then filter recent changes by tag [3] (we might have other uses for such an extension too mwhoahahaha) Conrad.Irwin14:59, 31 January 2010 (UTC)[reply]
Well, it has a few dangerous features; if we were to install it I would ask that we did not activate the feature to automatically de-sysop people, (or to automatically indefinitely block them); I feel that the potential for abuse using those is rather high, and we have from time to time had less-than-trustworthy administrators. Over all though, it might be a useful tihng to have. Conrad.Irwin09:34, 3 February 2010 (UTC)[reply]
Currently, we require a vote to mark edits as bot-edits. This would be a big change to that policy. (Just pointing it out in case someone didn't notice.) That's not to say I'd vote against it. We'd need clear guidelines on the use of the flag. (Should an admin only give it to himself? only to others (as a sanity check)? either one? (I'm tending toward the latter option, fwiw.) Perhaps there should be a rule that a flag should not be granted for more than a certain number of edits or a certain length of time: require a bot vote for that. Perhaps there should be a rule that only certain types of edits (which?) should be allowed with a flood flag. Perhaps there should be a rule that a flag removed by a second admin should not be restored by the admin who granted it in the first place or by the person whose edits are flagged. Etc.)—msh210℠16:16, 4 February 2010 (UTC)[reply]
I'm not sure I completely agree with your first sentence. When we vote on a bot, are we voting on whether it can mark its edits as bot-edits, or are we voting on whether it can be run at all? I mean, if I create a bot and start a vote for the bot flag, and the vote fails, it's not like I can go ahead and just run it without the flag, is it? (I guess there are a lot of hypothetical distinctions that could be made, e.g. between "a computer program is performing the edits" and "an account with the bot flag is performing the edits", that we haven't historically worried about too much, because we mostly forbid the symmetric difference.) —RuakhTALK19:58, 4 February 2010 (UTC)[reply]
I always considered the difference to be that a voted on bot's edits were not being reviewed continually by a human. On the assumption that the user was doing something by hand, or using some technology that allows them to see in real time what is happening (WT:ACCEL or WT:AWB for example), it should be fine to run it under a flood flag. Conrad.Irwin20:17, 4 February 2010 (UTC)[reply]
Adminship on another account?
Could this account get admin privileges? I only use it when I'm doing audio pronunciations, but sometimes I see vandalism etc. and it's tiresome to log out and back in as Equinox and lose the thread of what I was doing. Xonique03:21, 31 January 2010 (UTC)[reply]
WONDERFOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOL (just to get it out the way). Equinox, just get commons to delete the other Equinox; it's a mere six steps of bureaucracy and a slab of patience away. Conrad.Irwin11:11, 31 January 2010 (UTC)[reply]
I can't. He has two "useful" edits (from 2006), and "usurpation will not be allowed in any event where the account has been reserved by somebody else as a Single Unified Login (SUL) account", which it has been which it appears to have been — but then how come I was able to get Equinox on here? I dunno. Too fiddly and boring for me to bother with really. Never mind. Xonique19:12, 1 February 2010 (UTC)[reply]
When we add FL citations to an FL entry, should it be translated to English? The {{cite}} template does not have parameters for translator, year of foreign language edition, translated title, translated text, a link to translated source. I tried my best to add all this using the current parameters: Citations:szökőár. Is this format acceptable? --Panda1015:01, 31 January 2010 (UTC)[reply]
Translations of foreign-language citations needn’t be “official” (though it is better if they are). The translations should be further indented and italicised. See δημοκρατία#Noun for a couple of fairly well-formatted examples. If I were you, I just wouldn’t bother with those citing templates. †﴾(u):Raifʻhār(t):Doremítzwr﴿19:25, 31 January 2010 (UTC)[reply]
Problem with Google Books
Does anyone know how to actually give feedback on google Books? All I can find is a tab with two options, neither of can adequately convey that "You mistakenly identified this copy of Oliver Twist as Wuthering Heights." [4]. --EncycloPetey22:25, 31 January 2010 (UTC)[reply]
I don't know if they are actually making corrections. I reported an error where a work's bibliographic information was wrong, the year being off by a century where a "19" was read instead of "18," and I got back a standard non-reply: "I appreciate your taking the time to offer us this feedback and encourage you to continue to let us know how we can improve Google Books. As this is still a young program, new features are under consideration and your feedback is very helpful." Dominic·t05:46, 1 February 2010 (UTC)[reply]
This is defined as an obsolete spelling of "several". In fact, it is nothing of the sort. That is simply what s looked like before the nineteenth century except at the ends of words. Unless hundreds of thousands of other such "spelling variants" that use the non-terminal s are to be admitted, it makes little sense to allow this one. At the moment ſeveral is the only word listed among English obsolete spellings. Also allowing this entry would by implication disqualify any entries that have only been recorded with the ſ. --82.0.9.2317:06, 14 January 2010 (UTC)[reply]
I broadly agree, we should probably have a policy against these, in the same way that we have equus, but not EQUUS, EQVVS, or even eqvvs. Possibly move this to the Beer Parlour, then delete it when it is official a 'bad entry title', unless we already have such a policy. Mglovesfun (talk) 17:11, 14 January 2010 (UTC)[reply]
What do we do about words in e.g. Arabic that have a different spelling when they are "final forms" (if that makes any sense — I don't know much about it)? This seems possibly analogous: it's a typographical variant rather than a different word. Equinox◑17:33, 14 January 2010 (UTC)[reply]
I believe that Arabic final forms in fact stay, as do Hebrew final forms, because while those may be a typographical variant, they are in fact a fairly universally recognised standard. In Hebrew, you simply don't use נ at the end of the word, and you do not use ן at the beginning or in the middle. I'm thinking that the Arabic form-letters are the same. This is very different from the English nonterminal S, because in English now, around the world, we don't actually /use/ it. --Neskayacontribs – talk?09:34, 15 January 2010 (UTC)[reply]
My feeling is that typographical variants should be hard redirects. (At least in the case of long esses; conceivably other cases might be different.) Ƿidsiþ17:44, 14 January 2010 (UTC)[reply]
I am also of the opinion that typographical variants with ſ should be redirects. There is very little use in spending effort in maintenance of such a large class of typographical variants. The way I see it, the only conceivable way that a user might want to know what ſeveral means is if they copied and pasted it from a web site, not knowing that ſ means the same as s. In such a case, they'd figure it out just as easily if they were redirected. --Internoob (Disc.•Cont.) 03:56, 15 January 2010 (UTC)[reply]
FWIW I have created some words like coeur as "nonstandard spellings" in French. AFAICT œ is disappearing from French as it doesn't appear on an AZERTY keyboard. Mglovesfun (talk) 08:31, 15 January 2010 (UTC)[reply]
Question: are there cases where hard redirects would cause problems, such as an orthographic variant for one word in one language being a valid spelling for another word in another language? If so, then redirects will not solve this problem. Answer: YEs, there are many such situations, especially where Latin and its derivatives are concerned, such as Latin iuncus which cannot be redirected from juncus, since that is a valid English word. Likewise the pairs io / jo and iota / jota have this problem as do many more words. Additionally, such hard redirects would break the capitalization redirects currently in place when someone typing "justus" would be redirected to iustus and not Justus. The proposed solution simply creates different problems instead of solving the current one. --EncycloPetey05:05, 17 January 2010 (UTC)[reply]
It's true that in Latin hard redirects can't be the solution, however, I think i/j and u/v are the only real cases where an orthographical variant has gone on to become a separate letter in its own right. Certainly in the case of long esses, they are no longer used in any language. So I see nothing wrong with taking this on a case-by-case basis. Ƿidsiþ06:45, 17 January 2010 (UTC)[reply]
So, what do you advocate for an extinct language that did regularly use long esses, like Middle English? You can't use the "redirect because it's no longer used" argument because none of the words in the language are being used now. The issue can't be handled on a case-by-case basis because this one decision can affect whole languages. --EncycloPetey16:45, 17 January 2010 (UTC)[reply]
But a redirect is only a problem if these other language, such as Middle English, treated the two esses as semantically different. I don't believe they did, but I could be wrong. If it was still a typographic variant in those languages, a redirect is fine (not the case for the u/v i/j in Latin since other languages treat the letters as semantically different). --Bequw → ¢ • τ06:14, 20 January 2010 (UTC)[reply]
Can someone explain what problems are caused by having {{obsolete spelling of}} entries for terms written with the long ess? In re Neskaya’s point: I don’t see why present currency matters; we catalogue all manner of obsolete and extinct language forms. Text written with long esses remains written with long esses irrespective of when it is read. If these are to be hard redirects only, how will we explain to baffled users (the ones Internoob alludes to) what they just saw? †﴾(u):Raifʻhār(t):Doremítzwr﴿05:22, 17 January 2010 (UTC)[reply]
People who come across the long ess will generally come across many of them within the same short text. It would nearly always be self-explanatory. An editor wishing to quote a single word will use modern typographic conventions. The problem is, there are tens of thousands of such typographic variants (probably hundreds of thousands if foreign entries are to be accepted). Unless a large proportion of these is added, it's very unlikely to help anyone. I suggest a much better solution would be to supply example quotations in the redirected entries that use the long ess. Besides, presumably this sort of rationale could be used to justify the creation of separate entries for script a like "ɑbɑcus". "Text written with long esses remains written with long esses irrespective of when it is read." - But not really, surely. Modern editions of Milton or Shakespeare don't retain their long esses. --81.108.129.9414:25, 17 January 2010 (UTC)[reply]
Again, I don’t see what harm their presence causes, only that done by their exclusion. FWIW, I wouldn’t mind soft-redirect entries for terms written with ɑ (as long as they’re attested), though I think they’d be virtually useless, since they’re so ordinary-looking as not to be registered as distinct by most readers (something which is patently untrue of ſ). Besides, use of the long ess does not seem to be a purely typographical consideration; for the sake of example: an s occurring at the end of a non-terminal morpheme of a compound word will, in many cases, not be written as ſ; also, note the unusual usage in Ralph Cudworth’s True Intellectual System of the Universe wherein hypostasis is written Hypostaſis in the singular but Hypoſtaſes in the plural. Whether people read modernised editions of many early works does not change how the original works were written (i.e., those ancient dead-tree editions in libraries aren’t magically updated in accordance with the development of our typographical conventions). †﴾(u):Raifʻhār(t):Doremítzwr﴿03:29, 18 January 2010 (UTC)[reply]
They're only ordinary looking for native speakers. Try reading Cyrillic in both a standard font and italics, and you'll quickly become confused, since the italicized (or script) versions sometimes do not resemble their standard printed counterparts. I can find The or THE used in the titles of books and articles, and we assume that a reader will know to decapitalize to find the entry. We need not create redirects for such typographical variants, and the same is true for other variant script forms. --EncycloPetey16:29, 18 January 2010 (UTC)[reply]
Let’s consider, in practice, what most readers first encountering the long ess are likely to think: read the Edward Lisle quotation at cyons, and consider the likely consequences of the passage being read by someone unfamiliar with ſ. I’d guess that the same confusion arises far less often with a and ɑ (in which case the latter might, on the rare occasions that it is misread, be confused for o, I suppose). As for the case of Cyrillic, is it as closely analogous that there are separately-encoded glyphs for its italic forms? If so, it seems entirely reasonable to have soft-redirected entries for {{form of|Italic rendering|word|lang=ru}}. Are there any search statistics we can draw on for guidance here? I was thinking of the search frequencies of terms that are searched for where substituting their ‘f’s for ‘s’s yields legitimate terms in languages that use or once used ſ and of the search frequencies of terms that are searched for where substituting their ‘o’s for ‘ɑ’s yields legitimate terms in languages that use or once used ɑ; were we to have that information, then perhaps we could starting making some reasoned decisions informed by evidence. Of course, we’d have to correct the relative frequencies, given the fact that ɑ, in absolute terms, will be encountered far more often than ſ by a general audience. Do we have these statistics? †﴾(u):Raifʻhār(t):Doremítzwr﴿19:59, 31 January 2010 (UTC)[reply]
Someone first encountering the long s on paper is likely to be out of luck, and only someone encountering the long s in computerized form for the very first time will need to look it up. We work with Unicode, which is explicitly not a glyph encoding; in this case there aren't separate characters encoded for the italic forms of Russian characters. With the exception of one rare African language, no languages use the Unicode character ɑ; many might have used the same glyph as ɑ for LATIN LOWERCASE LETTER A (a), but that's a glyph issue, not a character issue.--Prosfilaes00:28, 10 February 2010 (UTC)[reply]
All the Middle French texts I have read use the long s, but I always enter with with a 'modern s' because, well, I don't even have to explain, do I? Mglovesfun (talk) 14:30, 17 January 2010 (UTC)[reply]
Well, I read the 'unicode' versions. It depends how you turn the printed/handwritten original into something computer readable. That being said, not really, no. Mglovesfun (talk) 20:09, 31 January 2010 (UTC)[reply]
Hmmm. Well, it seems at variance with our descriptive ethos for us to have terms listed spelt in ways that would never be encountered in that form in the original corpora, even when there are no technical restrictions upon faithfully representing the original forms; it is, IMO, analogous with listing Ancient Greek terms using (perfectly bijective) Roman transliteration. I see the merits (though not always the primacy) of lemmatising the easier-to-input spellings (i.e., with s substituted for ſ), but not to the point where the real forms actually encountered in print are reduced to imperceptible redirects, or worse, are completely banned from being listed. †﴾(u):Raifʻhār(t):Doremítzwr﴿20:23, 31 January 2010 (UTC)[reply]
But do we list Ancient Greek terms in all capital letters? Wikipedia says "Modern editions of Ancient Greek texts are usually written with accents and breathing marks, interword spacing, modern punctuation, and sometimes mixed case, but these were all introduced later." Is that the tradition we're following here?--Prosfilaes01:40, 9 February 2010 (UTC)[reply]
IMO, we should list the forms as are actually used in the original works, given that those are the forms that would actually be encountered in the corpus. That said, we should lemmatise the standardised forms. By the same reasoning, the standard laid out at Wiktionary:About Latin#Orthography for Latin entries is just fine, and I’d support that kind of standardisation to make page titles easier to guess; however, I think it goes too far in disallowing even redirecting entries for the numerous variant forms that exist (especially those from Mediæval Latin). †﴾(u):Raifʻhār(t):Doremítzwr﴿14:48, 9 February 2010 (UTC)[reply]
I'm not sure what you mean by corpus. A linguistic corpus would usually obliterate the long-s. The vast majority of the books our users would encounter in Middle English don't use the long s.--Prosfilaes00:28, 10 February 2010 (UTC)[reply]
I think ſeveral meets several of our criteria for inclusion. It's a word in a language, it's attestable with a given meaning. There are other pretty similar cases; notably ligatures, variable capitalization. These are better described as typographical variants rather than spelling ones. However Doremítzwr makes a good point. These forms are used in print, and people may want to know what they mean. So do we exclude them, keep them, or keep them only as redirects? æ seems about the same as ſ is this context. Mglovesfun (talk) 20:47, 31 January 2010 (UTC)[reply]
I didn’t have the time to explain this fully earlier… The relation between æ / œ and ae / oe is not bijective, in fact, it is neither injective not surjective. In layman’s terms, this means that not all cases of æ / œ can be written as ae / oe and not all cases of ae / oe can be written as æ / œ. For examples of the former, neither the aforementioned Icelandic-derived Æsir nor the surname of the Norwegian deep œcologist Arne Næss can be correctly written digraphically, as *Aesir and *Naess, respectively; the same is (or at least was) the case with œthels taken from French — à contrecœur and coup d’œil, for example, ought not to be written with their œthels supplanted with oe digraphs. Vice versa, for Latinate and Græcian words, æ and œ can only be used for Latin ae and oediphthongs and for Ancient Greek αι(ai, “alpha-iota”) and οι(oi, “omicron-iota”) diphthongs; in cases where the Latin ae or oe represent two distinct vowels (as in aëneolithic and poëm) or the digraphs derive from any of the Greek digraphs αε(ae, “alpha-epsilon”), αη(aē, “alpha-eta”), οε(oe, “omicron-epsilon”), οη(oē, “omicron-eta”), ωι(ōi, “omega-iota”), ωε(ōe, “omega-epsilon”), or ωη(ōē, “omega-eta”), using a ligature is impermissible; in old-school typography, those various diphthongs that represent vowel pairs rather than diphthongs were written with disambiguating diæreses on the e as aë and oë. (I don’t really know whether all those digraphs actually occur in Ancient Greek, but that’s irrelevant to our considerations, anyway.) That whole spiel’s just a long and drawn-out way of saying that the use of the æsc and œthel ligatures in English is far more an etymological spelling consideration than it is a typographical one. †﴾(u):Raifʻhār(t):Doremítzwr﴿02:08, 1 February 2010 (UTC)[reply]
Please note that sometimes long ſ actually contrasts with round s (i. e. they refer to semantically different words). There are several examples in German language. -- Prince Kassad21:01, 31 January 2010 (UTC)[reply]
In such cases, when spelling is involved, they should be accepted. But in most cases, it's only a typographic issue, not a spelling issue. It's the same for ligatures: some are typographic issues (e.g. the ligature between s and t, very common in old books), some are spelling issues (e.g. in French, you spell œil (ligature between o and e), but moelle (using a ligature would be a misspelling)). Lmaltier21:13, 31 January 2010 (UTC)[reply]
In formal terms, then, for German (and some other languages?), ſ and s (as formerly distinguished) exist in a surjective relationship with the present-day use of s only; this means that conversion from ſ + s to only s results in a loss of information. As you’ve said, where this distinction exists, the long ess should be accepted, so we’re already agreed on that. The question is, what shall we do in cases where there is a word written with ſ in a language where the distinction is merely typographical which is spelt identically with a word written with ſ in another language where the distinction can be semantic, as in the case with German? †﴾(u):Raifʻhār(t):Doremítzwr﴿02:08, 1 February 2010 (UTC)[reply]
I propose to include words as they are spelled. This principle is simple. If spelling is not involved in a typographic variant, this variant should not be included (but it may be useful in some cases to comment inside the page), except possibly as a redirect. Lmaltier19:00, 1 February 2010 (UTC)[reply]
It's simple: I propose not to create pages such as ester with a ligature between s and t, because this ligature is purely typographic, and the actual spelling is E + S + T + E + R. But to create œil and moelle with the ligature in œil but not in moelle because they are the actual spellings, and spelling another way would be an error. The long s falls in the first case (pure typography), as far as I know, because it's the same letter as s, and does not deserve entries (except if there are rare exceptions where writing with an ordinary s would be a misspelling, but it seems surprising). Lmaltier22:13, 1 February 2010 (UTC)[reply]
But as Prince Kassad has already noted, the long ess sometimes makes semantic distinctions in German, so it falls into the second category, not the first. By contrast, ſt, ff, fi, fl, ffi, and ffl are genuinely examples of your first case, being ligatures purely typographical in nature. †﴾(u):Raifʻhār(t):Doremítzwr﴿01:20, 4 February 2010 (UTC)[reply]
I don't really understand what "it makes semantic distinction" means. If this dictinction was only in printed works, not in hand-written works, then it may be considered as purely typographic, and not a spelling issue. In such a case, it might deserve an explanatory note, not a normal page. Lmaltier06:36, 4 February 2010 (UTC)[reply]
An example is the German Wachstube, which can be spelt Wachſtube only in the sense of guardhouse, because the s in Wachstube(“tube of wax”) is the last letter of a component word of the compound word. I see no reason why the same distinction would not be / have been made in handwriting, and it’s quite likely that some English writers made such a distinction for compound words as well. Of additional import is the fact that ſ is not entirely obsolete in German (at least not in fraktur typefaces), retained as it is in the proprietary name Jägermeiſter, on this sign in Berlin, and so on. †﴾(u):Raifʻhār(t):Doremítzwr﴿11:01, 4 February 2010 (UTC)[reply]
OK, typographic rules were taking such differences into account, for printing. But this does not make a difference in spelling. Lmaltier22:12, 4 February 2010 (UTC)[reply]
The typographic difference is linked to the etymology, OK. But the spelling is the same, and has always been the same, I think. In Wachstube, I would remove the "Alternative spellings" section and add a note explaining that, when the long s is used in the typographic style of the text, the meanings can be distinguished by typography. Lmaltier19:09, 5 February 2010 (UTC)[reply]
What about the analogy with u–v and i–j? Those used to be typographically indistinct, but they became distinct with time. And what about s–c–k? Some (of the more extreme) spelling reformers want to change all uses of c (except in (deprecated template usage)ch, I think) to s (when pronounced s) and k (when pronounced k); in such a case, any traditional uses of c would be etymologically-based typographic differences. Is *ivstise a mere typographical variant of justice? †﴾(u):Raifʻhār(t):Doremítzwr﴿14:25, 8 February 2010 (UTC)[reply]
Well, due to the confusion wrought by the ambiguities of indentation, I have no idea what assertion of mine your posting is meant to be in response to. †﴾(u):Raifʻhār(t):Doremítzwr﴿16:48, 8 February 2010 (UTC)[reply]
Random section break for easier editing and reading. While we accept "all words in all languages", we don't necessarily accept all typographical variants. I like EP's point, The does indeed seem to be a typographical variant of the when the first word of a sentence. Are we ready for some sort of vote? AFAICT the long s is not used in any language currently. Could we therefore eliminate it all together? What are the other options? Allow anything that's attested, or have them all redirect to a message explaining what the long s is. Frankly, I feel uneasy about all three of those. Mglovesfun (talk) 14:46, 8 February 2010 (UTC)[reply]
As I said above (in my post, timestamped: 11:01, 4 February 2010), “ſ is not entirely obsolete in German (at least not in fraktur typefaces)”. However, even if the long ess were not used at all nowadays, it was used in many languages, and given that we document not only current languages, but also ancient, extinct ones, our self-imposed duty to describe language accurately as it is used would seem to require us to include the long ess in the entry titles of, for example, Middle French words (cf. EncycloPetey’s post, timestamped: 16:45, 17 January 2010) — this counters your point that “AFA[Y]CT the long s is not used in any language currently”. I imagine that this argument could be applied to other typographical variants, but in practice ſ is just so palpably different as to make it exceptional simply on the basis of its unfamiliarity. †﴾(u):Raifʻhār(t):Doremítzwr﴿16:48, 8 February 2010 (UTC)[reply]
It's not as it is used; modern publication of Middle English never uses the long-s. The long-s was but one of many, many alternate letter forms, ligatures and scribal marks used in the manuscript tradition; when transcribing that into modern typeset material, it is most faithful to merge the distinction between ſ and s the same way the distinction between õ and on is merged. Given that any word ending in "on" or "om" could be written with õ if space was needed on the line, it seems pointless to record those and anyone who can read the scribal tradition already knows this, and would silently convert the long-s.
More importantly to me, roughly one third of the words from the Unix words file would use the long-s. If we assume that the 42,000 entries in Johnson's dictionary represent the entirely of pre-1800 English, then there should be 14,000 English entries using the long-s. They're so low value that I hate to see a half-assed job; it serves no goal.
I can speak for English and Middle English, and I suspect the same is true for Middle French. As for German, where the usage of long-s was more complex and computerized use exists, I think the right answer might be different.--Prosfilaes01:36, 9 February 2010 (UTC)[reply]
What about the original texts (they’re the primary sources)? What about people who don’t know the scribal tradition? –The further we depart from how a text appears, the more difficult it is for a user to use this resource in conjunction with it. It may be a lot of work, but so is creating form-of entries for inflected and conjugated forms. And whereas no one (past a few weeks’ learning) has difficulty with English’s present participles or whatever, even native speakers can be tripped up by ſ. What are your thoughts about how we should treat ſ in German? †﴾(u):Raifʻhār(t):Doremítzwr﴿14:48, 9 February 2010 (UTC)[reply]
This won't matter at all to people who don't know the scribal tradition. Look at File:Yale University Library MS 365 Folio 33r.png; do you honestly think that the reader is going to be able to get "They fettyn forthe naylys ſtrong," for the first line, but not be able to figure out that the last word is strong? Both the scribal tradition and early printing made heavy use of abbreviations; see this page of Latin abbreviations for some examples, most of which Unicode doesn't have any way to type. Where as it may take weeks to get present participles, ſ should take about 15 seconds for any intelligent English speaker to learn; "it's a letter that looks like f without the full crossbar that's used in the place of s (except at the ends of words, and occasionally elsewhere) in early English texts." There's certainly cases in German where different words are spelled identically except for the long s. It's rare; the Wachstube example is the one example that comes up over and over again. My general opinion on this is that we should installed a (probably hypothetical) MediaWiki extension that redirects the long-s to the short s where ever found, and put a note on any word in any language where the long-s was used in more complex ways (i.e. German, but possibly Norwegian, etc.) giving the spelling of words with the long-s.--Prosfilaes00:28, 10 February 2010 (UTC)[reply]
OK, you’ve persuaded me. I reckon the best solution would be to have ſ → s autoredirection (as you suggest), and when that happens, to have some kind of notice (similar to the one that is displayed when one’s talk page is edited, except less intrusive) stating something like “You have been automatically redirected from [former pagename], a page title containing the largely obsolete character ſ (the long ess). For an explanation of the origins and rules of use of the long ess, see Appendix:Long ess.” In cases (like with Wachſtube) where the long ess makes a semantic distinction, we can have an entry for it, which would preclude such a notice. The same could be done for ɑ (vs. a), except for the case of that rare African language you mentioned above. What do you think of that way of doing things? Does anyone know how technically feasible this is? †﴾(u):Raifʻhār(t):Doremítzwr﴿14:51, 10 February 2010 (UTC)[reply]
I generally agree. I think it would be more useful to have the long-s spelling(s) on each German entry, but whatever. I think ɑ was more of a debating example; unlike the long-s, using it would be simply wrong from the Unicode perspective, and I don't think we'll ever find any live examples of English text using it, outside IPA and distinctly new orthographies for English.--Prosfilaes04:51, 14 February 2010 (UTC)[reply]
OK, well, since we're agreed on that, I suppose the only remaining consideration is the technical feasability of that character-dependent, autoredirection-triggered notice. I'll add {{look}} hereto and post in the Grease pit, to see if anyone is able to share some expertise. ∿’d: RaifʻhārDoremítzwr05:43, 15 February 2010 (UTC)[reply]
What am I supposed to do about these entries (you can see many more at Category:Mandarin_nouns)? Not only are they incorrectly formatted (noticed how the user has capitalised the pinyin sorting so they all go to the top of the list), but they are dubiously constructed (some SoP, others weird mixes of pinyin and English). I've already talked about the formatting numerous times at User_talk:123abc but the user appears to be ignoring me. The user's actions, IMO, constitute a kind of vandalism, as other users and I are going to have to cleanup all of this mess ourselves. Moreover, it's just going to get worse unless some kind of action is taken. Tooironic23:44, 15 January 2010 (UTC)[reply]
Yeah, I've been watching 123abc for a while, and one of the first things he did was start adding toneless pinyin, which (I believe) other Wiktionarians had displayed that they didn't want toneless pinyin on the English Wiktionary. This was a couple of months ago. Lately, I think that 123abc has been branching out and has been either trying to tackle words that are too hard for him to understand, or that are incredibly complex to format correctly. These are just my thoughts, so don't that too much away from this, as they are just someone's opinion. Razorflame00:03, 16 January 2010 (UTC)[reply]
He seems to have stopped creating toneless pinyin entries; however, he's decided to start tagging Tooironic's entries for deletion and correcting the changes I've made to toneless pinyin entries from nonstandard spelling to alternative spelling. Tooironic and I have both agreed to block this user as he's creating way too much hassle for us. He's been editing anonymously lately under the IP: 91.106.42.25. Jamesjiao → T◊C05:51, 2 March 2010 (UTC)[reply]
Transliteration decision
Hello there all. The Telugu words that are currently in the translation tables are the non-scientific transliterations of the words, which is why there are capital letters. I believe that that suits the Telugu entries because it does not seem that they have the ṅ in their transliterations. Basically, the ṅ is the equivalent of ŋ in Kannada. What I am requesting is that we figure out how we should have the Kannada transliterations set up so that we can make it standardized throughout the rest of the time that people are/will edit in the language and add translations. Therefore, there are two choices:
We can use the non-scientific transliteration method, which would basically be the same as the Telugu language, only with a lot more different letters with diacritics over them.
Seeing as how the Kannada language has several addition characters which would prove hard to transliterate in a non-scientific manner, I would like to kindly request that we use the scientific transliteration method, because that way, the characters with the diacritics and other things would make it quite a bit easier instead of making it really, really hard to decipher, however, we can discuss which one that you think would be better. Cheers, Razorflame23:29, 21 January 2010 (UTC)[reply]
There are not many Kannada or Telugu contributors here, any contributions are welcome and any consistent transliteration must be welcome too. Some languages have their own rules in Wiktionary or use some random ones - dependent on the contributor and the source of the transliteration. Eventually it must be standardised, of course. Please use About:<language> pages to describe any decision you make. Anatoli01:38, 27 January 2010 (UTC)[reply]
Since I've been making Kannada entries using the scientific transliteration method, I think we should adopt that for Kannada transliterations in the future. Since the majority of our entries are formatted in scientific transliteration, it seems like a good idea to make that the standard. Razorflame01:22, 22 February 2010 (UTC)[reply]
Reversible words
need name of the word that means, thenword spells the same forward and backwards please?
User:Unfree has provided us with heading-by-heading comments on gleeman at Talk:gleeman. This is the kind of thing that would cost a couple of hundred dollars (delivered) in the real world, in lots of 30 or more. How could we get more users of different types to make such comments? DCDuringTALK19:49, 23 January 2010 (UTC)[reply]
The template currently contains a boatload of information, and is rather complicated for a simple inflection line template. Now that declension table templates {{nl-decl-adj}} and {{nl-decl-adj-nc}} have been made, I believe it would be best to simplify the inflection line template to show just the comparative and superlative, in the style of {{de-adj}}. The remaining information can then be displayed in a separate table, which allows for more information to be displayed. And the more expanded view given in the declension table would also be more useful to those less familiar with Dutch grammar. Thoughts? --CodeCat20:33, 31 January 2010 (UTC)[reply]
