Re: [Wikimedia-l] Africa / Gender gaps (was Re: Gender gap on "classical" encyclopedias)

2016-04-21 Thread Anders Wennersten 


Den 2016-04-21 kl. 08:21, skrev Gerard Meijssen:

Hoi,

Anders, have you looked into the ArticlePlaceholder? Could it serve you
well instead of adding articles using the bot?
Thanks,
   GerardM
I do not know, as  I am not directly involved. What I do know is that 
Sverker, Lsjbot creator is one of the most clever persons I ever met, 
and that he knows well of Wikidata and that all expertise on Wikidata on 
svwp has been involved in the design


I would suggest you ask Sverker directly (and offlist) I think you 
clever guys on Wikidata can learn a lot from his insights


Anders







On 21 April 2016 at 07:48, Anders Wennersten 
wrote:


Our traditional way of creating article is based on the interests of the
contributors. This produces skewed total result, and this becomes more
evident on a smaller version like svwp, then on bigger. We have long come
to the conclusion that we will never be able to fill categories like towns
in Mali and basketballplayer in Brazil, where we have had something like
less then 10% of entries then the same categories on enwp (or frwp)

Wikidata can be of help evening out, but on svwp we have (also since long)
said we must work and have a more systematic and deliberate approach to
fill out "empty spaces"

We therefore love Lsjbot which now generates several million good and
comprehensive articles on geographical entities all over the world from the
most complete database that exists, and where areas like Africa is getting
exactly the same attention like a Nordic country. It is completely
unrealistic to think that the few contributors on svwp would ever create
the now existing  25 article on entities on Canada or 16500 entries in
Antartica. But the bias in the source means Djibuti only gets 4000 and
Camerun 9000 but it is none the less a huge improvement.

For articles on woman/men project are being run by wmse and it now exist a
group of dedicated contributors generating articles on women. I am all
fascinated of sources being used, specially to get entries of women from
middle of 19-th century. All famous ballet dancers in Copenhagen in 1850.
All women who had local fame in Finland around 1860, including local
healers etc. They have now created many thousands articles and getting the
rate of articles up to 20% of total (25% of the number for men)

So I believe skewness is becoming an important issue and that we need to
adress oit even if it means to let go the "holy rule" only manual created
article are "real" articles, it is the need of our readers who should have
priority.

Anders

[1] list of article generated by country this far:
https://sv.wikipedia.org/wiki/Kategori:Robotskapade_geografiartiklar
[2] latest article generated just now a river in Fiji
https://sv.wikipedia.org/wiki/Mbau_Creek_%28vattendrag,_lat_-16,50,_long_179,08%29


Den 2016-04-20 kl. 23:30, skrev Florence Devouard:


Hello

Sorry for highjacking your thread, but reading your message, I wanted to
share with you a small page I made a few days ago, to quantify the double
gap Gender/Africa.

http://www.wikiloveswomen.org/about-the-project/mind-the-gap/

If anyone has additional links or studies that could be useful to further
illustrate that double gap... I am interested.

Also, if anyone is interested in further exploring this data-wise, please
raise your hand ;)

Florence


Le 20/04/16 09:39, alexhin...@gmail.com a écrit :


Hi, as some of you may know, the Wikipedia gender indicator [1] tells us
how many articles are biographies about women x language/country/culture.

In order to compare these numbers...Does anyone knows if there is an
existing comparison with gender balance in classical encyclopedias?
(Britannica, Larousse...) or, if not, could someone prepare a WD query
about it?

I think it could be a good argument for us to use: e.g "at cawiki 12% of
bios are about women, compared to 5% in GEC, Our most famous encyclopedia".

We could compare it also for temathic encyclopedias or other databases
existing in projects like Mix and match.

Can someone help? thanks in advance


[1]http://wigi.wmflabs.org/


Àlex Hinojo
User:Kippelboy
Amical Wikimedia Programme manager
___
Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,





___
Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,




___
Wikimedia-l mailing list, guidelines at:

Re: [Wikimedia-l] Africa / Gender gaps (was Re: Gender gap on "classical" encyclopedias)

2016-04-21 Thread Gerard Meijssen 
Hoi,

Anders, have you looked into the ArticlePlaceholder? Could it serve you
well instead of adding articles using the bot?
Thanks,
  GerardM

On 21 April 2016 at 07:48, Anders Wennersten 
wrote:

> Our traditional way of creating article is based on the interests of the
> contributors. This produces skewed total result, and this becomes more
> evident on a smaller version like svwp, then on bigger. We have long come
> to the conclusion that we will never be able to fill categories like towns
> in Mali and basketballplayer in Brazil, where we have had something like
> less then 10% of entries then the same categories on enwp (or frwp)
>
> Wikidata can be of help evening out, but on svwp we have (also since long)
> said we must work and have a more systematic and deliberate approach to
> fill out "empty spaces"
>
> We therefore love Lsjbot which now generates several million good and
> comprehensive articles on geographical entities all over the world from the
> most complete database that exists, and where areas like Africa is getting
> exactly the same attention like a Nordic country. It is completely
> unrealistic to think that the few contributors on svwp would ever create
> the now existing  25 article on entities on Canada or 16500 entries in
> Antartica. But the bias in the source means Djibuti only gets 4000 and
> Camerun 9000 but it is none the less a huge improvement.
>
> For articles on woman/men project are being run by wmse and it now exist a
> group of dedicated contributors generating articles on women. I am all
> fascinated of sources being used, specially to get entries of women from
> middle of 19-th century. All famous ballet dancers in Copenhagen in 1850.
> All women who had local fame in Finland around 1860, including local
> healers etc. They have now created many thousands articles and getting the
> rate of articles up to 20% of total (25% of the number for men)
>
> So I believe skewness is becoming an important issue and that we need to
> adress oit even if it means to let go the "holy rule" only manual created
> article are "real" articles, it is the need of our readers who should have
> priority.
>
> Anders
>
> [1] list of article generated by country this far:
> https://sv.wikipedia.org/wiki/Kategori:Robotskapade_geografiartiklar
> [2] latest article generated just now a river in Fiji
> https://sv.wikipedia.org/wiki/Mbau_Creek_%28vattendrag,_lat_-16,50,_long_179,08%29
>
>
> Den 2016-04-20 kl. 23:30, skrev Florence Devouard:
>
>> Hello
>>
>> Sorry for highjacking your thread, but reading your message, I wanted to
>> share with you a small page I made a few days ago, to quantify the double
>> gap Gender/Africa.
>>
>> http://www.wikiloveswomen.org/about-the-project/mind-the-gap/
>>
>> If anyone has additional links or studies that could be useful to further
>> illustrate that double gap... I am interested.
>>
>> Also, if anyone is interested in further exploring this data-wise, please
>> raise your hand ;)
>>
>> Florence
>>
>>
>> Le 20/04/16 09:39, alexhin...@gmail.com a écrit :
>>
>>> Hi, as some of you may know, the Wikipedia gender indicator [1] tells us
>>> how many articles are biographies about women x language/country/culture.
>>>
>>> In order to compare these numbers...Does anyone knows if there is an
>>> existing comparison with gender balance in classical encyclopedias?
>>> (Britannica, Larousse...) or, if not, could someone prepare a WD query
>>> about it?
>>>
>>> I think it could be a good argument for us to use: e.g "at cawiki 12% of
>>> bios are about women, compared to 5% in GEC, Our most famous encyclopedia".
>>>
>>> We could compare it also for temathic encyclopedias or other databases
>>> existing in projects like Mix and match.
>>>
>>> Can someone help? thanks in advance
>>>
>>>
>>> [1]http://wigi.wmflabs.org/
>>>
>>>
>>> Àlex Hinojo
>>> User:Kippelboy
>>> Amical Wikimedia Programme manager
>>> ___
>>> Wikimedia-l mailing list, guidelines at:
>>> https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
>>> New messages to: Wikimedia-l@lists.wikimedia.org
>>> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
>>> 
>>>
>>>
>>
>>
>> ___
>> Wikimedia-l mailing list, guidelines at:
>> https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
>> New messages to: Wikimedia-l@lists.wikimedia.org
>> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
>> 
>>
>
>
> ___
> Wikimedia-l mailing list, guidelines at:
> https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
> New messages to: Wikimedia-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
>

Re: [Wikimedia-l] Africa / Gender gaps (was Re: Gender gap on "classical" encyclopedias)

2016-04-20 Thread Anders Wennersten 
Our traditional way of creating article is based on the interests of the 
contributors. This produces skewed total result, and this becomes more 
evident on a smaller version like svwp, then on bigger. We have long 
come to the conclusion that we will never be able to fill categories 
like towns in Mali and basketballplayer in Brazil, where we have had 
something like less then 10% of entries then the same categories on enwp 
(or frwp)


Wikidata can be of help evening out, but on svwp we have (also since 
long) said we must work and have a more systematic and deliberate 
approach to fill out "empty spaces"


We therefore love Lsjbot which now generates several million good and 
comprehensive articles on geographical entities all over the world from 
the most complete database that exists, and where areas like Africa is 
getting exactly the same attention like a Nordic country. It is 
completely unrealistic to think that the few contributors on svwp would 
ever create the now existing  25 article on entities on Canada or 
16500 entries in Antartica. But the bias in the source means Djibuti 
only gets 4000 and Camerun 9000 but it is none the less a huge improvement.


For articles on woman/men project are being run by wmse and it now exist 
a group of dedicated contributors generating articles on women. I am all 
fascinated of sources being used, specially to get entries of women from 
middle of 19-th century. All famous ballet dancers in Copenhagen in 
1850. All women who had local fame in Finland around 1860, including 
local healers etc. They have now created many thousands articles and 
getting the rate of articles up to 20% of total (25% of the number for men)


So I believe skewness is becoming an important issue and that we need 
to  adress oit even if it means to let go the "holy rule" only manual 
created article are "real" articles, it is the need of our readers who 
should have priority.


Anders

[1] list of article generated by country this far: 
https://sv.wikipedia.org/wiki/Kategori:Robotskapade_geografiartiklar
[2] latest article generated just now a river in Fiji 
https://sv.wikipedia.org/wiki/Mbau_Creek_%28vattendrag,_lat_-16,50,_long_179,08%29


Den 2016-04-20 kl. 23:30, skrev Florence Devouard:

Hello

Sorry for highjacking your thread, but reading your message, I wanted 
to share with you a small page I made a few days ago, to quantify the 
double gap Gender/Africa.


http://www.wikiloveswomen.org/about-the-project/mind-the-gap/

If anyone has additional links or studies that could be useful to 
further illustrate that double gap... I am interested.


Also, if anyone is interested in further exploring this data-wise, 
please raise your hand ;)


Florence


Le 20/04/16 09:39, alexhin...@gmail.com a écrit :
Hi, as some of you may know, the Wikipedia gender indicator [1] tells 
us how many articles are biographies about women x 
language/country/culture.


In order to compare these numbers...Does anyone knows if there is an 
existing comparison with gender balance in classical encyclopedias? 
(Britannica, Larousse...) or, if not, could someone prepare a WD 
query about it?


I think it could be a good argument for us to use: e.g "at cawiki 12% 
of bios are about women, compared to 5% in GEC, Our most famous 
encyclopedia".


We could compare it also for temathic encyclopedias or other 
databases existing in projects like Mix and match.


Can someone help? thanks in advance


[1]http://wigi.wmflabs.org/


Àlex Hinojo
User:Kippelboy
Amical Wikimedia Programme manager
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines

New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: 
https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, 







___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines

New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, 




___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,