Toytoy's proposed policy for wiki closure

From Meta, a Wikimedia project coordination wiki
Jump to: navigation, search
Historical This page is kept for historical interest. Any policies mentioned may be obsolete. If you want to revive the topic, you can use the talk page or start a discussion on the community forum.


This page is an archive of the proposal below. As per the vote in Talk:Toytoy's proposed policy for wiki closure#Move to close debate, I'm marking this debate closed. No further edits should be made to this page.

Raise your wiki-stake
Light up your wiki-torch
This is your wiki-land
Do whatever your wiki-do -- Toytoy 16:19, 27 July 2005 (UTC)
People complains to me that I used too many "we" in this proposal. I'll fix it when I have time. Don't forget to vote on the talk page. Have a nice day. -- Toytoy 16:37, 26 July 2005 (UTC)
This proposal is personal. I am responsible for it. Please do not change it because you are not me. Please use the talk page or make another proposal of your own. I'll move other people's edits to the talk page. -- Toytoy 02:13, 26 July 2005 (UTC)
Disclaimer: As a user of English and Chinese wikis, my unorthodox view does not reflect the general ideas of these communities. If you don't like my view. You are not alone. They hated me too. -- Toytoy 14:48, 24 July 2005 (UTC)

This is a proposed policy for closing wikis that are not up to expectation. If you dislike my idea, please use the talk page.

Contents

[edit] Background

As of this writing, there are at least 204 non-English Wikipedias. 70 of them have more than 1,000 articles. The rest of them are either inactive or infrequently used. There has to be a reasonable means to weed out inactive wikis. Each version of Wikipedia is supposed to be a good enough w:encyclopedia that can be used by people. It is good PR to have more than 100 non-functioning wikis hosted on the Wikipedia server. It costs almost nothing while it creates a false sense of cosmopolitanism. However, it serves possibly no one. And no one is thankful.

An article is not per se worthy to keep. That's why we have rules dealing with deletion and speedy deletion. The same standard shall be applicable to a whole Wikipedia. If it is not a good encyclopedia, it shall not be kept indefinitely. A good encyclopedia is supposed to be reasonably up-to-date, credible and accountable. That's why I make this proposal.

[edit] Goal

According to en:Wikipedia, there are 92 "active" Wikipedias as of March 2005 and the rest of them are not so active. I propose that we close, very possibly, all inactive ones and also many "active" but non-functioning ones as well.

[edit] Rationales

Wikipedia was not created with Pope's blessing. It has been criticized and challenged. Its predecessor, w:Nupedia, closed its door forever because its idea was not working. There are other encylopedias or encyclopedia-like websites such as w:Encyclopædia Britannica, w:Encarta, w:h2g2 and w:Everything2, so far, Wikipedia is considered highly successful by most. Face it, if there's an even better and more popular web project, how many of you would stay here and face other people's scrutiny?

Success is the only justification.

Based on w:Criticism of Wikipedia, Wikipedia has been questioned on many fronts. A functioning wiki shall be able to face the minimum amount of internal scrutiny. Some indicators may be used to raise a w:prima facie concern. If a wiki in question cannot reasonably justify its perceived failure, we'd better close it.

[edit] Tests

In my opinion, there are many usable tests to tell if a wiki is worthy to keep even if you cannot read a single word of that language. A particular wiki may not meet all of them, however, failure to meet most of them shall be a w:prima facie no-no.

  • Market test: A good wiki shall be copied by other for-profit websites.
  • Openness tests:
    • A good wiki shall attract new users.
    • A good wiki shall attract IP users.
  • Interactiveness tests:
    • A good wiki shall generate a significant amount of discussion.
    • A good wiki shall not be written by just a handful of people.
  • Usability tests:
    • A good wiki shall be cited by books, newspapers or periodicals.
    • A good wiki shall be translated by other wikis.

If a wiki fails to meet most of these tests without an acceptable excuse, we shall not host it any longer. Wikipedia has to have a minimum quality standard. It must not host inferior materials. Inactiveness shall not be the sole reason to close a wiki. Can we afford to spread questionable information to those people just because we cannot read their languages? A wiki shall not be worthy to keep per se.

The word "w:wiki" is Hawaiian for "fast". A slow wiki is an unfortunate oxymoron that we are not supposed to breed. Hopeless wikis, in my opinion, shall be closed. We can always keep their files and supply them with MediaWiki software so someone may recycle the information somewhere. This is nothing to do with their freedom of speech.

[edit] Case study

There are many inactive or zombie-like wikis. Here I am showing you many good-looking wikis are also in trouble.

[edit] The one-man-band symptoms

Based on StatisticsUsers.csv (last update: May 27, 2005), I have found many "alive and kicking" wikis possibly are one-man-bands.

This is a list of top_user_edits/all_edits of some more-or-less functioning wikis, I cannot rule out all bots (there are bots), artificial languages are in italics for comparison's sake:

  1. Esperanto - eo - 25328 articles (8560/112841 = 7.6%)
  2. Simple English - simple - 3922 articles (1769/17111 = 10%)
  3. Klingon - tlh - 61 articles (86/444 = 19%)
  4. Arabic (العربية) - ar - 3222 articles (4475/20314 = 22%)
  5. Welsh (Cymraeg) - cy - 3037 articles (3331/14373 = 23%)
  6. Toki Pona - tokipona - (371/1412 = 26%)
  7. Afrikaans - af - 3890 articles (5034/19467 = 26%)
  8. Greek (Ελληνικά) - el - 4321 articles (3658/13420 = 27%)
  9. Catalan (Català) - ca - 14863 articles (21996/72220 = 30%)
  10. Indonesian (Bahasa Indonesia) - id - 10422 articles (8005/26417 = 30%)
  11. Icelandic (Íslenska) - is - 2687 articles (6874/22248 = 31%)
  12. Estonian (Eesti) - et - 10534 articles (20721/64082 = 32%)
  13. Faroese (Føroyskt) - fo - 636 articles (647/1656 = 39%)
  14. Persian (فارسی) - fa - 2911 articles (4110/9920 = 41%)
  15. Asturian (Asturianu) - ast - 3170 articles (4167/9168 = 45%)
  16. Walloon (Walon) - wa - 5035 articles (12340/23178 = 53%) (See talk page)
  17. Min Nan (Bân-lâm-gú) - zh-min-nan - 979 articles (4052/7478 = 54%) (See talk page)
  18. Ido - io - 4072 articles (6382/10313 = 61%)
  19. Sicilian (Sicilianu) - scn - 1210 articles (2384/3226 = 74%)
  20. Armenian (Հայերեն) Hayeren - hy - 544 articles (1340/1507 = 89%)

As you can see, many wikis are contributed by very few people. And even one user may dominate the wiki. I really don't think this is a good idea to have a wiki created by a only few. I have even seen one wiki has a big error on its Main Page for about five months. Can we rely on a wiki edited by a handful of people? Don't we make mistakes?

  • I really don't know much about bots. It will be worthwhile to subtract all known bot edits from the edit counts and do the above list again. I also do not have the whole database. If any of you knows how to do it, you may learn lots more from analysing the database contents.

Now let me show you some statements in Chinese, with translations in English:

  • 一加一等於三。 --- One plus one equals three.
  • 愛因斯坦是希臘人。 --- Einstein was a Greek person.
  • 狗有六隻腳。 --- Dogs have 6 feet.

Each statement has exactly one mistake in it. Can you correct them?

A wiki is generally trustworthy because errors are almost always corrected. But what if only five people are working for the Chinese wiki? Can they correct all mistakes before it's too late?

[edit] Examples

As a native Chinese speaker living in Taiwan, I can show you some problems created with unneeded dialect versions. There can be more problems with other languages, I simply cannot spot them. I propose that we examine all wikis and close the ones that are not functioning well.

[edit] zh-min-nan (a dream that never come true)

There are millions active speakers of Min Nan dialect(s) worldwide. And zh-min-nan has been here since May 2004 (I guess). The end result is totally pathetic in my opinion. Even though zh-min-nan now has over 1,000 articles, some of them are pretty lengthy, most of these articles are created and maintained by just a couple of regular users. I don't think there is hope of an increased user base in the foreseeable future.

The creator of zh-min-nan is also its top writer. Her edits account for 54% (4052/7478 = 54%) of all edits (May 27, 2005). Do we setup a wiki for just a few persons? Do we setup a wiki so a single willing contributor can spend several hours a day writing it to justify its existence? If that wiki is working, it'll generate new users on its own. No one needs to waste too much time on it just to keep it running. Yes, zh-min-nan now has over 1,000 articles. Most of them are contributed by just a couple of the regular five users. Do you trust such a wiki?

The top five users account for about 86% edits (4052+749+738+480+452). This wiki simply fails to be diverse and active. It becomes a test to a determined writer's endurance. If she, for whatever reason, gives up writing, this wiki will be practically dead.

The day-in-day-out devotion also fails to keep this wiki from making gross mistakes. Several greetins are posted on its Main Page. Two side-by-side greetings contain simply contradictory statements for over five months without anyone's correction. The questionable statements read:

  • 世界頂上儕儂講啲語言當中，閩南語排第22名。...（Wu dialect; added by A-giâu on 16:17 Feb. 27, 2005）
  • ... 全世界數千種語言當中，閩南語人口排名第二十一名。（Mandarin Chinese; added by A-giâu on 01:34 Feb. 11, 2005）
  • My rough translation: Among all languages, Min Nan's population is on the 22nd/21st place.

Can you imagine such a glaringly undisputed visible mistake sitting on the Main Page of any functioning Wikipedia without being corrected in five minutes? I cannot. However, the small user base of zh-min-nan fails to correct it for a long time. Shall we ask them to work harder? Or shall we close that wiki? Remember, this innocent-looking error could be the tip of an iceberg. I can speak a little Min Nan. I cannot read their script. Fact is, I am living in Taiwan, I have never seen their POJ script being used anywhere outside some churchs. Most educated Min Nan speakers write standard written Chinese. Almost none of them use POJ. That's why I don't think they'll gain an adequately large user base in the foreseeable future. This is a hopeless venture that fails to defy gravity.

For my merciless attacks against zh-min-nan, please see (in Chinese):

[edit] Test-WP/zh-yue/ (hopelessly vandalized)

The Yue dialect(s) is spoken mainly in Canton province. Its test pages are being vandalized for months without any hope of improvement. Not unlike other Chinese dialects, an educated Yue speaker uses Chinese or another mainstream language as his/her tool of learning and writing.

So far the Yue test wiki has only 12 pages (including sub-stubs). I have found three of them vandalized but uncorrected for months:

Political vandalism:

There's a page with outrageously incorrect information in its first sentence:

  • 番梘 (Feb. 17, 2005 -): w:soap
    • Original text: 差唔多所有種類嘅番梘都係含有硫酸鹽嘅化合物。
    • My rough translation: About every kind of soap contains en:sulfates. (Possibly badly translated from en:surfactant)

For more information, please see (in Chinese):

[edit] za (an one-man-band with a self-invented writing system)

Zhuang language is unrelated to Chinese. I don't know anything about it.

As far as I know, the only contributor of za wiki invented a Latin-based writing system for his/her own local dialect. The commonly accepted Zhuang language has been written in unaccented Latin alphabets. His/her unorthodox system contains accents.

For more information, please see: zh:Wikipedia:聊天#壮语Wikipedia (in Chinese, and partly translated into English).

[edit] Suggestions

We have to close many non-functioning wikis. Wikipedia is not a language ICU. If a language is dying. We are not supposed to save it. We shall not give too much and pay too little attention. Otherwise, we will have garbage materials in almost every minority language. No information is better than bad information.--User:Toytoy

  • User:Mosesofmason's comments and my reply moved to the talk page.
  • Edits made by 168.209.98.35 (Arguments against a policy for wiki closure) moved to the talk page because this proposal is highly personal. The questions will be answered by me. -- Toytoy 02:04, 26 July 2005 (UTC)

[edit] Consequences

What will it be if we kill some wikis?

I really don't know. I don't know how many wikis shall be closed based on a generally acceptible standard. But I know this shall not be the end of human civilization.

Face it. There are other wikis. Their files are not gone. They can always start a wiki somewhere else if they could find people to join them. The point is they have to find some others to join them.

I personally believe a successful wiki shall be rooted in its own culture. It is a bad idea to rely on a foreign wiki established by a foreign owner with foreign money if you are proud of your own culture. Wikipedia is good. But how many of your people would regularly visit a foreign website? Personalli, I take it as a reason why so many wikis are not growing as quickly as it should be.

What if Mr. 張大砲, an imaginary Chinese businessman, setup the world's first wiki knowledgebase in Beijing? Would people in the U.S. jump on his bandwagon? Then how about en.wikipedia.co.jp? Sometimes it is more than just a domain name. You can register wikipedia anywhere in the world and point it to your own wiki. The point is no one will advertise it if it's not your own business.

To me, this "let's setup a wiki and one day they will come" mindset is bloody U.S.-centered. It's a loser's mindset. Next time when you think about it, don't forget my words.

[edit] See also

Retrieved from "http://meta.wikimedia.org/w/index.php?title=Toytoy%27s_proposed_policy_for_wiki_closure&oldid=3546530"
Personal tools

Variants
Actions
Navigation
Community
Beyond the Web
Print/export
Toolbox