[Foundation-l] Promoting non-en Wikipedias (was In defence of Google)

Gerard Meijssen gerard.meijssen at gmail.com
Sun Jan 21 21:29:58 UTC 2007

Birgitte SB schreef:
> I have wondered that it would be better to encourage
> translations to purposely "stubify" articles rather
> than focus on full Featured Article translations. 
> This would give the less developed langages something
> where they see a need to "edit", eliminate most
> cultural bias without a detailed study, and make the
> limited translation reasources go farther to creating
> a interlinked web of articles.  This idea was brought
> up in BOF at Wikimania but I do not know if anything
> was ever done in this line of thinking.
> BirgitteSB
There is no one size fits all answer to this question. It depends very 
much on what you want to achieve. When you have translators, you can not 
"stubify" because that is very much against their ethos. It also 
requires them to concentrate on the technical aspects of wikipedia ie 
wikification and categories. This means that you do not use translators 
and or that you have to train them considerably.

Creating much content in stubs is something that is done on many 
projects; they are all the towns of country xxx. This is of particular 
interest when they are also the towns of the country where this language 
is spoken. There are as far as I know no such lists with the information 
for countries like Nigeria, Kazakhstan or Ecuador. The same is true for 
the years information; it is very much biased to its source. There are 
for instance few African or South American historical figures in the 
English Wikipedia.. Translating this as the starting information is 
problematic as a consequence because of the bias that you import. For a 
language like Neapolitan, the Italian years information was used ...

The advantage of writing background information to the news is that you 
work on content that is of current relevance and certainly when you base 
it on what can be found in newspapers of that language, you may even 
find articles like "Big Brother" in Indian wikipedias. :)

When the aim is to make investments count, it is important to create 
relevance. Sex, sport and news is what people read. When the aim is to 
get more people to contribute to a Wikipedia, it is eye balls that you 
want. I am completely in favour of working on wikification because this 
can be one way of enticing people to try and edit. I think it should be 
a healthy mix of these activities.

One problem in several projects is that people control how people can 
and cannot write. This leads often to political fights that prevent a 
healthy community from happening. One bad example is the Belarus 
wikipedia where the orthography insisted on is not the one used by 
people who live in Belarus. This is in my opinion contrary to what we 
hope to achieve with our projects. Wikipedia is not to create new 
languages nor is it to force particular orthographies.

My conclusion is therefore, for every language, for every project there 
are different things to consider. There is no silver bullet or a one 
size fits all solution. It pays off when there is sufficient time to 
prepare for a content creation project.



More information about the foundation-l mailing list