One of the best responses to some of the hyperbole out there about the closing, failure, end of WP is the figure of how many articles are actually locked down in any way, however, this is a difficult figure to authoritatively find/claim. There's Main and Featured [1] of course, about 11 protected articles [2], and then 785 semi-protected [3].
So are those the right numbers? If so can we claim about .0026% of pages are protected from editing by "anyone" and .00004% of pages are protected from Wikipedians (i.e., you've signed up for an account and haven't done anything stupid for a few days.)
How many pages (BPL + ?) are likely to fall under Flagged Protection?
[1]:http://en.wikipedia.org/wiki/Wikipedia:List_of_indefinitely_protected_pages [2]:http://en.wikipedia.org/wiki/Category:Wikipedia_protected_pages [3]:http://en.wikipedia.org/wiki/Category:Wikipedia_semi-protected_pages
Sorry, no.
A quick look at the protection log shows many more protections of articles as well as other pages; listing in the protected pages categories almost seems an exception when these are clicked on.
As well a wide range of pages are "salted" - deleted then protected to prevent recreation. Those don't appear in categories either.
It looks like you'd need to do a check on actual status of mainspace pages via the toolserver to get accurate statistics.
FT2
On Fri, Sep 4, 2009 at 2:00 PM, Joseph Reagle reagle@mit.edu wrote:
One of the best responses to some of the hyperbole out there about the closing, failure, end of WP is the figure of how many articles are actually locked down in any way, however, this is a difficult figure to authoritatively find/claim. There's Main and Featured [1] of course, about 11 protected articles [2], and then 785 semi-protected [3].
So are those the right numbers? If so can we claim about .0026% of pages are protected from editing by "anyone" and .00004% of pages are protected from Wikipedians (i.e., you've signed up for an account and haven't done anything stupid for a few days.)
How many pages (BPL + ?) are likely to fall under Flagged Protection?
WikiEN-l mailing list WikiEN-l@lists.wikimedia.org To unsubscribe from this mailing list, visit: https://lists.wikimedia.org/mailman/listinfo/wikien-l
Okay, found out why.
You need to account for [[Category:Wikipedia pages protected due to dispute]] and other protection categories, as well. Pages such as "Russell's teapot" and "Developed country" are in there, protected, but not tagged.
The root cause seems to be that the category isn't itself a subcategory of some "protected pages" category. Specifically, there are protection templates http://en.wikipedia.org/wiki/Category:Protection_templates such as "Pp-dispute" that don't also include the page in one of the main "protected pages" categories you name.
FT2
On Fri, Sep 4, 2009 at 2:29 PM, FT2 ft2.wiki@gmail.com wrote:
Sorry, no.
A quick look at the protection log shows many more protections of articles as well as other pages; listing in the protected pages categories almost seems an exception when these are clicked on.
As well a wide range of pages are "salted" - deleted then protected to prevent recreation. Those don't appear in categories either.
It looks like you'd need to do a check on actual status of mainspace pages via the toolserver to get accurate statistics.
FT2
On Fri, Sep 4, 2009 at 2:00 PM, Joseph Reagle reagle@mit.edu wrote:
One of the best responses to some of the hyperbole out there about the closing, failure, end of WP is the figure of how many articles are actually locked down in any way, however, this is a difficult figure to authoritatively find/claim. There's Main and Featured [1] of course, about 11 protected articles [2], and then 785 semi-protected [3].
So are those the right numbers? If so can we claim about .0026% of pages are protected from editing by "anyone" and .00004% of pages are protected from Wikipedians (i.e., you've signed up for an account and haven't done anything stupid for a few days.)
How many pages (BPL + ?) are likely to fall under Flagged Protection?
WikiEN-l mailing list WikiEN-l@lists.wikimedia.org To unsubscribe from this mailing list, visit: https://lists.wikimedia.org/mailman/listinfo/wikien-l
Passed on to WP:AN
http://en.wikipedia.org/wiki/Wikipedia:Administrators%27_noticeboard#Protect...
FT2
On Fri, Sep 4, 2009 at 2:36 PM, FT2 ft2.wiki@gmail.com wrote:
Okay, found out why.
You need to account for [[Category:Wikipedia pages protected due to dispute]] and other protection categories, as well. Pages such as "Russell's teapot" and "Developed country" are in there, protected, but not tagged.
The root cause seems to be that the category isn't itself a subcategory of some "protected pages" category. Specifically, there are protection templates http://en.wikipedia.org/wiki/Category:Protection_templates such as "Pp-dispute" that don't also include the page in one of the main "protected pages" categories you name.
FT2
On Fri, Sep 4, 2009 at 2:29 PM, FT2 ft2.wiki@gmail.com wrote:
Sorry, no.
A quick look at the protection log shows many more protections of articles as well as other pages; listing in the protected pages categories almost seems an exception when these are clicked on.
As well a wide range of pages are "salted" - deleted then protected to prevent recreation. Those don't appear in categories either.
It looks like you'd need to do a check on actual status of mainspace pages via the toolserver to get accurate statistics.
FT2
On Fri, Sep 4, 2009 at 2:00 PM, Joseph Reagle reagle@mit.edu wrote:
One of the best responses to some of the hyperbole out there about the closing, failure, end of WP is the figure of how many articles are actually locked down in any way, however, this is a difficult figure to authoritatively find/claim. There's Main and Featured [1] of course, about 11 protected articles [2], and then 785 semi-protected [3].
So are those the right numbers? If so can we claim about .0026% of pages are protected from editing by "anyone" and .00004% of pages are protected from Wikipedians (i.e., you've signed up for an account and haven't done anything stupid for a few days.)
How many pages (BPL + ?) are likely to fall under Flagged Protection?
WikiEN-l mailing list WikiEN-l@lists.wikimedia.org To unsubscribe from this mailing list, visit: https://lists.wikimedia.org/mailman/listinfo/wikien-l
Try this instead to refute the "Wikipedia is dying" argument. Wikipedia's featured picture program started in May 2004. It took until 30 December 2007 to reach 1000 featured pictures. We're on track to reach number 2000 within a week: currently at 1973 FPs with 63 active nominations.
It would be interesting if someone wrote a tool to check article citations. Footnoting has been getting more and more commonplace, as well as more extensive.
-Durova
On Fri, Sep 4, 2009 at 7:02 AM, FT2 ft2.wiki@gmail.com wrote:
Passed on to WP:AN
http://en.wikipedia.org/wiki/Wikipedia:Administrators%27_noticeboard#Protect...http://en.wikipedia.org/wiki/Wikipedia:Administrators%27_noticeboard#Protection_template_issue
FT2
On Fri, Sep 4, 2009 at 2:36 PM, FT2 ft2.wiki@gmail.com wrote:
Okay, found out why.
You need to account for [[Category:Wikipedia pages protected due to dispute]] and other protection categories, as well. Pages such as
"Russell's
teapot" and "Developed country" are in there, protected, but not tagged.
The root cause seems to be that the category isn't itself a subcategory
of
some "protected pages" category. Specifically, there are protection templates http://en.wikipedia.org/wiki/Category:Protection_templates such as "Pp-dispute" that don't also include the page in one of the main "protected pages" categories you name.
FT2
On Fri, Sep 4, 2009 at 2:29 PM, FT2 ft2.wiki@gmail.com wrote:
Sorry, no.
A quick look at the protection log shows many more protections of
articles
as well as other pages; listing in the protected pages categories almost seems an exception when these are clicked on.
As well a wide range of pages are "salted" - deleted then protected to prevent recreation. Those don't appear in categories either.
It looks like you'd need to do a check on actual status of mainspace
pages
via the toolserver to get accurate statistics.
FT2
On Fri, Sep 4, 2009 at 2:00 PM, Joseph Reagle reagle@mit.edu wrote:
One of the best responses to some of the hyperbole out there about the closing, failure, end of WP is the figure of how many articles are
actually
locked down in any way, however, this is a difficult figure to authoritatively find/claim. There's Main and Featured [1] of course,
about
11 protected articles [2], and then 785 semi-protected [3].
So are those the right numbers? If so can we claim about .0026% of
pages
are protected from editing by "anyone" and .00004% of pages are
protected
from Wikipedians (i.e., you've signed up for an account and haven't
done
anything stupid for a few days.)
How many pages (BPL + ?) are likely to fall under Flagged Protection?
[1]:
http://en.wikipedia.org/wiki/Wikipedia:List_of_indefinitely_protected_pages
[3]:
http://en.wikipedia.org/wiki/Category:Wikipedia_semi-protected_pages
WikiEN-l mailing list WikiEN-l@lists.wikimedia.org To unsubscribe from this mailing list, visit: https://lists.wikimedia.org/mailman/listinfo/wikien-l
WikiEN-l mailing list WikiEN-l@lists.wikimedia.org To unsubscribe from this mailing list, visit: https://lists.wikimedia.org/mailman/listinfo/wikien-l
On Friday 04 September 2009, Joseph Reagle wrote:
One of the best responses to some of the hyperbole out there about the closing, failure, end of WP is the figure of how many articles are actually locked down in any way, however, this is a difficult figure to authoritatively find/claim. There's Main and Featured [1] of course, about 11 protected articles [2], and then 785 semi-protected [3].
OK, so the protected categories aren't reliable, after some digging, here's some figures:
[[ The recent focus on Wikipedia "failing" or being "closed" merit some figures and explanation. On the afternoon of Sept 04, 2009 the English Wikipedia with 3,024,063 articles.
The [Special:ProtectedPages][1] for the Article namespace tells us:
* 5,137 articles are protected (that's 0.17% of all articles). * The majority of those, (3,553 articles or 69% of protected articles), are semi-protected, meaning that while they aren't editable by anonymous users, they are by Wikipedians (i.e., those that sign up for an account and don't do anything stupid). * Therefore, only 1,583 articles (.05%) are fully protected, and not available to editing by non-administrative Wikipedians. * Of all the articles being protected, 1337 of them (26%) are set to expired, most within a month or two.
That's the status quo. Yet, some means of flagging a vetted version of an article has been [discussed since 2005][2]. The current widely [discussed idea][3] is to conduct a two month experiment in which [biographies of living people][4] (402,672 articles, about 13% of the English Wikipedia) or more likely *some subset* thereof are "flag protected" which means anyone *can still edit* but the public (not Wikipedians) see the last reviewed version. This doesn't necessarily replace the existing protection mechanisms, but could be a good alternative to semi-protection. The experiment will helpfully give guidance on who should be a "Reviewer" and how long it takes time to review and flag a newer version. Another part of the experiment is "partrolled revisions" which would apply to a wider swath of articles and permit vandalism fighters to bookmark a known good version so they can easily evaluate subsequent contributions, but it won't affect who can edit or what the public sees.
The goal of this, and other features, is to maximize the benefits of open collaboration while limiting the damage from disruptive edits. This has always been the case and Wikipedia continues to experiment with achieving the best balance.
[1]:http://en.wikipedia.org/wiki/Special:ProtectedPages [2]:http://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2009-08-31/Flagged... [3]:http://en.wikipedia.org/wiki/Wikipedia:Flagged_protection_and_patrolled_revi... [4]:http://en.wikipedia.org/wiki/Category:Living_people
]]
Does that sound right?
On Fri, Sep 4, 2009 at 12:54 PM, Joseph Reaglereagle@mit.edu wrote:
On Friday 04 September 2009, Joseph Reagle wrote:
One of the best responses to some of the hyperbole out there about the closing, failure, end of WP is the figure of how many articles are actually locked down in any way, however, this is a difficult figure to authoritatively find/claim. There's Main and Featured [1] of course, about 11 protected articles [2], and then 785 semi-protected [3].
OK, so the protected categories aren't reliable, after some digging, here's some figures:
[[ The recent focus on Wikipedia "failing" or being "closed" merit some figures and explanation. On the afternoon of Sept 04, 2009 the English Wikipedia with 3,024,063 articles.
The [Special:ProtectedPages][1] for the Article namespace tells us:
- 5,137 articles are protected (that's 0.17% of all articles).
- The majority of those, (3,553 articles or 69% of protected articles), are semi-protected, meaning that while they aren't editable by anonymous users, they are by Wikipedians (i.e., those that sign up for an account and don't do anything stupid).
- Therefore, only 1,583 articles (.05%) are fully protected, and not available to editing by non-administrative Wikipedians.
- Of all the articles being protected, 1337 of them (26%) are set to expired, most within a month or two.
That's the status quo. Yet, some means of flagging a vetted version of an article has been [discussed since 2005][2]. The current widely [discussed idea][3] is to conduct a two month experiment in which [biographies of living people][4] (402,672 articles, about 13% of the English Wikipedia) or more likely *some subset* thereof are "flag protected" which means anyone *can still edit* but the public (not Wikipedians) see the last reviewed version. This doesn't necessarily replace the existing protection mechanisms, but could be a good alternative to semi-protection. The experiment will helpfully give guidance on who should be a "Reviewer" and how long it takes time to review and flag a newer version. Another part of the experiment is "partrolled revisions" which would apply to a wider swath of articles and permit vandalism fighters to bookmark a known good version so they can easily evaluate subsequent contributions, but it won't affect who can edit or what the public sees.
The goal of this, and other features, is to maximize the benefits of open collaboration while limiting the damage from disruptive edits. This has always been the case and Wikipedia continues to experiment with achieving the best balance.
]]
Does that sound right?
Would it be possible for you to do a comparison with Wikipedia just before semiprotection was enabled? I've long wanted to know whether the argument that semiprotections would replace full protections holds any water.
This would also seem to be quite important to know for flagged, inasmuch as that argument has been recycled for flagging pages...
I'm undertaking to have all article and talk page semiprotections on Wikipedia reviewed. The process I'm using is to enter a brief proposal on the article talk page and contact the protecting sysop. The idea is that we discuss whether to unprotect the article or talk page and watch it vigilantly.
This has already met considerable success, with more 30% of the proposals I've made this evening being enacted upon. There appear to be a lot of semiprotections that have simply been forgotten by the original sysop.
I'll keep this up until I either run out of articles to review or get bored. Since there are several thousand semiprotected article the latter is more likely to happen first.
Gwern Branwen wonders whether semiprotections have taken over from protections. Well one cannot really compare the current Wikipedia with the Wikipedia of 2005. Then we had no real way of dealing with biographies of living persons, and little awareness of the problem, and as for the protected articles, they numbered dozens at the most, and certainly not thousands. It's important to strike a balance. While many of the semiprotected pages may actually be redirects that we wouldn't normally want to see edited by unregistered users, I suspect many are not. It's always a good idea to review the situation regularly.
On Fri, Sep 4, 2009 at 4:55 PM, Tony Sidawaytonysidaway@gmail.com wrote:
I'm undertaking to have all article and talk page semiprotections on Wikipedia reviewed. Â The process I'm using is to enter a brief proposal on the article talk page and contact the protecting sysop. The idea is that we discuss whether to unprotect the article or talk page and watch it vigilantly.
This has already met considerable success, with more 30% of the proposals I've made this evening being enacted upon. Â There appear to be a lot of semiprotections that have simply been forgotten by the original sysop.
I'll keep this up until I either run out of articles to review or get bored. Â Since there are several thousand semiprotected article the latter is more likely to happen first.
Gwern Branwen wonders whether semiprotections have taken over from protections. Â Well one cannot really compare the current Wikipedia with the Wikipedia of 2005. Â Then we had no real way of dealing with biographies of living persons, and little awareness of the problem, and as for the protected articles, they numbered dozens at the most, and certainly not thousands. Â It's important to strike a balance. While many of the semiprotected pages may actually be redirects that we wouldn't normally want to see edited by unregistered users, I suspect many are not. Â It's always a good idea to review the situation regularly.
Excellent idea. Some problem areas just are and will remain so, but a lot of problems were one particular set of editors beating on each other and not a general social or topic issue. Those go away over time.
Thanks for the effort in doing that, Tony.
Tony gets the Gary Cooper award for this week. Or in particular the "Meet John Doe" award http://knol.google.com/k/chair-potato/gary-cooper-movies-on-youtube/hyujx7mc...
-----Original Message----- From: Tony Sidaway tonysidaway@gmail.com To: English Wikipedia wikien-l@lists.wikimedia.org Sent: Fri, Sep 4, 2009 4:55 pm Subject: Re: [WikiEN-l] Putting some perspective on the end of Wikipedia
I'm undertaking to have all article and talk page semiprotections on Wikipedia reviewed. The process I'm using is to enter a brief proposal on the article talk page and contact the protecting sysop. The idea is that we discuss whether to unprotect the article or talk page and watch it vigilantly.
This has already met considerable success, with more 30% of the proposals I've made this evening being enacted upon. There appear to be a lot of semiprotections that have simply been forgotten by the original sysop.
I'll keep this up until I either run out of articles to review or get bored. Since there are several thousand semiprotected article the latter is more likely to happen first.
Gwern Branwen wonders whether semiprotections have taken over from protections. Well one cannot really compare the current Wikipedia with the Wikipedia of 2005. Then we had no real way of dealing with biographies of living persons, and little awareness of the problem, and as for the protected articles, they numbered dozens at the most, and certainly not thousands. It's important to strike a balance. While many of the semiprotected pages may actually be redirects that we wouldn't normally want to see edited by unregistered users, I suspect many are not. It's always a good idea to review the situation regularly.
_______________________________________________ WikiEN-l mailing list WikiEN-l@lists.wikimedia.org To unsubscribe from this mailing list, visit: https://lists.wikimedia.org/mailman/listinfo/wikien-l
Tony Sidaway wrote:
I'm undertaking to have all article and talk page semiprotections on Wikipedia reviewed. The process I'm using is to enter a brief proposal on the article talk page and contact the protecting sysop. The idea is that we discuss whether to unprotect the article or talk page and watch it vigilantly.
This has already met considerable success, with more 30% of the proposals I've made this evening being enacted upon. There appear to be a lot of semiprotections that have simply been forgotten by the original sysop.
I'll keep this up until I either run out of articles to review or get bored. Since there are several thousand semiprotected article the latter is more likely to happen first.
Gwern Branwen wonders whether semiprotections have taken over from protections. Well one cannot really compare the current Wikipedia with the Wikipedia of 2005. Then we had no real way of dealing with biographies of living persons, and little awareness of the problem, and as for the protected articles, they numbered dozens at the most, and certainly not thousands. It's important to strike a balance. While many of the semiprotected pages may actually be redirects that we wouldn't normally want to see edited by unregistered users, I suspect many are not. It's always a good idea to review the situation regularly.
I agree that indefinite semi-protection is inimical to the purpose of the encyclopedia and should be subject to periodical review, so I regard this as a beneficial initiative. I wonder, however, how many of those articles are then edited by (a) editors who set up accounts and become auto-confirmed in order to do so, or (b) subject of {{editprotected}} requests on their talk pages. I took a quick look the other day at the categories of unsourced articles, which go back to December 2006; to be honest, I don't currently have the time or will myself to trawl through what is a Sisyphean task. Even limiting that to BLP articles is more than enough to tax the stamina of most volunteer editors. It's easy enough to begin a stub, and as easy to tag as unsourced, but it does take some commitment to take the bricks and fashion a mansion, which I think we should be doing.
On 9/5/09, Phil Nash pn007a2145@blueyonder.co.uk wrote:
I took a quick look the other day at the categories of unsourced
articles, which go back to December 2006; to be honest, I don't currently have the time or will myself to trawl through what is a Sisyphean task. Even limiting that to BLP articles is more than enough to tax the stamina of most volunteer editors. It's easy enough to begin a stub, and as easy to tag as unsourced, but it does take some commitment to take the bricks and fashion a mansion, which I think we should be doing.
I feel strongly that biographies of living people without sources should be deleted on sight. They can always be recreated by someone who possesses at least one reliable source.
Tony is right that these lists of long-term and indefinitely protected or semi-protected pages should be reviewed periodically. The place to find this information is here: http://en.wikipedia.org/wiki/Wikipedia:Database_reports
There are about 3000 indefinitely permanently protected talk pages; they are almost all user talk pages and were protected at the time that the account was blocked. Most of those can be unprotected. They run back to 2006.
There are 39 indefinitely fully protected article titles, the vast majority of which are soft redirects to Wiktionary or pages salted to prevent recreation. For the others, most are quite recent, and it would probably be appropriate to ask the protecting admin to review and, at minimum, set an end-date. In addition, there are 1478 indefinitely protected redirects, many of them to prevent forking.
There are 1900+ indefinitely semiprotected articles, with many of them indicating they have been repeated vandalism targets. These include articles on recent US presidents, certain high profile musicians, politically charged subjects, and those with a wide and opinionated fandom. These should, of course, be periodically reviewed; however, if someone decides to unprotect many of these articles, I would hope they don't just keep it on their watchlist but actively review new edits regularly for a few weeks afterward.
There are also 300+ indefinitely semiprotected redirects, which include repeatedly recreated articles previously deemed inappropriate, and titles associted with attempts to fork articles. These might bear review as well, either with a move up to full protection or semiprotection lifted on a trial basis, but again they would need to be monitored closely if they are unprotected.
Of the approximately 400 talk pages and talk page redirects that are indefinitely semi-protected, almost all are user talk pages, many of admins who carry out antivandal work. There were about 30 article talk pages indefinitely semi-protected before Tony carried out his review, and there are quite a bit fewer now.
There are some opportunities to improve practices here, and to really take a look and decide which articles (and rarely, article talk pages) need this indefinite protection. At the same time, I really do believe that if an admin is going to reduce protection on a page with an extensive history of problems, he or she has a responsibility to keep an eye on the page for at least a couple of weeks afterward to ensure there isn't a fresh outbreak of inappropriate behaviour. Since so many of the articles involved are BLPs, and even on non-BLPs the problems were related to inappropriate addition of information about LPs, this is an area where special sensitivity is required.
Risker
On 9/5/09, Risker risker.wp@gmail.com wrote:
There are some opportunities to improve practices here, and to really take a look and decide which articles (and rarely, article talk pages) need this indefinite protection. At the same time, I really do believe that if an admin is going to reduce protection on a page with an extensive history of problems, he or she has a responsibility to keep an eye on the page for at least a couple of weeks afterward to ensure there isn't a fresh outbreak of inappropriate behaviour. Since so many of the articles involved are BLPs, and even on non-BLPs the problems were related to inappropriate addition of information about LPs, this is an area where special sensitivity is required.
I've done a tiny bit of work by examining some 60 semiprotected pages in article-space, most of which turned out to be redirects. There are some obvious articles to keep semiprotected: those that are magnets for vandalism by their nature, those that have been protected under an OTRS ticket, and those that are known to be targeted for long term abuse.
Of the remainder, I've initiated reviews of 9 semiprotected articles, contacting the protecting admin and starting a discussion on the article talk page. One review has been completed with the decision to retain semiprotection because the vandal is known to be still around and unblockable because of dynamic IP issues.
I was away and missed the FR discussions, but I have to say this: the vanishing point is nowhere in sight!
Charles
On Sat, Sep 5, 2009 at 11:39 AM, Charles Matthewscharles.r.matthews@ntlworld.com wrote:
I was away and missed the FR discussions, but I have to say this: the vanishing point is nowhere in sight!
"vanishing point"?
Carcharoth
On 9/5/09, Charles Matthews charles.r.matthews@ntlworld.com wrote:
I was away and missed the FR discussions, but I have to say this: the vanishing point is nowhere in sight!
FR?
(Racks brains).
I assume you mean flagged revisions?
Tony Sidaway wrote:
On 9/5/09, Charles Matthews charles.r.matthews@ntlworld.com wrote:
I was away and missed the FR discussions, but I have to say this: the vanishing point is nowhere in sight!
FR?
(Racks brains).
I assume you mean flagged revisions?
Got it in one! Oh, and "vanishing point" is a term in perspective drawing. Just ignore me if the opacity get unbearable, though.
Charles
Risker wrote:
There are some opportunities to improve practices here, and to really take a look and decide which articles (and rarely, article talk pages) need this indefinite protection. At the same time, I really do believe that if an admin is going to reduce protection on a page with an extensive history of problems, he or she has a responsibility to keep an eye on the page for at least a couple of weeks afterward to ensure there isn't a fresh outbreak of inappropriate behaviour.
Agree with both points, naturally. But the discussion as a whole seems to indicate that "protection" has become one of our more Byzantine concepts. Some work ought to go on, simplifying it from a hypertext stance (categorisation and tagging), so that what happens is more transparent. Anyone interested in reviewing the system and writing an on-site essay?
Charles
2009/9/5 Gwern Branwen gwern0@gmail.com:
Would it be possible for you to do a comparison with Wikipedia just before semiprotection was enabled? I've long wanted to know whether the argument that semiprotections would replace full protections holds any water.
Such a comparison should be possible. We have protection logs, we should be able to go through them and work out how many articles were (semi-)protected at any given time. A graph of that data would reveal any obvious impact of the introduction of semi protection (work out a trend from the data before the introduction and see if the data after it fits it or not).
Is there a machine readable version of the protection log anywhere?
On 9/5/09, Thomas Dalton thomas.dalton@gmail.com wrote:
Is there a machine readable version of the protection log anywhere?
You could probably use the MediaWiki API to do this:
http://www.mediawiki.org/wiki/API
Alternatively find some of the following regular reports easy to parse:
2009/9/6 Tony Sidaway tonysidaway@gmail.com:
On 9/5/09, Thomas Dalton thomas.dalton@gmail.com wrote:
Is there a machine readable version of the protection log anywhere?
You could probably use the MediaWiki API to do this:
http://www.mediawiki.org/wiki/API
Alternatively find some of the following regular reports easy to parse:
I've just found that there is a database dump of all the logs. It is nearly a gig for enwiki, but I can manage that. I'm downloading it now and will try and get some useful stats out of it.
About two-thirds of the 30 semiprotected talk pages (and in some case the associated articles) have been unprotected. One of the articles was reprotected after immediate spamming (to be fair, it *is* called "poop", so I was obviously very naive there!) The others are still unprotected.
I've made slow, steady progress on the semiprotection reviews. You can see the results on my talk page.
http://en.wikipedia.org/wiki/User_talk:Tony_Sidaway#Review_of_semiprotected_...
At the moment I'm dealing with the older indefinite semiprotections--those that have survived for two years--so quite a high proportion turn out to be obvious vandalism magnets or biographies of living persons that have encountered severe enough problems to become semiprotected.
Of the remainder, about 18 out of 100 articles looked due for review so I put a note on each talk page and contacted the protecting admins. Of those 17 reviews, I withdrew one on further information, one has been declined by the sysop, three quickly reached consensus to keep protection, and four have been unprotected. leaving ten articles still under review.
This is very slow work but I think we should probably do it from time to time, as part of the principle of keeping Wikipedia an open encyclopedia.
On 9/6/09, Thomas Dalton thomas.dalton@gmail.com wrote:
2009/9/6 Tony Sidaway tonysidaway@gmail.com:
On 9/5/09, Thomas Dalton thomas.dalton@gmail.com wrote:
Is there a machine readable version of the protection log anywhere?
You could probably use the MediaWiki API to do this:
http://www.mediawiki.org/wiki/API
Alternatively find some of the following regular reports easy to parse:
I've just found that there is a database dump of all the logs. It is nearly a gig for enwiki, but I can manage that. I'm downloading it now and will try and get some useful stats out of it.
WikiEN-l mailing list WikiEN-l@lists.wikimedia.org To unsubscribe from this mailing list, visit: https://lists.wikimedia.org/mailman/listinfo/wikien-l
On Friday 04 September 2009, Gwern Branwen wrote:
Would it be possible for you to do a comparison with Wikipedia just before semiprotection was enabled? I've long wanted to know whether the argument that semiprotections would replace full protections holds any water.
This would also seem to be quite important to know for flagged, inasmuch as that argument has been recycled for flagging pages...
I'm not a DB/server-tools whiz, so I just did my analysis from a wiki page to get a gloss and hopefully prompt someone who knows what they are doing to correct and extend it. My analysis was prompted because of all the discussion of flagging and I very much second your question and hope someone will step forth.
I'm also cc:ing wiki-research and will refer to my brief analysis [1] in the hopes that a DB/quant person would be interested in the question.
[1]:http://reagle.org/joseph/blog/social/wikipedia/protection-figures