User talk:DCDuring/2010 QIV

Header images

Since you had commented on some of the related BP dicussions, I was wondering if you could lend your thoughts to WT:RFFF#Pre-language images. Thanks. --Bequw → τ 19:35, 1 October 2010 (UTC)

Replacing derived term lists with categories

Hello DC -- Are you sure this is a good idea? Isn't it a pretty radical departure from current practice? -- Ghost of WikiPedant 16:47, 3 October 2010 (UTC)

It could be viewed as radical. In what way isn't it a good idea? What specifically is wrong with it other than radicality?

I don't think it is all that radical. Without objection I had used categories similarly numerous times for affix DT sections. Others had hard-coded enormous lists of DTs and RTs for an apparently random assortment of words. What is new so far is only:

1. using categories for compounds and for bases in affixation
2. using etymology-located templates to autocategorize words into the categories.

I had never understood:

1. why the category approach was limited to affixes,
2. why we confounded historic and morphological derivations,
3. why we accepted variabilty in completeness of RT and DTs.
4. why we accepted inconsistency between parent and child views of derivation. (ie, Why should populating a Descendants section of a Latin word not take advantage of work done in the etymology sections of various descendant languages?)

I really like the idea of having complete listings of derived terms both from and morphological and a historical perspective and in comparing the two, where that is possible. In many cases, the one sheds light on the other. I am fascinated by the way productive prefixes arise, often by a kind of back-formation.

Procedurally, I had surfaced the matter at WT:BP, met no opposition, and little interest. With help I have an architecture of templates and categories that is sufficient for demonstration. I have been running what amounts to a prototyping exercise. In implementing, I am surfacing some practical issues and associated conceptual ones.

Feel free to take a look at the more densely populated categories in Category:English derived terms for examples of various approaches. The use of categories is probably less dramatic than the presentation of "grammatical" and/or morphological information in Etymology sections. Soon more "Etymology" sections will approach the screen space taken up by Pronunciation sections. I am not at all settled on the right way to present the morphological derivation data at the leafy end of the tree. There are many inconsistencies/experiments at present. DCDuring TALK 17:29, 3 October 2010 (UTC)

Radical? Nah. I only wish irony were made of a nice wine, I'd be getting shitface drunk. — [ R·I·C ] opiaterein — 16:57, 3 October 2010 (UTC)

I learned how much freedom one has to act without building consensus from that vote. DCDuring TALK 17:29, 3 October 2010 (UTC)

Daniel. works mostly in templates, appendices and not especially controversial categories. What you're doing is far more radical and I expect will be considerably more controversial. — [ R·I·C ] opiaterein — 18:01, 3 October 2010 (UTC)

He continues to do his work without documentation. He and his enablers generate many useful behavioral precedents. The general implication: One may as well be bold. DCDuring TALK 18:11, 3 October 2010 (UTC)

So DCDuring feels the inclination to be bold. How interesting. By the way, I have created and edited documentations for various Wiktionarian templates during the course of my life. Which other, undocumented, templates do you wish to be documented? Asking for the documentations or perhaps doing them yourself seem better alternatives than insulting the programmer. --Daniel. 11:30, 4 October 2010 (UTC)

DC, Getting back to the topic, what bothers me about the practice of shoveling derived terms onto separate category pages is that I don't think it is user-friendly. From an end-user's view, categories are not part of the entry and are (I strongly suspect) much more likely to be ignored or left unexplored than a (collapsible) section within the entry. I'm sure it's easy for an admin to automate this method of categorization, but that doesn't mean it is appealing or logical from an end-user's point of view. According to WT:ELE, which I see as a thoughtful guideline for how to structure a dictionary for end-users, derived terms are main-entry-level information, not poor cousins to be shunted off to an ancillary page. (PS: The "Daniel" vote was what it was, but I don't think any reasonable Wiktionarian should think a precedent has now been established giving every editor license to change the structure of the dictionary without consensus.) -- Ghost of WikiPedant 04:28, 4 October 2010 (UTC)

Re: "shoveling ... onto ... category pages". Users can add items under the category. I take your point that the form is somewhat discouraging. There might be advantage in having the bare category under a show-hide bare. Even better would be a line under the show-hide that invited users to add any missing related terms.

The tradeoff seems to be between having something that can be maintained as new entries are added at the expense only of work on the new entry. Without this kind our utilization of morphological information, I don't think the morphological information is worth the effort, which is why is one reason why it has languished as a vague hope. The other reason has to do with the ambiguity of the meaning of "derived terms" (morphological/synchronic vs historic/diachronic). — This unsigned comment was added by DCDuring (talk • contribs).

I for one disagree with moving derived terms into separate category pages as it was done in this edit to "stage". Can someone point me to a BP discussion that proposed this? --Dan Polansky 09:51, 4 October 2010 (UTC)

I, personally, agree with the idea of categorizing entries by their various roots, like how [[Category:English words derived from: load (noun)]] currently lists certain etymologically related entries.

As an initial suggestion, instead of simply linking from the entries to the categories, the current collapsible format from -ity (linking to [[Category:English words suffixed with -ity]]) looks better to me.

I noticed how the current category naming scheme that includes simultaneously "English derivations", "English derived terms", "English words suffixed with ness", "Hungarian derivations", "Hungarian nouns suffixed with -g" and "English words derived from: coach (noun)" is bound to be very confusing.

However, the existence of English words derived from: -er Etymology 1 implies that the common suffixal categories are being deprecated by you in the process of developing a plethora of etymological categories.

Since you are apparently going to rename more than 3500 categories, not counting the basic format of "de:Italian derivations" and "fr:Portuguese derivations", I strongly advise you to explain and seek feedback from the community about your vast project. Notably, I'm very curious about how derivations between two languages would be categorized, and if they would. --Daniel. 11:30, 4 October 2010 (UTC)

I appreciate the suggestion and had considered it. What I like about {{suffixsee}} and its sibling(s) is that the information appears quickly and on the same page, facilitating comparisons with other material on the same page. What I don't like about them is that they present the information in a one-column list, which limits the very advantage. I understand that it is not possible to make it a two- or three-column presentation. Secondarily, we would not have addressed GhostofWP's concern with casual-user-accessiblity. I have some thoughts on that point.

• Is my understanding about multi-column presentation not being available correct?

I am agnostic on the naming scheme. Any naming scheme that allowed for both morphological and historical derivations would be fine. For languages that are well studied it would be useful to enable or, at least, not foreclose the distinction between the two kinds of derivations even for within-language derivations.

re: -er {{suffix}} and siblings do not support Etymology/sense-level discrimination of category membership (nor the morphological/historical distinction).

-er was the first suffix that I ran across in experimenting/prototyping that had multiple etymologies that I was familiar with. I am not sure whether it would prove to be a good example. It is only a singular experiment. To some extent, it has succeeded in raising the question of this limitation of {{suffix}} et al. I would much rather use/integrate the infrastructure of the affix templates than replace it.

I know that I need feedback. I wanted to both have something to display that didn't leave too much to the imagination. There are issues of presentation of the morphological information in the Etymology sections that we haven't touched on.

I have intentionally limited myself to within-language derivations in English so that the specifics of such could be addressed. English seems to have more use of DT and RT headings than other languages AFAICT. Therefore the maintenance and completeness issues arise in English before they do in other languages. My investment of time in deploying the scheme has not been so great that I would object to radical alteration in support of the desiderata it can help with. I think those are:

1. Clarification of the meaning of derivation (morphological vs historical).
2. Integration of information between the trunk and twigs of derivational chains.
3. Presentation of morphological information.
4. Completeness and maintainability of DT and RT lists.

Thanks for taking the time to address this so thoughtfully. DCDuring TALK 12:55, 4 October 2010 (UTC)

Making {{prefixsee}} and {{suffixsee}} present their lists in multiple columns seems a very reasonable goal, so I simply have edited both templates to implement this function (specifically, I decided to split the lists into three columns).

I, differently, believe that these two templates do address GWP's concerns expressed at this discussion. The derived terms listed by prefixsee and suffixsee are transcluded at each entry; they don't require going to a separate page to see the category.

Per WT:CFI and other conventions, a list of related terms at beautify may display together beauty, beautiful, beautifulness and beautifully; it may even contain beauty mark and beauty sleep, but maybe it would be incomplete without various derivations of "beauty" and of "-ify". Perhaps the best choice would be comprehensively listing all the derived terms of beauty and all the derived terms of -ify together as related terms of beautify.

Your proposals, as I understood them (including the implied prospects of categorizing derivations of any language, and between different languages), when completely effected, would necessarily deprecate the systems of Portuguese nouns suffixed with -a and pt:Etymology. Therefore, a consistent and logical naming scheme would necessarily cover them all. Let me number a few related thoughts:

1. By comparison with English terms spelled with Á, I personally see either colons or quotation marks as unelegant and unnecessary, in names of etymological categories.
2. The categories English words derived from: collective (adjective), English words derived from: bass Etymology 1 and English words derived from: bill (noun) Etymology 3 are named very similarly: To discern between the root and the part-of-speech, parentheses are used. On the other hand, to discern between the root and the etymology, a certain capital letter is used. I believe this system of similar causes with different effects is particularly user-unfriendly in essence. It could even be particularly distracting in possible categories with more capital letters, like English words derived from: Big Brother Etymology 1. I then suggest using only parentheses to keep the POS and the etymology together.
3. Let me express a simple logical analysis based on simplicity as a goal: Unless there is any reason to keep both "derived words" and "derivations" in etymological categories, only one of them should be kept. I, personally, prefer "derivations". In addition, I think there is a pretty solid consensus on choosing "terms", not "words", in names of categories whose members may include possible multi-word entries.
4. Finally, based on this discussion as of now, the final category tree may include: Category:English derivations from Japanese 漢字 (noun), Category:Portuguese derivations from English own (verb) and Category:English derivations from bill (noun, etymology 3).

No worries; improving Wiktionary is my pleasure. --Daniel. 15:53, 4 October 2010 (UTC)

The change to {{suffixsee}} is beautiful.

re: GofWP. Before seeing this post, I had posted on this talk page to ask for his opinion about the new, improved {{tem[|suffixsee}}.

I am reluctant to include derivations based on affixes and clitics in related terms for single-word entries because it seems to draw in too many terms to make the related terms listing useful. Analogously, for multiword entries I would be reluctant to include any entries whose relatedness was based on prepositions, contractions, the close set of adverbs, and sometimes even the open sets of adjectives and adverbs. I am not sure what related terms should look like for any but the very simplest of multi-word entries.

Some other questions are:

1. how far to go up the same-language derivation tree to collect "related terms". My inclination is that what constitutes enough and too much varies greatly by entry. Some words
2. whether to include terms that are derived from related cognates.

I agree it would be useful to have a naming scheme that did not limit the application of the general scheme. I am not sure that every language would want to opt in, however. CJKV languages seem to have completely distinct naming conventions and needs for example. I don't know about agglutinative languages like Finnish and Hungarian and, for that matter, the special needs of English pharmaceutical and other chemical names.

I was very opportunistic and incremental in the prototyping stage about many things. I regretted choosing "word" at an early point. I like to visually differentiate the "term" from the preceding text and the following differentiators (PoS and/or Ety, so far). The differentiators need some separation from each other. One convention would be to have the word/term in italics and the differentiators in parentheses, separated by commas:

English terms derived from [langname] langword (etymology 1, noun, gloss)

[langname] would be omitted for same-language derivations. Italics would be unnecessary for non-Roman script. You have been dealing with such matters and probably have faced issues I have no idea about.

As to "final" in final category naming, I don't think that we can limit ourselves to two differentiators in the architecture. In Template talk:derv I mentioned this. Differentiation based on sense (for affixes especially) and possibly position/grammatical role (eg attributive role vs head role for nouns in noun-noun compounds) have already occurred to me. I don't think that exhausts the possibilities. For example, in Latin, there are the Pronunciation N headers, which are sometimes a substitute for Etymology. It would be unwise, IMHO, to not have some mode of extending the structure a bit.

I think there are very good reasons never to make simplicity a goal in itself rather than a important desideratum, a figure of merit, for architectural decisions.

• We have not discussed one very important conceptual issue. In this case there is a fundamental distinction between historical and morphological derivation. The problem with not respecting it is the confusion caused. To take a very simple example in English, it is useful to know both that Marxism#English is historically derived from Marxisme#French and is morphologically Marx + -ism.

It is my understanding that some languages have no major potential for historical etymology. In contrast, in well studied languages we expect to be able to find etymologies for almost every base in the language. But even in well studied languages the historical and morphological derivations may be meaningful for the same word, as in the Marxism example. There are also cases such as disagreement where the morphological derivation has two possibilities, but we would probably seek to select a single one for historical derivation purposes. Shouldn't disagree and agreement both have both historical and morphological "derived terms" categories? That would mean the potential in the naming systems for having both kinds of categories would have to be retained. Perhaps morphological derivation should be the default for many languages. For English I don't know. DCDuring TALK 17:26, 4 October 2010 (UTC)

The possibility of easily seeing a list of terms derived from the roots of a word (like adding {{dervcat|throw out}}, {{dervcat|throw}}, {{dervcat|the}}, {{dervcat|baby}}, {{dervcat|out}}, {{dervcat|with}} and {{dervcat|bathwater}} at the Related terms of throw the baby out with the bathwater) is not entirely devoid of merit.

Yet, a limit must be set, otherwise many or all English words would conceivably be listed at the Related terms sections of all entries of that language. Your proposal of not adding related terms based on "affixes and clitics", "prepositions, contractions, the close set of adverbs, and sometimes even the open sets of adjectives and adverbs." seems very reasonable, and I'm inclined to support it.

I believe the phenomenon of listing only few related terms excludes categorization as a tool. That is, for instance, Category:English words derived from beauty would list too many derivations, thus it would not appear in the Related terms section of beautify, beautifulness and so on. I would also probably oppose the creation of Category:English words related to beauty or other similarly obscure categories. Therefore, lists of related terms would be written per entry, individually.

Can you please tell me why Category:English words derived from: load (noun) does not contain the word load?

As for the title of the category, it seems we agree on the format:

English terms derived from [langname] langword (etymology 1, noun, gloss)

However, I believe italics are not possible in MediaWiki category names.

Despite I favoring "derivations" over "derived terms" in a previous message, both versions are fine to me, so I may as well support your decision of "derived terms".

I am aware that different languages may have different characteristics, causing the categorization to be affected. Perhaps the best initial decision would be simply to perfect the English categorization (including common pairs of languages that include English, such as maybe Category:English terms derived from Old Portuguese zevra), to subsequently provide a basis for introducing other languages into the overall system. I particularly like how the parenthesized text may be eventually adapted or expanded to contain additional information, including possibly (pronunciation 2, participle, gloss).

Can Marxism be categorized into English terms morphologically derived from Marx, English terms morphologically derived from -ism and English terms historically derived from French Marxisme?

Or, alternatively, can Marxim be categorized into English terms indirectly derived from Marx, English terms indirectly derived from -ism and English terms derived from French Marxisme? — This unsigned comment was added by Daniel. (talk • contribs).

It may be that we should first work on the derived terms and defer related terms, on which there is less apparent agreement. "Derivation" may have two or three possible meanings. "Related" is more vague, with numerous possible meanings. "Cognates" can be a very large class indeed. For related terms we may be able to somehow learn from and piggyback on what we work out and gain consensus on with regards to derived terms.

Also, there is no point in getting everything worked out in detail before others have a chance to consider the ideas and contribute. There may be some surprises.

In addition, I would like to slowly roll out any proposal in one language, preferably with entries not now well covered with Etymology and Derived terms and possibly only with respect to English. When we address cross-language derivation, Middle English, Anglo-Norman, Old and Middle French, and, possibly, Old English would be desirable languages for derivations into English.

OTOH I don't want to do anything that is not potentially extensible to a wide range of languages. Certainly all category and template naming should ba consistent with as wide a range of languages as possible. DCDuring TALK 04:46, 6 October 2010 (UTC)

I support all the proposals from your last message. Thanks for your work and for your thoughtfulness. --Daniel. 15:50, 7 October 2010 (UTC)

• Could this perhaps be moved to the Beer Parlour? I was also taken aback when I saw this going on and only now find a conversation about it. Ƿidsiþ 15:59, 4 October 2010 (UTC)

Of course it can. The whole effort emerged from BP discussions and the initial steps in the prototype/experiment were presented there without objection or even response. DCDuring TALK 17:26, 4 October 2010 (UTC)

I have a Question

Hi, thank you for the advice on my talk page. I'd like to ask - is there any use of references on Wiktionary, like on Wikipedia, or do you folks use common sense? I've seen the 'Citations' tab, but usually they're red links, so?
Thanks
--George2001hi (Discussion) 18:20, 4 October 2010 (UTC)

Here, citations and references refer to two different things.

"References" is a heading appearing at the bottom of a Language section in an entry. Below it are the authoritative sources that have been consulted in creating and editing the entry and/or that the user may wish to consult. They may be associated with footnote numbers, but usually aren't.

"Citations" refers to cited quotations from "durably archived sources" that attest to the usage of the headword of the entry. The Citations tab is one place such cited quotations might be. The preferred place is in the entry itself under the specific sense that has been attested by the citations. On some entries you will see in small type "[quotations]". Clicking on this will show the quotations for the sense.

"Citations" tend to command more respect than references, partially because copyright law makes it difficult to frequently quote definitions from the small number of high-quality unabridged dictionaries that are available, even in English. The wording of out-of-copyright dictionaries such as Websters 1913 and Century 1911/2 tends to be quite dated BTW. Hope that helps. DCDuring TALK 18:52, 4 October 2010 (UTC)

Thank you, it does help.
Thanks
--George2001hi (Discussion) 20:13, 4 October 2010 (UTC)

Category:Forest products

Do you need the empty [[:Category:Forest products]] for anything? There is already Category:Forestry. Could perhaps [[:Category:Forest products]] get deleted? --Dan Polansky 19:39, 5 October 2010 (UTC)

Where do you think Category:Paper sizes should go? For some reason it was in Category:Art. Forest products seems a subcategory of Forestry, but I don't get the principles of our topical classification scheme. Is there an explanation somewhere? DCDuring TALK 19:47, 5 October 2010 (UTC)

I don't know where Category:Paper sizes should go, but if you create a new category, you should populate it with at least five to seven members, I think. Wikipedia's W:Forest product says "A forest product is any material derived from a forest for commercial use, such as lumber, paper, or forage for livestock". So "paper", "lumber" and even "forage" could be members. To me, the category seems too specific or specialized for a dictionary semantic classification. I would probably never create the category "forest products" myself, but I have no strong stakes at getting the category deleted. When I have posted here yesterday, the category was empty; only now it is non-empty after you have added Category:Paper sizes to it.

I don't know of any explanation in Wiktionary of how topical categories are supposed to work. I have figured them out from looking around and seeking patterns and regularities. --Dan Polansky 10:02, 6 October 2010 (UTC)

Do you think you could take a run at a Wiktionary page on Topic categories. Such basic points as the differentiation between lexical and topical categories need some illumination. Principles and practices of naming categories, too. From the fact that you were willing to take on Wikisaurus, it seems you have an ambitious and systematic approach to things, which makes you a good candidate to get this kind of thing going.

Such a page would be a way for people to grasp what you are doing, try to help, and avoid inadvertently undoing what you are trying to do. Some may have some specific constructive advice on some particulars or even sources of knowledge on similar matters. Based on my experience in trying to understand the implications of integrating Etymology and Derived terms, morphological vs historical derivation, and related matters, I understand how hard it is to make a complete proposal or even to decide on the proper scope of a proposal. As this is a wiki - and a relatively collaborative, non-political one - it is supposed to be OK to have incomplete pages for an extended period of time.

This just reminded myself of a book that has been sitting unread on my shelves:

Though the authors use health classification as their leading example (shades of [[respiratory disease]]), they attempt to generalize and have a 30-page bibliography. DCDuring TALK 12:02, 6 October 2010 (UTC)

(unindent) I have rewritten the page Wiktionary:Categorization a bit. The page was already there, explaining what major classes of categories there are in Wiktionary. Topical categories are dealt with in Wiktionary:Categorization#Topic, a section that is redirected to from Wiktionary:Topical category and Wiktionary:Topic category. I have extended the section "Topic" a bit; a further extension may be needed. --Dan Polansky 12:28, 8 October 2010 (UTC)

Thanks. Wow! Is that ever out of date! For example, it doesn't mention the consensus practice on autocategorizing tags for obsolescence, rarity, and archaicness; register; dialect; and usage context, and with the less-generally accepted categorizion by grammatical category. The first trio, register, and dialect cannot readily be confused with topics, but grammatical categories and usage context can be and, apparently, are. We may need to clarify the distinctions explicitly. The widespread use of derivational categories, such as Category:English words suffixed with -ism, is not mentioned either. You may recall that MZajac even campaigned against topical categories as being outside of the purview of a dictionary. I suspect that we would benefit from some principles to guide our decision making about creating, naming, deleting, and structuring categories for sufficient consistency to help users and contributors. DCDuring TALK 19:52, 8 October 2010 (UTC)

-idae

Any idea what to do with the translations? Or should we make the entry Translingual? --Bequw → τ 01:19, 6 October 2010 (UTC)

For the translations how about WT:RE:fr and WT:RE:zh? As for Translingual, I wouldn't really want to go against EP's reasoned thoughts in this area, despite the sluggishness in getting coverage in Latin for New Latin. I suppose that users would expect us to have these terms covered.

Perhaps we need some kind of Appendix that explained enough about NL. taxonomic naming practices to allow someone not expert in Latin to make basic entries for the numerous missing terms. That would suggest some templates to mark terms that are only used in NL. and to mark classical terms used in taxonomic names. There is something strange about trying to keep nearly 2500 years of Latin under one language name. DCDuring TALK 05:07, 6 October 2010 (UTC)

user talk:msh210

I've responded to your posts at my talkpage, q.v., about the math-etymology book and about {{morph}}.​—msh210℠ (talk) 16:55, 11 October 2010 (UTC)

Feedback on books and lists

Hey! Thanks for the feedback at the community portal. I have started creating vocabulary lists at my user page. Do you know of any templates I can use to separate columns horizontally over the user page or anything that can help? (I don't want to end with an enormous vertical list).

Thanks! --Smoken Flames 00:54, 17 October 2010 (UTC)

See {{top}}, {{top2}}, {{top3}}, and {{top4}}, all used with {{mid}} and {{bottom}}, and {{rel-top}} (2 column, with show-hide button), used with {{rel-mid}} and {{rel-bottom}}. DCDuring TALK 01:14, 17 October 2010 (UTC)

Chunks, Collocations, etc.

Hi DC Here is an interesting link It is the third of three blogs on chunks and collocations from the Macmillan dict. web site. The body text contains links to the first two blogs, plus some other interesting links, such as to Macmillan’s new collocations dictionary. It touches on a point of interest for us, demonstrating that 1. collocations affect the meanings of words and 2. that chunks can be at least as common, statistically, as much ordinary single word vocab. All of which takes me back to the proposal to have a new tab on the entries for common collocations and lexical bundles containing the target word. Tell me your opinion when you get a moment. Cheers. -- ALGRIF talk 09:37, 17 October 2010 (UTC)

Thanks for the explanatory blog links. They provide a good summary but fall short of being evidence. Has a linguist published on this concept yet? (I'd prefer a book to academic articles.) That Macmillan finds it worthwhile to publish a collocations dictionary is interesting. I wonder how it is selling. That context statistically alters the distributions of possible meanings would not, to my mind, warrant much effort on chunking. I find the production of unnatural phrases by language learners a better rationale. BrettR's objection to giving the matter high priority in language learning seems valid to me, but I have little experience with language learning. I am not sure that I could see how user would take advantage of collocations in their learning. Where is the advantage of having a dictionary of collocations over participating in conversations or watching television?

Do we need to make any distinctions between "chunks" that are the result of real-world "collocations" of things, their attributes, and their actions (eg, birds, songs, berries, insects, worms, perches, roosts, and nests; fly, perch, roost, peck, feed, flock, migrate) and those that are more purely linguistic? The real-world context activates a set of associated terms that are more useful for conversations about the context. An out-of-physical-context conversation requires some cues to put a listener "in mind of" the real-world context.

How would you envisage us populating the proposed tab? Do we have a single Appendix that demonstrates the application of the notion to a word like "crime" (ie, one not too common, but common enough)? DCDuring TALK 10:45, 17 October 2010 (UTC)

Thanks for your input. Food for thought as always. I am basically interested in finding a suitable place in the Wikt where learners and users can quickly find typical collocations which in turn leads to examples of how a word is used, particularly the words with multiple definitions. The how is almost as important as the meaning, and way more important than the metadata the entries now contain. Furthermore, the lexical bundles are very useful, I'm sure you would agree. There is a grey line dividing the collocations and lexical bundles that can be accepted as entries in their own right, and those that are rejected as SoP etc. I feel that a tab of some kind would be a good repository for all these items that are so common that many people think they could be separate entries. I am coming to terms with your idea of placing derived term lists in categories pages. I believe that you and I are thinking about the same subject, but on either side of the grey division line. I will try to clarify my own thoughts by putting together some example Appendix entries, as you suggest. -- ALGRIF talk 11:03, 19 October 2010 (UTC)

I'm warming up to the tab/namespace idea for collocations. I suppose that it is not all that important for your intended purpose to make a distinction between real-world-knowledge-based collocations and more linguistically driven ones. I suppose that the academic rationale is the old idea of a semantic field. We are still cursed with the need to maintain some correspondence between our definitions and other parts of the entry using software ill-suited for the purpose. If more of our English-language entries were of higher quality, they would be more stable and glosses that helped users navigate would be more stable and reliable. OTOH, a rich population of collocations might facilitate the improvement of definitions for words and the identification of truly lexical collocations. (We still lack so many common lexical items: eg, yesterday I entered as follows.)

It would be very nice to be able to buy and deploy en masse collocations from COCA and BNC. How good are your grantsmanship skills? It is in the interests of the UK and US governments to facilitate worldwide learning of English, which includes lowering the cost of such learning. I wonder whether government or foundation grants might be available for such a purpose. If we don't go that route I don't see how we can rapidly (or ever, actually) populate such a collocation space. Do you have any ideas about how "crowd-sourcing" could be made to work for this purpose? DCDuring TALK 15:31, 19 October 2010 (UTC)

BNYA

I was about to speedy this as a clear protologism, but then I saw that you added the etymology. So, is this real? Should I RFV it? —Ruakh_TALK 02:22, 5 November 2010 (UTC)

I believe I just formatted the material present. I won't object to its deletion. As we have no explicit standards for inclusion of abbreviations, I don't ever (well hardly ever) challenge them. I suppose a basic requirement might be the includability of the abbreviendum (!). I haven't found any evidence at Google Groups searching for "burned now you are" alone or for "BNYA" with any of a list of common English words. We have many questionable senses among our abbreviations and need some efficient way to winnow them. DCDuring TALK 14:19, 5 November 2010 (UTC)

/*Romace Etymologies */

Hi, what about the lacks in the etymologies in the deriving latin words? You sent me a message if you remember, we could collaborate :). write soon --Zoologo 18:28, 11 November 2010 (UTC)

swing shift

I don't understand why you added a second definition. How is it different from the one that was already there? Ƿidsiþ 19:29, 16 November 2010 (UTC)

I found the existing one mostly unintelligible, wrong in at least one particular, and without the salient detail of most likely timing, As I know the term and as the quote shows there is nothing irregular about it in most cases. I did find a sense in one other dictionary about some kind of shift that had an irregular schedule designed to achieve some kind of labor-contract-driven scheduling objectives, but I am not sufficiently familiar to improve the wording of the previous sense to make it fit. I certainly didn't see how one definition could fit both or how the existing sense fit what I knew. DCDuring TALK 22:53, 16 November 2010 (UTC)

Well, I will have to accept the responsibility for any unintelligibility. But as for the rest...I called it a ‘work shift between the standard day shift and night shift’, you called it ‘A work shift between a day shift and a night shift’; I said ‘especially one lasting from afternoon to late evening‘, you said ‘such as from 4PM to midnight’. I'm not seeing any distinction here. If you think I got something wrong, just replace the definition, don't add a new one. But please stick a {{US}} tag in – it's not used here. Ƿidsiþ 21:35, 17 November 2010 (UTC)

I apologize for charging this entry with all the confusion I had with shift work, graveyard shift, night shift, and day shift. In this case, it was really the "irregular" and the other definition I found that came close to the "irregular" sense that threw me. DCDuring TALK 22:40, 17 November 2010 (UTC)

No worries, it looks good now anyway. Ƿidsiþ 06:20, 18 November 2010 (UTC)

Cleanup categories

I've been wanting some sort of clarity on this for ages. Do you consider bad categories to be clean up categories? Just in some WT:RFDO deletion debates you seem to want to keep some categories, not inspite of the fact they're bad/off topic, but because of it. What I understand from your comments (without being very sure) is that once the entries in the category are fixed, the category should be deleted. If I'm wrong, feel free to correct me. Mglovesfun (talk) 00:36, 21 November 2010 (UTC)

As a general rule, yes. I think that most categories have some information content which should be exploited even when the category is a bad structure for the information. Sometimes I can't put my finger on what the content is, sometimes not. Sometimes I am capable and willing to extract the content, sometimes not. DCDuring TALK 03:00, 21 November 2010 (UTC)

Wiktionary:Requests for verification#para-

What should I do with this? —Ruakh_TALK 16:27, 26 November 2010 (UTC)

I suppose there is suggestive evidence in favor, though the DTs haven't been entered. DCDuring TALK 17:06, 26 November 2010 (UTC)

Thanks. —Ruakh_TALK 17:15, 26 November 2010 (UTC)

Back on the wagon?  ;-) ​—msh210℠ (talk) 17:45, 2 December 2010 (UTC)

Phrasebook template placement

I think the standard practice is to place the template at the top, just under the language header. That's how it has mostly been used so far, anyway. —CodeCat 11:14, 28 November 2010 (UTC)

It squanders too much screen space for such a placement, IMO. DCDuring TALK 13:09, 28 November 2010 (UTC)

It's a big, gaudy template. I think that's deliberate to an extent to make it visible, but a compromise would be to tone it down a bit - a bit smaller and a bit less gaudy. Mglovesfun (talk) 13:36, 28 November 2010 (UTC)

very well

I've posted thoughts at [[WT:RFD#very well]] that you might be the best person to comment on.​—msh210℠ (talk) 17:45, 2 December 2010 (UTC)

I'm still thinking. I am not getting my head around "[modal verb] very well [bare infinitive]". I think it is the "clearly" sense. DCDuring TALK 19:34, 2 December 2010 (UTC)

to death

Hi DCD. Could you have a look at to death when you have a moment? I want to make sure I'm on the right track, and would value your input. Thanks, Pingku 23:37, 4 December 2010 (UTC)

Good catch. Looks good to me. I suppose someone could argue that it is just a figurative or hyperbolic sense of death, but it still seems idiomatic.

bust ass cold

As a native US English speaker, is this idiomatic? It looks to me like bust ass + cold. We'd need a definition for bust ass, meaning "very". Mglovesfun (talk) 14:27, 8 December 2010 (UTC)

Exactly as you say. DCDuring TALK 15:32, 8 December 2010 (UTC)

Luso- et al.

Should Luso-, Russo-, Sino-, et al. be in a category of their own? (Note, we have category:Demonyms (which should be [[:category:English Demonyms|English demonyms]], but that's another story) and category:Nationalities (which should probably be category:Nationality or deleted, but that, too is another story), but I was thinking of one just for these prefixes. What do you think?​—msh210℠ (talk) 18:34, 9 December 2010 (UTC)

Ah! I just discovered Appendix:Nationality prefixes. My question remains, though.​—msh210℠ (talk) 18:38, 9 December 2010 (UTC)

More structure doesn't hurt. But this is a kind of mixed topical - morphological category that doesn't have much precedent AFAIK. I've thought that there is some rationale for special treatment of language names because this is a multilingual dictionary. That rationale could extend to speakers, but it seems a stretch. Other -nyms are Category:English eponyms and retronyms, for which a linguistic rationale exists. We also have English names, with two subcategories, and English toponyms. Most of our English categories have more linguistic rationale than this. An appendix that contained all Demonyms and language names and had the associated prefixes seems much easier.

BTW, Category:English language could use a little structure itself. DCDuring TALK 21:51, 9 December 2010 (UTC)

All right, thanks for your input. I'll leave off categorizing, and link to the appendix from each entry bluelinked in it.​—msh210℠ (talk) 22:24, 9 December 2010 (UTC)

Fact

Ah sorry, I was not aware it wasn't used. Do we use the ((unreferenced)) one? I just figured using fact would take up less space and it was easier to remember. I'll try to remember the ones you put in, difficult to memorize them I guess. Dictabeard 00:29, 10 December 2010 (UTC)

Our procedures differ from WP's. Here are the main tags:

1. RfV - Prove this is a real word.
2. RfV-sense - Prove this word really has this definition.
3. RfD - This does meet our standards for inclusion, even if it is used.
4. RfD-sense - This definition does not meet our standards for inclusion.
5. RfC - Clean up.

- HTH DCDuring TALK 00:43, 10 December 2010 (UTC)

haltuunsa

You added an attention tag without explanation to this and a few other of my recent entries. What's the problem? --Hekaheka 21:55, 17 December 2010 (UTC)

Probably because there is no category. Isn't it a postposition form? Mglovesfun (talk) 21:56, 17 December 2010 (UTC)

Exactly. Sometimes I can take a stab at using {{infl}} to address the issue. I simply don't know enough in these cases to even make a suggestion. DCDuring TALK 22:29, 17 December 2010 (UTC)

Yes, I suppose this could be categorized as postposition form, because that's what it is. The problem is that the fi-form of -template has not really been designed for postpositions. I made a new attempt for formulation. Please check and comment. --Hekaheka 06:14, 21 December 2010 (UTC)

I see what you mean about {{fi-form of}}. One can use {{infl}}, in the inflection line, and/or {{form of}}, in the sense line when the better, more specific templates don't cover it. Someone who understands both the grammar and templates could check for all the instances of the use of these general-purpose templates and replace them with more specific ones or extend the existing ones as seems warranted. Is this a matter for Wiktionary Talk:About_Finnish? DCDuring TALK 12:04, 21 December 2010 (UTC)

I changed the header back to "Postposition" because somebody had tagged it non-standard. I think the current format is ok, at least I'm happy with it. I think it is a good solution to handle possessive suffixes and similar such as the interrogative suffix (deprecated template usage) -ko, which can be added to any noun (and in the case of -ko, to all words except conjunctions) in the etymology section. Somebody who knows how should still create a category for "Finnish postposition forms". --Hekaheka 08:21, 23 December 2010 (UTC)

I think that Kassadbot is doing what Autoformat used to do. So PrinceKassad might be able to take Postpositions off the bad header list, at least for Finnish. Jyril was someone who seemed to know Finnish and technical matters. Is he still active? DCDuring TALK 11:35, 23 December 2010 (UTC)

Sorry, I meant "Postposition form" was tagged nonstandard. "Postposition" is accepted as header. --Hekaheka 23:53, 23 December 2010 (UTC)

university

Thanks for your help! :D I noticed you changed the etymology as well. Good work, you're amazing! 81.68.255.36 14:00, 31 December 2010 (UTC)