New projects opened

classic Classic list List threaded Threaded
73 messages Options
1234
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Yann Forget-2
Andre Engels wrote:

> On Thu, Aug 20, 2009 at 5:22 AM, Lars Aronsson<[hidden email]> wrote:
>
>> Of these 270 languages of Wikipedia, only 41 have more than 50,000
>> articles and only 69 had more than 1 million page views in July of
>> 2009.  The 69th most used Wikipedia is Swahili. This East African
>> language has 50 million speakers, which is huge, but less than
>> 13,000 Wikipedia articles.  Can poverty and illiteracy alone
>> explain why the Swahili Wikipedia is so far behind?
>
> Poverty, or better said, lack of internet access, is probably the main
> factor. Here in Europe and North America, we are used to having fast
> internet from the home 24/7. In those countries it may well be (I am
> not sure, never having been there) that dial-up speeds paid per minute
> at some internet cafe is the norm. That would considerably lessen
> people's interest in writing the material, and if it is not written,
> people will not read it either.
>
> But another issue could be a lack of expectancy of having material in
> the own language. I have heard this plays a role with the languages
> from India, and it may well have the same, or even stronger so, with
> the African ones: the daily language for speaking is the local
> language, but when one is writing or looking for something on the
> internet, one is more likely to use English (or in other parts of
> Africa, French). It may well be that many Swahili speakers use English
> when they are on the internet - either because that is the language
> they learned reading and writing in (although people for which that is
> true are probably not the generation using internet the most), or
> because they found that they can get so much more information (on the
> internet as a whole) in English than in Swahili, that it well
> outweighs the linguistic disadvantage. They come to the English
> Wikipedia, not the Swahili one, and when they find that here too there
> is much more in English, that's where they stick.

This explains the situation very well.
In the case of languages not using the Latin alphabet, there is one more
obstacle: you need a localized computer, i.e. for reading, at least the
proper fonts are needed, and for writing an adapted keyboard is also
needed. For what I have seen, this is rarely the case in India. Every
computer is sold with an English keyboard only, and the fonts must be
installed by the user himself.

> In the case of Swahili there is yet another factor, namely that
> Swahili itself is rarely a mother tongue and much more often a second
> language. Because of that, the relative size of the disadvantage of
> using English is even smaller.

Right. This is also the case for Hindi, the second or third language for
more than 200 M speakers (native Assamese, Bengali, Bihari, Gujarati,
Kashmiri, Marathi, Oriya or Punjabi speakers and more).

Yann

>> But Swahili is far from the worst.  Swahili has twice as many
>> speakers as the West African language Yoruba (50 vs 25 M, both are
>> huge languages) and twice the number of articles (13 k vs 6.3 k),
>> but the Swahili Wikipedia had 6 times as many page views (1.0 M vs
>> 172 k).  Somebody with knowledge of Africa should study this in
>> more detail.  For the speakers of these languages, in which
>> proportions do they read (newspapers) or listen (to radio
>> broadcasts) to get news and knowledge?  Do they ever use (printed)
>> encyclopedias?
>
> Taking a look at Wikipedia, I see
> http://en.wikipedia.org/wiki/List_of_newspapers_in_Nigeria and
> http://en.wikipedia.org/wiki/List_of_newspapers_in_Kenya. For Nigeria
> about 32 newspapers are given - from their titles, 80% seem to be in
> English. The 3 or 4 mentioned for Kenya are all in English, and
> although the articles mention some of the papers have Swahili sister
> publications, the English language newspapers seem to have by far the
> greatest market share. This I think confirms my hypothesis above, that
> another reason for African languages to do so poorly is that in the
> countries and regions where they are spoken, there is a large
> competition from the languages of the former colonizers - especially
> in the area of written communication.

--
http://www.non-violence.org/ | Site collaboratif sur la non-violence
http://www.forget-me.net/ | Alternatives sur le Net
http://fr.wikisource.org/ | Bibliothèque libre
http://wikilivres.info | Documents libres

_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Marcus Buck-2
In reply to this post by Andre Engels
Andre Engels hett schreven:

> On Thu, Aug 20, 2009 at 5:22 AM, Lars Aronsson<[hidden email]> wrote:
>
>  
>> Of these 270 languages of Wikipedia, only 41 have more than 50,000
>> articles and only 69 had more than 1 million page views in July of
>> 2009.  The 69th most used Wikipedia is Swahili. This East African
>> language has 50 million speakers, which is huge, but less than
>> 13,000 Wikipedia articles.  Can poverty and illiteracy alone
>> explain why the Swahili Wikipedia is so far behind?
>>    
>
> Poverty, or better said, lack of internet access, is probably the main
> factor. Here in Europe and North America, we are used to having fast
> internet from the home 24/7. In those countries it may well be (I am
> not sure, never having been there) that dial-up speeds paid per minute
> at some internet cafe is the norm. That would considerably lessen
> people's interest in writing the material, and if it is not written,
> people will not read it either.
>
> But another issue could be a lack of expectancy of having material in
> the own language.
Another important factor: If your language has no localized version of
Windows, of Office, of Google or of equivalent softwares, this almost
excludes all people not speaking at least one foreign language from
using a computer. If understanding a foreign language is a prerequisite
to using computers, there are no native-onlys - who have the most
interest in native content - to write native content, and there are no
native-onlys to read the native content.
The bilingual people have less interest in creating content. And then in
many societies which have bilingualism between the people's languages
and a non-native official language, there is some amount of elitarism.
Good knowledge of the official language and good education provide you a
certain social status. Educating the native masses could endanger this
social advantage. The more social and general insecurity exists in an
area, the more elitarist are the educated.

And creating content for the benefit of everybody is a leisure time
activity. Poor people rather try to earn money instead of writing
content for free. And rich people in under-developed countries ususally
won't contribute too, cause to become rich in a poor country, you must
be rather callous and not be too "social".

I guess, it would be possible to greatly improve the number of
contributions to several of our Wikipedias, if we established some kind
of reward system, in which contributors get paid for their work. E.g.
Burundi has a per capita income of less than 150 $ a year. If it would
be possible to make some dollars a day by writing Wikipedia articles,
you could easily gain some full-time editors with just a few thousand
dollars. Rundi Wikipedia article count would surely skyrocket if the
Foundation would provide let's say 100,000 $ for a project like that (of
course a native Rundi project manager would be needed to ensure the
quality of the contributions). Wouldn't it be great if the Wikimedia
Foundation could go to the Bill & Melinda Gates Foundation and say "Hey,
with 100,000 $ you can help us to create a 100,000 entry encyclopedia
for 10 million speakers of Chichewa, where before there was exactly _no_
encyclopedia-like content in that language!"?

Marcus Buck
User:Slomox

_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Lars Aronsson
In reply to this post by Marcus Buck-2
Marcus Buck wrote:

> What I want to say: please everybody get away from calling
> projects "failure", "worse", "weak" or whatever. It's all
> subjective. And it's entirely meaningless,

I disagree, it's neither subjective nor meaningless.  Wikipedia
has a mission to disseminate free knowledge.  It's an important
mission and a powerful project.  The general public and mainstream
media have a geniune interest in knowing how we are doing.  The 3
millionth article in the English Wikipedia was a global news item,
as was the PARC research that showed Wikipedia might not be
growing so fast anymore.  The problem is that both reports are
based on article counts, as if all articles were equal, and they
aren't.

For Wikipedia's future growth, we learned early on to use a
wishlist, a list of red links to not yet existing articles.  But
the items on that list are not equally important.  And the
improvement of some existing articles can be more important than
the addition of any new article.  We need better tools to help us
understand which improvements are needed.  And we need to know how
much we improved Wikipedia, even if no new articles were created.
This is meaningful.

We might have to go out to the people in Nigeria (or New York) and
ask them what knowledge they need, and what tools are best suited.
Perhaps it's the English Wikipedia that is best for them.  Then we
might conclude that the Yoruba Wikipedia was a failed attempt,
that never even reached 10,000 articles, and instead of 270
languages we should only have 269 (or 41) languages of Wikipedia.
Or on the other hand, we might discover some basic mistake that we
did with the Yoruba Wikipedia, and once we fix that mistake its
size and usefulness will start to grow faster.

> If 988 people had no interest in looking up Michael Jackson,
> then that's okay. We still served the 12 who had.

Sure, but it's not likely that the interest for Michael Jackson is
far lower in Denmark than in neighboring Sweden and Germany.  I
still think the Danish Wikipedia has some trivial flaw that can be
fixed.  I just don't know what it is.



--
  Lars Aronsson ([hidden email])
  Aronsson Datateknik - http://aronsson.se

_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Chad
On Thu, Aug 20, 2009 at 2:35 PM, Lars Aronsson<[hidden email]> wrote:

> Marcus Buck wrote:
>
>> What I want to say: please everybody get away from calling
>> projects "failure", "worse", "weak" or whatever. It's all
>> subjective. And it's entirely meaningless,
>
> I disagree, it's neither subjective nor meaningless.  Wikipedia
> has a mission to disseminate free knowledge.  It's an important
> mission and a powerful project.  The general public and mainstream
> media have a geniune interest in knowing how we are doing.  The 3
> millionth article in the English Wikipedia was a global news item,
> as was the PARC research that showed Wikipedia might not be
> growing so fast anymore.  The problem is that both reports are
> based on article counts, as if all articles were equal, and they
> aren't.
>
> For Wikipedia's future growth, we learned early on to use a
> wishlist, a list of red links to not yet existing articles.  But
> the items on that list are not equally important.  And the
> improvement of some existing articles can be more important than
> the addition of any new article.  We need better tools to help us
> understand which improvements are needed.  And we need to know how
> much we improved Wikipedia, even if no new articles were created.
> This is meaningful.
>
> We might have to go out to the people in Nigeria (or New York) and
> ask them what knowledge they need, and what tools are best suited.
> Perhaps it's the English Wikipedia that is best for them.  Then we
> might conclude that the Yoruba Wikipedia was a failed attempt,
> that never even reached 10,000 articles, and instead of 270
> languages we should only have 269 (or 41) languages of Wikipedia.
> Or on the other hand, we might discover some basic mistake that we
> did with the Yoruba Wikipedia, and once we fix that mistake its
> size and usefulness will start to grow faster.
>
>> If 988 people had no interest in looking up Michael Jackson,
>> then that's okay. We still served the 12 who had.
>
> Sure, but it's not likely that the interest for Michael Jackson is
> far lower in Denmark than in neighboring Sweden and Germany.  I
> still think the Danish Wikipedia has some trivial flaw that can be
> fixed.  I just don't know what it is.
>
>
>
> --
>  Lars Aronsson ([hidden email])
>  Aronsson Datateknik - http://aronsson.se
>
> _______________________________________________
> foundation-l mailing list
> [hidden email]
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>

I agree wholeheartedly. We need to get away from this idea that "more
projects in more languages is better." It's not. It's lead to the issue we
see now: dead projects lying around until somebody bothers to clean it
up or close it.

We tout the "Wikipedia in 270 languages" statistic quite often, and it's
something that is seen as an accomplishment. I would rather see how
many Wikipedias we have that are successes--measured in terms of
growth and a supportive community (of both readers and writers).

What good is a Revised-Lower-Eastern-Phoenician Wikipedia if
nobody uses it?

-Chad

_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Gerard Meijssen-3
In reply to this post by Lars Aronsson
Hoi,
Lars I completely agree that the failure of a Wikipedia IS meaningful. But
it is only meaningful if we are interested in learning what causes these
failures, what we can do to remedy these situations and when we are willing
to act upon our findings.

I mentioned earlier that the Danish localisation is less developed compared
with the Norwegian, Nynorsk and Swedish Wikipedias. What should the
consequence be for the WMF when this is the one significant factor that is
different ? Would it be reasonable to remedy such issues because it will
lower the inflection point where projects take off autonomously? When we do
not invest in language support and language technonlogy would it make sense
to enrich the content about a country in the English Wikipedia?

My question is: is it reasonable to dissect our "failed" projects and not
act upon our findings? Is it reasonable to leave finding solutions to
volunteers when there are none at present? Is it reasonable to expect people
to volunteer for complex tasks when they typically exist only in out more
mature communitees?
Thanks,
      GerardM


2009/8/20 Lars Aronsson <[hidden email]>

> Marcus Buck wrote:
>
> > What I want to say: please everybody get away from calling
> > projects "failure", "worse", "weak" or whatever. It's all
> > subjective. And it's entirely meaningless,
>
> I disagree, it's neither subjective nor meaningless.  Wikipedia
> has a mission to disseminate free knowledge.  It's an important
> mission and a powerful project.  The general public and mainstream
> media have a geniune interest in knowing how we are doing.  The 3
> millionth article in the English Wikipedia was a global news item,
> as was the PARC research that showed Wikipedia might not be
> growing so fast anymore.  The problem is that both reports are
> based on article counts, as if all articles were equal, and they
> aren't.
>
> For Wikipedia's future growth, we learned early on to use a
> wishlist, a list of red links to not yet existing articles.  But
> the items on that list are not equally important.  And the
> improvement of some existing articles can be more important than
> the addition of any new article.  We need better tools to help us
> understand which improvements are needed.  And we need to know how
> much we improved Wikipedia, even if no new articles were created.
> This is meaningful.
>
> We might have to go out to the people in Nigeria (or New York) and
> ask them what knowledge they need, and what tools are best suited.
> Perhaps it's the English Wikipedia that is best for them.  Then we
> might conclude that the Yoruba Wikipedia was a failed attempt,
> that never even reached 10,000 articles, and instead of 270
> languages we should only have 269 (or 41) languages of Wikipedia.
> Or on the other hand, we might discover some basic mistake that we
> did with the Yoruba Wikipedia, and once we fix that mistake its
> size and usefulness will start to grow faster.
>
> > If 988 people had no interest in looking up Michael Jackson,
> > then that's okay. We still served the 12 who had.
>
> Sure, but it's not likely that the interest for Michael Jackson is
> far lower in Denmark than in neighboring Sweden and Germany.  I
> still think the Danish Wikipedia has some trivial flaw that can be
> fixed.  I just don't know what it is.
>
>
>
> --
>  Lars Aronsson ([hidden email])
>  Aronsson Datateknik - http://aronsson.se
>
> _______________________________________________
> foundation-l mailing list
> [hidden email]
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>
_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Marcus Buck-2
In reply to this post by Chad
Chad hett schreven:
> I agree wholeheartedly. We need to get away from this idea that "more
> projects in more languages is better." It's not. It's lead to the issue we
> see now: dead projects lying around until somebody bothers to clean it
> up or close it.
>  
More projects in more languages _is_ better. They just need to be cared
about. At the moment Wikimedia just sets up the wikis and waits for
articles flowing in. The amount of work invested by the Foundation after
the initial setup of the wiki is exactly zero. Languages of societies
with much leisure time easily gained enough momentum by themselves. But
other language versions from societies with educational and social
hardships don't gain momentum by themselves. They don't reach the
critical mass to sustain active wiki work. Therefore they need support
by Wikimedia. Support like hiring somebody who is fluent in several
African languages, sending him to Africa and let him promote Wikipedia
participation at universities for example. Enthuse a handful of people
and let them spread interest in Wikipedia collaboration. Perhaps soon
you'll have a stable community. Even if my ideas may be naive, I don't
know, at least the foundation could consider and explore projects like that.

I don't think that there are generally too few people interested in
those languages. It's just hard to make the start. It's immensely
frustrating to work on a wiki all alone, writing article for article,
and after a year, you maybe have 100 or 200 articles and your Wikipedia
is still just a little heap of disjunct articles with hardly any blue
links and you realize that it will take years (or decades) until you
have written enough articles to establish a resource, that is
interconnected through blue links and covering all basic concepts. Most
users won't stay for more than some months under circumstances like
that. They realize, that they can't achieve the goal all alone and give up.
Therefore these projects need starting help. We should aid them until a
little community is established and the basic articles are written. Once
Rundi Wikipedia is at 100,000 articles, I'm sure, they won't need help
anymore, cause at that moment it will be a useful resource, actually
used by the people, and it will be fun for Rundi speakers to be part of
the community and to add even more articles. Unlike the 38-article wiki
we have now at which contributing is _not_ fun.

Marcus Buck
User:Slomox

_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Jussi-Ville Heiskanen
Marcus Buck wrote:
>
> I don't think that there are generally too few people interested in
> those languages. It's just hard to make the start. It's immensely
> frustrating to work on a wiki all alone, writing article for article,
> and after a year, you maybe have 100 or 200 articles and your Wikipedia
> is still just a little heap of disjunct articles with hardly any blue
> links and you realize that it will take years (or decades) until you
> have written enough articles to establish a resource, that is
> interconnected through blue links and covering all basic concepts.
I think in this situation a useful page that Danny Wool
and a few of his friends thought up a few years ago,
and has been improved upon subsequently by diverse
hands, might help.

I am of course thinking about the list of 1000 articles
each wikipedia should have. Just completing a
significant part of that list is an accomplishment for
a tiny pool of editors, but is within reach, and
can serve as a useful incentive.

BTW, I understand there is some work being done
currently to define a tinier subset of that list, which
could be even better for projects with fewer contributors,
which would define what the really really really core
encyclopaedia articles are.


Yours,

Jussi-Ville Heiskanen



_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Kåre Thor Olsen
In reply to this post by Gerard Meijssen-3
On Thu, 20 Aug 2009 09:14:14 +0200
Gerard Meijssen <[hidden email]> wrote:

> One of the reasons why Danish has been sluggish may be that the
> localisation of Danish was not optimal; in Februari 83.66% of the
> MediaWiki messages and 14.11% of the WMF used extensions were
> localised. This has improved to 100.00%  and 59.30% respectively ...
> compare this with Norwegian 100.00% 96.92% Nynorsk 100.00% 84.81%
> and Swedish 100.00% 99.33%..

Gerard, that reasoning is rather far-fetched.  As you know, MediaWiki contains many messages which regular users will never be exposed to.  Even with admin (and more) access, I rarely come across anything needing translation.  We would surely have complaints galore if missing interface translations turned people away from the Danish language Wikipedia, and I've never seen a complaint about that.

I have a hunch why there is/was a comparatively low number of Danish MediaWiki message translations, but am not going to open old wounds.  Old wounds aside, another reason could be that people focus on other aspects; since most message translators to Danish are from the Danish language Wikipedia, they may not care much about messages/extensions not relevant in Wikipedia, as they feel Wikipedia is their project, not MediaWiki or Translatewiki.

What I think is the primary reason for the Danish Wikipedia being much smaller than the "neighbouring" languages is that Danes generally are internationally minded and pride themselves on being good at English - people may simply prefer to use/edit Wikipedia in that language (even I did that when first attracted to Wikipedia).

--
Regards, Kaare

_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Jussi-Ville Heiskanen
In reply to this post by Gerard Meijssen-3
Gerard Meijssen wrote:
> Hoi,
> Lars I completely agree that the failure of a Wikipedia IS meaningful. But
> it is only meaningful if we are interested in learning what causes these
> failures, what we can do to remedy these situations and when we are willing
> to act upon our findings.
>  

Interesting data points outside of Wikipedia entirely, are:

http://en.wikipedia.org/wiki/Den_Store_Danske_Encyklop%C3%A6di

and;

http://en.wikipedia.org/wiki/Nationalencyklopedin

with a further interesting contrast offered by:

http://en.wikipedia.org/wiki/Susning.nu

The Danish commercially published (but with direct funding
subsidy from the government) Ensyklopedia did not excite
the Danish public. It is very hard to credit this lack of interest
20 years before wikipedia came into existence to the lack
of adequate MediaWiki localizations.

In contrast the Swedish Nationalencylopedin was gobbled
up by the public at large, and even was able to pay back
the government in full all the money loaned for the production
as  a guarantee against losses.

There must be some difference in national character or
something else to explain this. But I certainly don't
claim to have the answer, except to state that since this all
happened before there was a Wikipedia, the "problem"
must also predate Wikipedia.

A further wrinkle in the current situation is that currently
Nationalencyklopedin (the Swedish one) is behind a paywall
with the nearest competitor to Wikipedia in the "free as in
beer" stakes being susning.nu, run by Lars Aronsson; while
the Danish "Den Store Danske Encyklopaedi" has been
liberated and can be freely accessed. Thus, quite unlike
the current Swedish situation, Wikipedia and the formerly
government subsidized but commercially published (and
very professionally edited) encyclopaedia are nearly level
pegging on the internet, in terms of amount of content
and ease of access, as far as the Danish encyclopaedia
reading public are concerned. This can't but have an
effect.


Yours,

Jussi-Ville Heiskanen









_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Lars Aronsson
In reply to this post by Kåre Thor Olsen
Kaare Olsen wrote:

> What I think is the primary reason for the Danish Wikipedia
> being much smaller than the "neighbouring" languages is that
> Danes generally are internationally minded and pride themselves
> on being good at English - people may simply prefer to use/edit
> Wikipedia in that language (even I did that when first attracted
> to Wikipedia).

I find it hard to believe that this would be a major difference
between Denmark and Sweden. But it would be really interesting if
we could somehow trace the use of the English Wikipedia to users
of various mother tongues (for Northern Europe, country or IP
address range might be a good enough approximation for mother
tongue).  Perhaps Swedish users stay on the Swedish Wikipedia to
read about sports, but go to the English to read about music.

For each IP address range, we could (well, Domas could) analyze
which language of Wikipedia those users primarily go to.  If users
from 130.236.xxx.yyy mostly visit the English and Swedish
Wikipedia, we can assume that it constitutes a Swedish-speaking
community.  If no conclusive pattern is shown on the /16 (class B)
range, each /24 (class C) net can be analyzed individually.


--
  Lars Aronsson ([hidden email])
  Aronsson Datateknik - http://aronsson.se

_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Gerard Meijssen-3
In reply to this post by Kåre Thor Olsen
Hoi,
Given that on Februari first 96.07% of the most used messages were
localised, it is clear that some of the most used messages were not even
localised. Consequently your puh puh reaction that only the rare messages
are affected is not correct.
Thanks,
       GerardM

2009/8/20 Kaare Olsen <[hidden email]>

> On Thu, 20 Aug 2009 09:14:14 +0200
> Gerard Meijssen <[hidden email]> wrote:
>
> > One of the reasons why Danish has been sluggish may be that the
> > localisation of Danish was not optimal; in Februari 83.66% of the
> > MediaWiki messages and 14.11% of the WMF used extensions were
> > localised. This has improved to 100.00%  and 59.30% respectively ...
> > compare this with Norwegian 100.00% 96.92% Nynorsk 100.00% 84.81%
> > and Swedish 100.00% 99.33%..
>
> Gerard, that reasoning is rather far-fetched.  As you know, MediaWiki
> contains many messages which regular users will never be exposed to.  Even
> with admin (and more) access, I rarely come across anything needing
> translation.  We would surely have complaints galore if missing interface
> translations turned people away from the Danish language Wikipedia, and I've
> never seen a complaint about that.
>
> I have a hunch why there is/was a comparatively low number of Danish
> MediaWiki message translations, but am not going to open old wounds.  Old
> wounds aside, another reason could be that people focus on other aspects;
> since most message translators to Danish are from the Danish language
> Wikipedia, they may not care much about messages/extensions not relevant in
> Wikipedia, as they feel Wikipedia is their project, not MediaWiki or
> Translatewiki.
>
> What I think is the primary reason for the Danish Wikipedia being much
> smaller than the "neighbouring" languages is that Danes generally are
> internationally minded and pride themselves on being good at English -
> people may simply prefer to use/edit Wikipedia in that language (even I did
> that when first attracted to Wikipedia).
>
> --
> Regards, Kaare
>
> _______________________________________________
> foundation-l mailing list
> [hidden email]
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>
_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Jussi-Ville Heiskanen
Gerard Meijssen wrote:
> Hoi,
> Given that on Februari first 96.07% of the most used messages were
> localised, it is clear that some of the most used messages were not even
> localised. Consequently your puh puh reaction that only the rare messages
> are affected is not correct.
>  
Not all of the messages which are not included in "most used"
are rare, and not all messages that are in the "most used"
messages pile are ones that beginning editors are bound to
encounter.

So I do think Kaare is spot on to suggest it is an extraordinary
claim that lack of localizations is driving away Danish editors,
and as such requires extraordinary proof! Have there been
beginning Danish wiki-editors complaining about the poor
localization level?


Yours,

Jussi-Ville Heiskanen


_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Gregory Maxwell
In reply to this post by Lars Aronsson
On Thu, Aug 20, 2009 at 9:22 PM, Lars Aronsson<[hidden email]> wrote:

> Kaare Olsen wrote:
>
>> What I think is the primary reason for the Danish Wikipedia
>> being much smaller than the "neighbouring" languages is that
>> Danes generally are internationally minded and pride themselves
>> on being good at English - people may simply prefer to use/edit
>> Wikipedia in that language (even I did that when first attracted
>> to Wikipedia).
>
> I find it hard to believe that this would be a major difference
> between Denmark and Sweden. But it would be really interesting if
> we could somehow trace the use of the English Wikipedia to users
> of various mother tongues (for Northern Europe, country or IP
> address range might be a good enough approximation for mother
> tongue).  Perhaps Swedish users stay on the Swedish Wikipedia to
> read about sports, but go to the English to read about music.
>
> For each IP address range, we could (well, Domas could) analyze
> which language of Wikipedia those users primarily go to.  If users
> from 130.236.xxx.yyy mostly visit the English and Swedish
> Wikipedia, we can assume that it constitutes a Swedish-speaking
> community.  If no conclusive pattern is shown on the /16 (class B)
> range, each /24 (class C) net can be analyzed individually.

I published a very simple GEO vs Project readership report a couple of
years back. I could dig up the data, but it's old now.  It's not
terribly hard to run, and the old script should still work.


It was generally the case that for much of the world English Wikipedia
was accessed Wikipedia by readers with roughly comparable frequency to
the 'expected' language, and in some cases far more so… though there
were some significant exceptions: For example the Italians stuck to
itwiki and the Japanese stuck to jawiki.  Much of Europe was more
mixed.


There is also this old data:
http://meta.wikimedia.org/wiki/Edits_by_project_and_country_of_origin


How many messages need to be translated to make mediawiki basically
usable?  My own belief was that you only needed a few dozens to make
the software basically usable, at least enough to bootstrap usage.

_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Svip
In reply to this post by Jussi-Ville Heiskanen
2009/8/21 Jussi-Ville Heiskanen <[hidden email]>:

> Gerard Meijssen wrote:
>> Hoi,
>> Given that on Februari first 96.07% of the most used messages were
>> localised, it is clear that some of the most used messages were not even
>> localised. Consequently your puh puh reaction that only the rare messages
>> are affected is not correct.
>>
> Not all of the messages which are not included in "most used"
> are rare, and not all messages that are in the "most used"
> messages pile are ones that beginning editors are bound to
> encounter.
>
> So I do think Kaare is spot on to suggest it is an extraordinary
> claim that lack of localizations is driving away Danish editors,
> and as such requires extraordinary proof! Have there been
> beginning Danish wiki-editors complaining about the poor
> localization level?

I am would not claim myself to be a "beginning Danish wiki-editor",
but I am certainly complaining about the localisation of the Danish
MediaWiki.  If it wasn't obvious spelling mistakes, it was often odd
direct translations from English, without much consideration of
context.

If I were a beginning Danish wiki-editor, I would see this as
unprofessional - something which I already do in my current position -
and as a result, probably leave, or contribute less than I had
intended.

It's unfortunate to see less and less Danes taking Danish serious
enough, with English's slow integration into the Danish language.
Many things can be translated, but not necessarily everything should.

It's a balance.

/Svip

Furthermore, I think Carthage must be destroyed.

> Yours,
>
> Jussi-Ville Heiskanen
>
>
> _______________________________________________
> foundation-l mailing list
> [hidden email]
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>

_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Gerard Meijssen-3
In reply to this post by Gregory Maxwell
Hoi,
We are not talking about bootstrap usage. The Danish Wikipedia is obviously
way past that point. We are talking about usability and the acceptance of
MediaWiki as a proper platform for a language. Basically usage is not the
same as being accepted as an environment that provides proper functionality
for a language.  Ask people who deal with usability the importance of
internationalisation and localisation and you will learn that the notion of
"bootstrap usage" is exactly that.
Thanks,
      GerardM

2009/8/21 Gregory Maxwell <[hidden email]>

>
> How many messages need to be translated to make mediawiki basically
> usable?  My own belief was that you only needed a few dozens to make
> the software basically usable, at least enough to bootstrap usage.
>
>
_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Jussi-Ville Heiskanen
In reply to this post by Svip
Just to clarify, are you saying that in your view, too
few messages are translated to Danish, or are you
saying that too many messages are translated to the
Danish language?


Yours,

Jussi-Ville Heiskanen


_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Svip
2009/8/21 Jussi-Ville Heiskanen <[hidden email]>:
> Just to clarify, are you saying that in your view, too
> few messages are translated to Danish, or are you
> saying that too many messages are translated to the
> Danish language?

Unfortunately; neither.  Messages that shouldn't be translated to
Danish, have unfortunately been, while messages that should have been
translated to Danish, haven't.

But that's without mentioning the horrible state of the localisation
in general:  Wrong context translations, just wrong translations and
many spelling errors.

The Danish Wikipedia itself is in a pretty bad state to.  Too many
articles on it are close to laughable, and you can often find better
articles on Danish subjects on the English Wikipedia than the Danish
one.

I my mind, the combination of the poor localisation and the bad state
the Danish Wikipedia is in, scares many Danish editors off.

And it should be mentioned, that many Danes have a pretty good
conception of English, and the interest in Danish is unfortunately
lessen as well.

I am trying to do my part, by translating the pages on Meta wiki and
the Strategy wiki to Danish.

/Svip

_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Gregory Maxwell
On Fri, Aug 21, 2009 at 1:36 AM, Svip<[hidden email]> wrote:
> But that's without mentioning the horrible state of the localisation
> in general:  Wrong context translations, just wrong translations and
> many spelling errors.

Contextual errors I can understand, figuring out all the right
contexts for a message can be tricky.

How were the spelling errors and wrong translations introduced?

_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Gerard Meijssen-3
Hoi,
At translatewiki.net many of the messages include information about the
context. The coverage of this information has been improving steadily. This
information is not available when messages are localised on the local wiki.

So there are two places where localisations can originate; local and
translatewiki.net. There is still a problem with localising at translatewiki
in that localisations do not become available in a timely fashion. The
LocalisationUpdate extension that is currently being assessed by Brion will
fix that problem. It will reduce the amount of time before localisations
become available to maximally two days.

Once this is done, it is important to remove as many local localisations as
possible because they override the messages provided from translatewiki.net.
The only local localisations should be the ones that provide local
information like policies.
Thanks,
      GerardM


2009/8/21 Gregory Maxwell <[hidden email]>

> On Fri, Aug 21, 2009 at 1:36 AM, Svip<[hidden email]> wrote:
> > But that's without mentioning the horrible state of the localisation
> > in general:  Wrong context translations, just wrong translations and
> > many spelling errors.
>
> Contextual errors I can understand, figuring out all the right
> contexts for a message can be tricky.
>
> How were the spelling errors and wrong translations introduced?
>
> _______________________________________________
> foundation-l mailing list
> [hidden email]
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>
_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Reply | Threaded
Open this post in threaded view
|

Re: New projects opened

Jussi-Ville Heiskanen
In reply to this post by Svip
Svip wrote:

> 2009/8/21 Jussi-Ville Heiskanen <[hidden email]>:
>  
>> Just to clarify, are you saying that in your view, too
>> few messages are translated to Danish, or are you
>> saying that too many messages are translated to the
>> Danish language?
>>    
>
> Unfortunately; neither.  Messages that shouldn't be translated to
> Danish, have unfortunately been, while messages that should have been
> translated to Danish, haven't.
>
> But that's without mentioning the horrible state of the localisation
> in general:  Wrong context translations, just wrong translations and
> many spelling errors.
>
> The Danish Wikipedia itself is in a pretty bad state to.  Too many
> articles on it are close to laughable, and you can often find better
> articles on Danish subjects on the English Wikipedia than the Danish
> one.
>
>  
I find your comments very significant. This suggests to
me quite strongly that it may in fact be that emphasizing
the *quantity* and *completeness* of localizations, over
the _quality_ of the same, can be a double edged sword.

The thought also is hard to escape, that there is a definite
"late mover disadvantage" to the creation of a wikipedia.
Anyone who has experience of the growth process of a
LOTE-wikipedia (to borrow a four-letter acronym from a
different free content context -- 'Language Other Than
English'), will inevitably recognize the feeling of seeing
something better written on the en-wikipedia than at
their own language version, and as the en-wikipedia
matures, this disadvantage is not going away, or losing
effect, sadly.

I could write more on these subjects, but in the interest
of briefness, choose not to, but will instead mull my
thoughts over and reflect...

> I my mind, the combination of the poor localisation and the bad state
> the Danish Wikipedia is in, scares many Danish editors off.
>
> And it should be mentioned, that many Danes have a pretty good
> conception of English, and the interest in Danish is unfortunately
> lessen as well.
>
> I am trying to do my part, by translating the pages on Meta wiki and
> the Strategy wiki to Danish.
>
>  

Yours,

Jussi-Ville Heiskanen



_______________________________________________
foundation-l mailing list
[hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
1234