These are fascinating experiments, I hope that the Waray-waray and Cebuano communities will at some point report back to the wider community as to how this worked out. My fear is that too fast a growth rate could overwhelm whatever community we have in those languages leading to burn out of existing editors dealing with too many newbies at once, my suspicion is that this will vary by language depending on such variables as the ratio of PC users to smartphone users, and the ease with which editors can access the necessary character sets.
We have long known that bot creation of stubs that are of interest to speakers of a language is a way to recruit readers, and that some readers become editors. What I think we don't yet know is the maximum growth rate that a wiki community can cope with.
There is also a sustainability angle, though hopefully we can mitigate that by bot replacing of articles where the source has changed but they haven't been edited on the Cebuano or Waray-waray Wikipedias. Otherwise within a decade we could have pedias that look very dated, for example various record holders whose articles in other languages show their records have been surpassed, and villages
WereSpielChequers
I can probably speak for those communities. On the whole, the logic behind the Lsjbot experiment was simple: build it and they will come.
So far though, this hasn’t happened. We from the Tagalog Wikipedia were also approached for this experiment, but we know what happens when bot-generated articles are made: the community is overwhelmed. Out of that fear, we declined to participate.
One of the concerns some editors in the Philippines have (and these are sentiments I share) is that these two Wikipedias turn us into a laughingstock, willing to increase article numbers at any cost. At one point, the Cebuano Wikipedia was described as a Wikipedia of French communes, not content relevant to Cebu or Cebuanos. I don’t think we’d like that with other Wikipedias in the Philippines or elsewhere.
Regards,
Josh
Wiadomość napisana przez WereSpielChequers werespielchequers@gmail.com w dniu 6 lip 2015, o godz. 04:52:
These are fascinating experiments, I hope that the Waray-waray and Cebuano communities will at some point report back to the wider community as to how this worked out. My fear is that too fast a growth rate could overwhelm whatever community we have in those languages leading to burn out of existing editors dealing with too many newbies at once, my suspicion is that this will vary by language depending on such variables as the ratio of PC users to smartphone users, and the ease with which editors can access the necessary character sets.
We have long known that bot creation of stubs that are of interest to speakers of a language is a way to recruit readers, and that some readers become editors. What I think we don't yet know is the maximum growth rate that a wiki community can cope with.
There is also a sustainability angle, though hopefully we can mitigate that by bot replacing of articles where the source has changed but they haven't been edited on the Cebuano or Waray-waray Wikipedias. Otherwise within a decade we could have pedias that look very dated, for example various record holders whose articles in other languages show their records have been surpassed, and villages
WereSpielChequers _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
JAMES JOSHUA G. LIM Bachelor of Arts in Political Science Class of 2013, Ateneo de Manila University Quezon City, Metro Manila, Philippines
jamesjoshualim@yahoo.com mailto:jamesjoshualim@yahoo.com | +63 (915) 321-7582 Facebook/Twitter: akiestar | Wikimedia: Sky Harbor http://about.me/josh.lim http://about.me/josh.lim
Indeed, as Josh points out, there are also costs (even if only perceived or reputational costs) to populating a tiny Wikipedia with next to no active editors with hundreds of thousands of bot-generated stubs. Is having stubs on all French communes in Cebuano better than having nothing in Cebuano? Probably, yes. And by increasing pageviews (which is measurable), one increases the likelihood of "organic" conversion of readers into editors (which is *still* the most effective way to make Wikipedians, albeit not the easiest to directly control).
But, again as Josh says, that increase in *editorship* is yet to be attained. The Waray Wikipedia (btw, "Waray-Waray" is, it turns out, objectionable to Waray speakers, and is mildly derogatory) is still largely edited by *one* committed individual, User:JinJian[1], as the stats plainly show. Given that the bot was run *with* JinJian's consent, there can be no objection to its operation.
As Milos suggests, there seems to be an emotional response to those Wikipedias appearing in the top 10 view. This should be divorced from those communities' sovereign decisions to run or not run the bot. If the top 10 inclusion truly bothers people, and there's a strong consensus that Wikipedias largely populated by bot-generated stubs "should" not be included, a discussion could be had on what this view *should* mean, precisely, if not plainly the top 10 Wikipedias by article count. And whatever refined definition is agreed upon (e.g. thresholds like a minimum number of active editors, or some formula involving the "article depth" figure, or whatever) can then be made the basis for the list, or indeed, for a different list, that would be more satisfying for those who are displeased with being "under" these Wikipedias on the list.
A.
[1] http://stats.wikimedia.org/EN/TablesWikipediaWAR.htm#wikipedians
On Mon, Jul 6, 2015 at 2:04 PM, Josh Lim jamesjoshualim@yahoo.com wrote:
I can probably speak for those communities. On the whole, the logic behind the Lsjbot experiment was simple: build it and they will come.
So far though, this hasn’t happened. We from the Tagalog Wikipedia were also approached for this experiment, but we know what happens when bot-generated articles are made: the community is overwhelmed. Out of that fear, we declined to participate.
One of the concerns some editors in the Philippines have (and these are sentiments I share) is that these two Wikipedias turn us into a laughingstock, willing to increase article numbers at any cost. At one point, the Cebuano Wikipedia was described as a Wikipedia of French communes, not content relevant to Cebu or Cebuanos. I don’t think we’d like that with other Wikipedias in the Philippines or elsewhere.
Regards,
Josh
Wiadomość napisana przez WereSpielChequers werespielchequers@gmail.com
w dniu 6 lip 2015, o godz. 04:52:
These are fascinating experiments, I hope that the Waray-waray and
Cebuano
communities will at some point report back to the wider community as to
how
this worked out. My fear is that too fast a growth rate could overwhelm whatever community we have in those languages leading to burn out of existing editors dealing with too many newbies at once, my suspicion is that this will vary by language depending on such variables as the ratio
of
PC users to smartphone users, and the ease with which editors can access the necessary character sets.
We have long known that bot creation of stubs that are of interest to speakers of a language is a way to recruit readers, and that some readers become editors. What I think we don't yet know is the maximum growth rate that a wiki community can cope with.
There is also a sustainability angle, though hopefully we can mitigate
that
by bot replacing of articles where the source has changed but they
haven't
been edited on the Cebuano or Waray-waray Wikipedias. Otherwise within a decade we could have pedias that look very dated, for example various record holders whose articles in other languages show their records have been surpassed, and villages
WereSpielChequers _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
JAMES JOSHUA G. LIM Bachelor of Arts in Political Science Class of 2013, Ateneo de Manila University Quezon City, Metro Manila, Philippines
jamesjoshualim@yahoo.com mailto:jamesjoshualim@yahoo.com | +63 (915) 321-7582 Facebook/Twitter: akiestar | Wikimedia: Sky Harbor http://about.me/josh.lim http://about.me/josh.lim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Hello,
I belong to the group of Josh and Ilario and others who have strong objections against the inundation of "pseudo articles" (one sentence-articles, bot creations based on database information etc.).
The people who justify their bot creations make two wrong assumptions: a) about the contributors: "Later, someone will come and improve the stubs." No, that is not true, at least not in the large majority of the language versions, and not within the 5 or 10 years we have experience with this phenomenon. b) about the readers: "No article is better than none." No, bad articles create a poor impression about a wiki. If you lure someone to your website, because Google indicated an article about a topic, and if the article is far from the expections, you don't give readers a reason to come back. They'll keep preferring Wikipedia in English.
I can imagine that bot articles about your own topics can make sence: like Frisian villages in Frisian Wikipedia (but certainly not Hungarian villages in Basque Wikipedia). There is a realistic chance that those articles will be expanded.
In general, if a Wikipedia language version does not have an article about a peticular village in a far far away country, or a star in a far far galaxy, then the reader should be sent to Wikidata or Reasonator.
A friend of mine is an expert on the Corsican language. He told me about Corsican Wikipedia: All those mini articles on French or Italian villages, that's nonsense. Those Wikipedians should better concentrate on important articles such as "History of Corsica" or "Corsican literature", they are still very poorly written. (This was in a conversation from a few years ago, I apologize if the situation is different now.)
Kind regards Ziko
Am Montag, 6. Juli 2015 schrieb Asaf Bartov :
Indeed, as Josh points out, there are also costs (even if only perceived or reputational costs) to populating a tiny Wikipedia with next to no active editors with hundreds of thousands of bot-generated stubs. Is having stubs on all French communes in Cebuano better than having nothing in Cebuano? Probably, yes. And by increasing pageviews (which is measurable), one increases the likelihood of "organic" conversion of readers into editors (which is *still* the most effective way to make Wikipedians, albeit not the easiest to directly control).
But, again as Josh says, that increase in *editorship* is yet to be attained. The Waray Wikipedia (btw, "Waray-Waray" is, it turns out, objectionable to Waray speakers, and is mildly derogatory) is still largely edited by *one* committed individual, User:JinJian[1], as the stats plainly show. Given that the bot was run *with* JinJian's consent, there can be no objection to its operation.
As Milos suggests, there seems to be an emotional response to those Wikipedias appearing in the top 10 view. This should be divorced from those communities' sovereign decisions to run or not run the bot. If the top 10 inclusion truly bothers people, and there's a strong consensus that Wikipedias largely populated by bot-generated stubs "should" not be included, a discussion could be had on what this view *should* mean, precisely, if not plainly the top 10 Wikipedias by article count. And whatever refined definition is agreed upon (e.g. thresholds like a minimum number of active editors, or some formula involving the "article depth" figure, or whatever) can then be made the basis for the list, or indeed, for a different list, that would be more satisfying for those who are displeased with being "under" these Wikipedias on the list.
A.
[1] http://stats.wikimedia.org/EN/TablesWikipediaWAR.htm#wikipedians
On Mon, Jul 6, 2015 at 2:04 PM, Josh Lim <jamesjoshualim@yahoo.com javascript:;> wrote:
I can probably speak for those communities. On the whole, the logic behind the Lsjbot experiment was simple: build it and they will come.
So far though, this hasn’t happened. We from the Tagalog Wikipedia were also approached for this experiment, but we know what happens when bot-generated articles are made: the community is overwhelmed. Out of
that
fear, we declined to participate.
One of the concerns some editors in the Philippines have (and these are sentiments I share) is that these two Wikipedias turn us into a laughingstock, willing to increase article numbers at any cost. At one point, the Cebuano Wikipedia was described as a Wikipedia of French communes, not content relevant to Cebu or Cebuanos. I don’t think we’d like that with other Wikipedias in the Philippines or elsewhere.
Regards,
Josh
Wiadomość napisana przez WereSpielChequers <
werespielchequers@gmail.com javascript:;>
w dniu 6 lip 2015, o godz. 04:52:
These are fascinating experiments, I hope that the Waray-waray and
Cebuano
communities will at some point report back to the wider community as to
how
this worked out. My fear is that too fast a growth rate could overwhelm whatever community we have in those languages leading to burn out of existing editors dealing with too many newbies at once, my suspicion is that this will vary by language depending on such variables as the
ratio
of
PC users to smartphone users, and the ease with which editors can
access
the necessary character sets.
We have long known that bot creation of stubs that are of interest to speakers of a language is a way to recruit readers, and that some
readers
become editors. What I think we don't yet know is the maximum growth
rate
that a wiki community can cope with.
There is also a sustainability angle, though hopefully we can mitigate
that
by bot replacing of articles where the source has changed but they
haven't
been edited on the Cebuano or Waray-waray Wikipedias. Otherwise within
a
decade we could have pedias that look very dated, for example various record holders whose articles in other languages show their records
have
been surpassed, and villages
WereSpielChequers _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l@lists.wikimedia.org javascript:; Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org javascript:;
?subject=unsubscribe>
JAMES JOSHUA G. LIM Bachelor of Arts in Political Science Class of 2013, Ateneo de Manila University Quezon City, Metro Manila, Philippines
jamesjoshualim@yahoo.com javascript:; <mailto:jamesjoshualim@yahoo.com
javascript:;> | +63 (915)
321-7582 Facebook/Twitter: akiestar | Wikimedia: Sky Harbor http://about.me/josh.lim http://about.me/josh.lim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org javascript:; Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org javascript:;
?subject=unsubscribe>
-- Asaf Bartov Wikimedia Foundation http://www.wikimediafoundation.org
Imagine a world in which every single human being can freely share in the sum of all knowledge. Help us make it a reality! https://donate.wikimedia.org _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org javascript:; ?subject=unsubscribe>
Are there any Cebuano or Waray community members on the list to offer an opinion? On 6 Jul 2015 23:47, "Ziko van Dijk" zvandijk@gmail.com wrote:
Hello,
I belong to the group of Josh and Ilario and others who have strong objections against the inundation of "pseudo articles" (one sentence-articles, bot creations based on database information etc.).
The people who justify their bot creations make two wrong assumptions: a) about the contributors: "Later, someone will come and improve the stubs." No, that is not true, at least not in the large majority of the language versions, and not within the 5 or 10 years we have experience with this phenomenon. b) about the readers: "No article is better than none." No, bad articles create a poor impression about a wiki. If you lure someone to your website, because Google indicated an article about a topic, and if the article is far from the expections, you don't give readers a reason to come back. They'll keep preferring Wikipedia in English.
I can imagine that bot articles about your own topics can make sence: like Frisian villages in Frisian Wikipedia (but certainly not Hungarian villages in Basque Wikipedia). There is a realistic chance that those articles will be expanded.
In general, if a Wikipedia language version does not have an article about a peticular village in a far far away country, or a star in a far far galaxy, then the reader should be sent to Wikidata or Reasonator.
A friend of mine is an expert on the Corsican language. He told me about Corsican Wikipedia: All those mini articles on French or Italian villages, that's nonsense. Those Wikipedians should better concentrate on important articles such as "History of Corsica" or "Corsican literature", they are still very poorly written. (This was in a conversation from a few years ago, I apologize if the situation is different now.)
Kind regards Ziko
Am Montag, 6. Juli 2015 schrieb Asaf Bartov :
Indeed, as Josh points out, there are also costs (even if only perceived
or
reputational costs) to populating a tiny Wikipedia with next to no active editors with hundreds of thousands of bot-generated stubs. Is having
stubs
on all French communes in Cebuano better than having nothing in Cebuano? Probably, yes. And by increasing pageviews (which is measurable), one increases the likelihood of "organic" conversion of readers into editors (which is *still* the most effective way to make Wikipedians, albeit not the easiest to directly control).
But, again as Josh says, that increase in *editorship* is yet to be attained. The Waray Wikipedia (btw, "Waray-Waray" is, it turns out, objectionable to Waray speakers, and is mildly derogatory) is still
largely
edited by *one* committed individual, User:JinJian[1], as the stats
plainly
show. Given that the bot was run *with* JinJian's consent, there can be
no
objection to its operation.
As Milos suggests, there seems to be an emotional response to those Wikipedias appearing in the top 10 view. This should be divorced from those communities' sovereign decisions to run or not run the bot. If the top 10 inclusion truly bothers people, and there's a strong consensus
that
Wikipedias largely populated by bot-generated stubs "should" not be included, a discussion could be had on what this view *should* mean, precisely, if not plainly the top 10 Wikipedias by article count. And whatever refined definition is agreed upon (e.g. thresholds like a
minimum
number of active editors, or some formula involving the "article depth" figure, or whatever) can then be made the basis for the list, or indeed, for a different list, that would be more satisfying for those who are displeased with being "under" these Wikipedias on the list.
A.
[1] http://stats.wikimedia.org/EN/TablesWikipediaWAR.htm#wikipedians
On Mon, Jul 6, 2015 at 2:04 PM, Josh Lim <jamesjoshualim@yahoo.com javascript:;> wrote:
I can probably speak for those communities. On the whole, the logic behind the Lsjbot experiment was simple: build it and they will come.
So far though, this hasn’t happened. We from the Tagalog Wikipedia
were
also approached for this experiment, but we know what happens when bot-generated articles are made: the community is overwhelmed. Out of
that
fear, we declined to participate.
One of the concerns some editors in the Philippines have (and these are sentiments I share) is that these two Wikipedias turn us into a laughingstock, willing to increase article numbers at any cost. At one point, the Cebuano Wikipedia was described as a Wikipedia of French communes, not content relevant to Cebu or Cebuanos. I don’t think we’d like that with other Wikipedias in the Philippines or elsewhere.
Regards,
Josh
Wiadomość napisana przez WereSpielChequers <
werespielchequers@gmail.com javascript:;>
w dniu 6 lip 2015, o godz. 04:52:
These are fascinating experiments, I hope that the Waray-waray and
Cebuano
communities will at some point report back to the wider community as
to
how
this worked out. My fear is that too fast a growth rate could
overwhelm
whatever community we have in those languages leading to burn out of existing editors dealing with too many newbies at once, my suspicion
is
that this will vary by language depending on such variables as the
ratio
of
PC users to smartphone users, and the ease with which editors can
access
the necessary character sets.
We have long known that bot creation of stubs that are of interest to speakers of a language is a way to recruit readers, and that some
readers
become editors. What I think we don't yet know is the maximum growth
rate
that a wiki community can cope with.
There is also a sustainability angle, though hopefully we can
mitigate
that
by bot replacing of articles where the source has changed but they
haven't
been edited on the Cebuano or Waray-waray Wikipedias. Otherwise
within
a
decade we could have pedias that look very dated, for example various record holders whose articles in other languages show their records
have
been surpassed, and villages
WereSpielChequers _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l@lists.wikimedia.org javascript:; Unsubscribe:
https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org javascript:;
?subject=unsubscribe>
JAMES JOSHUA G. LIM Bachelor of Arts in Political Science Class of 2013, Ateneo de Manila University Quezon City, Metro Manila, Philippines
jamesjoshualim@yahoo.com javascript:; <mailto:
jamesjoshualim@yahoo.com
javascript:;> | +63 (915)
321-7582 Facebook/Twitter: akiestar | Wikimedia: Sky Harbor http://about.me/josh.lim http://about.me/josh.lim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org javascript:; Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org javascript:;
?subject=unsubscribe>
-- Asaf Bartov Wikimedia Foundation http://www.wikimediafoundation.org
Imagine a world in which every single human being can freely share in the sum of all knowledge. Help us make it a reality! https://donate.wikimedia.org _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org javascript:; ?subject=unsubscribe>
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
I don't think any bot is subscribed to this list.
Il 07/07/2015 01:08, Richard Symonds ha scritto:
Are there any Cebuano or Waray community members on the list to offer an opinion? On 6 Jul 2015 23:47, "Ziko van Dijk" zvandijk@gmail.com wrote:
Hello,
I belong to the group of Josh and Ilario and others who have strong objections against the inundation of "pseudo articles" (one sentence-articles, bot creations based on database information etc.).
The people who justify their bot creations make two wrong assumptions: a) about the contributors: "Later, someone will come and improve the stubs." No, that is not true, at least not in the large majority of the language versions, and not within the 5 or 10 years we have experience with this phenomenon. b) about the readers: "No article is better than none." No, bad articles create a poor impression about a wiki. If you lure someone to your website, because Google indicated an article about a topic, and if the article is far from the expections, you don't give readers a reason to come back. They'll keep preferring Wikipedia in English.
I can imagine that bot articles about your own topics can make sence: like Frisian villages in Frisian Wikipedia (but certainly not Hungarian villages in Basque Wikipedia). There is a realistic chance that those articles will be expanded.
In general, if a Wikipedia language version does not have an article about a peticular village in a far far away country, or a star in a far far galaxy, then the reader should be sent to Wikidata or Reasonator.
A friend of mine is an expert on the Corsican language. He told me about Corsican Wikipedia: All those mini articles on French or Italian villages, that's nonsense. Those Wikipedians should better concentrate on important articles such as "History of Corsica" or "Corsican literature", they are still very poorly written. (This was in a conversation from a few years ago, I apologize if the situation is different now.)
Kind regards Ziko
Am Montag, 6. Juli 2015 schrieb Asaf Bartov :
Indeed, as Josh points out, there are also costs (even if only perceived
or
reputational costs) to populating a tiny Wikipedia with next to no active editors with hundreds of thousands of bot-generated stubs. Is having
stubs
on all French communes in Cebuano better than having nothing in Cebuano? Probably, yes. And by increasing pageviews (which is measurable), one increases the likelihood of "organic" conversion of readers into editors (which is *still* the most effective way to make Wikipedians, albeit not the easiest to directly control).
But, again as Josh says, that increase in *editorship* is yet to be attained. The Waray Wikipedia (btw, "Waray-Waray" is, it turns out, objectionable to Waray speakers, and is mildly derogatory) is still
largely
edited by *one* committed individual, User:JinJian[1], as the stats
plainly
show. Given that the bot was run *with* JinJian's consent, there can be
no
objection to its operation.
As Milos suggests, there seems to be an emotional response to those Wikipedias appearing in the top 10 view. This should be divorced from those communities' sovereign decisions to run or not run the bot. If the top 10 inclusion truly bothers people, and there's a strong consensus
that
Wikipedias largely populated by bot-generated stubs "should" not be included, a discussion could be had on what this view *should* mean, precisely, if not plainly the top 10 Wikipedias by article count. And whatever refined definition is agreed upon (e.g. thresholds like a
minimum
number of active editors, or some formula involving the "article depth" figure, or whatever) can then be made the basis for the list, or indeed, for a different list, that would be more satisfying for those who are displeased with being "under" these Wikipedias on the list.
A.[1] http://stats.wikimedia.org/EN/TablesWikipediaWAR.htm#wikipedians
On Mon, Jul 6, 2015 at 2:04 PM, Josh Lim <jamesjoshualim@yahoo.com javascript:;> wrote:
I can probably speak for those communities. On the whole, the logic behind the Lsjbot experiment was simple: build it and they will come.
So far though, this hasn’t happened. We from the Tagalog Wikipedia
were
also approached for this experiment, but we know what happens when bot-generated articles are made: the community is overwhelmed. Out of
that
fear, we declined to participate.
One of the concerns some editors in the Philippines have (and these are sentiments I share) is that these two Wikipedias turn us into a laughingstock, willing to increase article numbers at any cost. At one point, the Cebuano Wikipedia was described as a Wikipedia of French communes, not content relevant to Cebu or Cebuanos. I don’t think we’d like that with other Wikipedias in the Philippines or elsewhere.
Regards,
Josh
Wiadomość napisana przez WereSpielChequers <
werespielchequers@gmail.com javascript:;>
w dniu 6 lip 2015, o godz. 04:52:
These are fascinating experiments, I hope that the Waray-waray and
Cebuano
communities will at some point report back to the wider community as
to
how
this worked out. My fear is that too fast a growth rate could
overwhelm
whatever community we have in those languages leading to burn out of existing editors dealing with too many newbies at once, my suspicion
is
that this will vary by language depending on such variables as the
ratio
of
PC users to smartphone users, and the ease with which editors can
access
the necessary character sets.
We have long known that bot creation of stubs that are of interest to speakers of a language is a way to recruit readers, and that some
readers
become editors. What I think we don't yet know is the maximum growth
rate
that a wiki community can cope with.
There is also a sustainability angle, though hopefully we can
mitigate
that
by bot replacing of articles where the source has changed but they
haven't
been edited on the Cebuano or Waray-waray Wikipedias. Otherwise
within
a
decade we could have pedias that look very dated, for example various record holders whose articles in other languages show their records
have
been surpassed, and villages
WereSpielChequers _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l@lists.wikimedia.org javascript:; Unsubscribe:
https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org javascript:;
?subject=unsubscribe>
JAMES JOSHUA G. LIM Bachelor of Arts in Political Science Class of 2013, Ateneo de Manila University Quezon City, Metro Manila, Philippines
jamesjoshualim@yahoo.com javascript:; <mailto:
jamesjoshualim@yahoo.com
javascript:;> | +63 (915)
321-7582 Facebook/Twitter: akiestar | Wikimedia: Sky Harbor http://about.me/josh.lim http://about.me/josh.lim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org javascript:; Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org javascript:;
?subject=unsubscribe>
-- Asaf Bartov Wikimedia Foundation http://www.wikimediafoundation.org
Imagine a world in which every single human being can freely share in the sum of all knowledge. Help us make it a reality! https://donate.wikimedia.org _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org javascript:; ?subject=unsubscribe>
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
On Mon, Jul 6, 2015 at 4:08 PM, Richard Symonds < richard.symonds@wikimedia.org.uk> wrote:
Are there any Cebuano or Waray community members on the list to offer an opinion?
There is no likely productive outcome of sharing more opinions on the bot issue, even by members of those communities.
The point of my earlier post was an attempt to:
1. derail the rehashing of the Lsjbot-good-or-evil discussion, as it's pointless and, quite simply, Not Up To You, as Wikimedia-l subscribers, but up to each specific community, as would be discussed in each community's on-wiki discussion spaces. 2. separate the "top 10" issue that was the original trigger for this renewed discussion, and suggest a constructive direction in which that anxiety could be fruitfully addressed, quite apart from whether and which communities decide to employ Lsjbot to populate their wikis.
I hope that's clearer now.
A.
Completely agree with you Asaf. I don't think it's up to us Wikimedia-l subscribers, but up to each specific community. We can learn from their decisions but are on difficult ground if "we" judge them for it. Your third paragraph was a bit complex for me to "succinctly reword while agreeing with it", but it's a good starting point. But rather than a "top" list, perhaps we should be looking at "number of editors/number of speakers"?
Richard Symonds Wikimedia UK 0207 065 0992
Wikimedia UK is a Company Limited by Guarantee registered in England and Wales, Registered No. 6741827. Registered Charity No.1144513. Registered Office 4th Floor, Development House, 56-64 Leonard Street, London EC2A 4LT. United Kingdom. Wikimedia UK is the UK chapter of a global Wikimedia movement. The Wikimedia projects are run by the Wikimedia Foundation (who operate Wikipedia, amongst other projects).
*Wikimedia UK is an independent non-profit charity with no legal control over Wikipedia nor responsibility for its contents.*
On 7 July 2015 at 00:14, Asaf Bartov abartov@wikimedia.org wrote:
On Mon, Jul 6, 2015 at 4:08 PM, Richard Symonds < richard.symonds@wikimedia.org.uk> wrote:
Are there any Cebuano or Waray community members on the list to offer an opinion?
There is no likely productive outcome of sharing more opinions on the bot issue, even by members of those communities.
The point of my earlier post was an attempt to:
- derail the rehashing of the Lsjbot-good-or-evil discussion, as it's
pointless and, quite simply, Not Up To You, as Wikimedia-l subscribers, but up to each specific community, as would be discussed in each community's on-wiki discussion spaces. 2. separate the "top 10" issue that was the original trigger for this renewed discussion, and suggest a constructive direction in which that anxiety could be fruitfully addressed, quite apart from whether and which communities decide to employ Lsjbot to populate their wikis.
I hope that's clearer now.
A.
Asaf Bartov Wikimedia Foundation <http://www.wikimediafoundation.org>Imagine a world in which every single human being can freely share in the sum of all knowledge. Help us make it a reality! https://donate.wikimedia.org _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
On Mon, Jul 6, 2015 at 4:36 PM, Richard Symonds < richard.symonds@wikimedia.org.uk> wrote:
it's a good starting point. But rather than a "top" list, perhaps we should be looking at "number of editors/number of speakers"?
We (well, some of us, I guess), have indeed been looking at that figure (one might call it "editorship penetration", as in "Internet penetration"). Erik Zachte's stats site helpfully provides that metric in the "summary" view for a language. E.g. Bulgarian Wikipedia has 20 editors-per-million speakers[1], German has 31[2], Hebrew has 147[3], and Waray has 4[4]. Other useful, robust figures are the aforementioned active editor count (and very-active editor count).
Speaking at least for the Community Resources (formerly Grantmaking) team at WMF, I can say we have been paying a lot more attention to these figures than to article counts.
A.
[1] http://stats.wikimedia.org/EN/SummaryBG.htm [2] http://stats.wikimedia.org/EN/SummaryDE.htm [3] http://stats.wikimedia.org/EN/SummaryHE.htm [4] http://stats.wikimedia.org/EN/SummaryWAR.htm
The figures for Cebano is: 20 M speakers (60th biggest language in world) and 13 active contributors (compare to Dutch 28 M speakers 1183 active)
Waray-Waray has 3,1 M speakers 11 active contributors (but 4 new ones!) (compare to Slovene 2,4 M speakers, 141 active)
Sverker (who runs lsjbot) has met the few editors and their frustration in wanting to make their language established but are so extremely few to make this happen.
On the Quality side, we have identified (and corrected) around 0,02 % of the Lsjbot generated articles from error occurring from COL (mostly like given a the same specie a name ending with -a alternatively -um) and we estimate a total of around 0,1% of the articles generated have errors like this. In manually created articles we have an error rate in entering of data from sources of between 1-3%, and of mere sever character.
Also I wonder how many has looked into the quality of our language versions with fewer then 15 active contributors? I have and the findings are harirraising. Many of them have more unreverted vandal edits then serious edits and may many have fewer then a hundred serious edits a week and many fewer then 10.
I believe we should respect that the reality for our language version are looks significantly different.
And I would like to applaud the editors on Cebuano and Waray-waray Wikipedia in their effort against odds to get a viable version up and running, in whatever way they find being of value for them.
And to be judgemental to their (all serious) choices to work with their Wikiedpiaversions I believe is highly unfair.
We should all support these good struggling effort, like Sverkar has done in person and with a lot of time and effort.
And these people have never had as an ambition to appear on this list, it is also highly unfair indicating that they are to blamed in any way for what they have done
Anders
Asaf Bartov skrev den 2015-07-07 01:46:
On Mon, Jul 6, 2015 at 4:36 PM, Richard Symonds < richard.symonds@wikimedia.org.uk> wrote:
it's a good starting point. But rather that contributorsyn a "tootsp" list, perhaps we should be looking at "number of editors/number of speakers"?
We (well, some of us, I guess), have indeed been looking at that figure (one might call it "editorship penetration", as in "Internet penetration"). Erik Zachte's stats site helpfully provides that metric in the "summary" view for a language. E.g. Bulgarian Wikipedia has 20 editors-per-million speakers[1], German has 31[2], Hebrew has 147[3], and Waray has 4[4]. Other useful, robust figures are the aforementioned active editor count (and very-active editor count).
Speaking at least for the Community Resources (formerly Grantmaking) team at WMF, I can say we have been paying a lot more attention to these figures than to article counts.
A.[1] http://stats.wikimedia.org/EN/SummaryBG.htm [2] http://stats.wikimedia.org/EN/SummaryDE.htm [3] http://stats.wikimedia.org/EN/SummaryHE.htm [4] http://stats.wikimedia.org/EN/SummaryWAR.htm
On 07.07.2015 01:14, Asaf Bartov wrote:
On Mon, Jul 6, 2015 at 4:08 PM, Richard Symonds < richard.symonds@wikimedia.org.uk> wrote:
Are there any Cebuano or Waray community members on the list to offer an opinion?
- separate the "top 10" issue that was the original trigger for this
renewed discussion, and suggest a constructive direction in which that anxiety could be fruitfully addressed, quite apart from whether and which communities decide to employ Lsjbot to populate their wikis.
I hope that's clearer now.
A.
I don't have any objection in this point as I have no objection in other languages that have increased a lot their number of articles running bots.
At the opposite I am asking to look forward and to implement something that can be helpful for any linguistic projects (and surely also for Cebuan and Waray where it seems that few members have in charge the big workload to keep the pages updated).
Regards
There is already a consensus on enwiki (please, hold your rotten tomatoes) that projects like this which have inflated article counts due to extensive botting rather than through having a lively community not be included on the main page. I think a lot of the comments here about a huge article count attracting communities to curate that content are somewhat disingenous, it seems that despite having lots of articles there is only one active user on Waray Wikipedia, who is responsible for more than 99% of total edits. As Milos has alluded to, "number of articles" is a poor metric for understanding how useful a particular project is to speakers of that language.
Speaking here as a speaker of a minority language myself, I understand the temptation of quickly creating lots of articles to have some sort of demonstrable impact, and I believe there is a place for some bot generation of articles on any project. But after hitting "Random" a few times on Waray, and seeing what came back, I'm not really sure how this is a more useful resource for speakers of the language than just going into Wikidata with the interface set to Waray. I believe the time honoured, if slower way of creating a Wikipedia, lovingly handcrafting it article by article, is far more likely to lead to a positive impact for people.
Cheers, Craig
On 7 July 2015 at 07:55, Asaf Bartov abartov@wikimedia.org wrote:
Indeed, as Josh points out, there are also costs (even if only perceived or reputational costs) to populating a tiny Wikipedia with next to no active editors with hundreds of thousands of bot-generated stubs. Is having stubs on all French communes in Cebuano better than having nothing in Cebuano? Probably, yes. And by increasing pageviews (which is measurable), one increases the likelihood of "organic" conversion of readers into editors (which is *still* the most effective way to make Wikipedians, albeit not the easiest to directly control).
But, again as Josh says, that increase in *editorship* is yet to be attained. The Waray Wikipedia (btw, "Waray-Waray" is, it turns out, objectionable to Waray speakers, and is mildly derogatory) is still largely edited by *one* committed individual, User:JinJian[1], as the stats plainly show. Given that the bot was run *with* JinJian's consent, there can be no objection to its operation.
As Milos suggests, there seems to be an emotional response to those Wikipedias appearing in the top 10 view. This should be divorced from those communities' sovereign decisions to run or not run the bot. If the top 10 inclusion truly bothers people, and there's a strong consensus that Wikipedias largely populated by bot-generated stubs "should" not be included, a discussion could be had on what this view *should* mean, precisely, if not plainly the top 10 Wikipedias by article count. And whatever refined definition is agreed upon (e.g. thresholds like a minimum number of active editors, or some formula involving the "article depth" figure, or whatever) can then be made the basis for the list, or indeed, for a different list, that would be more satisfying for those who are displeased with being "under" these Wikipedias on the list.
A.
[1] http://stats.wikimedia.org/EN/TablesWikipediaWAR.htm#wikipedians
On Mon, Jul 6, 2015 at 2:04 PM, Josh Lim jamesjoshualim@yahoo.com wrote:
I can probably speak for those communities. On the whole, the logic behind the Lsjbot experiment was simple: build it and they will come.
So far though, this hasn’t happened. We from the Tagalog Wikipedia were also approached for this experiment, but we know what happens when bot-generated articles are made: the community is overwhelmed. Out of
that
fear, we declined to participate.
One of the concerns some editors in the Philippines have (and these are sentiments I share) is that these two Wikipedias turn us into a laughingstock, willing to increase article numbers at any cost. At one point, the Cebuano Wikipedia was described as a Wikipedia of French communes, not content relevant to Cebu or Cebuanos. I don’t think we’d like that with other Wikipedias in the Philippines or elsewhere.
Regards,
Josh
Wiadomość napisana przez WereSpielChequers <
werespielchequers@gmail.com>
w dniu 6 lip 2015, o godz. 04:52:
These are fascinating experiments, I hope that the Waray-waray and
Cebuano
communities will at some point report back to the wider community as to
how
this worked out. My fear is that too fast a growth rate could overwhelm whatever community we have in those languages leading to burn out of existing editors dealing with too many newbies at once, my suspicion is that this will vary by language depending on such variables as the
ratio
of
PC users to smartphone users, and the ease with which editors can
access
the necessary character sets.
We have long known that bot creation of stubs that are of interest to speakers of a language is a way to recruit readers, and that some
readers
become editors. What I think we don't yet know is the maximum growth
rate
that a wiki community can cope with.
There is also a sustainability angle, though hopefully we can mitigate
that
by bot replacing of articles where the source has changed but they
haven't
been edited on the Cebuano or Waray-waray Wikipedias. Otherwise within
a
decade we could have pedias that look very dated, for example various record holders whose articles in other languages show their records
have
been surpassed, and villages
WereSpielChequers _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
JAMES JOSHUA G. LIM Bachelor of Arts in Political Science Class of 2013, Ateneo de Manila University Quezon City, Metro Manila, Philippines
jamesjoshualim@yahoo.com mailto:jamesjoshualim@yahoo.com | +63 (915) 321-7582 Facebook/Twitter: akiestar | Wikimedia: Sky Harbor http://about.me/josh.lim http://about.me/josh.lim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- Asaf Bartov Wikimedia Foundation http://www.wikimediafoundation.org
Imagine a world in which every single human being can freely share in the sum of all knowledge. Help us make it a reality! https://donate.wikimedia.org _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Hoi, Have you EVER set foot in Wikidata? Because the notion that its data can be readily understood is absurd. Reasonator can be used in stead it may even generate text and, this works rather well for English. It could work for other languages. One problem with Reasonator is that for ordinary users it is too slow. It is not cached and it is on Labs so it is not guaranteed to be available.
Finally, Wikipedia is not what the WMF aims to do. It is sharing in the sum of all knowledge. Wikipedia plays a large part but it is not the stated objective in and of itself. Many Wikipedia articles and subjects are painful to read as they do not provide easily understood information and are often the result of painful compromises.
Random in Waray-Waray may give you information when you understand the language. The question is, do you find alternate sources for the same information in the Waray language?... Try random in Reasonator.. You will find a lot of information that does not exist in English Wikipedia.
The question is; do we really care about providing information? Thanks, GerardM
On 7 July 2015 at 12:21, Craig Franklin cfranklin@halonetwork.net wrote:
There is already a consensus on enwiki (please, hold your rotten tomatoes) that projects like this which have inflated article counts due to extensive botting rather than through having a lively community not be included on the main page. I think a lot of the comments here about a huge article count attracting communities to curate that content are somewhat disingenous, it seems that despite having lots of articles there is only one active user on Waray Wikipedia, who is responsible for more than 99% of total edits. As Milos has alluded to, "number of articles" is a poor metric for understanding how useful a particular project is to speakers of that language.
Speaking here as a speaker of a minority language myself, I understand the temptation of quickly creating lots of articles to have some sort of demonstrable impact, and I believe there is a place for some bot generation of articles on any project. But after hitting "Random" a few times on Waray, and seeing what came back, I'm not really sure how this is a more useful resource for speakers of the language than just going into Wikidata with the interface set to Waray. I believe the time honoured, if slower way of creating a Wikipedia, lovingly handcrafting it article by article, is far more likely to lead to a positive impact for people.
Cheers, Craig
On 7 July 2015 at 07:55, Asaf Bartov abartov@wikimedia.org wrote:
Indeed, as Josh points out, there are also costs (even if only perceived
or
reputational costs) to populating a tiny Wikipedia with next to no active editors with hundreds of thousands of bot-generated stubs. Is having
stubs
on all French communes in Cebuano better than having nothing in Cebuano? Probably, yes. And by increasing pageviews (which is measurable), one increases the likelihood of "organic" conversion of readers into editors (which is *still* the most effective way to make Wikipedians, albeit not the easiest to directly control).
But, again as Josh says, that increase in *editorship* is yet to be attained. The Waray Wikipedia (btw, "Waray-Waray" is, it turns out, objectionable to Waray speakers, and is mildly derogatory) is still
largely
edited by *one* committed individual, User:JinJian[1], as the stats
plainly
show. Given that the bot was run *with* JinJian's consent, there can be
no
objection to its operation.
As Milos suggests, there seems to be an emotional response to those Wikipedias appearing in the top 10 view. This should be divorced from those communities' sovereign decisions to run or not run the bot. If the top 10 inclusion truly bothers people, and there's a strong consensus
that
Wikipedias largely populated by bot-generated stubs "should" not be included, a discussion could be had on what this view *should* mean, precisely, if not plainly the top 10 Wikipedias by article count. And whatever refined definition is agreed upon (e.g. thresholds like a
minimum
number of active editors, or some formula involving the "article depth" figure, or whatever) can then be made the basis for the list, or indeed, for a different list, that would be more satisfying for those who are displeased with being "under" these Wikipedias on the list.
A.
[1] http://stats.wikimedia.org/EN/TablesWikipediaWAR.htm#wikipedians
On Mon, Jul 6, 2015 at 2:04 PM, Josh Lim jamesjoshualim@yahoo.com
wrote:
I can probably speak for those communities. On the whole, the logic behind the Lsjbot experiment was simple: build it and they will come.
So far though, this hasn’t happened. We from the Tagalog Wikipedia
were
also approached for this experiment, but we know what happens when bot-generated articles are made: the community is overwhelmed. Out of
that
fear, we declined to participate.
One of the concerns some editors in the Philippines have (and these are sentiments I share) is that these two Wikipedias turn us into a laughingstock, willing to increase article numbers at any cost. At one point, the Cebuano Wikipedia was described as a Wikipedia of French communes, not content relevant to Cebu or Cebuanos. I don’t think we’d like that with other Wikipedias in the Philippines or elsewhere.
Regards,
Josh
Wiadomość napisana przez WereSpielChequers <
werespielchequers@gmail.com>
w dniu 6 lip 2015, o godz. 04:52:
These are fascinating experiments, I hope that the Waray-waray and
Cebuano
communities will at some point report back to the wider community as
to
how
this worked out. My fear is that too fast a growth rate could
overwhelm
whatever community we have in those languages leading to burn out of existing editors dealing with too many newbies at once, my suspicion
is
that this will vary by language depending on such variables as the
ratio
of
PC users to smartphone users, and the ease with which editors can
access
the necessary character sets.
We have long known that bot creation of stubs that are of interest to speakers of a language is a way to recruit readers, and that some
readers
become editors. What I think we don't yet know is the maximum growth
rate
that a wiki community can cope with.
There is also a sustainability angle, though hopefully we can
mitigate
that
by bot replacing of articles where the source has changed but they
haven't
been edited on the Cebuano or Waray-waray Wikipedias. Otherwise
within
a
decade we could have pedias that look very dated, for example various record holders whose articles in other languages show their records
have
been surpassed, and villages
WereSpielChequers _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l@lists.wikimedia.org Unsubscribe:
https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
JAMES JOSHUA G. LIM Bachelor of Arts in Political Science Class of 2013, Ateneo de Manila University Quezon City, Metro Manila, Philippines
jamesjoshualim@yahoo.com mailto:jamesjoshualim@yahoo.com | +63 (915) 321-7582 Facebook/Twitter: akiestar | Wikimedia: Sky Harbor http://about.me/josh.lim http://about.me/josh.lim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- Asaf Bartov Wikimedia Foundation http://www.wikimediafoundation.org
Imagine a world in which every single human being can freely share in the sum of all knowledge. Help us make it a reality! https://donate.wikimedia.org _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Thanks to have summarized my position.
As speaker than more than one minor language I agree that there is no sense to inflate articles over the possibility of the small community to manage them. Not in opposition of automated generated articles but having in my hands the experience of project management of IT where the big challenge is not to produce a software but to keep it updated and efficient. Il 07/Lug/2015 12:22, "Craig Franklin" cfranklin@halonetwork.net ha scritto:
There is already a consensus on enwiki (please, hold your rotten tomatoes) that projects like this which have inflated article counts due to extensive botting rather than through having a lively community not be included on the main page. I think a lot of the comments here about a huge article count attracting communities to curate that content are somewhat disingenous, it seems that despite having lots of articles there is only one active user on Waray Wikipedia, who is responsible for more than 99% of total edits. As Milos has alluded to, "number of articles" is a poor metric for understanding how useful a particular project is to speakers of that language.
Speaking here as a speaker of a minority language myself, I understand the temptation of quickly creating lots of articles to have some sort of demonstrable impact, and I believe there is a place for some bot generation of articles on any project. But after hitting "Random" a few times on Waray, and seeing what came back, I'm not really sure how this is a more useful resource for speakers of the language than just going into Wikidata with the interface set to Waray. I believe the time honoured, if slower way of creating a Wikipedia, lovingly handcrafting it article by article, is far more likely to lead to a positive impact for people.
Cheers, Craig
On 7 July 2015 at 07:55, Asaf Bartov abartov@wikimedia.org wrote:
Indeed, as Josh points out, there are also costs (even if only perceived
or
reputational costs) to populating a tiny Wikipedia with next to no active editors with hundreds of thousands of bot-generated stubs. Is having
stubs
on all French communes in Cebuano better than having nothing in Cebuano? Probably, yes. And by increasing pageviews (which is measurable), one increases the likelihood of "organic" conversion of readers into editors (which is *still* the most effective way to make Wikipedians, albeit not the easiest to directly control).
But, again as Josh says, that increase in *editorship* is yet to be attained. The Waray Wikipedia (btw, "Waray-Waray" is, it turns out, objectionable to Waray speakers, and is mildly derogatory) is still
largely
edited by *one* committed individual, User:JinJian[1], as the stats
plainly
show. Given that the bot was run *with* JinJian's consent, there can be
no
objection to its operation.
As Milos suggests, there seems to be an emotional response to those Wikipedias appearing in the top 10 view. This should be divorced from those communities' sovereign decisions to run or not run the bot. If the top 10 inclusion truly bothers people, and there's a strong consensus
that
Wikipedias largely populated by bot-generated stubs "should" not be included, a discussion could be had on what this view *should* mean, precisely, if not plainly the top 10 Wikipedias by article count. And whatever refined definition is agreed upon (e.g. thresholds like a
minimum
number of active editors, or some formula involving the "article depth" figure, or whatever) can then be made the basis for the list, or indeed, for a different list, that would be more satisfying for those who are displeased with being "under" these Wikipedias on the list.
A.
[1] http://stats.wikimedia.org/EN/TablesWikipediaWAR.htm#wikipedians
On Mon, Jul 6, 2015 at 2:04 PM, Josh Lim jamesjoshualim@yahoo.com
wrote:
I can probably speak for those communities. On the whole, the logic behind the Lsjbot experiment was simple: build it and they will come.
So far though, this hasn’t happened. We from the Tagalog Wikipedia
were
also approached for this experiment, but we know what happens when bot-generated articles are made: the community is overwhelmed. Out of
that
fear, we declined to participate.
One of the concerns some editors in the Philippines have (and these are sentiments I share) is that these two Wikipedias turn us into a laughingstock, willing to increase article numbers at any cost. At one point, the Cebuano Wikipedia was described as a Wikipedia of French communes, not content relevant to Cebu or Cebuanos. I don’t think we’d like that with other Wikipedias in the Philippines or elsewhere.
Regards,
Josh
Wiadomość napisana przez WereSpielChequers <
werespielchequers@gmail.com>
w dniu 6 lip 2015, o godz. 04:52:
These are fascinating experiments, I hope that the Waray-waray and
Cebuano
communities will at some point report back to the wider community as
to
how
this worked out. My fear is that too fast a growth rate could
overwhelm
whatever community we have in those languages leading to burn out of existing editors dealing with too many newbies at once, my suspicion
is
that this will vary by language depending on such variables as the
ratio
of
PC users to smartphone users, and the ease with which editors can
access
the necessary character sets.
We have long known that bot creation of stubs that are of interest to speakers of a language is a way to recruit readers, and that some
readers
become editors. What I think we don't yet know is the maximum growth
rate
that a wiki community can cope with.
There is also a sustainability angle, though hopefully we can
mitigate
that
by bot replacing of articles where the source has changed but they
haven't
been edited on the Cebuano or Waray-waray Wikipedias. Otherwise
within
a
decade we could have pedias that look very dated, for example various record holders whose articles in other languages show their records
have
been surpassed, and villages
WereSpielChequers _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l@lists.wikimedia.org Unsubscribe:
https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
JAMES JOSHUA G. LIM Bachelor of Arts in Political Science Class of 2013, Ateneo de Manila University Quezon City, Metro Manila, Philippines
jamesjoshualim@yahoo.com mailto:jamesjoshualim@yahoo.com | +63 (915) 321-7582 Facebook/Twitter: akiestar | Wikimedia: Sky Harbor http://about.me/josh.lim http://about.me/josh.lim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- Asaf Bartov Wikimedia Foundation http://www.wikimediafoundation.org
Imagine a world in which every single human being can freely share in the sum of all knowledge. Help us make it a reality! https://donate.wikimedia.org _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Hoi, You will appreciate that when text is generated and cached. It may change when additional statements are made. So it will improve as and when improved information becomes available.
As someone who knows the IT branch as well, the maintenance of data is a challenge but you will agree with me that the solution the Wikimedia Foundation has in its projects is utterly different from what we know professionally.
In what I do for Wikidata I aim to bring information that is more than can be found in a single project. For instance, the laureates of awards may have an item because of an article in "another" Wikipedia. Quite often the laureates do not have an article in any language even though the award is prestigious internationally. In this way I hope we will get to the tipping point where these discussions are only of historic interest in the same way as many discussions about the quality of Wikipedia. Thanks, GerardM
On 7 July 2015 at 13:05, Ilario Valdelli valdelli@gmail.com wrote:
Thanks to have summarized my position.
As speaker than more than one minor language I agree that there is no sense to inflate articles over the possibility of the small community to manage them. Not in opposition of automated generated articles but having in my hands the experience of project management of IT where the big challenge is not to produce a software but to keep it updated and efficient. Il 07/Lug/2015 12:22, "Craig Franklin" cfranklin@halonetwork.net ha scritto:
There is already a consensus on enwiki (please, hold your rotten
tomatoes)
that projects like this which have inflated article counts due to
extensive
botting rather than through having a lively community not be included on the main page. I think a lot of the comments here about a huge article count attracting communities to curate that content are somewhat disingenous, it seems that despite having lots of articles there is only one active user on Waray Wikipedia, who is responsible for more than 99%
of
total edits. As Milos has alluded to, "number of articles" is a poor metric for understanding how useful a particular project is to speakers
of
that language.
Speaking here as a speaker of a minority language myself, I understand
the
temptation of quickly creating lots of articles to have some sort of demonstrable impact, and I believe there is a place for some bot
generation
of articles on any project. But after hitting "Random" a few times on Waray, and seeing what came back, I'm not really sure how this is a more useful resource for speakers of the language than just going into
Wikidata
with the interface set to Waray. I believe the time honoured, if slower way of creating a Wikipedia, lovingly handcrafting it article by article, is far more likely to lead to a positive impact for people.
Cheers, Craig
On 7 July 2015 at 07:55, Asaf Bartov abartov@wikimedia.org wrote:
Indeed, as Josh points out, there are also costs (even if only
perceived
or
reputational costs) to populating a tiny Wikipedia with next to no
active
editors with hundreds of thousands of bot-generated stubs. Is having
stubs
on all French communes in Cebuano better than having nothing in
Cebuano?
Probably, yes. And by increasing pageviews (which is measurable), one increases the likelihood of "organic" conversion of readers into
editors
(which is *still* the most effective way to make Wikipedians, albeit
not
the easiest to directly control).
But, again as Josh says, that increase in *editorship* is yet to be attained. The Waray Wikipedia (btw, "Waray-Waray" is, it turns out, objectionable to Waray speakers, and is mildly derogatory) is still
largely
edited by *one* committed individual, User:JinJian[1], as the stats
plainly
show. Given that the bot was run *with* JinJian's consent, there can
be
no
objection to its operation.
As Milos suggests, there seems to be an emotional response to those Wikipedias appearing in the top 10 view. This should be divorced from those communities' sovereign decisions to run or not run the bot. If
the
top 10 inclusion truly bothers people, and there's a strong consensus
that
Wikipedias largely populated by bot-generated stubs "should" not be included, a discussion could be had on what this view *should* mean, precisely, if not plainly the top 10 Wikipedias by article count. And whatever refined definition is agreed upon (e.g. thresholds like a
minimum
number of active editors, or some formula involving the "article depth" figure, or whatever) can then be made the basis for the list, or
indeed,
for a different list, that would be more satisfying for those who are displeased with being "under" these Wikipedias on the list.
A.
[1] http://stats.wikimedia.org/EN/TablesWikipediaWAR.htm#wikipedians
On Mon, Jul 6, 2015 at 2:04 PM, Josh Lim jamesjoshualim@yahoo.com
wrote:
I can probably speak for those communities. On the whole, the logic behind the Lsjbot experiment was simple: build it and they will come.
So far though, this hasn’t happened. We from the Tagalog Wikipedia
were
also approached for this experiment, but we know what happens when bot-generated articles are made: the community is overwhelmed. Out
of
that
fear, we declined to participate.
One of the concerns some editors in the Philippines have (and these
are
sentiments I share) is that these two Wikipedias turn us into a laughingstock, willing to increase article numbers at any cost. At
one
point, the Cebuano Wikipedia was described as a Wikipedia of French communes, not content relevant to Cebu or Cebuanos. I don’t think
we’d
like that with other Wikipedias in the Philippines or elsewhere.
Regards,
Josh
Wiadomość napisana przez WereSpielChequers <
werespielchequers@gmail.com>
w dniu 6 lip 2015, o godz. 04:52:
These are fascinating experiments, I hope that the Waray-waray and
Cebuano
communities will at some point report back to the wider community
as
to
how
this worked out. My fear is that too fast a growth rate could
overwhelm
whatever community we have in those languages leading to burn out
of
existing editors dealing with too many newbies at once, my
suspicion
is
that this will vary by language depending on such variables as the
ratio
of
PC users to smartphone users, and the ease with which editors can
access
the necessary character sets.
We have long known that bot creation of stubs that are of interest
to
speakers of a language is a way to recruit readers, and that some
readers
become editors. What I think we don't yet know is the maximum
growth
rate
that a wiki community can cope with.
There is also a sustainability angle, though hopefully we can
mitigate
that
by bot replacing of articles where the source has changed but they
haven't
been edited on the Cebuano or Waray-waray Wikipedias. Otherwise
within
a
decade we could have pedias that look very dated, for example
various
record holders whose articles in other languages show their records
have
been surpassed, and villages
WereSpielChequers _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l@lists.wikimedia.org Unsubscribe:
https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
JAMES JOSHUA G. LIM Bachelor of Arts in Political Science Class of 2013, Ateneo de Manila University Quezon City, Metro Manila, Philippines
jamesjoshualim@yahoo.com mailto:jamesjoshualim@yahoo.com | +63
(915)
321-7582 Facebook/Twitter: akiestar | Wikimedia: Sky Harbor http://about.me/josh.lim http://about.me/josh.lim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe:
https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- Asaf Bartov Wikimedia Foundation http://www.wikimediafoundation.org
Imagine a world in which every single human being can freely share in
the
sum of all knowledge. Help us make it a reality! https://donate.wikimedia.org _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
What gives you the right to be judgemental how they act on their version? Is that your idea of the movement values and vision, to talk badly of other efforts?
and I know for a fact they did not to this to get into this list you are upset of. It is untrue when you state "like this which have inflated article counts"
And I also know as a fact they are very happy with this effort because it has energized their small community. You talk of big increased I think of how many communities of this size that implodes which is a more common scenario.
As I have already stated I have no problem that you (and others) have another view of the benefits of botgenrated arciels.
But please be supportive to the very small communities, who do their best to survive and grow
Anders
Craig Franklin skrev den 2015-07-07 12:21:
There is already a consensus on enwiki (please, hold your rotten tomatoes) that projects like this which have inflated article counts due to extensive botting rather than through having a lively community not be included on the main page. I think a lot of the comments here about a huge article count attracting communities to curate that content are somewhat disingenous, it seems that despite having lots of articles there is only one active user on Waray Wikipedia, who is responsible for more than 99% of total edits. As Milos has alluded to, "number of articles" is a poor metric for understanding how useful a particular project is to speakers of that language.
Speaking here as a speaker of a minority language myself, I understand the temptation of quickly creating lots of articles to have some sort of demonstrable impact, and I believe there is a place for some bot generation of articles on any project. But after hitting "Random" a few times on Waray, and seeing what came back, I'm not really sure how this is a more useful resource for speakers of the language than just going into Wikidata with the interface set to Waray. I believe the time honoured, if slower way of creating a Wikipedia, lovingly handcrafting it article by article, is far more likely to lead to a positive impact for people.
Cheers, Craig
On 7 July 2015 at 07:55, Asaf Bartov abartov@wikimedia.org wrote:
Indeed, as Josh points out, there are also costs (even if only perceived or reputational costs) to populating a tiny Wikipedia with next to no active editors with hundreds of thousands of bot-generated stubs. Is having stubs on all French communes in Cebuano better than having nothing in Cebuano? Probably, yes. And by increasing pageviews (which is measurable), one increases the likelihood of "organic" conversion of readers into editors (which is *still* the most effective way to make Wikipedians, albeit not the easiest to directly control).
But, again as Josh says, that increase in *editorship* is yet to be attained. The Waray Wikipedia (btw, "Waray-Waray" is, it turns out, objectionable to Waray speakers, and is mildly derogatory) is still largely edited by *one* committed individual, User:JinJian[1], as the stats plainly show. Given that the bot was run *with* JinJian's consent, there can be no objection to its operation.
As Milos suggests, there seems to be an emotional response to those Wikipedias appearing in the top 10 view. This should be divorced from those communities' sovereign decisions to run or not run the bot. If the top 10 inclusion truly bothers people, and there's a strong consensus that Wikipedias largely populated by bot-generated stubs "should" not be included, a discussion could be had on what this view *should* mean, precisely, if not plainly the top 10 Wikipedias by article count. And whatever refined definition is agreed upon (e.g. thresholds like a minimum number of active editors, or some formula involving the "article depth" figure, or whatever) can then be made the basis for the list, or indeed, for a different list, that would be more satisfying for those who are displeased with being "under" these Wikipedias on the list.
A.[1] http://stats.wikimedia.org/EN/TablesWikipediaWAR.htm#wikipedians
On Mon, Jul 6, 2015 at 2:04 PM, Josh Lim jamesjoshualim@yahoo.com wrote:
I can probably speak for those communities. On the whole, the logic behind the Lsjbot experiment was simple: build it and they will come.
So far though, this hasn’t happened. We from the Tagalog Wikipedia were also approached for this experiment, but we know what happens when bot-generated articles are made: the community is overwhelmed. Out of
that
fear, we declined to participate.
One of the concerns some editors in the Philippines have (and these are sentiments I share) is that these two Wikipedias turn us into a laughingstock, willing to increase article numbers at any cost. At one point, the Cebuano Wikipedia was described as a Wikipedia of French communes, not content relevant to Cebu or Cebuanos. I don’t think we’d like that with other Wikipedias in the Philippines or elsewhere.
Regards,
Josh
Wiadomość napisana przez WereSpielChequers <
werespielchequers@gmail.com>
w dniu 6 lip 2015, o godz. 04:52:
These are fascinating experiments, I hope that the Waray-waray and
Cebuano
communities will at some point report back to the wider community as to
how
this worked out. My fear is that too fast a growth rate could overwhelm whatever community we have in those languages leading to burn out of existing editors dealing with too many newbies at once, my suspicion is that this will vary by language depending on such variables as the
ratio
of
PC users to smartphone users, and the ease with which editors can
access
the necessary character sets.
We have long known that bot creation of stubs that are of interest to speakers of a language is a way to recruit readers, and that some
readers
become editors. What I think we don't yet know is the maximum growth
rate
that a wiki community can cope with.
There is also a sustainability angle, though hopefully we can mitigate
that
by bot replacing of articles where the source has changed but they
haven't
been edited on the Cebuano or Waray-waray Wikipedias. Otherwise within
a
decade we could have pedias that look very dated, for example various record holders whose articles in other languages show their records
have
been surpassed, and villages
WereSpielChequers _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
JAMES JOSHUA G. LIM Bachelor of Arts in Political Science Class of 2013, Ateneo de Manila University Quezon City, Metro Manila, Philippines
jamesjoshualim@yahoo.com mailto:jamesjoshualim@yahoo.com | +63 (915) 321-7582 Facebook/Twitter: akiestar | Wikimedia: Sky Harbor http://about.me/josh.lim http://about.me/josh.lim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- Asaf Bartov Wikimedia Foundation http://www.wikimediafoundation.org
Imagine a world in which every single human being can freely share in the sum of all knowledge. Help us make it a reality! https://donate.wikimedia.org _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
The best evaluation is to understand the evolution and the trend.
In the last months in waray for instance I have seen less than 10 edits in the overall project in one month.
This is not revitalization. I agree with the enthusiasm of the community members but I am personally in favor of comparison of numbers in a long time perspective.
I am happy to see more data and to monitor them to know if this is a model to be replied but the numbers are not really supportive.
Regards Il 07/Lug/2015 13:12, "Anders Wennersten" mail@anderswennersten.se ha scritto:
What gives you the right to be judgemental how they act on their version? Is that your idea of the movement values and vision, to talk badly of other efforts?
and I know for a fact they did not to this to get into this list you are upset of. It is untrue when you state "like this which have inflated article counts"
And I also know as a fact they are very happy with this effort because it has energized their small community. You talk of big increased I think of how many communities of this size that implodes which is a more common scenario.
As I have already stated I have no problem that you (and others) have another view of the benefits of botgenrated arciels.
But please be supportive to the very small communities, who do their best to survive and grow
Anders
Craig Franklin skrev den 2015-07-07 12:21:
There is already a consensus on enwiki (please, hold your rotten tomatoes) that projects like this which have inflated article counts due to extensive botting rather than through having a lively community not be included on the main page. I think a lot of the comments here about a huge article count attracting communities to curate that content are somewhat disingenous, it seems that despite having lots of articles there is only one active user on Waray Wikipedia, who is responsible for more than 99% of total edits. As Milos has alluded to, "number of articles" is a poor metric for understanding how useful a particular project is to speakers of that language.
Speaking here as a speaker of a minority language myself, I understand the temptation of quickly creating lots of articles to have some sort of demonstrable impact, and I believe there is a place for some bot generation of articles on any project. But after hitting "Random" a few times on Waray, and seeing what came back, I'm not really sure how this is a more useful resource for speakers of the language than just going into Wikidata with the interface set to Waray. I believe the time honoured, if slower way of creating a Wikipedia, lovingly handcrafting it article by article, is far more likely to lead to a positive impact for people.
Cheers, Craig
On 7 July 2015 at 07:55, Asaf Bartov abartov@wikimedia.org wrote:
Indeed, as Josh points out, there are also costs (even if only perceived
or reputational costs) to populating a tiny Wikipedia with next to no active editors with hundreds of thousands of bot-generated stubs. Is having stubs on all French communes in Cebuano better than having nothing in Cebuano? Probably, yes. And by increasing pageviews (which is measurable), one increases the likelihood of "organic" conversion of readers into editors (which is *still* the most effective way to make Wikipedians, albeit not the easiest to directly control).
But, again as Josh says, that increase in *editorship* is yet to be attained. The Waray Wikipedia (btw, "Waray-Waray" is, it turns out, objectionable to Waray speakers, and is mildly derogatory) is still largely edited by *one* committed individual, User:JinJian[1], as the stats plainly show. Given that the bot was run *with* JinJian's consent, there can be no objection to its operation.
As Milos suggests, there seems to be an emotional response to those Wikipedias appearing in the top 10 view. This should be divorced from those communities' sovereign decisions to run or not run the bot. If the top 10 inclusion truly bothers people, and there's a strong consensus that Wikipedias largely populated by bot-generated stubs "should" not be included, a discussion could be had on what this view *should* mean, precisely, if not plainly the top 10 Wikipedias by article count. And whatever refined definition is agreed upon (e.g. thresholds like a minimum number of active editors, or some formula involving the "article depth" figure, or whatever) can then be made the basis for the list, or indeed, for a different list, that would be more satisfying for those who are displeased with being "under" these Wikipedias on the list.
A.[1] http://stats.wikimedia.org/EN/TablesWikipediaWAR.htm#wikipedians
On Mon, Jul 6, 2015 at 2:04 PM, Josh Lim jamesjoshualim@yahoo.com wrote:
I can probably speak for those communities. On the whole, the logic
behind the Lsjbot experiment was simple: build it and they will come.
So far though, this hasn’t happened. We from the Tagalog Wikipedia were also approached for this experiment, but we know what happens when bot-generated articles are made: the community is overwhelmed. Out of
that
fear, we declined to participate.
One of the concerns some editors in the Philippines have (and these are sentiments I share) is that these two Wikipedias turn us into a laughingstock, willing to increase article numbers at any cost. At one point, the Cebuano Wikipedia was described as a Wikipedia of French communes, not content relevant to Cebu or Cebuanos. I don’t think we’d like that with other Wikipedias in the Philippines or elsewhere.
Regards,
Josh
Wiadomość napisana przez WereSpielChequers <
werespielchequers@gmail.com>
w dniu 6 lip 2015, o godz. 04:52:
These are fascinating experiments, I hope that the Waray-waray and
Cebuano
communities will at some point report back to the wider community as to
how
this worked out. My fear is that too fast a growth rate could overwhelm whatever community we have in those languages leading to burn out of existing editors dealing with too many newbies at once, my suspicion is that this will vary by language depending on such variables as the
ratio
of
PC users to smartphone users, and the ease with which editors can
access
the necessary character sets.
We have long known that bot creation of stubs that are of interest to speakers of a language is a way to recruit readers, and that some
readers
become editors. What I think we don't yet know is the maximum growth
rate
that a wiki community can cope with.
There is also a sustainability angle, though hopefully we can mitigate
that
by bot replacing of articles where the source has changed but they
haven't
been edited on the Cebuano or Waray-waray Wikipedias. Otherwise within
a
decade we could have pedias that look very dated, for example various
record holders whose articles in other languages show their records
have
been surpassed, and villages
WereSpielChequers _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
JAMES JOSHUA G. LIM Bachelor of Arts in Political Science Class of 2013, Ateneo de Manila University Quezon City, Metro Manila, Philippines
jamesjoshualim@yahoo.com mailto:jamesjoshualim@yahoo.com | +63 (915) 321-7582 Facebook/Twitter: akiestar | Wikimedia: Sky Harbor http://about.me/josh.lim http://about.me/josh.lim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- Asaf Bartov Wikimedia Foundation http://www.wikimediafoundation.org
Imagine a world in which every single human being can freely share in the sum of all knowledge. Help us make it a reality! https://donate.wikimedia.org _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
I am not saying this should be repeated. I am saying we should respect their choice, and not as outsiders criticize their effort. or put erroneous bad faith assumptions on why they did this choice.
Anders
Ilario Valdelli skrev den 2015-07-07 13:21:
The best evaluation is to understand the evolution and the trend.
In the last months in waray for instance I have seen less than 10 edits in the overall project in one month.
This is not revitalization. I agree with the enthusiasm of the community members but I am personally in favor of comparison of numbers in a long time perspective.
I am happy to see more data and to monitor them to know if this is a model to be replied but the numbers are not really supportive.
Regards Il 07/Lug/2015 13:12, "Anders Wennersten" mail@anderswennersten.se ha scritto:
What gives you the right to be judgemental how they act on their version? Is that your idea of the movement values and vision, to talk badly of other efforts?
and I know for a fact they did not to this to get into this list you are upset of. It is untrue when you state "like this which have inflated article counts"
And I also know as a fact they are very happy with this effort because it has energized their small community. You talk of big increased I think of how many communities of this size that implodes which is a more common scenario.
As I have already stated I have no problem that you (and others) have another view of the benefits of botgenrated arciels.
But please be supportive to the very small communities, who do their best to survive and grow
Anders
Craig Franklin skrev den 2015-07-07 12:21:
There is already a consensus on enwiki (please, hold your rotten tomatoes) that projects like this which have inflated article counts due to extensive botting rather than through having a lively community not be included on the main page. I think a lot of the comments here about a huge article count attracting communities to curate that content are somewhat disingenous, it seems that despite having lots of articles there is only one active user on Waray Wikipedia, who is responsible for more than 99% of total edits. As Milos has alluded to, "number of articles" is a poor metric for understanding how useful a particular project is to speakers of that language.
Speaking here as a speaker of a minority language myself, I understand the temptation of quickly creating lots of articles to have some sort of demonstrable impact, and I believe there is a place for some bot generation of articles on any project. But after hitting "Random" a few times on Waray, and seeing what came back, I'm not really sure how this is a more useful resource for speakers of the language than just going into Wikidata with the interface set to Waray. I believe the time honoured, if slower way of creating a Wikipedia, lovingly handcrafting it article by article, is far more likely to lead to a positive impact for people.
Cheers, Craig
On 7 July 2015 at 07:55, Asaf Bartov abartov@wikimedia.org wrote:
Indeed, as Josh points out, there are also costs (even if only perceived
or reputational costs) to populating a tiny Wikipedia with next to no active editors with hundreds of thousands of bot-generated stubs. Is having stubs on all French communes in Cebuano better than having nothing in Cebuano? Probably, yes. And by increasing pageviews (which is measurable), one increases the likelihood of "organic" conversion of readers into editors (which is *still* the most effective way to make Wikipedians, albeit not the easiest to directly control).
But, again as Josh says, that increase in *editorship* is yet to be attained. The Waray Wikipedia (btw, "Waray-Waray" is, it turns out, objectionable to Waray speakers, and is mildly derogatory) is still largely edited by *one* committed individual, User:JinJian[1], as the stats plainly show. Given that the bot was run *with* JinJian's consent, there can be no objection to its operation.
As Milos suggests, there seems to be an emotional response to those Wikipedias appearing in the top 10 view. This should be divorced from those communities' sovereign decisions to run or not run the bot. If the top 10 inclusion truly bothers people, and there's a strong consensus that Wikipedias largely populated by bot-generated stubs "should" not be included, a discussion could be had on what this view *should* mean, precisely, if not plainly the top 10 Wikipedias by article count. And whatever refined definition is agreed upon (e.g. thresholds like a minimum number of active editors, or some formula involving the "article depth" figure, or whatever) can then be made the basis for the list, or indeed, for a different list, that would be more satisfying for those who are displeased with being "under" these Wikipedias on the list.
A.[1] http://stats.wikimedia.org/EN/TablesWikipediaWAR.htm#wikipedians
On Mon, Jul 6, 2015 at 2:04 PM, Josh Lim jamesjoshualim@yahoo.com wrote:
I can probably speak for those communities. On the whole, the logic
behind the Lsjbot experiment was simple: build it and they will come.
So far though, this hasn’t happened. We from the Tagalog Wikipedia were also approached for this experiment, but we know what happens when bot-generated articles are made: the community is overwhelmed. Out of
that
fear, we declined to participate.
One of the concerns some editors in the Philippines have (and these are sentiments I share) is that these two Wikipedias turn us into a laughingstock, willing to increase article numbers at any cost. At one point, the Cebuano Wikipedia was described as a Wikipedia of French communes, not content relevant to Cebu or Cebuanos. I don’t think we’d like that with other Wikipedias in the Philippines or elsewhere.
Regards,
Josh
Wiadomość napisana przez WereSpielChequers < werespielchequers@gmail.com> w dniu 6 lip 2015, o godz. 04:52:
These are fascinating experiments, I hope that the Waray-waray and
Cebuano
communities will at some point report back to the wider community as to
how
this worked out. My fear is that too fast a growth rate could overwhelm whatever community we have in those languages leading to burn out of existing editors dealing with too many newbies at once, my suspicion is that this will vary by language depending on such variables as the
ratio of
PC users to smartphone users, and the ease with which editors can
access the necessary character sets.
We have long known that bot creation of stubs that are of interest to speakers of a language is a way to recruit readers, and that some
readers become editors. What I think we don't yet know is the maximum growth rate that a wiki community can cope with.
There is also a sustainability angle, though hopefully we can mitigate
that
by bot replacing of articles where the source has changed but they
haven't
been edited on the Cebuano or Waray-waray Wikipedias. Otherwise within
a decade we could have pedias that look very dated, for example various
record holders whose articles in other languages show their records
have been surpassed, and villages
WereSpielChequers _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
JAMES JOSHUA G. LIM Bachelor of Arts in Political Science Class of 2013, Ateneo de Manila University Quezon City, Metro Manila, Philippines
jamesjoshualim@yahoo.com mailto:jamesjoshualim@yahoo.com | +63 (915) 321-7582 Facebook/Twitter: akiestar | Wikimedia: Sky Harbor http://about.me/josh.lim http://about.me/josh.lim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- Asaf Bartov Wikimedia Foundation http://www.wikimediafoundation.org
Imagine a world in which every single human being can freely share in the sum of all knowledge. Help us make it a reality! https://donate.wikimedia.org _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Hi Anders,
While I would normally agree with your logic, take note that for both the Cebuano and Waray Wikipedias, there were only 1-2 editors who said "yes" to the endeavor. For Waray, that was JinJian. For Cebuano, that happened to also be JinJian and one other editor who is currently not active.
I think Sverker meant well when he proposed (and implemented) bot-generated articles for both Wikipedias, but other Filipino Wikipedians have also seen first-hand what bot-generated articles do. As I mentioned previously, the Tagalog Wikipedia also had tens of thousands of bot-generated articles (our article count went up from near-20,000 to over 60,000 as a result), which was stopped only because of community opposition. The biggest question thus is how will a small community be able to maintain all these articles to begin with?
Now, as I’ve understood Sverker has given Wikipedia editing workshops in Cebu, as has Wikimedia Philippines in Samar and Leyte (where Waray is spoken). At this point, we’re figuring out how to grow our community in those regions, though if you ask me we’re still at a loss as to how we can sustainably grow communities in the Philippines without us having to look at them too closely. It’s not just a matter of content not being there—we’re also dealing with a whole gamut of socio-cultural factors as to why after all our attempts at introducing Wikipedia to people and showing them how to edit, people still don’t choose to do so.
That said, while I also think the Wikipedians concerned meant well, I hope this serves as a lesson to those considering bot-generated articles: bot-generated articles won’t jump-start a community, as the Cebuano and Waray examples have shown.
Regards,
Josh
Wiadomość napisana przez Anders Wennersten mail@anderswennersten.se w dniu 7 lip 2015, o godz. 04:33:
I am not saying this should be repeated. I am saying we should respect their choice, and not as outsiders criticize their effort. or put erroneous bad faith assumptions on why they did this choice.
Anders
Ilario Valdelli skrev den 2015-07-07 13:21:
The best evaluation is to understand the evolution and the trend.
In the last months in waray for instance I have seen less than 10 edits in the overall project in one month.
This is not revitalization. I agree with the enthusiasm of the community members but I am personally in favor of comparison of numbers in a long time perspective.
I am happy to see more data and to monitor them to know if this is a model to be replied but the numbers are not really supportive.
Regards Il 07/Lug/2015 13:12, "Anders Wennersten" mail@anderswennersten.se ha scritto:
What gives you the right to be judgemental how they act on their version? Is that your idea of the movement values and vision, to talk badly of other efforts?
and I know for a fact they did not to this to get into this list you are upset of. It is untrue when you state "like this which have inflated article counts"
And I also know as a fact they are very happy with this effort because it has energized their small community. You talk of big increased I think of how many communities of this size that implodes which is a more common scenario.
As I have already stated I have no problem that you (and others) have another view of the benefits of botgenrated arciels.
But please be supportive to the very small communities, who do their best to survive and grow
Anders
Craig Franklin skrev den 2015-07-07 12:21:
There is already a consensus on enwiki (please, hold your rotten tomatoes) that projects like this which have inflated article counts due to extensive botting rather than through having a lively community not be included on the main page. I think a lot of the comments here about a huge article count attracting communities to curate that content are somewhat disingenous, it seems that despite having lots of articles there is only one active user on Waray Wikipedia, who is responsible for more than 99% of total edits. As Milos has alluded to, "number of articles" is a poor metric for understanding how useful a particular project is to speakers of that language.
Speaking here as a speaker of a minority language myself, I understand the temptation of quickly creating lots of articles to have some sort of demonstrable impact, and I believe there is a place for some bot generation of articles on any project. But after hitting "Random" a few times on Waray, and seeing what came back, I'm not really sure how this is a more useful resource for speakers of the language than just going into Wikidata with the interface set to Waray. I believe the time honoured, if slower way of creating a Wikipedia, lovingly handcrafting it article by article, is far more likely to lead to a positive impact for people.
Cheers, Craig
On 7 July 2015 at 07:55, Asaf Bartov abartov@wikimedia.org wrote:
Indeed, as Josh points out, there are also costs (even if only perceived
or reputational costs) to populating a tiny Wikipedia with next to no active editors with hundreds of thousands of bot-generated stubs. Is having stubs on all French communes in Cebuano better than having nothing in Cebuano? Probably, yes. And by increasing pageviews (which is measurable), one increases the likelihood of "organic" conversion of readers into editors (which is *still* the most effective way to make Wikipedians, albeit not the easiest to directly control).
But, again as Josh says, that increase in *editorship* is yet to be attained. The Waray Wikipedia (btw, "Waray-Waray" is, it turns out, objectionable to Waray speakers, and is mildly derogatory) is still largely edited by *one* committed individual, User:JinJian[1], as the stats plainly show. Given that the bot was run *with* JinJian's consent, there can be no objection to its operation.
As Milos suggests, there seems to be an emotional response to those Wikipedias appearing in the top 10 view. This should be divorced from those communities' sovereign decisions to run or not run the bot. If the top 10 inclusion truly bothers people, and there's a strong consensus that Wikipedias largely populated by bot-generated stubs "should" not be included, a discussion could be had on what this view *should* mean, precisely, if not plainly the top 10 Wikipedias by article count. And whatever refined definition is agreed upon (e.g. thresholds like a minimum number of active editors, or some formula involving the "article depth" figure, or whatever) can then be made the basis for the list, or indeed, for a different list, that would be more satisfying for those who are displeased with being "under" these Wikipedias on the list.
A.[1] http://stats.wikimedia.org/EN/TablesWikipediaWAR.htm#wikipedians
On Mon, Jul 6, 2015 at 2:04 PM, Josh Lim jamesjoshualim@yahoo.com wrote:
I can probably speak for those communities. On the whole, the logic
behind the Lsjbot experiment was simple: build it and they will come.
So far though, this hasn’t happened. We from the Tagalog Wikipedia were also approached for this experiment, but we know what happens when bot-generated articles are made: the community is overwhelmed. Out of
that
fear, we declined to participate.
One of the concerns some editors in the Philippines have (and these are sentiments I share) is that these two Wikipedias turn us into a laughingstock, willing to increase article numbers at any cost. At one point, the Cebuano Wikipedia was described as a Wikipedia of French communes, not content relevant to Cebu or Cebuanos. I don’t think we’d like that with other Wikipedias in the Philippines or elsewhere.
Regards,
Josh
Wiadomość napisana przez WereSpielChequers < werespielchequers@gmail.com> w dniu 6 lip 2015, o godz. 04:52:
> These are fascinating experiments, I hope that the Waray-waray and > Cebuano
> communities will at some point report back to the wider community as to > how
> this worked out. My fear is that too fast a growth rate could overwhelm > whatever community we have in those languages leading to burn out of > existing editors dealing with too many newbies at once, my suspicion is > that this will vary by language depending on such variables as the > ratio of
> PC users to smartphone users, and the ease with which editors can > access the necessary character sets. > We have long known that bot creation of stubs that are of interest to > speakers of a language is a way to recruit readers, and that some > readers become editors. What I think we don't yet know is the maximum growth rate that a wiki community can cope with. > There is also a sustainability angle, though hopefully we can mitigate > that
> by bot replacing of articles where the source has changed but they > haven't
> been edited on the Cebuano or Waray-waray Wikipedias. Otherwise within > a decade we could have pedias that look very dated, for example various > record holders whose articles in other languages show their records > have been surpassed, and villages > WereSpielChequers > _______________________________________________ > Wikimedia-l mailing list, guidelines at: > https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
> Wikimedia-l@lists.wikimedia.org > Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, > mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
JAMES JOSHUA G. LIM Bachelor of Arts in Political Science Class of 2013, Ateneo de Manila University Quezon City, Metro Manila, Philippines
jamesjoshualim@yahoo.com mailto:jamesjoshualim@yahoo.com | +63 (915) 321-7582 Facebook/Twitter: akiestar | Wikimedia: Sky Harbor http://about.me/josh.lim http://about.me/josh.lim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- Asaf Bartov Wikimedia Foundation http://www.wikimediafoundation.org
Imagine a world in which every single human being can freely share in the sum of all knowledge. Help us make it a reality! https://donate.wikimedia.org _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
JAMES JOSHUA G. LIM Bachelor of Arts in Political Science Class of 2013, Ateneo de Manila University Quezon City, Metro Manila, Philippines
jamesjoshualim@yahoo.com mailto:jamesjoshualim@yahoo.com | +63 (915) 321-7582 Facebook/Twitter: akiestar | Wikimedia: Sky Harbor http://about.me/josh.lim http://about.me/josh.lim
Allow me to correct myself: my last line should read "bot-generated articles ALONE won’t jump-start a community, as the Cebuano and Waray examples have shown".
Josh
Wiadomość napisana przez Josh Lim jamesjoshualim@yahoo.com w dniu 7 lip 2015, o godz. 10:01:
Hi Anders,
While I would normally agree with your logic, take note that for both the Cebuano and Waray Wikipedias, there were only 1-2 editors who said "yes" to the endeavor. For Waray, that was JinJian. For Cebuano, that happened to also be JinJian and one other editor who is currently not active.
I think Sverker meant well when he proposed (and implemented) bot-generated articles for both Wikipedias, but other Filipino Wikipedians have also seen first-hand what bot-generated articles do. As I mentioned previously, the Tagalog Wikipedia also had tens of thousands of bot-generated articles (our article count went up from near-20,000 to over 60,000 as a result), which was stopped only because of community opposition. The biggest question thus is how will a small community be able to maintain all these articles to begin with?
Now, as I’ve understood Sverker has given Wikipedia editing workshops in Cebu, as has Wikimedia Philippines in Samar and Leyte (where Waray is spoken). At this point, we’re figuring out how to grow our community in those regions, though if you ask me we’re still at a loss as to how we can sustainably grow communities in the Philippines without us having to look at them too closely. It’s not just a matter of content not being there—we’re also dealing with a whole gamut of socio-cultural factors as to why after all our attempts at introducing Wikipedia to people and showing them how to edit, people still don’t choose to do so.
That said, while I also think the Wikipedians concerned meant well, I hope this serves as a lesson to those considering bot-generated articles: bot-generated articles won’t jump-start a community, as the Cebuano and Waray examples have shown.
Regards,
Josh
Wiadomość napisana przez Anders Wennersten <mail@anderswennersten.se mailto:mail@anderswennersten.se> w dniu 7 lip 2015, o godz. 04:33:
I am not saying this should be repeated. I am saying we should respect their choice, and not as outsiders criticize their effort. or put erroneous bad faith assumptions on why they did this choice.
Anders
Ilario Valdelli skrev den 2015-07-07 13:21:
The best evaluation is to understand the evolution and the trend.
In the last months in waray for instance I have seen less than 10 edits in the overall project in one month.
This is not revitalization. I agree with the enthusiasm of the community members but I am personally in favor of comparison of numbers in a long time perspective.
I am happy to see more data and to monitor them to know if this is a model to be replied but the numbers are not really supportive.
Regards Il 07/Lug/2015 13:12, "Anders Wennersten" <mail@anderswennersten.se mailto:mail@anderswennersten.se> ha scritto:
What gives you the right to be judgemental how they act on their version? Is that your idea of the movement values and vision, to talk badly of other efforts?
and I know for a fact they did not to this to get into this list you are upset of. It is untrue when you state "like this which have inflated article counts"
And I also know as a fact they are very happy with this effort because it has energized their small community. You talk of big increased I think of how many communities of this size that implodes which is a more common scenario.
As I have already stated I have no problem that you (and others) have another view of the benefits of botgenrated arciels.
But please be supportive to the very small communities, who do their best to survive and grow
Anders
Craig Franklin skrev den 2015-07-07 12:21:
There is already a consensus on enwiki (please, hold your rotten tomatoes) that projects like this which have inflated article counts due to extensive botting rather than through having a lively community not be included on the main page. I think a lot of the comments here about a huge article count attracting communities to curate that content are somewhat disingenous, it seems that despite having lots of articles there is only one active user on Waray Wikipedia, who is responsible for more than 99% of total edits. As Milos has alluded to, "number of articles" is a poor metric for understanding how useful a particular project is to speakers of that language.
Speaking here as a speaker of a minority language myself, I understand the temptation of quickly creating lots of articles to have some sort of demonstrable impact, and I believe there is a place for some bot generation of articles on any project. But after hitting "Random" a few times on Waray, and seeing what came back, I'm not really sure how this is a more useful resource for speakers of the language than just going into Wikidata with the interface set to Waray. I believe the time honoured, if slower way of creating a Wikipedia, lovingly handcrafting it article by article, is far more likely to lead to a positive impact for people.
Cheers, Craig
On 7 July 2015 at 07:55, Asaf Bartov <abartov@wikimedia.org mailto:abartov@wikimedia.org> wrote:
Indeed, as Josh points out, there are also costs (even if only perceived
or reputational costs) to populating a tiny Wikipedia with next to no active editors with hundreds of thousands of bot-generated stubs. Is having stubs on all French communes in Cebuano better than having nothing in Cebuano? Probably, yes. And by increasing pageviews (which is measurable), one increases the likelihood of "organic" conversion of readers into editors (which is *still* the most effective way to make Wikipedians, albeit not the easiest to directly control).
But, again as Josh says, that increase in *editorship* is yet to be attained. The Waray Wikipedia (btw, "Waray-Waray" is, it turns out, objectionable to Waray speakers, and is mildly derogatory) is still largely edited by *one* committed individual, User:JinJian[1], as the stats plainly show. Given that the bot was run *with* JinJian's consent, there can be no objection to its operation.
As Milos suggests, there seems to be an emotional response to those Wikipedias appearing in the top 10 view. This should be divorced from those communities' sovereign decisions to run or not run the bot. If the top 10 inclusion truly bothers people, and there's a strong consensus that Wikipedias largely populated by bot-generated stubs "should" not be included, a discussion could be had on what this view *should* mean, precisely, if not plainly the top 10 Wikipedias by article count. And whatever refined definition is agreed upon (e.g. thresholds like a minimum number of active editors, or some formula involving the "article depth" figure, or whatever) can then be made the basis for the list, or indeed, for a different list, that would be more satisfying for those who are displeased with being "under" these Wikipedias on the list.
A.[1] http://stats.wikimedia.org/EN/TablesWikipediaWAR.htm#wikipedians http://stats.wikimedia.org/EN/TablesWikipediaWAR.htm#wikipedians
On Mon, Jul 6, 2015 at 2:04 PM, Josh Lim <jamesjoshualim@yahoo.com mailto:jamesjoshualim@yahoo.com> wrote:
I can probably speak for those communities. On the whole, the logic > behind the Lsjbot experiment was simple: build it and they will come. > > So far though, this hasn’t happened. We from the Tagalog Wikipedia were > also approached for this experiment, but we know what happens when > bot-generated articles are made: the community is overwhelmed. Out of > that
> fear, we declined to participate. > > One of the concerns some editors in the Philippines have (and these are > sentiments I share) is that these two Wikipedias turn us into a > laughingstock, willing to increase article numbers at any cost. At one > point, the Cebuano Wikipedia was described as a Wikipedia of French > communes, not content relevant to Cebu or Cebuanos. I don’t think we’d > like that with other Wikipedias in the Philippines or elsewhere. > > Regards, > > Josh > > Wiadomość napisana przez WereSpielChequers < > werespielchequers@gmail.com mailto:werespielchequers@gmail.com> > w dniu 6 lip 2015, o godz. 04:52: > >> These are fascinating experiments, I hope that the Waray-waray and >> > Cebuano > >> communities will at some point report back to the wider community as to >> > how > >> this worked out. My fear is that too fast a growth rate could overwhelm >> whatever community we have in those languages leading to burn out of >> existing editors dealing with too many newbies at once, my suspicion is >> that this will vary by language depending on such variables as the >> > ratio > of > >> PC users to smartphone users, and the ease with which editors can >> > access > the necessary character sets. >> We have long known that bot creation of stubs that are of interest to >> speakers of a language is a way to recruit readers, and that some >> > readers > become editors. What I think we don't yet know is the maximum growth > rate > that a wiki community can cope with. >> There is also a sustainability angle, though hopefully we can mitigate >> > that > >> by bot replacing of articles where the source has changed but they >> > haven't > >> been edited on the Cebuano or Waray-waray Wikipedias. Otherwise within >> > a > decade we could have pedias that look very dated, for example various >> record holders whose articles in other languages show their records >> > have > been surpassed, and villages >> WereSpielChequers >> _______________________________________________ >> Wikimedia-l mailing list, guidelines at: >> > https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines > >> Wikimedia-l@lists.wikimedia.org >> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, >> > mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe > > JAMES JOSHUA G. LIM > Bachelor of Arts in Political Science > Class of 2013, Ateneo de Manila University > Quezon City, Metro Manila, Philippines > > jamesjoshualim@yahoo.com mailto:jamesjoshualim@yahoo.com | +63 (915) > 321-7582 > Facebook/Twitter: akiestar | Wikimedia: Sky Harbor > http://about.me/josh.lim http://about.me/josh.lim > _______________________________________________ > Wikimedia-l mailing list, guidelines at: > https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines > Wikimedia-l@lists.wikimedia.org > Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, > mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe >
>
Asaf Bartov Wikimedia Foundation <http://www.wikimediafoundation.org <http://www.wikimediafoundation.org/>>Imagine a world in which every single human being can freely share in the sum of all knowledge. Help us make it a reality! https://donate.wikimedia.org https://donate.wikimedia.org/ _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org mailto:Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org mailto:Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
JAMES JOSHUA G. LIM Bachelor of Arts in Political Science Class of 2013, Ateneo de Manila University Quezon City, Metro Manila, Philippines
jamesjoshualim@yahoo.com mailto:jamesjoshualim@yahoo.com | +63 (915) 321-7582 Facebook/Twitter: akiestar | Wikimedia: Sky Harbor http://about.me/josh.lim http://about.me/josh.lim
JAMES JOSHUA G. LIM Bachelor of Arts in Political Science Class of 2013, Ateneo de Manila University Quezon City, Metro Manila, Philippines
jamesjoshualim@yahoo.com mailto:jamesjoshualim@yahoo.com | +63 (915) 321-7582 Facebook/Twitter: akiestar | Wikimedia: Sky Harbor http://about.me/josh.lim http://about.me/josh.lim
Wiadomość napisana przez Anders Wennersten mail@anderswennersten.se w dniu 7 lip 2015, o godz. 04:11:
What gives you the right to be judgemental how they act on their version? Is that your idea of the movement values and vision, to talk badly of other efforts?
and I know for a fact they did not to this to get into this list you are upset of. It is untrue when you state "like this which have inflated article counts"
And I also know as a fact they are very happy with this effort because it has energized their small community. You talk of big increased I think of how many communities of this size that implodes which is a more common scenario.
As I have already stated I have no problem that you (and others) have another view of the benefits of botgenrated arciels.
But please be supportive to the very small communities, who do their best to survive and grow
Anders
Craig Franklin skrev den 2015-07-07 12:21:
There is already a consensus on enwiki (please, hold your rotten tomatoes) that projects like this which have inflated article counts due to extensive botting rather than through having a lively community not be included on the main page. I think a lot of the comments here about a huge article count attracting communities to curate that content are somewhat disingenous, it seems that despite having lots of articles there is only one active user on Waray Wikipedia, who is responsible for more than 99% of total edits. As Milos has alluded to, "number of articles" is a poor metric for understanding how useful a particular project is to speakers of that language.
Speaking here as a speaker of a minority language myself, I understand the temptation of quickly creating lots of articles to have some sort of demonstrable impact, and I believe there is a place for some bot generation of articles on any project. But after hitting "Random" a few times on Waray, and seeing what came back, I'm not really sure how this is a more useful resource for speakers of the language than just going into Wikidata with the interface set to Waray. I believe the time honoured, if slower way of creating a Wikipedia, lovingly handcrafting it article by article, is far more likely to lead to a positive impact for people.
Cheers, Craig
On 7 July 2015 at 07:55, Asaf Bartov abartov@wikimedia.org wrote:
Indeed, as Josh points out, there are also costs (even if only perceived or reputational costs) to populating a tiny Wikipedia with next to no active editors with hundreds of thousands of bot-generated stubs. Is having stubs on all French communes in Cebuano better than having nothing in Cebuano? Probably, yes. And by increasing pageviews (which is measurable), one increases the likelihood of "organic" conversion of readers into editors (which is *still* the most effective way to make Wikipedians, albeit not the easiest to directly control).
But, again as Josh says, that increase in *editorship* is yet to be attained. The Waray Wikipedia (btw, "Waray-Waray" is, it turns out, objectionable to Waray speakers, and is mildly derogatory) is still largely edited by *one* committed individual, User:JinJian[1], as the stats plainly show. Given that the bot was run *with* JinJian's consent, there can be no objection to its operation.
As Milos suggests, there seems to be an emotional response to those Wikipedias appearing in the top 10 view. This should be divorced from those communities' sovereign decisions to run or not run the bot. If the top 10 inclusion truly bothers people, and there's a strong consensus that Wikipedias largely populated by bot-generated stubs "should" not be included, a discussion could be had on what this view *should* mean, precisely, if not plainly the top 10 Wikipedias by article count. And whatever refined definition is agreed upon (e.g. thresholds like a minimum number of active editors, or some formula involving the "article depth" figure, or whatever) can then be made the basis for the list, or indeed, for a different list, that would be more satisfying for those who are displeased with being "under" these Wikipedias on the list.
A.
[1] http://stats.wikimedia.org/EN/TablesWikipediaWAR.htm#wikipedians
On Mon, Jul 6, 2015 at 2:04 PM, Josh Lim jamesjoshualim@yahoo.com wrote:
I can probably speak for those communities. On the whole, the logic behind the Lsjbot experiment was simple: build it and they will come.
So far though, this hasn’t happened. We from the Tagalog Wikipedia were also approached for this experiment, but we know what happens when bot-generated articles are made: the community is overwhelmed. Out of
that
fear, we declined to participate.
One of the concerns some editors in the Philippines have (and these are sentiments I share) is that these two Wikipedias turn us into a laughingstock, willing to increase article numbers at any cost. At one point, the Cebuano Wikipedia was described as a Wikipedia of French communes, not content relevant to Cebu or Cebuanos. I don’t think we’d like that with other Wikipedias in the Philippines or elsewhere.
Regards,
Josh
Wiadomość napisana przez WereSpielChequers <
werespielchequers@gmail.com>
w dniu 6 lip 2015, o godz. 04:52:
These are fascinating experiments, I hope that the Waray-waray and
Cebuano
communities will at some point report back to the wider community as to
how
this worked out. My fear is that too fast a growth rate could overwhelm whatever community we have in those languages leading to burn out of existing editors dealing with too many newbies at once, my suspicion is that this will vary by language depending on such variables as the
ratio
of
PC users to smartphone users, and the ease with which editors can
access
the necessary character sets.
We have long known that bot creation of stubs that are of interest to speakers of a language is a way to recruit readers, and that some
readers
become editors. What I think we don't yet know is the maximum growth
rate
that a wiki community can cope with.
There is also a sustainability angle, though hopefully we can mitigate
that
by bot replacing of articles where the source has changed but they
haven't
been edited on the Cebuano or Waray-waray Wikipedias. Otherwise within
a
decade we could have pedias that look very dated, for example various record holders whose articles in other languages show their records
have
been surpassed, and villages
WereSpielChequers _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
JAMES JOSHUA G. LIM Bachelor of Arts in Political Science Class of 2013, Ateneo de Manila University Quezon City, Metro Manila, Philippines
jamesjoshualim@yahoo.com mailto:jamesjoshualim@yahoo.com | +63 (915) 321-7582 Facebook/Twitter: akiestar | Wikimedia: Sky Harbor http://about.me/josh.lim http://about.me/josh.lim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- Asaf Bartov Wikimedia Foundation http://www.wikimediafoundation.org
Imagine a world in which every single human being can freely share in the sum of all knowledge. Help us make it a reality! https://donate.wikimedia.org _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
JAMES JOSHUA G. LIM Bachelor of Arts in Political Science Class of 2013, Ateneo de Manila University Quezon City, Metro Manila, Philippines
jamesjoshualim@yahoo.com mailto:jamesjoshualim@yahoo.com | +63 (915) 321-7582 Facebook/Twitter: akiestar | Wikimedia: Sky Harbor http://about.me/josh.lim http://about.me/josh.lim
Craig - are you volunteering to "lovingly handcraft" the Waray encyclopedia article by article?
Suggesting people go use Wikidata seems crazy. How will Waray speakers even know that Wikidata exists? How will they know how to manipulate the interface? More readable information available in the language they speak seems a clear net gain, regardless of whether the information meets some imaginary Platonic ideal of an encyclopedia.
Before Wikipedia, the time honored, if slow, way of building an encyclopedia was for paid experts to lovingly handcraft each entry under the supervision of paid publishing staff. The reins of tradition, custom and nostalgia should have little power to retard progress in our community.
On Tue, Jul 7, 2015 at 6:21 AM, Craig Franklin cfranklin@halonetwork.net wrote: I believe the time honoured, if slower way of creating a Wikipedia, lovingly handcrafting it article by article, is far more likely to lead to a positive impact for people. Cheers, Craig
wikimedia-l@lists.wikimedia.org