User talk:Nbarth/Archive 2009

This is an archive of past discussions. Please do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

← Archive 2008 |

Archive 2009

| Archive 2010 →

Template:lang

See: User_talk:Robert_Ullmann#lang.2FHans

This is a WP atrocity we want to get rid of; should have been shot quite a while ago.

The lang identification is to be done within the script templates and CSS.

{{Hans}} already identifies the language as "zh-cn" etc ... problems should be resolved within; Michael is sorting this. Robert Ullmann 00:22, 15 January 2009 (UTC)

I've fixed {{Hant}}, {{Hans}}; tell me if you are getting a good display Robert Ullmann 13:41, 15 January 2009 (UTC)

絵文字

See: User_talk:Bendono#Etymology_of_.E7.B5.B5.E6.96.87.E5.AD.97

etyl = "Etymology language"

In case you're wondering: etym/ety = "Etymology", etyl = "Etymology language". Not that it matters. We go a bit overboard with the abbreviations. :-)   —Ruakh_TALK 02:12, 18 February 2009 (UTC)

Pendula

I still believe that the plural pendula is falsely pedantic, because there never was a Latin noun pendulum. The English word was derived from the Latin adjective, so forms a normal English plural. Pedantry implies correctness, which cannot be the case here, hence my claim of falsehood. I've put this view on the talk page of pendula, and I'm happy to discuss this, and to defer to experts in the dead language. Dbfirs 21:53, 27 May 2009 (UTC)

Hi Dbfirs,

Thanks for the note! As discussed at Talk:pendula, I’ve marked (deprecated template usage) pendula as hypercorrect, which is the standard term for these constructions, and elaborated the etymology – hope you like it!

—Nils von Barth (nbarth) (talk) 17:06, 28 May 2009 (UTC)

Template:term

See User talk:EncycloPetey#Template:l

This template should not be used in bulleted lists. It should only be used for in-line mentions of words. The template you want for bulleted lists is {{l}}, although it may not support transcriptions. --EncycloPetey 01:25, 31 August 2008 (UTC)

Some people have been pushing to get those featured added to {{l}}, so you're not alone in wanting them. I recommend asking for those features in the Grease Pit. No, {{term}} should not be used in any list. It's not just about italicization, but about standardizing the way we do things. At some point, someone may decide that format in bulleted lists should look different, and the template that formats in bulletted lists gets changed as a result. Likewise, the {{term}} template is designed to allow words mentioned within text to be displayed diferently, and has the potential to let a user customize his display. If the same template is used in other situations, then that breaks the customization. If we aren't consistent about where we use the templates, this creates problems. --EncycloPetey 01:40, 31 August 2008 (UTC)

The other consideration is that (in the past) we've always wanted to avoid glosses in lists. So, if you want to gloss words in lists, then that's a whole separate discussion about changing the way we format lists. --EncycloPetey 01:51, 31 August 2008 (UTC)

It's been a long time. I'm not sure I could find the old discussions at this point. I do reccommend raising the issue though, because I understand about character lists and how the glosses could be useful. It's a valid point that needs to be considered, even if it goes against past format standards. --EncycloPetey 02:06, 31 August 2008 (UTC)

Stroke order

See: User talk:Robert Ullmann#Japanese_stroke_order.3F

modus

The "Derived terms" you added are all English legal expressions that are Latinate, and are not entries in Latin. They should thus not be listed as "Derived terms" since they are not considered Latin for Wiktionary purposes. The Derived terms section is only for listing entries in the same language. --EncycloPetey 00:09, 10 July 2009 (UTC)

Ok, got it – where should they be?

(Pedantically, they are legal and logical terms.)

• “Descendants” (of Latin word)?
• “Derived terms” (of English word)?
• elsewhere?

—Nils von Barth (nbarth) (talk) 10:36, 10 July 2009 (UTC)

Digging deeper (multi-language Wikipedia and Google searching), these terms are used in multiple European languages (e.g., Kant uses “modus ponens”, German police and Italian hackers use “modus operandi”, etc.), so it seems correct to class these as Late Latin/New Latin terms, no?

—Nils von Barth (nbarth) (talk) 11:01, 10 July 2009 (UTC)

Not being specific to English doesn't necessary mean they're New Latin. Taxonomic names, for example, are Translingual. The question is whether the term is used primarily within Latin sentences, or whether it's used in many different languages. Kant's use of the term in a German context would not make it Latin, but it could mean that it's Translingual (if it's used in many many langages). We'd need to decide the language before we decide where it gets listed. --EncycloPetey 19:17, 15 July 2009 (UTC)

Ok, so if I understand correctly, the distinction is:

• New Latin = Latin as used post 1500 (e.g., Newton’s Principia) – a scholarly language :: terms placed in “Derived terms” section of root word
• “Translingual Latin(ate) phrases” = phrases formed from Latin words and used in multiple languages, though not used in Classical or New Latin proper :: terms placed in ?? section of root word

…while it’s also worth contrasting with:

• Classical compounds = native words formed from Latin (or Greek) roots :: terms placed in “Descendants” section of root word

Assuming that “modus ponens” etc. are Translingual Latin phrases, where should they be put at modus?

Also, should I (or you) update Wiktionary:About Latin to clarify the above?

—Nils von Barth (nbarth) (talk) 22:22, 15 July 2009 (UTC)

Pretty much "yes" to everything except the last two (Assuming... and Also...). There isn't a community consensus or a precedent for placing Translingual compounds as "Descendants". I know some key editors here would certainly balk at the idea, but I can see the rationale for putting them there. I think, for now at least, that might be the best option, listing them as Translingual descendants under the Latin base word. For modus ponens I'd use modus as the souce word, since it functions as the noun. --EncycloPetey 22:34, 16 July 2009 (UTC)

admin?

See: Wiktionary:Votes/sy-2009-10/User:Nbarth for admin

Hello Nils. I would like to nominate you as administrator. Are you interested? Rising Sun 13:38, 11 October 2009 (UTC)

That’s very kind of you Rising Sun – I’d be happy to accept your nomination!

Were there particular areas or tools that you were hoping I could help with, or is this more “generally useful, keep doing what you’re doing, only better”?

—Nils von Barth (nbarth) (talk) 17:56, 11 October 2009 (UTC)

You're welcome. It is more for the "generally useful...", i.e. a user is generally useful, so they become admin. Please accept formally at Wiktionary:Votes/sy-2009-10/User:Nbarth for admin -Rising Sun 19:48, 11 October 2009 (UTC)

Done!

—Nils von Barth (nbarth) (talk) 00:56, 12 October 2009 (UTC)

Passed, implemented, and vote now archived. (Now to carefully use my new powers.) Thanks for the nod, Rising Sun!

—Nils von Barth (nbarth) (talk) 07:01, 15 November 2009 (UTC)

OK - this is what sysops (are supposed to) do. While doing all your normal editing tasks, have a second session open to "Recent Changes" and refresh it from time to time. Whenever you see an act of vandalism or stupidity, block the user (see later) and revert the vandalism using "rollback" or delete an inappropriate new entry. If you have time, mark good edits as "patrolled". How long you block someone is up to you - I normally use a short time for a first offence, and longer times for repeat offences (the block page shows you previous blocks). Cheers. SemperBlotto 08:39, 15 November 2009 (UTC)

And normally that session would be with patrolled edits hidden (so that you don't double check already-patrolled edits), and also you might wanna set your browser to auto-refresh that tab so that you don't waste you time on needless clicking (e.g. for Firefox browser that would be ReloadEvery plugin or sth similar) --Ivan Štambuk 08:44, 15 November 2009 (UTC)

Thanks for the tips both – will do!

—Nils von Barth (nbarth) (talk) 09:42, 15 November 2009 (UTC)

Transwiki

In addition to CM you might want to consult with User:Goldenrowley who has been more recently active than CM. If you can get CM back into things, that would be great from a good number of people's point of view. He was treated a bit roughly for what amounted to stubbornness. He had been a major contributor. DCDuring TALK 20:48, 19 November 2009 (UTC)

Will do – I noticed Goldenrowley’s work, and will reach out to CM (and be gentle).

—Nils von Barth (nbarth) (talk) 22:09, 19 November 2009 (UTC)

ARML.

I saw your name in the ARML contributions e-mail. I'm impressed. :-)   —Ruakh_TALK 23:24, 20 November 2009 (UTC)

(^.-)

At the time I was working for a company which generously matched (1 for 1) contributions, hence the “with match” qualifier. But yes, I enjoyed ARML (mid-90s) and wanted to give back; trust you had a good time too.

—Nils von Barth (nbarth) (talk) 23:58, 24 November 2009 (UTC)

Etymology

Please be careful with etymology entries. Lately I have fixed many entries with the same problems. For example, in ikura, you implied that the Japanese word is a proto Slavic and PIE borrowing. The Japanese word is borrowed from Russian, not from proto Slavic or PIE which Japanese had no contact with. The Russian word may derive from proto Slavic and ultimately from PIE, but that does not mean that the Japanese word is from proto Slavic. Please just leave the lang parameter off in such a case. Regards, Bendono 22:45, 9 December 2009 (UTC)

Hi Bendono,

Thanks for flagging this!

To not categorize the word when using {proto}, one actually must specify |lang= as otherwise it is implicitly categorized into (English) Proto-* derivations; I’ve fixed this at イクラ.

Regarding whether to include in categories prior to a borrowing, there appear to be differing opinions on this, similar to whether steps in borrowings should be included. I’ve flagged the issue at WT:Etyl: Whether to pursue borrowings? and in discussion at Wikt talk:Etyl: Borrowing step.

Current practice seems inconsistent, and leans (perhaps lazily) to including previous categories, as in (deprecated template usage) huitlacoche (from Nahuatl via Spanish) and (deprecated template usage) varenyky (analogous to ikura).

Perhaps we should raise discussion on Beer Parlour?

Thinking about it, my inclination is to include categories for languages prior to a borrowing, for two reasons:

• we do the same for derivations – English words that are from Old Norse via Old English are classified both as OE derivations and ON derivations, rather than the English word being classified as an OE derivation, and the OE wording being an ON derivation (or are you suggesting not doing this?)
• it seems useful information – one cares about all English words that ultimately trace their roots to Nahuatl, as evidenced by words of Nahuatl origin (and natural interest), so one would want some category for this. In principle one might distinguish: “Words ultimately from Nahuatl” and “Words borrowed directly from Nahuatl”, but we do not finely distinguish between etymology categories (borrowings vs. derivations).

Does this make sense?

—Nils von Barth (nbarth) (talk) 23:28, 9 December 2009 (UTC)

“Surface” etymologies

In re this revision of yours
Whilst what you added isn’t the true derivation of sesquipedalianism, I can see some usefulness in including these “surface” etymologies, although only if it is made clear that they are mere re-interpretations, rather than actual deriviations. For that purpose, what do you think of a template such as {{equivalent}}, which would allow this kind of morphemic breakdown and, hopefully, also autocategorise by affix? † ﴾^(u):Raifʻhār ^(t):Doremítzwr﴿ 05:53, 13 December 2009 (UTC)

Hi Raifʻhār Doremítzwr,

This is an excellent point, and one that had been troubling me lately (I’d just been fixing up (deprecated template usage) sesqui- words.)

I’ve written a note at WT:ELE: Surface etymologies that says “history =/= surface form”. More consistent practice would certainly be good, and if templates can help, so much the better!

For sesquipedalianism, I’ve listed it as “surface form analyzed as: …” which yields ok categorization of affixes, but you’re right, it could use work – any contributions welcome!

—Nils von Barth (nbarth) (talk) 08:49, 13 December 2009 (UTC)

I assume you meant WT:ETY#Surface etymologies (rather than ELE). Well, the ideal would be to allow categories to yield lists that could be transcluded to affixes’ Derived terms sections, and that those lists would be split between those terms that were actually formed using that affix (as a historical fact) and those terms that merely feature that affix (upon the analysis of their “surface forms”); I see no reason that this could not also be applied to whole words, with those lists transcluded to their Derived terms and Related terms sections, respectively. So, for example, I envision that the etymology for antidisestablishmentarianism would be written:
{{confix|anti-|disestablishmentarian|-ism}}; {{equivalent|anti-|dis-|establish|-ment|-arian|-ism}}
Which would yield:
anti- + disestablishmentarian + -ism; equivalent to anti- + dis- + establish + -ment + -arian + -ism
Which would add the entry to the derived-terms lists of (deprecated template usage) anti-, (deprecated template usage) disestablishmentarian, and (deprecated template usage) -ism, as well as adding the entry to the related-terms lists of (deprecated template usage) dis-, (deprecated template usage) establish, (deprecated template usage) -ment, and (deprecated template usage) -arian.
Issues to be dealt with that spring from this are:

1. How will such autocategorisation be possible when the lists yielded need to be split by language (as with (deprecated template usage) anti-), by etymology (as with -en), and/or by sense (as with -ish)?
2. How can we integrate transcriptions and glosses into these templates? (This probably won’t be difficult, considering {{confix}}’s numbered gloss#= and t#= parameters; for example, transcriptions should be easily accommodated with numbered tr#= parameters or somesuch.)
3. What do we do about unattested-but-analytically-hypothesised roots to which various inferred affixes attach?

As you might have gathered, I’ve been thinking about this for a while! :-) It’s not an easy thing to sort, but the potential benefits are considerable. How do you think we should proceed hence? † ﴾^(u):Raifʻhār ^(t):Doremítzwr﴿ 08:34, 14 December 2009 (UTC)

Oops – yes, WT:ETY.

It sounds like you’re doing several things:

1. Distinguishing derivations (via affixes) from analysis of surface forms – this involves:
   1. Creating separate categories.
   2. Distinguishing them in templates.
   3. …and then including them in the relevant Derived/Related terms sections (easy enough, in principle).
   4. You also mention a different distinction, which is “words that include an affix at some point in them”, as in dis-, above; this is yet a third category: “words that include dis- and then more prefixes besides”.
2. To your issue of “split by language (as with (deprecated template usage) anti-), by etymology (as with -en), and/or by sense (as with -ish)”:
   1. Language can be handled by a “lang=…” parameter, no?
   2. Etymology should presumably be handled as in the “ss=” parameter of {{term}}.
   3. Sense can be handled as via {{jump}}, which creates lots of anchors.

   You should talk to DCDuring, who’s facing similar categorization questions with -s (as used in adverbs such as backwards); see current: WT:BP#Categories for morphemes: un/productive

   That said, creating automatic categories from this sounds possibly very, very painful – “English words prefixed with a- in the sense of ‘intensifying’” – nailing jello to a tree. If we want categories that are not for human consumption, or are ok with ugly ones, then “English words prefixed with a-/Etymology 3” works (but makes the “link to sections” even more brittle than it already is).

3. Transcriptions and glosses are a doddle, as you indicate – just use gloss/t/tr.
4. There is a subtle issue of order in these long chains of affixes – as the detailed etymology of (deprecated template usage) antidisestablishmentarianism demonstrates, the etymology is not anti- + d… + -ism, but rather ( (anti + d… ) + arian ) + -ism, which is why (deprecated template usage) disestablishmentarian is a red link: many possible groupings are not valid words.

   In principle, the correct answer is to:

   1. give an order for affixes – notably, using {confix} with 3 args states that both were attached onto the root, rather than (a + b) + c or a + (b + c)
   2. mark unattested forms (say, if it’s a + (b + c) and (b + c) is not attested) with a * (and note as a parameter, say unattested2=1, or more briefly ua2=1 for “2nd possible for unattested – mark with * and don’t link”).
5. So I’d suggest:
   1. First think about what semantic distinctions and categories we want to make.
   2. Second think about how this can be done technically.
   3. Third, implement it.
6. …and at each stage, going to BP to chat about it.
7. I’d also suggest trying to do small steps: this is a potentially big and messy project; starting with distinguishing “words suffixed with -logy” (but maybe coming from Greek, like astrology) from “words derived by adding -logy” is already some work.

You’ve clearly thought about this in detail; those are my immediate thoughts – do they help?

—Nils von Barth (nbarth) (talk) 09:16, 14 December 2009 (UTC)

und vs unk.

I see that you deprecated {unk.} per Wiktionary_talk:Etymology#Deprecate_.7B.7Bunk..7D.7D_?. I don't think that's current as during Wiktionary:Beer_parlour_archive/2009/July#template:und Ruakh expressed in his last comment that the two have slightly different meanings. This last conversation prompted me to make this clarification on WT:ETY. Do you still think we should deprecate {{unk.}}, or leave them to mean different things? --Bequw → ¢ • τ 00:44, 19 December 2009 (UTC)

Thanks for pointing out that discussion!

Formally, I suggested deprecating {unk.} (hence ‘?’) – as you note, there is some debate, with I’ve now linked from the template talk page and WT:ETY. Perhaps we should bring this to BP?

—Nils von Barth (nbarth) (talk) 02:39, 19 December 2009 (UTC)

Formatting

Please add images immediately under the language header, and not within subsections of an entry. Also, please do not define terms listed as synonyms, coordinate terms, or the link. Definitions appear only within a part of speech section of an entry, or sometimes in an etymology section. Thank you for your additions! --EncycloPetey 19:27, 25 December 2009 (UTC)

Oops -- thanks for the pointers!

I'll place images correctly in future; I've listed correct placement at Wiktionary:Pictures (I've suggested under etymology if distinct etymologies exist).

Sorry for the def'ns in Coord Terms -- I'd thought to briefly gloss them, but you're correct, glosses belong in Wikisaurus or a separate Glossary, not as inline clutter.

—Nils von Barth (nbarth) (talk) 19:38, 25 December 2009 (UTC)