Wikipedia talk:Conlangs/Straw poll

The talk page of the previous (failed) poll has moved to Wikipedia talk:Conlangs/Old straw poll.

New poll form

Hello! I have just replaced the old poll text by a new one, based on my original alternative proposal. Please review, correct my English and my other errors! --IJzeren Jan 14:55, 7 September 2005 (UTC)

Opposing proposal in its entirety

I have significant objections to this proposal. They are discussed at Wikipedia:Conlangs/Alternative proposal#Please consider dropping the voting, it's probably a waste of time. / Alarm 19:54, 7 September 2005 (UTC)

• I am paying no attention to any alternative that is so complex as to require creation of its own subpage, but doesn't rate a description on this page. See next unbulleted contrib.
  --Jerzy•t 17:04, 3 October 2005 (UTC)
  

_ _ My objections are unqualified. We've apparently had about a conlang article a week, amounting to one for every 4,000 non-deleted articles, and compared to something like 1500 AfDs per week. This is not a problem worthy of this much attention.
_ _ And even if it deserved the attention, it is only by a gross overestimation of the significance of conlangs that anyone considers formulating a set of criteria embracing all conlangs but nothing else: Newspeak is a fictional conlang; Klingon and Vulcan are conlang realizations of fictional natural languages; other conlangs include toy languages for instructional purposes and for personal satisfaction; there are attempts at a universal language (so far notable for serving that role less well than various natural languages in their respective periods of ascendancy) and there's Toki Pona, which IIRC has been described as a tool for inducing a mental illness in oneself.
_ _ The disproportionate effort contemplated by the straw poll proposal, and i presume its predecessor and probably the alternative, may well reflect the misguided hope of formulating a Universal ConLang Evaluation System (UnCLES), whose results we don't need.
_ _ If the conlang situation is too murky, consider compiling a list of conlangs known to have been at least mentioned in WP articles or lists; subdivide it (or maybe just label the entries) according to some improvement upon my classification scheme above; note creation dates for those that have had articles; for each whose deletion has been weighed, note also AfD/VfD date, whether deleted, and rough percentage for deletion; sort one copy of the list by VfD date, or creation date if none, or date of first mention if none; sort another by percentage for deletion (with mention only = 103%, section of article only = 102%, speedy w/o undelete = 101%, and never VfD'd = -1%). The chronological version helps spot trends as editor population and article population have grown, and may suggest additional sorting schemes that weight the percentage to compensate for time variations of inclusionism level, editor-to-article ratio, broadness of interest in conlangs, and whatever. Such a tool requires no consensus to become useful, and IMO would be more practical than an ironclad notability standard.
--Jerzy•t 17:04, 3 October 2005 (UTC)

Length

I think this poll is too long and has too many criteria. --MacRusgail 13:52, 24 September 2005 (UTC)

• I think you could potentially argue that this poll is a good way to begin focusing what the Wikipedia community thinks is and is not a valid criteria for the notability of a conlang. The reason the poll is too long is precisely the reason why a poll is needed: In debating the value of various conlang pages on the merit of notability, everyone is using different standards of notability. Some focus may be desirable. On the flip-side: Some of the questions have far too many sub-questions. It would be more valuable, perhaps, to first narrow our focus to what general criteria we think are important and then start looking at exactly how we would assess some of those criteria in an objective fashion. Justin Bacon 15:56, 25 September 2005 (UTC)

• • I'm not against it, per se, but feel that the length may put some people off and also cloud the issue. I do feel, however, this is a hard issue to crack. I would say that Esperanto, Klingon, Quenya, Volapuk, Interlingua are all articles worthy of existence, but where to draw the line? --MacRusgail 14:54, 26 September 2005 (UTC)

General Thoughts

I've only just stumbled across this debate in the past couple of days, but my thoughts:

• Number of speakers and/or works composed in the language clearly denotes notability, for the same reason that the popularity or widespread use of any subject connotes notability.
• Completeness, in my opinion, is irrelevant when it comes to notability. Completeness may be indicative of widespread use (which is a useful indicator, but you can look at it directly rather than using completeness as an indicator), but it could also just mean that one guy has made it a labor of love. (I'm all for labors of love, but it takes more than that to merit an encyclopedia entry.) On the flip-side, I can think of many examples where incomplete conlangs could merit notability. For example, Vulcan (last time I checked) is an incomplete conlang (unlike Klingon), but its significance to Star Trek makes it notable.
• Uniqueness doesn't seem indicative of notability in any way, IMO, for similar "labor of love" reasons. (However, it's possible that a conlang could be notable for no other reason than creating a linguistic approach which was picked up by other conlangs which have notability themselves. Or by influencing the conlang community as a whole. But I'd consider that a secondary effect which would fall under Reputation.)
• Publication I agree with Mgml's logic on the straw poll page.
• Reputation. I agree with the logic of Average Earthman on the straw poll page, with my additional thought (above) that influencing the conlang community would be a notable form of reputation.
• Year of Creation. This would appear to fall under Reputation, IMO. Those older conlangs, by their very nature, have a reputation and influenced the conlang community as a whole.
• Notability by proxy. Being used in a notable way clearly connotes notability. Although I'd argue for a more intuitive description and broader category of Notable use
• Last question. The answer to this is entirely dependent on how many major criteria we end up with. Including it in this poll will only result in a meaningless result, IMO. Justin Bacon 15:56, 25 September 2005 (UTC)

• • I think it is worth pointing out that the number of speakers of any given conlang tend to be notoriously difficult to calculate. With Esperanto, there are wildly varying estimates for world usage. --MacRusgail 14:57, 26 September 2005 (UTC)
  • p.s. Agree about Vulcan, but would say perhaps only Vulcan and Klingon are worthy of articles.
  • p.p.s. Too many criteria like I say above, but perhaps "consistency" and "linguistic interest" are also worthy of consideration. A poorly constructed conlang is not of much interest.

Languages created by notable people.

Would deserve at least a redirect to that person's article, or similar. I'm thinking particularly of Boxen and Tolkien's childish languages, but I'm dsure there are more. The same would apply to fictional conlangs, Newspeak for example. Rich Farmbrough 15:20, 7 October 2005 (UTC)

• Comment I agree with this. --MacRusgail 19:38, 16 October 2005 (UTC)

The whole thing is not needed, and overly complicated.

If a conlang is verifiable, then we should keep it. My God. 200 articles. The horror. Trollderella 05:52, 28 October 2005 (UTC)

While I'm not a member of whatever that association is, this is clearly a case where we should just use common sense; no guideline page is needed. --DDerby-(talk) 20:48, 1 November 2005 (UTC)

The proposals for conlang inclusion are horrendously overcomplicated. In fact, every Wikipedia guideline is horrendously overcomplicated. Every single one of them should be thrown out. I don't know about anybody else, but I have stopped following Wikipedia guidelines altogether. There's too damn many of them. I assume -- and so far I've been proven correct -- that common sense will guide me adequately. If I were to spend my free-computer-use time boning up on current policy, I'd have no time left to edit articles! As it stands, I spend more time voting whether to keep or remove articles than actually editing them and making Wikipedia better. I'd like to put forward a modest proposal: Delete only those articles which qualify for "speedy deletion". Let the rest hang around taking up space. Who cares? There's a lot of space ("Wikipedia is not paper"). Everybody, spend your time improving the articles we need, rather than trying to kill the ones we don't need. It will be time better spent. Ravenswood 21:27, 28 November 2005 (UTC)

I think the argument in favor of deleting some types of articles is to retain Wikipedia's credibility. If Wikipedia is filled with nonsense articles, advertisements, vanities, and other silliness, then it decreases the value of the legitimate stuff. That's just my view, of course. | Klaw Talk 22:49, 28 November 2005 (UTC)

I agree, basically, with all of you. But:

• "Common sense" is a troublesome thing. It tends to differ from one person to another.
• "Wikipedia is not paper" is not exactly a commonly accepted point of view either. Personally, I agree with Ravenswood. But unfortunately, reality is that there are plenty of people who seem to carry some kind of grudge against conlangs in general, and that conlangs pop up regularly in the AFD discussions. And - I have to add this - there are thousands of conlangs, and surely we don't want to include all of them.
• I also agree with Klaw that we don't want nonsense articles/advertisements/vanities/other silliness, since they dó decrease WP's value in general. Therefore, some distinction needs to be made between valuable and non-valuable stuff. If we can achieve that with common sense and without guidelines, all for the better!
• At last, I also agree that the proposals for conlang inclusion are horrendously overcomplicated. But keep in mind, that my proposal for a straw poll was based on another proposal that was even more complicated. And, I'm sorry to say it, in my personal opinion it's all those reservations made by people (regarding the ISBN numbers of books, their sales figures, etc.) which make it really complicated. But don't forget what the ultimate purpose of this poll is: it is not supposed to turn into a consistent set of criteria with numbers, sales figures and the like that would require a spreadsheet to calculate the inclusibility of a conlang. It is supposed to give an impression about which points we have some consensus.

--IJzeren Jan _{In mij legge alle fogultjes een ij} 09:57, 29 November 2005 (UTC)

"there are thousands of conlangs, and surely we don't want to include all of them."

Why not? Would you oppose the creation of stub articles for barely-documented natural languages? I agree with Ravenswood here. Deletion on the basis of obscurity is spurious. Sure, don't feature some obscure conlang in a major list; don't clutter the central pages unless they explicitly aim to be comprehensive; we all agree on that I think. But as mentioned, this ain't paper. Having a stub article on every conlang in existence will not make it any more difficult to read the popular ones, or the historically / etc significant ones that are featured in major articles. Nor do they even clutter search, since they're sorted by rank. --Sai Emrys 20:22, 4 January 2006 (UTC)

I addressed this point above. Wikipedia's credibility suffers if every six-bored-teenagers conlang gets an article. Conlangs that never gained traction with speakers, that have no current speakers, and that have no significance in conlang history probably aren't worth including here. | Klaw ¡digame! 21:15, 4 January 2006 (UTC)

I wholeheartedly agree that they are not worth including in places that would give them an appearance of importance. That indeed should be for important conlangs, by which all the aforementioned standards of relevance etc etc etc apply. However, I do not see any way in which Wikipedia entries for important conlangs are "diulted" by ones for unimportant ones. The unimportants remain fringes - linked only in searches and exhaustive lists - and do not seriously affect the quality of the central articles. If someone happens to know some obscure conlang's name, by all means let them look it up. Or likewise if they happen across it by accident. But I don't see any way that it decreases the quality or credibility of Wikipedia in general, or its conlangs subsection in specific.. --Sai Emrys 01:02, 5 January 2006 (UTC)

Six Bored Teenagers

Revisiting the poll I find mixed responses to my voting rationale. To state things cogently, one of my primary concerns is to protect Wikipedia against vanity languages. Constructed languages appeal to a particular demographic. People who enjoy Star Trek and Lord of the Rings often have technical expertise, youth, and time on their hands. Internet trolls do engage in elaborate plans to disrupt legitimate online activities. It makes sense to consider this while Wikipedia establishes rules for including online languages.

I have known teenagers who created entire MMORPGs for the sole purpose of stealing other players' passwords on established MMORPGs. It takes considerably less effort to construct a conlang vocabulary of a few hundred words, spam some message boards, and create a fake trail of Google hits with the appearance of controversy. To my understanding, Wikipedia's aim is to include only academically and culturally significant conlangs. Durova 02:15, 12 December 2005 (UTC)

A suggestion

Why not start paring the list of conlang articles on Wikipedia? We can identify the least worthy article, AfD it, and once (if) it's deleted, move on to the next one. When we can't find another to delete, we're done. It's easier than a complicated set of rules, and it achieves the goal of removing non-notable and vanity conlangs. | Klaw ¡digame! 02:44, 12 December 2005 (UTC)

I think that's only productive if your only goal is to drastically reduce the number of articles in the section. It won't produce any guidelines for determining whether a conlang that doesn't appear in Wikipedia is worthy of being added, it will privilege conlangs that happen to be popular among established Wikipedians, and it will send a message to the conlang community not to contribute any new content to Wikipedia. I understand that it's slower and more complicated to come up with guidelines, but doing so addresses the long-term quality of the section, not just the short-term bloat. Cnoocy 03:55, 5 January 2006 (UTC)

I think your initial assertion is way off. The goal is to find the line dividing conlangs that should stay and those that should go, and this represents good way to find it in a practical manner rather than the abstract manner we're shooting for here. Furthermore, it's equally likely that very few conlang articles will be deleted. AfD requires a high bar for deletion, and there are a lot of editors on this page who are suggesting a low bar for conlang inclusion. After a few articles have gone through AfD, we can look back and see what variables seemed to be present on articles that stayed and absent on articles that were cut. | Klaw ¡digame! 04:04, 5 January 2006 (UTC)

I misunderstood your intentions, and I apologize. I had the impression that you were advocating deleting as many as possible, and declaring the issue settled. I do, however, think that your suggestion is just another way of conducting a straw poll, and one that will discourage possible new Wikipedians by sending the message that notablility is arbitrary.Cnoocy 14:35, 5 January 2006 (UTC)

Thanks, although your apology was not necessary. Although I'm generally a deletionist, my only strong feeling on conlangs is that vanity conlangs (Durova's "six bored teenagers" point) should be deleted. Should Babm, an obscure, never-spoken conlang with no historical descendents or significance have an article? I don't know. On the one hand, Wikipedia's not paper; on the other, Wikipedia is not an indiscriminate collection of information. I think that if we talk it out over specific articles, rather than an abstract class of articles, maybe we'll find we're all better able to articulate our views on what conlangs belong in Wikipedia. | Klaw ¡digame! 15:32, 5 January 2006 (UTC)

This is overspecified for a straw poll

If this is truly a straw poll, why are the mechanisms to identify the "winning" choices so precisely specified? Some of them, like taking the mean of a bunch of suggested numbers, are also very badly thought out. It should be possible to look for consensus without such calculations. But if the results of this vote really are going to create a precedent, don't call it a straw poll. That just sounds like a way to dodge the "Voting is evil!" people. rspeer / ɹəədsɹ 23:29, 26 December 2005 (UTC)

It basically was renamed as "straw poll" after being undeleted when a certain "voting is evil" administrator had deleted the original page with no discussion. I suspect you're probably right that it's excessively complex. --Jim Henry 17:05, 29 December 2005 (UTC)

Thanks for the explanation. Now that I've looked over the poll again, I've concluded that, if the rules of this poll are followed, the resulting guidelines are going to be nonsensical. This is an example of instruction creep completely obliterating common sense.

So I've withdrawn my votes except to vote "Throw out this 'straw' poll and use common sense" in the last section, and I encourage others who disagree with this process to do the same. rspeer / ɹəədsɹ 00:35, 31 December 2005 (UTC)

I suggested in the prior section that we use a more organic method of doing this: Pick the least encyclopedic conlang article in Wikipedia at the moment, propose it for deletion, and see how it goes. If it's deleted, we move to the next one. There are obviously enough people who care about the topic to make sure that the AfDs get a suitable debate. Anyone who's still reading this page should have a look at Constructed languages and Fictional languages and nominate one article that earns their vote as the least Wikipedic. My guess is we'll quickly settle on three or four such articles that we can probably agree aren't worthy of inclusion. | Klaw ¡digame! 03:10, 31 December 2005 (UTC)