User talk:Connel MacKenzie/archive-2007-11

back to http://en.wiktionary.orghttps://dictious.com/en/Special:Prefixindex/User_talk:Connel_MacKenzie/archive

Um, I recently closed an AfD on List of Internet slang phrases on en.wiki (where I'm an admin), and since it was already here I figured I'd move it out of the Transwiki: namespace. Dmcdevit explained to me that articles like this belong Appendix: namespace here, I understand this... but couldn't you have simply let me know about this and explain what I should properly do next time, rather than instantly blocking me indefinitely without even a warning? --Krimpet 08:06, 2 November 2007 (UTC)Reply

I did not realize the "next Wikisaurus" was quite so important to people. While I can appreciate the Wikipedia desire to get rid of it, dumping a vandal magnet on Wiktionary (pointedly incorrectly, pointedly out of process,) really isn't appreciated. --Connel MacKenzie 15:26, 5 November 2007 (UTC)Reply

I was trying to help out; if I did it incorrectly or out of process it was an honest mistake, I'd be glad to learn more about Wiktionary's processes and policies. To instantly accuse me of maliciousness and "dumping a vandal magnet" on Wiktionary and block me over a single edit seems very vindictive. If this kind of action towards new users is par for the course on Wiktionary, then perhaps this isn't the project for me. --Krimpet 07:50, 7 November 2007 (UTC)Reply

In Wiktionary:Votes/2007-10/Lemma entries, you said that we should eventually aim for complete entries everywhere, synchronized in an automated fashion. Since we don't yet have the ability to synchronize entries automatically, it seems important to minimize manual synchronization efforts by means of standards. Do you have a different idea for how to deal with the synchronization problem now, in lieu of automation? Rod (A. Smith) 17:41, 2 November 2007 (UTC)Reply

My observations about current entries is that they are very often manually synchronized (for things that we do synch, that is.) I remain steadfastly astounded at how many people devote a lot of time to manually synchronizing things. Very often, their manual efforts have much better results than automated relection would be. (To be fair: some small portion of mistakes also creep in. But being a wiki, they invariably are caught and fixed.)

The notion you mention: "seems important to minimize" is what really gets my gourd. There is absolutely no way you can predict what thousands of contributors here will do. Will two or three dozen dedicate themselves to manual synchronization of things, one language at a time, then way more manual synchronization (than even I imagine) could be accomplished.

While on one hand, that seems a waste of effort, it does allow newcomers to familiarize themselves with our general formats. That is one of the reasons TheCheatBot isn't run the moment a new XML dump comes out. But there, I'm talking about automation (again?)

The comment "seems important to minimize" means only one thing: that experimental synchronization bots are prohibited from existing. That is monstrous. We've spent four+ years discouraging automation. As the various tools stabilize and the pool of talent to draw on grows, it becomes more important to encourage automation experiments. Especially, when they are beneficial to the final product the readers see. That's what I'm talking about, when I say the proposal is headed in the wrong direction.

--Connel MacKenzie 15:43, 3 November 2007 (UTC)Reply

There are no definition synchronization bots today, but if somebody later solves the technical hurdles and creates a bot that can synchronize the definitions across inflected entries, that would be a welcome improvement even if Wiktionary:Votes/2007-11/Lemma entries 2 resolves our current limitations today. Rod (A. Smith) 21:17, 5 November 2007 (UTC)Reply

<pulling hair in frustration> Then why start a vote for it? You aren't recommending a change in policy. You are making (TLDR) ELE longer. You would however, be solidifying the wrong approach and if it took hold, would be "general-practice"-preventing the creation of that entire class of bots. It is not at all clear what "limitations" you are trying to "resolve." --Connel MacKenzie 21:37, 5 November 2007 (UTC)Reply

Ug. That is not the wrong approach. It is the right approach, indeed, the only approach taken by any translating dictionary. Since you don't understand the problem, you should not be opposing its solution. We lack an automated system of synchronizing translations, so we should show readers know where to find full translations. Rod (A. Smith) 23:32, 5 November 2007 (UTC)Reply

And it's not just that translations to and from non-lemma entries is hard or that nobody else does it. It is misleading and wrong. Rod (A. Smith) 23:36, 5 November 2007 (UTC)Reply

How dare you be away from keyboard. ;-) I'm on IRC now. Looking forward to chatting. Rod (A. Smith) 00:47, 6 November 2007 (UTC)Reply

Grr. You weren't around.  :-( So, continuing this here.... Do you understand why only foreign lemma forms should be given in English entry translation tables or do you think foreign conjugation/declension information is appropriate in English translation tables? Rod (A. Smith) 02:42, 6 November 2007 (UTC)Reply

/back for a few then back in an hour or so... --Connel MacKenzie 03:30, 6 November 2007 (UTC)Reply

Yes, of course I understand why it is easier to maintain lemmas-only in translation tables. I still think that years from now, people will find much more value in having the conjugation/declension information readily available. But that isn't what your proposal says. You are specifically calling out English examples. Did I misread that part? (I.e. the whole thing?) --Connel MacKenzie 03:33, 6 November 2007 (UTC)Reply

From the wording of your reply above, you don't seem to realize the real reasons for translating only between lemma forms. Rather, it seems that you think the restriction is just for convenience. Is that so? If so, I will need to explain further why translation of English words into a diverse set of other languages can be accurately done in terms of semantics and register but not in terms of inflection. (And further, I wil need to explain why attempting to gloss over those differences would mislead and confuse readers.) If you do understand that lemma-to-lemma is the only way to translate accurately, I will be able to continue to other parts of the proposal. Rod (A. Smith) 07:19, 6 November 2007 (UTC)Reply

You are right: I do not view that as a valid restriction. The enormity of completeness does not fit on paper (but of course, Wiktionary is not paper.) So yes, I do completely disagree; the various translation texts in existence today do limit themselves only for convenience. Logically, you are representing directed graphs of one-to-many as one-to-one (falsely) when you do so. --Connel MacKenzie 14:21, 6 November 2007 (UTC)Reply

It's not just because inflections don't fit well on paper that we limit them to the foreign language entry. Each entry here specializes in showing information about its headword. I expect you're familiar with the "one authoritative source" principle of information architecture. (If you're not familiar with that concept, let me know and I'll try to explain it.) Within this project, the authoritative source for all of the detailed information about any particular word is the entry whose title is that word. By adhering to that principle, we give readers a consistent place to read about (and contributors a consistent place to write about) any particular word.

So, readers and contributors are best served by our housing complete, authoritative inflection details for each foreign word in the foreign word's entry. I hope you agree with me on that point, but please speak up if you don't so I can rehash it in agonizing detail.

Next, should we create systems that synchronize certain details from the one authoritative source? Presumably, the imagined benefit for doing so with inflections is to provide easy access to foreign inflection details. I submit that to copy the contents from the one authoritative source (the foreign entry) to English entry translation tables, either manually or with bots, is to make the English entry overly complex without any benefit to readers or contributors. Readers of such an overly complex English entry would have to click a translation's "expand/collapse" link anyway to see foreign inflections, just as they can click a translation link today to see inflections in the foreign entry. Besides, a distant future with bots synchronizing data from foreign entries to semantically similar English entries is a short sighted means of achieving in-entry-translation-expansions. Better would be to run a UI on this project's database that lets translation links expand inline within the English translation table without actually copying the foreign entry data. In such a vision, any copying of details from foreign entries into English entries is at best a nuissance. So, if there may some day be a way to copy foreign inflection details into English entries, we neither have such a system today, nor should we necessarily welcome it.

So, it's not just for convenience and lack of bots that we restrict inflections and other details about a foreign word to the foreign entry. It's the best short-term and long-term design for this project following well-known principles of information architecture. We should update WT:ELE to educate contributors that the only appropriate place for foreign inflection details is in the foreign entry.

I am quite certain that foreign inflections have no place in English translation tables. If you accept that, we can move on to discuss other parts of the proposal about which I am less steadfast. If you don't accept that, let me know how you disagree and I'll try again to clarify. Rod (A. Smith) 17:51, 6 November 2007 (UTC)Reply

Wow. I am almost speechless. This is so completely at odds with the fundamentals of hyper-wiki-text that I am agape.

readers and contributors are best served by our housing complete, authoritative inflection details for each foreign word in the foreign word's entry. YES! I agree. EVERY WORD. Not just the ones that are designated as the "primary" or "lemma" words so that the information comprising the entirety of the relationship of language X to language Y can be (poorly) represented on 200 printed leaves of dead tree.

The conjugation templates are exactly the right structure to represent the information on every page for every form. (And the Spanish ones get it perfectly: there are specific templates for specific irregular verbs, that can be used for each form entry!)

Nobody suggested otherwise.

Similar things apply in different ways to other attributes; but the principles are the same. We are not that translating dictionary on your shelf, that is just barely useful. (Look up Swahili "tulisema", you'll have to know to take off the person (tu-), the tense (-li-) and look up "sema" (stem, the infinitive is "kusema"). Did you know that? No? Well then good luck using the dead-tree version of the book that is supposed to help you as an English speaker make sense of a bit of Swahili. But we can do a lot better, and there is utterly no excuse not to. If you want all those restrictions, buy the books: OUP makes lots of them, all very good. We are about doing what they cannot do, that we have precisely the tools for. Robert Ullmann 18:46, 6 November 2007 (UTC)Reply

Robert, I'm glad you agree that conjugation and declension tables only belong in foreign entries. Connel has not yet said he agrees with that, so stop side-tracking this conversation. You are arguing agains a position that nobody is taking here. I'm trying to establish common ground with regard to something simple: whether to show foreign inflections in English translation tables. Rod (A. Smith) 18:59, 6 November 2007 (UTC)Reply

(why would conjugation and declension tables occur in English entries? English only has a few conjugations, and only declines nouns to plural, etc.) Okay, to be specific: The forms (past, etc.) should have translations for languages where someone has found it useful to add them, manually or by bot. (Or check them likewise). No-one should get the impression that they don't belong there, or shouldn't be there, or that it is okay to remove them. The link(s) should be to a single form or stem as appropriate. (I don't get the nested tables business, I don't recall anyone suggesting that sort of thing even should be done. I think someone assumed that Connel or whomever was suggesting that complete inflections be put in the trans tables when they weren't suggesting that.) The thing is: the only "clarification" I think ELE needs is to stress that "non-lemmas" can and should be complete entries, and that removing content is unaccceptable. (Some people think they ought to remove content because "ELE says so". Even though it doesn't.) The only restriction is that we don't have Translations in FL entries. Robert Ullmann 19:24, 6 November 2007 (UTC)Reply

If English past tense verbs have translation tables, readers will get the mistaken impression that their search is over and editors will be unable to correct mistakes easily. That's a problem. Rod (A. Smith) 19:50, 6 November 2007 (UTC)Reply

Indeed, I do disagree with that statement. Editors will be able to correct mistakes just as easily then, as now, with approximately the same amount of errors introduced this way or that. --Connel MacKenzie 19:54, 6 November 2007 (UTC)Reply

What?! Do you really think it's just as easy to correct several inflected entries as it is to correct one entry? Rod (A. Smith) 19:57, 6 November 2007 (UTC)Reply

No, but I think such errors persist as errors, approximately as long. The benefit to the readers, though, outweigh that concern. --Connel MacKenzie 20:29, 6 November 2007 (UTC)Reply

Rod, your concern is valid: I do not agree with that premise. Obviously, I envision the expansion differently than you. HOWEVER the concept of including FL inflections within translation tables of English inflections is off topic. I am amazed at how far from the original complaint we are. Yes, I understand that a HEADWORD should be a complete entry. No, I do not agree that it should never duplicate material that can be found in other HEADWORD entries. Being a multilingual English dictionary, it is much more important for that duplication to exist. Transcluding that via templates is an option, but not a good one at all. Transcluding it through Javascript magic is also conceivable, but also not optimal (and error prone at irregularity points.) Transcluding/reduplicating at the MW software level is worst of all, as it wouldn't be flexible enough. Instead, I envision the duplication/synchronization happening at the Bot level. All the thrusts of your proposals so far this week have been at direct odds with that concept, hoping to prevent usable data from finding its place where it would be looked up. --Connel MacKenzie 19:33, 6 November 2007 (UTC)Reply

Off topic?! One of the proposed limitations is to translate only between lemma forms. The alternative is to put inflected forms into translation tables. I'm trying to determine what common ground we have, so I'm starting with the most obvious. If I cannot succeed in explaining to this community why we should avoid translating inflections, I may need to leave this doomed project. Rod (A. Smith) 20:27, 6 November 2007 (UTC)Reply

Isn't there enough drama without comments like that. Really, I'm not talking about anything imminent, but I am talking about future improvements. And most importantly, not preventing them! --Connel MacKenzie 20:32, 6 November 2007 (UTC)Reply

I'm not trying to prevent an improvement. I'm trying to establish common ground with respect to whether translations should be only from one lemma form to another or between any arbitrary inflected forms. You and Robert keep dragging in other arguments from other conversations here, effectively preventing us from establishing any common ground. Rod (A. Smith) 20:38, 6 November 2007 (UTC)Reply

This is getting too abstract (and far too indented.) Could you please try restating your opposition to translations being expanded, with an example? --Connel MacKenzie 20:38, 6 November 2007 (UTC)Reply

Sure. See below. Rod (A. Smith) 22:33, 6 November 2007 (UTC)Reply

Note about "bot level": Wehn a bot incorrectly duplicates something, it gets corrected by humans. This is much more true now, than it was a year or two ago, as there are more eyes watching. Systemic problems can be fixed before going too far off in left field. Those factors weigh unbelievably in favor of bot-level activities, over any of the alternates you implied. --Connel MacKenzie 19:36, 6 November 2007 (UTC)Reply

That last paragraph is completely false. I have tried for months to remove the inaccurate "Latin American second person" verb forms that TheDaveBot dumped here. They're wrong, but they seem to be stuck here without the opportunity for discussion. Rod (A. Smith) 19:50, 6 November 2007 (UTC)Reply

No, that isn't false: that is exactly the point: if that was done by software, it could not be fixed! Have you mentioned that item in WT:GP? Automatically entered entries can also use automation for their removal. --Connel MacKenzie 19:54, 6 November 2007 (UTC)Reply

As discussed above, Connel does not yet agree with limiting translations to lemma forms only. He asked for illustration of the problem by example. That example follows:

In an effort to help our project, a contributor adds the translation habla to (deprecated template usage) talks.

At first blush, that seems helpful, or at least it doesn't seem to hurt anything. Another editor sees that translation and, trying to be helpful, adds the following list of possible translations to the entry (deprecated template usage) talked: (yo) hablé, (tú) hablaste, (él/ella/usted) habló, (nosotros) hablamos, (vosotros) hablasteis, (ellos/ellas/ustedes) hablaron. Unfortunately, a reader who now visits (deprecated template usage) talked is mislead into thinking that (deprecated template usage) hablar in the past tense depends only on the person/number and so he or she doesn't click through to the Spanish entry for additional translation options. A contributor tries to correct that problem by adding other forms of (deprecated template usage) hablar to the English entry translation table for (deprecated template usage) talked: hablaba, hablabas, hablaba, hablábamos, hablabais, hablaban as well as hablaría, hablarías, hablaría, hablaríamos, hablaríais, hablarían and the subjunctive imperfect forms hablara, hablaras, hablara, habláramos, hablarais, hablaran and hablase, hablases, hablase, hablásemos, hablaseis, hablasen to the previously incomplete translation for (deprecated template usage) talked. Other contributors will start filling up the translation tables for (deprecated template usage) talked with translations into other languages. Some will use bots, others manual entry, but in the end, the translation table appears to become more complete but instead has become pretty much useless. Unfortunately, nobody may delete any of those so-called translations because we don't have policy restricting them.

What's even worse, though, is that a would-be contributor who wants to add (deprecated template usage) conversar as a translation is likely give up because it would be a ridiculous amount of work to add all the dozens of forms of (deprecated template usage) conversar to (deprecated template usage) talked, not to mention the need to synchronize translations between (deprecated template usage) talk, (deprecated template usage) talks, (deprecated template usage) talking, and (deprecated template usage) talked. So not only does that unmanaged mound of translations give readers misleading information about translation, but it effectively buries otherwise meaningful single-edit contributions under a mound of forms of (deprecated template usage) hablar scattered across our entries for (deprecated template usage) talk, (deprecated template usage) talks, (deprecated template usage) talking and (deprecated template usage) talked.

In an ideal situation, the contributor should see just a single pointer from the secondary entry (deprecated template usage) talked to the primary entry (deprecated template usage) talk, which should list just Spanish primary forms (e.g. hablar). I don't care whether that pointer is in the headword/inflection line, the definition line, or the translation table, but some contributors do seem to care about that. The contributor can then add conversar to (deprecated template usage) talk and, in one easy edit, has informed all future readers about the alternate translation.

Of course we can reopen discussions if somebody devises the incredibly complex synchronization system required to keep translations synchronized between all inflections across all languages. Hell, we can even make a note somewhere that welcomes any synchronization automation (including translations between two foreign languages). Until then, though, allowing ad hoc inflected translations creates confusion and a barrier to contribution. So, do we have common ground in the recommendation to give only translations of primary English entries into only primary foreign entries, and to encourage pointers from secondary entries to primary entries? (And Robert, I'm trying to establish a common ground, so don't go on about rabbit-holes or how I don't understand templates or how the vote proposal exceeds this scenario.) Rod (A. Smith) 22:33, 6 November 2007 (UTC)Reply

Hi, Connel. In response to feedback so far, I have taken a new approach to this initiative, providing more context and examples. Before I post it somewhere more public, though, I would appreciate your review. Please see User:Rodasmith/lexeme data. Is it still too abstract? Too wordy? Too vague? Rod (A. Smith) 22:32, 9 November 2007 (UTC)Reply

Well, first of all, thank you for un-indenting.  :-)   I very much would have preferred an example where a direct translation exists. (Hablar = speak, conversar = converse, discutir = discuss...none of which mean "talk" (verb) precisely.) From my decades in New York and San Diego (suburb of Tijuana) my Spanish "babel" is probably less than zero - the "Spanish" I know is generally wrong.  :-)

In your example above, person #1 is fine...well, no. Actually, because "hablar" is incorrect, that initial translation entry is wrong. But that is water under the bridge, as they say. And at least 80% of translations we have here are equally screwed up, when not blatantly misleading. I'll leave that aside for the moment.

Person #2's contribution seems helpful. Person #3's contribution seems helpful. I can understand the desire to mask the complete conjugation, but it should be accessible (yet still omitting the future senses, as you did.) How we'll ever get there technically, is something to ponder long and hard. Why you wish to prohibit it, remains a colossal mystery.

Person #4 is about to make the same mistake person #1 made, but again, that's water under the bridge. I think your assertion, that they will be deterred from making an entry, is already more true, by merit of the obscure template syntax we use in translation sections. The working assumption has to be that person #4 enters "conversar" and then person #5 fills it out further. If we get "Wizard" style data entry tools (Java or Javascript based entry tools that take a question-and-answer style approach to getting data from people) then person #4 will have entered "conversar" possibly without seeing the relation to the "hablar" inflections until after it is saved. At that point, it would be fine for them to throw up their hands and let person #5 fill it out further.  :-)

While I admit that I don't know how this should eventually look, I do know that the notion that they should be prohibited, is much worse. How would an AJAX'ed conjugation table from hablar mask out the future (etc.) senses when displayed under talked#translations? Wouldn't it be more reasonable to have that data in the entry and maked somehow, there? Yes, that makes some maintenance aspects more painful. But it is more painful for bots not painful for humans.

Yes, some bots propagate errors. (E.g. TheCheatBot.) Again, it is my experience that the overkill of trying to review those entries before they are propagated is a bandersnatch. When the error is caught, the error is caught and all three related entries are fixed. Could that potentially mean a half dozen translation table entries in your examples above? Possibly. But so what? There are a lot of very eager people around here who delight in finding minor secondary errors and correcting them. The amount of potential duplication we are talking about is very finite. If anything, it adds a degree of resilience to vandalism. And more to the point, simply everything about a dictionary, is redundant. Picking out one (finite) type of redundancy to prevent, still seems silly to me.

A lot of good things can come out of this, if approached more methodically. Having each of the Spanish conjugation tables broken down into individual line templates may make more sense, ultimately allowing for easier integration into form-of entry's translations. Who can guess what will become of it? But why prevent it? Paul and Eclecticology made that argument a couple years ago, but it never really made sense.

I guess I should be glad you didn't provide speak / spoke / spoken as your example. Irregulars, though, are much more of a reason to make our entries more complete. Not less complete and more ambiguous.

--Connel MacKenzie 01:37, 10 November 2007 (UTC)Reply

• You know, that many words would have done me a lot more good on http://www.nanowrimo.org/... --Connel MacKenzie 01:38, 10 November 2007 (UTC)Reply

Thanks for taking the time to consider and review my post. I don't agree with your answer (e.g., why you think (deprecated template usage) hablar and (deprecated template usage) talk are not exact translations of each other), but at least we tried to understand each other. I will think about your suggestions of how you'd prefer it to work (AJAX, Javascript, including foreign conjugations within English translation tables, etc.). In the meantime, although I realize this conversation has expended lots of space, did you also happen to read User:Rodasmith/lexeme data? Rod (A. Smith) 02:42, 10 November 2007 (UTC)Reply

Well, my Spanish "babel" is negative, remember?  :-)   From what I can tell, hablar translates to speak, not talk. But our entry says differently, so, what do I know. It is how it is usually translated, due to respective word frequency, but the direct translation is speak not talk. I guess.

For the various proposals, yes, you've now narrowed it down only to translations. That is progress. That is also quite opposite of what you had originally. Still, I think I've itemized above, why I think that (tiny remaining) limitation is conceptually flawed and anti-wiki-think. It too, is still far too broad; it doesn't limit itself to Nouns and Verbs only. Prepositions, Conjunctions...no...individual forms for those things always need direct, complete translation tables. (Nouns and verbs may be obvious to us, but not necessarily to all our readers.)

I've avoided expanding TheCheatBot to non-English entries. In hindsight, I am very glad I've avoided that, so far. The concept has only partly been fleshed out with Spanish and Italian. But almost every one of the form of entries in Spanish and Italian are, in my opinion, vastly incomplete right now. Identifying the FL lemma is helpful, but no where near as helpful as a complete entry would/will be.

Before embarking on a mega-synchronizer thing, I'd like to focus on putting together some of the previously mentioned "wizards." That of course, depends on having a usable entry-parser. But the format is (and shall remain) a moving target. So, it will be pretty challenging. No complaints at all, if someone decides to be less lazy than me. But I shall oppose notions that restrict content (especially content duplication) for a long time to come, yet. (I think there are tremendous similarities to the "possessives" vote here, that people didn't/couldn't recognize. That remains to be voted out, at some point in the future when it is clearer to more people how counter-wiki it is.)

--Connel MacKenzie 03:09, 10 November 2007 (UTC)Reply

So, you're not on board with any sort of distinction between lexicological details that apply to a specific grammatical form of a word and lexicological details that apply to an entire lexeme? Rod (A. Smith) 03:12, 10 November 2007 (UTC)Reply

That's a pretty broad generalization. I hear your frustration. In essence, I see more value in giving the information that is hidden away here, to the reader, than catering to the paper-dictionary concepts of truncation (in the name of "ease of editing.")

Perhaps another thing to consider, is the "language hiding" extension stuff that Hippietrail has been working on at wiktionarydev.leuksman.com. Picture Wiktionary a couple years from now, with most of the "top 18" or so languages, with pretty complete coverage. Most people will start turning off other specific languages, perhaps limiting what they see to only two or three languages. On an English form-of entry, wouldn't it be useful to see only the relevant lines of the conjugation table extracted from hablar? (A couple years in the future, the translation tables will either be much more elaborate or much less template oriented - but they won't be the mish-mash they are now.)

To me, while it may be a hard thing to say that expanded translations on forms is a good thing, it is very easy to recognize that putting limitations on how those translations can be expanded, is bad.

--Connel MacKenzie 04:16, 10 November 2007 (UTC)Reply

Hmm. Any such WTF (Wiktionary of The Future) will depend heavily on a proper data infrastructure. A component of that infrastructure is the division between lexicological details that apply to specific grammatical forms of words and those that apply to entire lexemes. It's obvious that you will not agree with me on this point, though, so I guess we agree to disagree.  :-( Rod (A. Smith) 04:39, 10 November 2007 (UTC)Reply

Hi Connel. Quetsion:- Why remove life of Reilly? It is a very common misspelling or even perhaps alternative spelling of life of Riley. That is why I put it in. No RFV???? Plenty of Google books hits and newspaper hits. Almost as many the one as the other. Algrif 14:55, 5 November 2007 (UTC)Reply

For a couple reasons. First of all, the message

• 10:07, November 1, 2007 Connel MacKenzie (Talk | contribs | block) deleted "life of Reilly" ‎ (Misspelling of life of Riley - --explanation of deletion--)

does appear if you arrive at that Wiktionary page now. So yes, it is identified as a misspelling for people, if they happen to look up the incorrect idiom.

Next, the entry I deleted was pooched moderately. That is, the headings were wrong, the inflection was wrong, the misspelling was identified as a valid alternate (that is what I assume you mean by "RFV"? - no, a comparative google search shows this is clearly a misspelling, not a valid alternate.) For misspelling entries it would need to not include any wikification (lest we incur accusations of bloating our article count by crap-loading misspellings) therefore, it would not be appropriate to format the inflection line in a Wiktionary compliant manner. Next, it wasn't a form of the idiom, it was a misspelling of a particular proper noun within an idiom. Therefore it shouldn't be a simple redirect of a valid alternate form, as the misspelling isn't some reinterpretation or conjugation or regional spelling or something - it is only an error.

Lastly, it had already triggered AF to tag it for additional cleanup. If left as an entry, it would have appeared on three or four different cleanup lists, after the next XML dump.

--Connel MacKenzie 15:20, 5 November 2007 (UTC)Reply

Nice, Connel; that was well explained. † ﴾^(u):Raifʻhār ^(t):Doremítzwr﴿ 15:04, 6 November 2007 (UTC)Reply

OK. Thanks. I need to know these things to understand how the project works. Algrif 18:26, 5 November 2007 (UTC)Reply

Hi Connel, I have over the weekend expanded the wiki-merge instructions based on most what we discussed so far and research on the license rules. I havent yet implemented our idea to differentiate "content merges" as a subcategory of merges so that they'd be automatically sorted but I plan to if we've got a consensus. You are welcome to check my work to date. Help_talk:Transwiki Goldenrowley 21:26, 5 November 2007 (UTC)Reply

Actually those were my own words, which is probably why they were not so great. For example, compare the version from the Free Dictionary:

Com`men`sal´i`ty n. 1. Fellowship at table; the act or practice of eating at the same table.

My version is considerably less elegant than that, for sure, but that is where I obtained my definition. I just rewrote it in my own words to avoid problems with copyright.

I also do not understand the format here very well yet so I apologize. --Filll 22:46, 6 November 2007 (UTC)Reply

Was removal of the plural intentional in this edit? If so, why? Rod (A. Smith) 08:03, 7 November 2007 (UTC)Reply

No, that was a byproduct of restore/editing the earlier version. --Connel MacKenzie 08:16, 7 November 2007 (UTC)Reply

What is your new syntax for the mentions within the ==Dictionary notes== section? --Connel MacKenzie 08:18, 7 November 2007 (UTC)Reply

OK. Mentions are thus. Good? Rod (A. Smith) 08:24, 7 November 2007 (UTC)Reply

Thank you! (For both corrections.) --Connel MacKenzie 08:38, 7 November 2007 (UTC)Reply

Hrm, is there a version that doesn't wikilink? (Normally, the link should be there, but in that comparative listing...it seems to give the bogus spelling undue emphasis by wikilinking it.) --Connel MacKenzie 08:42, 7 November 2007 (UTC)Reply

See my last edit. :-) —Ruakh_TALK 15:08, 7 November 2007 (UTC)Reply

Very clever - nicely done. Now I hope no one tries to "correct" that. --Connel MacKenzie 15:19, 7 November 2007 (UTC)Reply

I'm sorry. I didn't know. I mean, I'm not a wiktionary user.24.14.216.60 20:40, 10 November 2007 (UTC)Reply

Isn't co-operation just an alternative spelling of cooperation (and vice versa)? If they're not synonymous, I'll kick myself, apologize, and cordially notice that alt. spellings list on the cooperation entry is in that case misleading. But if they are, shouldn't {{alternative spelling of}} be enough in order to avoid repetition and propagation of errors? --Jyril 21:49, 10 November 2007 (UTC)Reply

No and a thousand times no. --Connel MacKenzie 21:53, 10 November 2007 (UTC)Reply

I'm wondering about PoSs for phrases. You have presented this one as a "Phrase". I would have presented it as a "Verb", using "en-verb", generating red links, but thereby also entering the category. The aesthetics of your approach are superior, but don't seem to fit with what I found in the documentation. What is the state of discussion on this? DCDuring 22:36, 10 November 2007 (UTC)Reply

I think the documentation doesn't/won't cover it, because it itself is too unique a term. (FWIW, the original contributor entered it as a ===Phrase===, didn't they?)

Reclassifying it as a verb, would, um. Hmmm. Gee, I don't know that that's right...is it? Kinda-sorta. But, hmmm.

Maybe send it to WT:TR / {{rft}} it? --Connel MacKenzie 22:51, 10 November 2007 (UTC)Reply

I didn't look closely at the history. Yes, it was the original contributor. I'll Rft it. DCDuring 23:23, 10 November 2007 (UTC)Reply

I should have moved this here a round or two ago, as I'm really only talking about you, not facade/façade. Anyway:

I didn't realize that the selective bowdlerizing above could still be so greatly misunderstood.

So you know: I didn't follow the whole thread in detail, so I wasn't misled by any of the "bowdlerization", whatever you mean by that. (But don't bother to explain; it really doesn't matter). As I said, the only reason I got into the discussion was that my eye was consistently caught by what looked like a whole lot of inappropriate POV pushing by you.

The initial vandalism itself wasn't caught as vandalism. ... When I pointed out that error, I got vitriolic personal attacks (that were edited/removed after I replied.) And this one example is not an isolated error. Do I think those tactics were malicious and intentional? Absolutely.

Fine. I didn't see those vitriolic personal attacks, but perhaps they'd been removed by the time I happened along. But -- and this is the important point -- no matter how bad and malicious and intentional they were, do they justify vitriolic personal attacks in return by you?

It continually shocks/shames me that you persist in saying things like "the misspelled entry façade" and "The trolls that wish to promote façade as a valid spelling". It reminds me of the times you've asserted that "colour is wrong" when what you really mean is that it's not correct in your dialect. Your continued near-fanatical distaste for any spelling that doesn't match your own American orthography makes you look like -- sorry for the harsh words here -- the worst kind of xenophobic Ugly American. I know you're not, I know you have only the best interests of the project at heart, so I really, really wish you could somehow adopt a more calm and tolerant stance towards these inevitable divergences in spelling.

Spelling is important, sure, but it's not a life-or-death issue, and in an inclusive and (I forget where you stand on this) descriptive dictionary, we should be focusing much more on right spellings than wrong spellings. That is, don't say "I think color is right and colour is wrong". Either say "color and colour are both right, depending", or, if you must, "I think color is right and Brits think colour is right." It may seem an insignificant distinction, but if you could somehow see it that way, if you could stop focusing on "wrong" and "nonstandard" so much, I think you'd get in a lot fewer arguments, and people wouldn't be put off by a perception of intolerance or demagoguery so often. —scs 02:14, 14 November 2007 (UTC)Reply

I blame myself for reacting to the trolls, as you say. But it is clearer with each round that they are trolls. Yes, of course I recognize that "colour" is the only valid British English spelling. I'm not convinced that "façade" is particular to the UK - it seems equally uncommon/unfavored in British contexts as American...that is to say, by an enormous margin. Taking more and more time off from Wiktionary, I view my own responses (to their ass-ertions that façade is the only valid spelling) unfavorably. The problem is, every time I leave it alone for a dozen hours, I return to find the conversation has gone full circle again, laced with more personal attacks. I am disappointed at my knee-jerk reactions to such idiocy.

The two times I suggested deleting or soft-redirecting façade, were just non-serious outlets of my exasperation at the pin-headed lunacy of leaving façade as a "main entry" for facade. Some of the trolls seem so thick-headed that they can't see that their promotion of their spelling is not descriptive of how the terms are used in English. I am beyond frustrated at that sick mentality. If you are going to be prescriptive about it, then for God's sake, at least prescribe the English normalization, not some uncommon/rare foreign-language spelling. Prescribe that the foreign language spelling be left in italics. If you are going to contribute at Wiktionary, then for Wiktionary's sake, be descriptive.

I've tried many different ways to convey that. I am disappointed with how my attempts came out and dissatisfied with each successive round of sickos POV pushing. I'm so disgusted by the troll-tactics, I'm tempted to try ignoring it - but each time I've done that in the past, the POV-pushers have taken that lack-of-militant-opposition as an endorsement of their ridiculousness.

Having read (just now) only this section of my talk page, after many hours away, I think I shall do Something Else for a little while, to at least calm down. The worst thing I could do now, would be to read what new nonsense is there in WT:BP and fire off more BS to feed the trolls (thereby making their attacks look justified.)

TTFN. --Connel MacKenzie 05:48, 14 November 2007 (UTC)Reply

I get where you’re coming from, Connel. The thing is, we advocate different bases for prescription — at the core of it, you have a strong naturalising–common-use preference, whereas I have an etymological–functional one. (Let’s be honest — we would both prefer to prescribe.) That’s probably not going to change, at least not significantly. I would hope that I am not counted amongst the “trolls” of whom you speak (having been involved in that discussion very little). A note for the future: people are more likely to favourably consider your view if you find references (the dictionary list now at facade and façade is a good start) that explicitly prescribe what you say (I’m sure you could find them; I’m sure some writers of style guides share your view, though they may express it more palatably). In the end, you’re more likely to get a balanced usage note than you are anything else — but isn’t that an acceptable modus vivendi? –Whilst I would also prefer things here to reflect my (like yours, minority) opinion, I can live with and coöperate in that modus vivendi. Honestly, try to be balanced and forbear flaming (a prime example: “façade should be listed as an inferior / offensive spelling”) — you had good points to make, but people find that very hard to concede when their egos are being attacked. (I tend to ignore such nowadays, but others (including you), as we have seen, do not.) I know we are diametrically opposed on most issues, but I still want you to stick around, doing the good work that you do on this project. Just try to be at least a little wiser in going about getting your way.

As a gestrue of good will, I’ll do something that I may regret:

Nota that The New York Times Manual of Style and Usage (1976; published by Quadrangle / The New York Times Book Company; ISBN 0–8129–0578–4), revised and edited by The New York Times News Editor Lewis Jordan, on page 75, prescribes the cedilla-less spelling thus:

• facade. No accent.

That’s the kind of evidence you need to present. † ﴾^(u):Raifʻhār ^(t):Doremítzwr﴿ 12:04, 14 November 2007 (UTC)Reply

If the purpose of your response was to see if my head would explode, I'll have you know you came pretty close.  :-)

Seriously though, I was not being prescriptive but rather descriptive when I said "inferior / offensive." Surely you see the distinction?

To respond to the larger issue you raised, I'm uncomfortable composing my response formally. But I can reply in a couple hundred words here, what, if cast formally, would take over the thousand. I will take great pains to edit out anything that might be misconstrued as insulting. But the tone of this response is candid/friendly/humorous - if you read anything mean, then you've misread it... Editing note: well, except the Unicode thing.

I do not understand how you can say "The thing is, we advocate different bases for prescription — at the core of it, you have a strong naturalising–common-use preference, whereas I have an etymological–functional one. (Let’s be honest — we would both prefer to prescribe.)"

Disclaimer: Yes, I would very much like to prescribe, but this is Wiktionary, so I don't.

See, if you wanted to prescribe "functionally" then you'd be vehemently advocating the naturalization of foreign terms. On the other hand, if you wanted to prescribe "etymologically" you would also strongly advocate the naturalized spelling, instead of recommending that people keep a foreign term in a foreign language, merely "borrowing" it in English. Imitation is the greatest compliment, eh? But wholesale copying is not imitation. Taking a good concept from another language and adapting it to one's own language is exactly what the English language is so very famous for doing. It is perhaps the central trait of the entire language. You aren't "paying homage" (or some-such) to French when you neglect the naturalized spelling, in favor of the French spelling...you are using a French word.

Now, there may in fact, be some erroneous publications like the OED that prescribe the wrong spelling for the wrong reasons. When I checked Oxford Reference Online (COED) a few minutes ago, it did not. Now, indeed, there are many demonstrations of people erroneously believing their Microsoft Word auto-correction. Without any doubt whatsoever, the "façade" spelling exists in English (whether the authors either intended to use the English word facade, or were trying to be extra-pompous, or just didn't notice the Microsoft error, is not clear.) But whichever the case, it is equally clear that the French spelling (when used in English) conveys something quite different from the English word.

You, Widsith, Scs, Ric (and perhaps several others) assert that you somehow intend to use the wrong spelling, or have been taught to use the wrong spelling, (or whatever,) for whatever reasons. Honestly, I cannot fathom why someone would think that the French spelling is correct in English, nor can I imagine someone teaching someone else to do it that way. I really can't. That notion just doesn't pass the most basic BS-tests. None of my computer keyboards have that character (nor that accent - if you wish to get picky.) None of my typewriter keys have that character. I know no person who knows how to enter it on their computer. I know no one that was taught to spell it that way (outside of a French class - for the French word.) Google searches show it is a common error. Scs noticed that MS word has that in its "autocorrect" list, (but backwards!) Normal dictionaries use the English spelling as their headwords, many listing the French spelling as an alternate (about as many, not listing the French spelling.) Checking on the claimed "authority" OED (granted, I only have easy access to the COED, which normally corresponds,) indicates some possible deception.

Now, if it weren't for computers, I might never have paid any attention at all. But the diacritics (collectively) have caused me many programming nightmares. Unicode is a very evil entity; at times I think that whole consortium should be lined up and shot. Unfortunately, there is nothing better out there. Often, I dream of an ASCII-only Wiktionary (foreign terms allowed only as romanizations,) - and indeed wake up from such dreams happy and refreshed.

Lastly, and perhaps most importantly, we are on the Internet here. This is a computing context, like it or not. When words in English are sorted, what collation is used? In a sorted list, where does "ç" appear in relation to "c"? Sometimes just after it, other times, miles away from it. Special:Prefixindex/fac, for example, doesn't list the French spelling. Note also, that most of the times I said "French spelling" above, it was because it is easier to type 15 characters, than it is to find the mouse, scroll up, select "façade", ALT-C, scroll back down, mouse over, then finally ALT-V.

So, yes, if I were being prescriptive, (which I'd like to be,) I would delete façade with the deletion summary "Misspelling of facade." You can rest easy - I have no intention of doing so. But I would like to. The further I analyze the situation, the more convinced I am, that that is the most reasonable description for it, that can be given. (Actually, the ==French== entry shares that entry title, so the other methods of indicating the misspelling in English would be preferable.) Relax - I am not suggesting that I will do so (I won't.) This is Wiktionary; I know that wouldn't fly. But I still don't understand the opposition to doing so. I've taken great pains to try to understand that point of view, but I just can't grok it.

--Connel MacKenzie 04:13, 15 November 2007 (UTC)Reply

1. OK, if you intended the “inferior / offensive” comment to be descriptive, then it would have been far better had you explained why. (That is, inferior for what reason (encoding?), offensive to whom (Francophobes?), and so on.) Decontextualised as that comment was, it just seemed like you were denigrating the façade spelling in an extreme, prejudiced, and baseless manner. (Do remember that misinterpretation is incredibly easy on-line, where vocal inflexion, facial gestures, and other factors which normally accompany spoken dialogue are absent.)
2. Intended tone noted. But don’t worry, I generally ignore inflamatory presentation — I’ve found that discussions flow a lot faster if I just keep my ego locked in a little cage, and refuse to get offended.
3. I don’t your claims that, according to my premises, I ought to be arguing a similar cause to yours. Regarding “functional” prescription: I’m using definition №2 — “seful; serving a purpose”; the façade spelling is functionally speaking better than the facade spelling because the cedilla in the former spelling serves the function of showing that the ‘c’ ought to be pronounced as , and not as . Regarding “etymological” prescription: I’m using definition №2 (which I just added) — “onsistent with its etymological characteristics (in historical usage and/or the source language)”; historically, façade has been the universally prefered spelling, and the retention of the cedilla shows greater consistency with the orthographical characteristics from the source language. This little semantic exercise aside, what I’ve described are the bases whereon I prescribe usage.
4. If you believe that the façade spelling has a different nuance from the facade spelling, then please provide evidence for this assertion (probably best to provide usage guides as references; analyzing the great many uses of both spellings for accurate comparison would just take too long).
5. I’ve already explained why I use the façade spelling. In Microsoft Word, I use “Ctrl + ,, c” to insert <ç>; on Wiktionary, I use the EditTools (Latin/Roman menu); elsewhere, I usually use the good old reliable copy & paste technique. Microsoft Word autocorrects a great many of the diacritical spellings I use to the “unadorned” forms (which is really annoying, not to mention illogical — who would add diacritical characters, which require longer keystroke chains, without intending to do so?). You’ll notice from the (admirably long) list of dictionaries we have that some dictionaries list only façade, some façade with facade as an alternative, some facade with façade as an alternative, and some just facade — there’s no real consensus anywhere. I’ll send you a screenshot of the OED entry when I visit my university library later on today (if I remember); that should be enough to dispell suspicions of deceit.
6. An ASCII-only Wiktionary would be a disaster. Technological constraints should never restrict our usefulness to humans to such an unacceptable degree. (I see whence you’re coming, but that oneiric proposal would go much too far.)
7. The ‘ç’ character should be sorted as ‘c’, immediately following it in cases of otherwise identical spelling.
8. Last point: From a descriptive stance, the only misspellings are the ones that stem from error or ignorance. Minority spellings, if intentional, and used by someone informed of the standard rules, are just that — minority spellings, not misspellings. For example, extreme feminists sometimes spell women as wymyn and history as herstory — these may be ridiculous and absurd, but they are not misspellings, as they arise from an intentional use. Do you understand that chain of argumentation? (It’s important that you do.)

† ﴾^(u):Raifʻhār ^(t):Doremítzwr﴿ 15:52, 15 November 2007 (UTC)Reply

I must second Raifʻhār on the importance of his point 8. It's the same point I was trying to make with the examples about the "proper" spelling of Connell, or the pronunciation of Houston Street.—scs 03:07, 16 November 2007 (UTC)Reply

I think part of what you don't grok is descriptivism.

Some nonzero number of people (myself included) spell the word "façade" when writing in English. It really doesn't matter why we do. The fact that some critical mass of us agrees that that's the way we spell it -- is what makes it right! That's how language is formed, and that's what descriptivism describes. The descriptivist finds it utterly nonsensical that you (Connel) could claim that a spelling used by lots of people is "wrong". It's a patently absurd claim, as if you accused me of pronouncing my name wrong, or I accused you of spelling your name wrong, or someone accused the people living on Houston Street in New York of pronouncing the name of their street wrong (because it's different from the pronunciation of Houston, Texas).

Most of the arguments that keep getting trotted out simply don't matter to us. We don't care that you don't know where to find the ç character on your computer or typewriter keyboards. We don't care that some number of other people spell the same word "facade". We don't care that you believe that ç cannot appear in the spelling of an English word. All those arguments mean is that "facade" is also a valid spelling. But they absolutely don't imply that "façade" is wrong. We rejected long ago (or never even had) the notion that there absolutely had to be exactly one officially correct spelling for a word.

I hope what I've written in the preceding paragraph doesn't too badly spoil the nice mood you were in after the last nice ASCII dream you had. I hope it doesn't cause your head to explode in anger. I hope you don't feel compelled to argue at length how wrong you think I am. Don't worry, you really don't have to, I already know you disagree with virtually every word I've said here, and I could probably give you the exact arguments you'll use (starting first and foremost with "How can we have any kind of standardized spelling if everybody's allowed to spell words however they like to?"). I don't have to convince you to agree with my notion of descriptivism, any more than I have to try to convince you that "facade" is wrong (which I'd obviously never do, becuase I don't think it is wrong!). Your belief that ç does not exist in English and that "facade" is spelled without it does not threaten my descriptivism, so long as you don't try to tell me that "façade" is wrong.

(P.S. It wasn't me who told you what Microsoft Word does with facade/façade, because I don't use it. I might have told you that my trusty American Heritage Dictionary lists façade first, because it does.)

(P.P.S. No, I don't see your distinction when you claim to be "descriptive" when you opine that "façade" is inferior or offensive. That's you stating your opinion; it's being neither prescriptive nor descriptive. You'd be prescriptive if you tried to tell other people how they should spell the word, and descriptive if you noticed and described how other people spelled the word in the absence of your trying to tell them how to. Do our entries on prescriptivism and descriptivism not get this right?) —scs 05:13, 15 November 2007 (UTC)Reply

1. My deepest apologies: it was Rod, not Scs who pointed out that detail.
2. Describing how a word is used is descriptivism, not simply listing every possible combination of letters, unannotated. Is our entry for describe inadequate?

--Connel MacKenzie 06:07, 15 November 2007 (UTC)Reply

By the way, thank you for your comments above. If you (plural - you all) really were so devoutly sold on descriptivism, wouldn't you (plural) have been the ones to jump to the "facade" entry's defense, when it was vandalized? --Connel MacKenzie 18:04, 15 November 2007 (UTC)Reply

By all means, if I'd seen it, but as I said, I joined the thread late. —scs 02:11, 16 November 2007 (UTC)Reply

I can't speak for anyone besides me, but as a halfway-descriptivist myself, I think it's reasonable to characterize "facade" and "façade" as alternative spellings, and as a programmer, I really hate the pointless and hard-to-maintain redundancy of giving full entries to both. You say "vandalized", I say "reduced to a single function call". ;-)   Unlike the anon, I'd have preferred that the full entry be at facade, firstly because it's easier to type and therefore more people are likely to try that spelling first, and secondly because it seems to be the more common spelling; but I don't feel strongly about it, because it doesn't seem to matter one way or the other. (Obviously, you feel differently.) —Ruakh_TALK 21:46, 15 November 2007 (UTC)Reply

I (almost) agree with you about the "pointless and hard-to-maintain redundancy". Having two separate entries is just stupefyingly wrong. It was only very reluctantly that I concluded (as I explain at Wiktionary:Beer parlour‎#my own conclusion) that for today's Wiktionary, they're nevertheless the only way. —scs 03:01, 16 November 2007 (UTC)Reply

I've one other question, User:Doremitzwr. You make a (valid) point, that the French accent might also be used in English to emphasize the French pronunciation of the otherwise-English-looking word. That point ignores the fact that particular words have particular pronunciations - but it is conceivable that someone might take that approach. Why then, doesn't the ==English== section of façade say that? For both strict descriptivism, as well as for naturally describing the "English" word façade, that distinction should be mentioned, right? Granted, no one is suggesting that façe or façility might be recognized as English words (or words at all) but still, the "pronunciation argument" seems to be partly plausible for façade. --Connel MacKenzie 15:55, 15 November 2007 (UTC)Reply

I agree entirely; shall I add such a usage note? BTW, nota that the general rule, both in French and in English, is that a ‘c’ is pronounced as before a consonant, an ‘a’, an ‘o’, or an ‘u’, whereas it is pronounced as before an ‘e’ or an ‘i’ (with a whole heap of exceptions, most notably when a ‘c’ is followed by an ‘h’, in which situations the resulting digraph is pronounced as replace ʧ with t͡ʃ, invalid IPA characters (ʧ), as in church, or as , as in loch). Therefore, a cedilla would be superfluous (under the rules of either language) in words such as face and facility. (However, a cedilla would serve a useful function in a word like accede (as açcede), as the spelling-intuitive pronunciation, /ækˈsiːd/, is different from the correct pronunciation, /æˈsiːd/; though, I have never seen such a (functional yet unetymological) spelling used.) † ﴾^(u):Raifʻhār ^(t):Doremítzwr﴿ 16:49, 15 November 2007 (UTC)Reply

Hmmm, perhaps we should be clearer on what the rule is, before adding the note. The ch digraph is usually listed as a higher-order (in terms of precedence/application) pronunciation, so it would be unfair to call those "exceptions" to the rule. In English, my understanding is that the "c" (or "cc") take on that pronunciation when preceding a vowel sound. At any rate, the pronunciation applies either way: the usage note should only describe the cedilla as emphasizing the pronunciation.

On another note, the IPA symbol "x" is usually given as an example of a sound that doesn't exist in English - every example I've seen is clear to distinguish that as the Scottish pronunciation of the word loch. (For that one word, the "x" pronunciation does sortof exist in English, but only imitatively/dramatically. I'm told my best GenAm approximation of the Scottish pronunciation comes no where near an actual Scottish pronunciation though.) --Connel MacKenzie 17:31, 15 November 2007 (UTC)Reply

Up above, Connel wrote:

Widsith, Scs, Ric (and perhaps several others) assert that you somehow intend to use the wrong spelling, or have been taught to use the wrong spelling, (or whatever,) for whatever reasons. Honestly, I cannot fathom why someone would think that the French spelling is correct in English, nor can I imagine someone teaching someone else to do it that way.

Me, I do tend to spell façade with a ç, and I dispute that this is "wrong" in English. But it's not that some strange person cruelly taught me this "misspelling"; it's just that after seeing it both ways, and now that we all finally do have computers that let us use glyphs like ç almost as easily as those boring old 95 plain ASCII ones, I decided I liked the spelling with a ç better.

Here's my model of how languages grow and evolve. (There's a certain amount of barely-informed speculation on my part here; real linguists might howl.)

The "proper" grammar, pronunciation, and spelling of a human language are all determined by consensus, over time, among all the speakers of that language. Everybody spells and pronounces words, and constructs sentences, according to what sounds right to them. But of course communication never happens in a vacuum; it takes at least two people to have a conversation. So we're always hearing the way other people speak, and reading what other people write. And if we hear or see something we like better, we may begin to adopt it ourselves, often quite subconsciously. More importantly, our children integrate over all the speakers they hear as they're learning their native tongue, and select what sounds best to them, without sticking out from the crowd of their peers too much.

Eventually, after enough generations go by, and although there may be rather extreme divergence between different dialects, there is also remarkable convergence among the speakers of what comes to be known as a single dialect.

Obviously, the mechanisms that mediate this consensus-building across all the far-flung speakers of a popular language are imperfect. Once upon a time, grammar and especially spelling were much more variable than they are today. But the rise of book printing, and other forms of mass communication, and of course dictionaries, have greatly aided the spread. For the most part, spelling is standardized now -- there are a few (quite noticeable!) exceptions, but for the vast majority of words in the language, there's really only one accepted spelling.

However: language does continue to evolve. Dictionaries are important arbiters of standardized spelling, but they're not absolutely imperative. If I really don't like the dictionary spelling of a word, once in a while I might deliberately ignore it, and spell something my own, "better" way. That's obviously heretical and sacrilegious, but on the other hand, unless you believe that language is static, there have to be some people somewhere doing this sort of thing, at some level, all the time, if the language is ever going to evolve at all. (I'm not saying I do this very often, and I'm certainly not saying you or anyone has to agree with me. It helps if I've got a peer group. Perfect example: the dictionary spelling of "judgmental" is just wrong. I always tend to spell it "judgemental", although I didn't start ignoring my spellchecker and leaving it that way until I came across an essay somewhere in which someone else argued that the spelling with the e was obviously preferable to them, too, and explained why. But don't worry, I'm not going to populate that redlink as an "alternative spelling" -- yet. :-) )

The bottom line is that at any given time, there are always some few words that don't have a single, accepted, "correct" standard spelling, either because they're too new for a consensus to have congealed, or because the collective has begun the process of changing its mind. —scs 03:50, 16 November 2007 (UTC)Reply

Heh - I assumed it was not you without even checking (but yes, it was a blue link when I first saw it here.  :-)   No worries - I understood what you were saying. --Connel MacKenzie 17:51, 16 November 2007 (UTC)Reply

RE: your post script, PLEASE! This is a much better place for venting, than WT:BP!!! Your candid comments are very much welcome here. --Connel MacKenzie 05:45, 16 November 2007 (UTC)Reply

That's a lot to absorb. Nothing new, though - but a very nice summary of where you stand.

Yes, truly alternate spellings exist. But you said it yourself above - language evolves. When you choose a different spelling, you are saying something subtly different, than you would if you used the "proper" spelling. Sometimes it is an eye dialect, sometimes it is spelling reform, sometimes is is grammar reform, sometimes it is regional nationalism, sometimes it is just regional. But sometimes it is a genuine mistake. "Descriptivism" doesn't mean any/every possible combination of letters - it still has to be describing something real. When Wiktionary gives a description of an alternate spelling, it is very rare, that the alternate spelling is "just because." (I doubt that situation truly exists, anyhow.) It is a million times more useful to describe what is different about a different spelling, than it is to just list it as an alternate. To not explain what is different about a spelling with a cedilla seems criminal. To not explain that some people will view it unfavorably seems a little mendacious.

"Descriptivism" evolved from traditional "prescriptivism" precisely why? To encourage new inventions? No, it evolved as a mechanism to account for new coinages, regional spellings and grammar evolution. "Descriptivism" is often misinterpreted to mean "include everything," but that would be useless - every "descriptivism" advocate knows that.

Now there are several conversations here that have been commingled. As I said, I'm not about to go running around, deleting stuff prescriptively. But I did try to make my prescriptive POV understood (conversationally) to those reading here, who were left wondering. As I said, I'm not about to go running around, deleting stuff prescriptively. I've explained why I'd like to - but I'm not about to.

Anyhow, I'm not even advocating that we move our "line in the sand" one way or the other, with regards to what "level" of descriptivism we maintain. I'm am rip-roaring-mad that #1) the proper spelling was vandalized, #2) the uncommon/specialty spelling was "promoted", #3) when it was pointed out, the so-called-descriptivists started defending the vandalism! Come off it! That doesn't comply with any version of Wiktionary policy, ever (and it certainly is neither "descriptivistic" nor descriptive.)

Now, you have your reasons for using the "wrong" spelling. I have my reasons for using the correct spelling. You have your reasons for not calling my spelling "wrong" (presumably, to be "descriptivistic" about it.) I have my reasons for calling the French spelling wrong in English (as explained several different times, several different ways.) What Wiktionary says about them, in the end, we've yet to see.

I do hope that you are not offended, that in my mind, your spelling is just wrong. But, would you rather not know that people won't like your spelling? When you write a book, do you really wish to unintentionally be off-putting to people? Would'nt you rather have a useful note (descriptively worded) that warns you?

--Connel MacKenzie 06:24, 16 November 2007 (UTC)Reply

No, don't worry, I'm not offended; my ego's way too big for that. Puzzled or dismayed, maybe, but not offended. (Because of course your attitude is just wrong. :-) )

I do, in fact, know that the cedillaful spelling might be distracting at least. I wouldn't necessarily use it in a book. (Actually, I see I didn't use the word "facade" in my book at all, and while there is one c-cedilla, it's in a Frenchman's name in the acknowledgements. It looks like I did use one "cliché", two "naïve"'s, one "naïvely", and an "à la", and it seems my copyeditor let me get away with it.)

I agree with you about the need for a usage note. See what you think of the one I just added. —scs 04:21, 21 November 2007 (UTC)Reply

Thanks for your concern, but there's no risk of vandalism from this account. I'm an enwiki administrator, Mediator, and OTRS agent. I've been using this same account across all WMF wikis since I signed up in 2005, so I see no reason to change, local policy or not. ^demon 22:19, 14 November 2007 (UTC)Reply

As in, I sent you one. Cheers! bd2412 T 08:11, 16 November 2007 (UTC)Reply

Thanks for the reminder - I forgot to send the response/results yesterday. --Connel MacKenzie 17:16, 16 November 2007 (UTC)Reply

We were having a discussion of how to format entries for verbs that can be both reflexive and non-reflexive over at Wiktionary talk:About Spanish#Reflexive verb formatting. The consensus had been to have all senses in the same entry. We mentioned having 1 POS verb header and 2 separate inflection lines, as well as having 2 separate POS verb headers. You and User:EncycloPetey were mentioned and I'm sure we'd appreciate your comments there. Thanks --Bequw 18:20, 28 November 2007 (UTC)Reply

FYI, I left this follow-up question for you. Rod (A. Smith) 18:05, 29 November 2007 (UTC)Reply

In the entry for antiship, I find {{en-adj|pos=]]|-}}. Is this syntax still in use, or should I change it when I come across it? What is/was it's purpose? Is it somehow trying to be a method of displaying alternate forms antiship and anti-ship? ('Cause if so, it doesn't display that way). -- Thisis0 19:59, 28 November 2007 (UTC)Reply

That inflection line is OK - I think. It is showing the components wiki-linked. There should also be {{see}} on line one and/or an ===Alternative spellings=== section (but that is quite separate from showing the individual components.) --Connel MacKenzie 20:04, 28 November 2007 (UTC)Reply

Perhaps it should have an ===Etymology=== section to make that clearer. --Connel MacKenzie 20:05, 28 November 2007 (UTC)Reply

So, would we prefer the Etym. section (from (deprecated template usage) anti- + (deprecated template usage) ship) because it's clearer? Obviously I didn't notice that antiship had two links breaking apart the word. What exactly does the "pos=" part in the template do, and when else is it useful? -- Thisis0 04:38, 30 November 2007 (UTC)Reply

pos= (for "positive", as opposed to comparative or superlative) is used to control the display of the headword in the inflection line; if you're familiar with {{en-noun}}'s sg=, this is {{en-adj}}'s answer to it. It's usually used when the adjective consists of multiple words separated by spaces or hyphens, so each word can be linkified. —Ruakh_TALK 05:47, 30 November 2007 (UTC)Reply

Yes, precisely. I do wish we had a standard name for that template variable. Thanks for the quick(er) response. --Connel MacKenzie 07:09, 30 November 2007 (UTC)Reply

Thanks all! -- Thisis0 05:12, 1 December 2007 (UTC)Reply

Hi Connel, I wanted to run this idea past someone who knows how things work before trying to hard in the implementation stage. Instead of the current Cookie based WT:PREFS, it would be possible to set up an XMLHttpRequest based auto editor for a user's JS and CSS files, this would allow the preferences to be included more easily and transferred between browsers. It would also be possible to not include the User:Connel MacKenzie/custom.js for every user which is what happens at the moment. An initial test I did suggests that this would work, however there are some possible security implications. Do you have any thoughts about this? (Also - would it be possible for you to leave a note on my talk page when/if you reply). Thanks in advance Conrad.Irwin 14:27, 30 November 2007 (UTC)Reply

Yes, that could work. The only fundamental difference is that Javascript is cached, XMLHttpRequest is not. Yes, the custom.js stuff was supposed to be moved copied to the Wiktionary: namespace a long time ago, but each person caught in the attempt (myself included) soon becomes overwhelmed by the gigantic refactoring task. (Pretty much everything in MW:Monobook.js, cmack:custom.js + all related .js files belong in MW:Common.js instead.) Every time I've set aside a weekend for the task, something else has come up. --Connel MacKenzie 14:43, 30 November 2007 (UTC)Reply

FYI: I blanked this temporarily so my code could automatically shoot a number of the old redirects. Will put it back ;-) Robert Ullmann 15:08, 30 November 2007 (UTC)Reply

No worries - it will get refreshed with the next XML dump. --Connel MacKenzie 15:21, 30 November 2007 (UTC)Reply