[Wikipedia-l] RFC: Principles of mass content adding on small Wikipedias

Sat Mar 4 18:29:42 UTC 2006

On 3/4/06, Andreas Vilén <andreas.vilen at gmail.com> wrote:
> Before rushing into this stub maniac, please think! Is it really a
> good idea to massadd substubs? Is the 'pedia of higher quality because
> it has more substubs about 20000 obscure regions than without them? If
> this is used, I believe it should aim towards making relevant articles
> that can be easily expanded by people speaking the language. We've
> seen what mass content adders have done to itwiki and plwiki, they
> simply rush to get many articles and completely forget about
> quality... Remember http://meta.wikimedia.org/wiki/Substub_disease !

There is no rush. I don't expect that we would add some relevant data
in the next couple of months. For example, if we want to make articles
about all countries in the world, we need a lot of work on
localization of templates, categories and a couple of sentences.

BUT, when we are talking about stubs on, for example, Swahili
Wikipedia, for example, article about Mongolian language which would
contain that: (1) How "Mongolian language" is written in Mongolian;
(2) it is spoken in [[China]], [[Kyrgyzstan]], [[Mongolia]],
[[Russia]]; (3) it is spoken by 5,7 million of humans; (4) genetic
classification of Mongolian language is: Altaic (disputed) -> Mongolic
-> Eastern -> Oirat-Khalkha -> Khalkha-Buryat -> Mongolian; (5)
language is official in Mongolia, China (Inner Mongolia), Russia
(Buryatia); (6) the language is not regulated; (7) ISO and Ethnologue
codes are ...; (8) consonants and vowels are... etc. -- I think that
it can be treated as a stub on English Wikipedia, but it is very
informative article for a person who knows Swahili, but doesn't know
English. And it is completely possible to do with bots and
localization.

Also, a lot of languages which don't have more then 50 millions of
speakers (I think that there are maybe 20 languages with more then 50
millions of speakers!) -- don't have encyclopedias like Britannica and
similar. For example, the biggest encyclopedia in Serbian language has
almost 100.000 articles. BUT, I saw that encyclopedia maybe two times
in my life (I saw Britannica a lot often even I am living in
Belgrade!). "Ordinary encyclopedia", which can be found in almost
every house, has between 30.000-50.000 very small articles. 90% of
such articles would be stubs on Serbian Wikipedia.

In other words: Maybe people who are native speakers of English,
German, Russian, Chinese, Spanish, French and Portuguese are able to
have "higher standards" about what article is stub and what is not;
they may remove article with one sentence about some Arabic dynasty,
but our encyclopedias are full of such articles AND this sentence is
very relevant encyclopedic information. And we are not talking about
one-two sentence articles, but about articles which would not be stubs
even for English Wikipedia. (I can make one page article about
Mongolian based on information from the article on English Wikipedia.)