http://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2009-07-06/Feature...
I couldn't help but notice: * Five articles were promoted to featured status this week * Four articles were delisted this week. * Twelve lists were promoted to featured status this week * Eight lists were delisted this week
What a lot of churn. So the overall rate was merely +1 FA, +4 FL (and also 3 topics and three images).
Is it always this bad?
Steve
On Sun, Jul 12, 2009 at 7:29 PM, Steve Bennettstevagewp@gmail.com wrote:
What a lot of churn. So the overall rate was merely +1 FA, +4 FL (and also 3 topics and three images).
Is it always this bad?
My long-time observation is that the people who obsess about FA over the long term want to keep the number of articles with that status approximately constant by making the standards more and more difficult to meet.
-Matt
2009/7/13 Matthew Brown morven@gmail.com:
My long-time observation is that the people who obsess about FA over the long term want to keep the number of articles with that status approximately constant by making the standards more and more difficult to meet.
Many have stated this directly on WT:FAC.
- d.
On Mon, Jul 13, 2009 at 5:01 PM, Matthew Brownmorven@gmail.com wrote:
My long-time observation is that the people who obsess about FA over the long term want to keep the number of articles with that status approximately constant by making the standards more and more difficult to meet.
Yeah, we see that on FPC (featured pics) - and RfA (admins) for that matter. There's probably a term for this somewhere. I don't think it's malicious, but a fact that when you constantly review stuff, you get jaded, and compare each item to all the great examples in the past. It's almost like a drug, you need bigger highs each time to register. Or maybe it's just perfectionism - it's very easy to quibble over tiny flaws, and miss the bigger picture.
Here's a great example: http://en.wikipedia.org/wiki/Wikipedia:Featured_picture_candidates/Face_of_a...
What an incredible image. This is a *wasp*, and we have great detail of the *hairs* on its forehead. Stunning sharpness, and this photo would not be out of place in a good science magazine. Yet two editors managed to oppose its promotion to "featured" on the basis of the tip of one antenna being obscured by an out of focus leaf fragment. Another, neutral, came up with "An amazing detail and sharpness...with a clumsy framing and cropping ruining an otherwise excellent picture. ... I will not support the promotion as I find little excuse for those flaws."
These would be perfectly apt comments if we were voting on National Geographic's "photo of the year". But Wikipedia "featured picture"? Whee.
Steve
On Mon, Jul 13, 2009 at 9:02 PM, Steve Bennett stevagewp@gmail.com wrote:.
Here's a great example:
http://en.wikipedia.org/wiki/Wikipedia:Featured_picture_candidates/Face_of_a...
What an incredible image. This is a *wasp*, and we have great detail of the *hairs* on its forehead. Stunning sharpness, and this photo would not be out of place in a good science magazine. Yet two editors managed to oppose its promotion to "featured" on the basis of the tip of one antenna being obscured by an out of focus leaf fragment. Another, neutral, came up with "An amazing detail and sharpness...with a clumsy framing and cropping ruining an otherwise excellent picture. ... I will not support the promotion as I find little excuse for those flaws."
These would be perfectly apt comments if we were voting on National Geographic's "photo of the year". But Wikipedia "featured picture"? Whee.
You should ask Durova about featured image reviews - she had a live one not long ago. Photograph of a moon (Eros perhaps?) that was the best that anyone could possibly take with current (government) technology, but it was opposed for reasons more suited to critiquing everyday items in posed situations.
Nathan
2009/7/14 Nathan nawrich@gmail.com:
On Mon, Jul 13, 2009 at 9:02 PM, Steve Bennett stevagewp@gmail.com wrote:.
Here's a great example:
http://en.wikipedia.org/wiki/Wikipedia:Featured_picture_candidates/Face_of_a...
What an incredible image. This is a *wasp*, and we have great detail of the *hairs* on its forehead. Stunning sharpness, and this photo would not be out of place in a good science magazine. Yet two editors managed to oppose its promotion to "featured" on the basis of the tip of one antenna being obscured by an out of focus leaf fragment. Another, neutral, came up with "An amazing detail and sharpness...with a clumsy framing and cropping ruining an otherwise excellent picture. ... I will not support the promotion as I find little excuse for those flaws."
These would be perfectly apt comments if we were voting on National Geographic's "photo of the year". But Wikipedia "featured picture"? Whee.
You should ask Durova about featured image reviews - she had a live one not long ago. Photograph of a moon (Eros perhaps?) that was the best that anyone could possibly take with current (government) technology, but it was opposed for reasons more suited to critiquing everyday items in posed situations.
Nathan
Titan the image was http://en.wikipedia.org/wiki/File:Titan_globe.jpg which rather runs into issues with the existence of the more recent http://en.wikipedia.org/wiki/File:Titan2005.jpg
Saturn's moon Triton; not my nomination. That delisting nomination was a particularly bad example of two trends: FPC reviewers failing to read the article for encyclopedic context, and the valued pictures program functioning as a parasitic growth upon the FP program. VP ought to be casting a broader net and building its own base of support, rather than trying to siphon the most encyclopedic images out of FP. At the same time as Triton was nominated, VP enthusiasts tried to delist the atomic bombing of Nagasaki.
-Durova
On Mon, Jul 13, 2009 at 6:38 PM, Nathan nawrich@gmail.com wrote:
On Mon, Jul 13, 2009 at 9:02 PM, Steve Bennett stevagewp@gmail.com wrote:.
Here's a great example:
http://en.wikipedia.org/wiki/Wikipedia:Featured_picture_candidates/Face_of_a...
What an incredible image. This is a *wasp*, and we have great detail of the *hairs* on its forehead. Stunning sharpness, and this photo would not be out of place in a good science magazine. Yet two editors managed to oppose its promotion to "featured" on the basis of the tip of one antenna being obscured by an out of focus leaf fragment. Another, neutral, came up with "An amazing detail and sharpness...with a clumsy framing and cropping ruining an otherwise excellent picture. ... I will not support the promotion as I find little excuse for those flaws."
These would be perfectly apt comments if we were voting on National Geographic's "photo of the year". But Wikipedia "featured picture"? Whee.
You should ask Durova about featured image reviews - she had a live one not long ago. Photograph of a moon (Eros perhaps?) that was the best that anyone could possibly take with current (government) technology, but it was opposed for reasons more suited to critiquing everyday items in posed situations.
Nathan _______________________________________________ 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 Mon, Jul 13, 2009 at 9:02 PM, Steve Bennettstevagewp@gmail.com wrote:
On Mon, Jul 13, 2009 at 5:01 PM, Matthew Brownmorven@gmail.com wrote:
My long-time observation is that the people who obsess about FA over the long term want to keep the number of articles with that status approximately constant by making the standards more and more difficult to meet.
Yeah, we see that on FPC (featured pics) - and RfA (admins) for that matter. There's probably a term for this somewhere. I don't think it's malicious, but a fact that when you constantly review stuff, you get jaded, and compare each item to all the great examples in the past. It's almost like a drug, you need bigger highs each time to register. Or maybe it's just perfectionism - it's very easy to quibble over tiny flaws, and miss the bigger picture.
Agreed -- I don't think the numerical counts of featured content mean very much in terms of measuring quality improvement overall, because of this effect. The showcase pieces, the ones that do get past the ever-increasing hurdles, are great -- and I'm glad we have a process for identifying them and bringing them into wider public view, both because the creators deserve the recognition and because the public ought to see it. But tracking the number doesn't give veyr much information except as a comment on the process itself.
-Kat
Kat Walsh wrote:
The showcase pieces, the ones that do get past the ever-increasing hurdles, are great -- and I'm glad we have a process for identifying them and bringing them into wider public view, both because the creators deserve the recognition and because the public ought to see it. But tracking the number doesn't give veyr much information except as a comment on the process itself.
Yes, in a way it's sad we have so much attention on gatekeeping and absolute standards (well, as has been said, high and upwardly mobile standards), and so little recognition on great added-value edits, the ones which take an article into a different class of usefulness. As far as I know the transition from B class to A class is still considered to be the most transforming, from the reader's point of view.
Charles
On Tue, Jul 14, 2009 at 10:38 AM, Charles Matthewscharles.r.matthews@ntlworld.com wrote:
Kat Walsh wrote:
The showcase pieces, the ones that do get past the ever-increasing hurdles, are great -- and I'm glad we have a process for identifying them and bringing them into wider public view, both because the creators deserve the recognition and because the public ought to see it. But tracking the number doesn't give veyr much information except as a comment on the process itself.
Yes, in a way it's sad we have so much attention on gatekeeping and absolute standards (well, as has been said, high and upwardly mobile standards), and so little recognition on great added-value edits, the ones which take an article into a different class of usefulness. As far as I know the transition from B class to A class is still considered to be the most transforming, from the reader's point of view.
Absolutely. And also the transition from unintelligibility (or poor stub or no article) to something with more frameworking and attempts at comprehensiveness. A *good* transition from stub to start-class can be vital to the future "health" of an article, just as a bad transition (or poor initial creation) can lead to trouble later.
Indeed, something looking at the traffic and flow of articles up this quality scale would be good. I think the main thing discouraging people from doing that is the unreliability of the assessments outside of FA, A and GA. But in any topic area, making clear what the "internal" standards are for stub, start, C, B, and A class articles and sorting articles into those categories is good, but it HAS to be followed up with attempts at improving the articles. Otherwise it has been an exercise in paperwork, in the (sometimes vain) hope that a writer will come along and improve the articles.
Maybe a wikiproject with a strong and reliable history of *both* assessment and improvement of articles could give a narrative and timeline of how their articles have improved?
Carcharoth
Carcharoth wrote
Indeed, something looking at the traffic and flow of articles up this quality scale would be good. I think the main thing discouraging people from doing that is the unreliability of the assessments outside of FA, A and GA. But in any topic area, making clear what the "internal" standards are for stub, start, C, B, and A class articles and sorting articles into those categories is good, but it HAS to be followed up with attempts at improving the articles.
I think you're probably right that a new departure needs to be made: we're at best mediocre at devising new "recognition mechanisms". How about a project aimed (since we are coming up to three million articles) at shifting the balance of stubs and other really substandard articles before we get to four million? We'll get to the end of the first decade of WP in 2011 before that happens. And I think we need some kind of two-dimensional plot, not single scale: urgency assessment as well as quality assessment.
I do spend more time on upgrading stubs than I used to, and I guess this will be true of anyone who is driven by what they find on the site. When we last discussed total article numbers, four million seemed a good enough guess for the "plateau".
Charles
2009/7/14 Charles Matthews charles.r.matthews@ntlworld.com:
I think you're probably right that a new departure needs to be made: we're at best mediocre at devising new "recognition mechanisms". How about a project aimed (since we are coming up to three million articles) at shifting the balance of stubs and other really substandard articles before we get to four million? We'll get to the end of the first decade of WP in 2011 before that happens. And I think we need some kind of two-dimensional plot, not single scale: urgency assessment as well as quality assessment.
I fear the first thing that would spring to the community's beautiful collective mind would be a mass deletion of all stubs.
- d.
David Gerard wrote:
2009/7/14 Charles Matthews charles.r.matthews@ntlworld.com:
I think you're probably right that a new departure needs to be made: we're at best mediocre at devising new "recognition mechanisms". How about a project aimed (since we are coming up to three million articles) at shifting the balance of stubs and other really substandard articles before we get to four million? We'll get to the end of the first decade of WP in 2011 before that happens. And I think we need some kind of two-dimensional plot, not single scale: urgency assessment as well as quality assessment.
I fear the first thing that would spring to the community's beautiful collective mind would be a mass deletion of all stubs.
That's the issue with one-dimensional thinking, certainly.
Charles
2009/7/14 David Gerard dgerard@gmail.com:
I fear the first thing that would spring to the community's beautiful collective mind would be a mass deletion of all stubs.
I have, interestingly, been noticing it moving in exactly the opposite direction; articles with a couple of paragraphs of text, a reference or two, an image or an infobox, being marked as "stubs". There's standards inflation at both ends of the rating system...
On Fri, Jul 17, 2009 at 5:06 AM, Andrew Grayandrew.gray@dunelm.org.uk wrote:
I have, interestingly, been noticing it moving in exactly the opposite
direction; articles with a couple of paragraphs of text, a reference
or two, an image or an infobox, being marked as "stubs". There's standards inflation at both ends of the rating system...
IMHO, this kind of thing is one of Wikipedia's greatest failings. We still can't even agree on a definition of things like "stub", and it seems to be in everyone's interest not to. People like stuff like that being subjective.
(FWIW, I think it's reasonable to have "stub" be relative to the expected content. Two paragraphs on a country would clearly be a "stub". Two paragraphs on an obscure medieval scribe might be the most comprehensive resource possible.)
Steve
Steve Bennett wrote:
On Fri, Jul 17, 2009 at 5:06 AM, Andrew Grayandrew.gray@dunelm.org.uk wrote:
I have, interestingly, been noticing it moving in exactly the opposite
direction; articles with a couple of paragraphs of text, a reference
or two, an image or an infobox, being marked as "stubs". There's standards inflation at both ends of the rating system...
IMHO, this kind of thing is one of Wikipedia's greatest failings. We still can't even agree on a definition of things like "stub", and it seems to be in everyone's interest not to. People like stuff like that being subjective.
(FWIW, I think it's reasonable to have "stub" be relative to the expected content. Two paragraphs on a country would clearly be a "stub". Two paragraphs on an obscure medieval scribe might be the most comprehensive resource possible.)
The stub business goes back almost forever, though. And the affection for grey areas is not the dominant trend: there are people who seem to have the MoS and its pickier points as bedtime reading. There has always been an adequate definition of stub, which relates to the idea that the article as stands has serious missing information, so is incomplete in an essential way. So Steve's FWIW is correct (no, I haven't looked up to see whether some genius has changed the definition of stub). I've never taken much notice of what is and isn't denominated a stub.
Charles
On 14/07/2009, Charles Matthews charles.r.matthews@ntlworld.com wrote:
I do spend more time on upgrading stubs than I used to, and I guess this will be true of anyone who is driven by what they find on the site. When we last discussed total article numbers, four million seemed a good enough guess for the "plateau".
It's looking to me like 3.5 million is about the plateau, since the curve is bang on that, but we might make 4 million *eventually*.
http://en.wikipedia.org/wiki/Wikipedia:Modelling_Wikipedia%27s_growth#Logist...
Charles
Ian Woollard wrote:
It's looking to me like 3.5 million is about the plateau, since the curve is bang on that, but we might make 4 million *eventually*.
http://en.wikipedia.org/wiki/Wikipedia:Modelling_Wikipedia%27s_growth#Logist...
We'll know more around the beginning of 2010. In my view something is likely to change in the direction of people valuing lists of "missing articles" more, when it is clearer that drive-by creation is getting drossier by the month (which is what that model implies). Of course I can't quantify that: I know it is still easy to come up with sets of 1000 topics that we don't cover at all well, and the total of redlinks is still large.
Charles
2009/7/14 Charles Matthews charles.r.matthews@ntlworld.com:
Ian Woollard wrote:
It's looking to me like 3.5 million is about the plateau, since the curve is bang on that, but we might make 4 million *eventually*.
http://en.wikipedia.org/wiki/Wikipedia:Modelling_Wikipedia%27s_growth#Logist...
We'll know more around the beginning of 2010. In my view something is likely to change in the direction of people valuing lists of "missing articles" more, when it is clearer that drive-by creation is getting drossier by the month (which is what that model implies). Of course I can't quantify that: I know it is still easy to come up with sets of 1000 topics that we don't cover at all well, and the total of redlinks is still large.
Charles
Redlinks in general perhaps. Redlinks in articles a significant number of people actually read less so.
On Tue, Jul 14, 2009 at 16:03, genigeniice@gmail.com wrote:
2009/7/14 Charles Matthews charles.r.matthews@ntlworld.com:
Ian Woollard wrote:
It's looking to me like 3.5 million is about the plateau, since the curve is bang on that, but we might make 4 million *eventually*.
http://en.wikipedia.org/wiki/Wikipedia:Modelling_Wikipedia%27s_growth#Logist...
We'll know more around the beginning of 2010. In my view something is likely to change in the direction of people valuing lists of "missing articles" more, when it is clearer that drive-by creation is getting drossier by the month (which is what that model implies). Of course I can't quantify that: I know it is still easy to come up with sets of 1000 topics that we don't cover at all well, and the total of redlinks is still large.
Charles
Redlinks in general perhaps. Redlinks in articles a significant number of people actually read less so.
Redlinks are likely to be a poor estimate of numbers of "missing" articles anyway. Some will be to articles that would be non-notable, and redlinks tend to be removed - in other words links that would be present if we had the article aren't there as redlinks.
-- geni
WikiEN-l mailing list WikiEN-l@lists.wikimedia.org To unsubscribe from this mailing list, visit: https://lists.wikimedia.org/mailman/listinfo/wikien-l
sineWAVE wrote:
Redlinks are likely to be a poor estimate of numbers of "missing" articles anyway. Some will be to articles that would be non-notable, and redlinks tend to be removed - in other words links that would be present if we had the article aren't there as redlinks.
Who are these people removing redlinks? They need a slap.
Charles
On Tue, Jul 14, 2009 at 5:08 PM, Charles Matthews < charles.r.matthews@ntlworld.com> wrote:
sineWAVE wrote:
Redlinks are likely to be a poor estimate of numbers of "missing" articles anyway. Some will be to articles that would be non-notable, and redlinks tend to be removed - in other words links that would be present if we had the article aren't there as redlinks.
Who are these people removing redlinks? They need a slap.
Charles
Depends where the links were going to.
geni wrote:
We'll know more around the beginning of 2010. In my view something is likely to change in the direction of people valuing lists of "missing articles" more, when it is clearer that drive-by creation is getting drossier by the month (which is what that model implies). Of course I can't quantify that: I know it is still easy to come up with sets of 1000 topics that we don't cover at all well, and the total of redlinks is still large.
Charles
Redlinks in general perhaps. Redlinks in articles a significant number of people actually read less so.
Well, now we come to it: one reason there may be less growth is the the nature of database use (people's queries tend to have less of a "long tail" than our entries). OTOH: I started the [[Oxford Professor of Poetry]] article, and had no idea there would be a media frenzy about it (last time had been Yevtushenko). I also started [[Ruth Padel]] ... when said frenzy arose I did fill in the redlinks as I could, including [[Joseph Trapp]], first ever Oxford Professor of Poetry (a few interesting things there, but for another time).
Charles
On Tue, Jul 14, 2009 at 10:07 AM, Ian Woollardian.woollard@gmail.com wrote:
It's looking to me like 3.5 million is about the plateau, since the curve is bang on that, but we might make 4 million *eventually*.
http://en.wikipedia.org/wiki/Wikipedia:Modelling_Wikipedia%27s_growth#Logist...
I don't think the bell-shaped articles/day curve of the logistic model is a good description of the trends. Since article creation peaked in 2007, the falloff in article creation has been much slower than than ramp-up. Rather than falling back to close to zero articles/day over the next 5 years or so (as the logistic model predicts), it looks like we're heading to an asymptote of (I'm eyeballing it here) around 1000 articles/day. I expect 4 million articles a lot sooner than *eventually*. ;)
The significant non-zero asymptote makes sense to me in terms of what kinds of articles remain to be written: *High-profile topics that aren't notable yet. The news generates a steady stream of newly notable topics that a wide variety of people want to write about. *The vast pool of notable topics (tens of millions, according to some back-of-the-envelope estimates that have made the rounds here before) for which article creation rate is limited by the number of interested/knowledgeable writers.
Between those classes, I'd be surprised if yearly growth fell below 300,000 any time soon.
-Sage
On 14/07/2009, Sage Ross ragesoss+wikipedia@gmail.com wrote:
I don't think the bell-shaped articles/day curve of the logistic model is a good description of the trends. Since article creation peaked in 2007, the falloff in article creation has been much slower than than ramp-up. Rather than falling back to close to zero articles/day over the next 5 years or so (as the logistic model predicts), it looks like we're heading to an asymptote of (I'm eyeballing it here) around 1000 articles/day. I expect 4 million articles a lot sooner than *eventually*. ;)
I don't see any evidence for an asymptote at all yet.
We're only about ~1300 per day now, and the trend is clearly downwards, on a *log* graph of *percentage* growth against time it's a straightish line downwards, and the size of the wiki seems to be plateauing; percentage growth is a quarter what it was two years ago, and the wiki is only 1/3 bigger.
IMO we're probably going to be under a 1000 per day by Christmas. If we do manage 4 million I don't think it will be in the next 4 years.
-Sage
2009/7/14 Ian Woollard ian.woollard@gmail.com:
I don't see any evidence for an asymptote at all yet. We're only about ~1300 per day now, and the trend is clearly downwards, on a *log* graph of *percentage* growth against time it's a straightish line downwards, and the size of the wiki seems to be plateauing; percentage growth is a quarter what it was two years ago, and the wiki is only 1/3 bigger. IMO we're probably going to be under a 1000 per day by Christmas. If we do manage 4 million I don't think it will be in the next 4 years.
Here's a question: how many articles are created and deleted within 24 hours?
- d.
On Tue, Jul 14, 2009 at 4:50 PM, David Gerarddgerard@gmail.com wrote:
2009/7/14 Ian Woollard ian.woollard@gmail.com:
I don't see any evidence for an asymptote at all yet. We're only about ~1300 per day now, and the trend is clearly downwards, on a *log* graph of *percentage* growth against time it's a straightish line downwards, and the size of the wiki seems to be plateauing; percentage growth is a quarter what it was two years ago, and the wiki is only 1/3 bigger. IMO we're probably going to be under a 1000 per day by Christmas. If we do manage 4 million I don't think it will be in the next 4 years.
Here's a question: how many articles are created and deleted within 24 hours?
Are you saying the numbers could go negative?? Contraction in real-terms? :-/
Carcharoth
2009/7/14 Carcharoth carcharothwp@googlemail.com:
Are you saying the numbers could go negative?? Contraction in real-terms? :-/
Carcharoth
It's happened at least once. Long term it would be unlikely since most deletions are of new articles.
On Tue, Jul 14, 2009 at 5:00 PM, genigeniice@gmail.com wrote:
2009/7/14 Carcharoth carcharothwp@googlemail.com:
Are you saying the numbers could go negative?? Contraction in real-terms? :-/
Carcharoth
It's happened at least once. Long term it would be unlikely since most deletions are of new articles.
Was it when all those algae articles got deleted?
Carcharoth
2009/7/14 Carcharoth carcharothwp@googlemail.com:
On Tue, Jul 14, 2009 at 5:00 PM, genigeniice@gmail.com wrote:
2009/7/14 Carcharoth carcharothwp@googlemail.com:
Are you saying the numbers could go negative?? Contraction in real-terms? :-/
Carcharoth
It's happened at least once. Long term it would be unlikely since most deletions are of new articles.
Was it when all those algae articles got deleted?
Yes. I doubt there are many other cases like that if any.
If it does finally plateau half the days will be negative of course; and they'll become more common before we reach the plateau just due to randomness. But if we start having negative weeks, stick a fork in her, she's probably done!
Do we have any plans for when we'll be taking the Wikipedia out of beta? ;-)
On 14/07/2009, geni geniice@gmail.com wrote:
2009/7/14 Carcharoth carcharothwp@googlemail.com:
On Tue, Jul 14, 2009 at 5:00 PM, genigeniice@gmail.com wrote:
2009/7/14 Carcharoth carcharothwp@googlemail.com:
Are you saying the numbers could go negative?? Contraction in real-terms? :-/
Carcharoth
It's happened at least once. Long term it would be unlikely since most deletions are of new articles.
Was it when all those algae articles got deleted?
Yes. I doubt there are many other cases like that if any.
-- geni
WikiEN-l mailing list WikiEN-l@lists.wikimedia.org To unsubscribe from this mailing list, visit: https://lists.wikimedia.org/mailman/listinfo/wikien-l
Ian Woollard wrote:
If it does finally plateau half the days will be negative of course; and they'll become more common before we reach the plateau just due to randomness. But if we start having negative weeks, stick a fork in her, she's probably done!
Do we have any plans for when we'll be taking the Wikipedia out of beta? ;-)
I think we should do a bot run with census data from 2001 for every tiny place in India first, and get those articles cleaned up, before we announce that the project has got to where it's going. Actually 3,750,000 articles sounds like a consensus figure for a couple of years out ... it will be interesting to see what trends start looking significant as we get closer.
Charles
On Tue, Jul 14, 2009 at 12:00 PM, genigeniice@gmail.com wrote:
2009/7/14 Carcharoth carcharothwp@googlemail.com:
Are you saying the numbers could go negative?? Contraction in real-terms? :-/
Carcharoth
It's happened at least once. Long term it would be unlikely since most deletions are of new articles.
Was that when the faulty bot-created algae articles were deleted? There were about 4000, so that would definitely be a net negative day if they were deleted at once.
-Sage
On Tue, Jul 14, 2009 at 5:03 PM, Sage Rossragesoss+wikipedia@gmail.com wrote:
On Tue, Jul 14, 2009 at 12:00 PM, genigeniice@gmail.com wrote:
2009/7/14 Carcharoth carcharothwp@googlemail.com:
Are you saying the numbers could go negative?? Contraction in real-terms? :-/
Carcharoth
It's happened at least once. Long term it would be unlikely since most deletions are of new articles.
Was that when the faulty bot-created algae articles were deleted? There were about 4000, so that would definitely be a net negative day if they were deleted at once.
As geni has confirmed, it seems that was the one.
Maybe it is time for another comedy scene depicting the first day of the wiki and the nascent conflict between deletionism and inclusionism?
Anyone remember where the old humour piece is?
Carcharoth
On Tue, Jul 14, 2009 at 5:06 PM, Carcharothcarcharothwp@googlemail.com wrote:
On Tue, Jul 14, 2009 at 5:03 PM, Sage Rossragesoss+wikipedia@gmail.com wrote:
On Tue, Jul 14, 2009 at 12:00 PM, genigeniice@gmail.com wrote:
2009/7/14 Carcharoth carcharothwp@googlemail.com:
Are you saying the numbers could go negative?? Contraction in real-terms? :-/
Carcharoth
It's happened at least once. Long term it would be unlikely since most deletions are of new articles.
Was that when the faulty bot-created algae articles were deleted? There were about 4000, so that would definitely be a net negative day if they were deleted at once.
As geni has confirmed, it seems that was the one.
Maybe it is time for another comedy scene depicting the first day of the wiki and the nascent conflict between deletionism and inclusionism?
Anyone remember where the old humour piece is?
Sideshow, sideshow only, remember.
Then back to the main thread.
Carcharoth
On Tue, Jul 14, 2009 at 5:06 PM, Carcharothcarcharothwp@googlemail.com wrote:
<snip>
Maybe it is time for another comedy scene depicting the first day of the wiki and the nascent conflict between deletionism and inclusionism?
http://en.wikipedia.org/wiki/User:Jaranda/Wikipedia%27s_first_IRC_chat
Took me long enough to find it! And it wasn't what I thought it was. No deletionism or inclusionism jokes there.
Carcharoth
On Tue, Jul 14, 2009 at 12:38 PM, Carcharothcarcharothwp@googlemail.com wrote:
http://en.wikipedia.org/wiki/User:Jaranda/Wikipedia%27s_first_IRC_chat
Took me long enough to find it! And it wasn't what I thought it was. No deletionism or inclusionism jokes there.
Maybe you were thinking of the sinister Deletionist Cabal in _Wikipedia The Movie_? https://secure.wikimedia.org/wikipedia/en/wiki/User:Raul654/Wikipedia_the_Mo...
On Tue, Jul 14, 2009 at 5:57 PM, Gwern Branwengwern0@gmail.com wrote:
On Tue, Jul 14, 2009 at 12:38 PM, Carcharothcarcharothwp@googlemail.com wrote:
http://en.wikipedia.org/wiki/User:Jaranda/Wikipedia%27s_first_IRC_chat
Took me long enough to find it! And it wasn't what I thought it was. No deletionism or inclusionism jokes there.
Maybe you were thinking of the sinister Deletionist Cabal in _Wikipedia The Movie_? https://secure.wikimedia.org/wikipedia/en/wiki/User:Raul654/Wikipedia_the_Mo...
No, but that was funny, especially the comment from number 7. Thank you!
It was good that there was a spoof of the LotR movies in the next scene:
"There may come a day when the strength of wikipedia fails, but it is not this day! A day when wiki becomes piki, and Pikmin conquer wikipedia, but it is not this day! This day we fight! Remember what you are fighting for: the random article, the wikiprojects, the featured content, and the use of square brackets everywhere!"
At least I think it was a spoof of Aragorn's speech at the Black Gate in the third LotR movie. That was itself probably a homage to something else, so it might be a spoof of that hypothetical original.
Carcharoth
PS. OMG, I just scrolled up to the top and saw the movie poster of Jimbo!
On Tue, Jul 14, 2009 at 11:50 AM, David Gerarddgerard@gmail.com wrote:
Here's a question: how many articles are created and deleted within 24 hours?
In early 2007, I did a quick and dirty estimate that about 2400 articles were deleted per day, at a time when the net gain per day was around 1800.
Activity of anons, new registered users, and established editors have all declined (roughly proportionally to each other) since then, so the ratio kept to deleted I would guess is similar. Therefore, by my utterly unscientific calculation, around 1750 newly created articles are deleted each day.
-Sage
On 14/07/2009, Sage Ross ragesoss+wikipedia@gmail.com wrote:
On Tue, Jul 14, 2009 at 10:07 AM, Ian Woollardian.woollard@gmail.com wrote:
It's looking to me like 3.5 million is about the plateau, since the curve is bang on that, but we might make 4 million *eventually*.
http://en.wikipedia.org/wiki/Wikipedia:Modelling_Wikipedia%27s_growth#Logist...
I don't think the bell-shaped articles/day curve of the logistic model is a good description of the trends. Since article creation peaked in 2007, the falloff in article creation has been much slower than than ramp-up. Rather than falling back to close to zero articles/day over the next 5 years or so (as the logistic model predicts), it looks like we're heading to an asymptote of (I'm eyeballing it here) around 1000 articles/day. I expect 4 million articles a lot sooner than *eventually*. ;)
http://commons.wikimedia.org/wiki/File:Enwikipediagrowth.PNG
We're already down to 1000/day growth on the unsmoothed graph as we fall off one of the two biannual growth peaks.
Looks like the Wikipedia is still bang-on for 3.5 million articles.
-Sage
On Tue, Nov 17, 2009 at 5:59 AM, Ian Woollard ian.woollard@gmail.com wrote:
On 14/07/2009, Sage Ross ragesoss+wikipedia@gmail.com wrote:
On Tue, Jul 14, 2009 at 10:07 AM, Ian Woollardian.woollard@gmail.com wrote:
It's looking to me like 3.5 million is about the plateau, since the curve is bang on that, but we might make 4 million *eventually*.
http://en.wikipedia.org/wiki/Wikipedia:Modelling_Wikipedia%27s_growth#Logist...
I don't think the bell-shaped articles/day curve of the logistic model is a good description of the trends. Since article creation peaked in 2007, the falloff in article creation has been much slower than than ramp-up. Rather than falling back to close to zero articles/day over the next 5 years or so (as the logistic model predicts), it looks like we're heading to an asymptote of (I'm eyeballing it here) around 1000 articles/day. I expect 4 million articles a lot sooner than *eventually*. ;)
http://commons.wikimedia.org/wiki/File:Enwikipediagrowth.PNG
We're already down to 1000/day growth on the unsmoothed graph as we fall off one of the two biannual growth peaks.
Looks like the Wikipedia is still bang-on for 3.5 million articles.
As long as history doesn't come to an end, and new people keep getting born and (annoyingly) becoming notable enough for a Wikipedia article, there will always be a need for new articles.
Carcharoth
On Tue, Nov 17, 2009 at 1:19 PM, Carcharoth carcharothwp@googlemail.com wrote:
On Tue, Nov 17, 2009 at 5:59 AM, Ian Woollard ian.woollard@gmail.com wrote:
On 14/07/2009, Sage Ross ragesoss+wikipedia@gmail.com wrote:
On Tue, Jul 14, 2009 at 10:07 AM, Ian Woollardian.woollard@gmail.com wrote:
It's looking to me like 3.5 million is about the plateau, since the curve is bang on that, but we might make 4 million *eventually*.
http://en.wikipedia.org/wiki/Wikipedia:Modelling_Wikipedia%27s_growth#Logist...
I don't think the bell-shaped articles/day curve of the logistic model is a good description of the trends. Since article creation peaked in 2007, the falloff in article creation has been much slower than than ramp-up. Rather than falling back to close to zero articles/day over the next 5 years or so (as the logistic model predicts), it looks like we're heading to an asymptote of (I'm eyeballing it here) around 1000 articles/day. I expect 4 million articles a lot sooner than *eventually*. ;)
http://commons.wikimedia.org/wiki/File:Enwikipediagrowth.PNG
We're already down to 1000/day growth on the unsmoothed graph as we fall off one of the two biannual growth peaks.
Looks like the Wikipedia is still bang-on for 3.5 million articles.
As long as history doesn't come to an end, and new people keep getting born and (annoyingly) becoming notable enough for a Wikipedia article, there will always be a need for new articles.
Carcharoth
There are even articles that were deleted without prejudice before, that *gasp* happen to do gain notability! (I happened to look over my own deletion logs yesterday, and a handfull of articles that were (IMO correctly) deleted without prejudice in the past, that are now decent articles).
Martijn
On Tue, Nov 17, 2009 at 12:19 PM, Carcharoth carcharothwp@googlemail.com wrote:
As long as history doesn't come to an end, and new people keep getting born and (annoyingly) becoming notable enough for a Wikipedia article, there will always be a need for new articles.
Not to mention people's irritating and continuing habit of publishing successful books, making notable films (running the risk of creating notable actors and other staff), creating successful companies with successful products, progressing with scientific enquiry, advancing technology, releasing new software...
At this rate we'll never finish the encyclopedia. Once Wikipedia has more cultural power I suggest we wield it to put a halt to all activity until we've caught up. Once that's achieved I suggest that all human endeavours are posted as requests through our OTRS and we can tell the actor/scientist/inventor whether we are willing to allow them to proceed (after we have assessed whether their activity is liable to create something that meet our notability criteria) or whether they must wait until we clear any current backlog.
Bod Notbod wrote:
On Tue, Nov 17, 2009 at 12:19 PM, Carcharoth carcharothwp@googlemail.com wrote:
As long as history doesn't come to an end, and new people keep getting born and (annoyingly) becoming notable enough for a Wikipedia article, there will always be a need for new articles.
Not to mention people's irritating and continuing habit of publishing successful books, making notable films (running the risk of creating notable actors and other staff), creating successful companies with successful products, progressing with scientific enquiry, advancing technology, releasing new software...
Yes. To be tedious and pedantic about it all, once we have finished Phase 1 of enWP, where we were playing catch-up with the obviously encyclopedic topics like chemical elements and US Presidents (etc.), we get to Phase 2, where the new articles fall into several distinct classes:
(1) articles about newly notable topics; (2) articles about fairly obviously encyclopedic topics, for which sources were available without too much trouble and which fit into existing coverage, but had been missed for whatever reason; (3) articles which are much like those in (2) to create, but only came to light after someone expanded existing coverage somewhere (new redlinks); (4) articles for redlinks where the supporting sources took a bit of quarrying out.
So we are really saying that (1) is generally speaking the 'reactive' class. To some extent the rate of creation is not under "our" control (these articles will be started in some form anyway). The others are the 'proactive' classes: (2) really just requires people to read the site and notice places where redlinks are or should be, and create good stubs that are not a huge effort (the traditional form of growth). (3) requires upgrading stubs to generate fuller coverage, and then we are back to (2). While (4) takes us back to the "librarian" discussion: deeper-cutting research skills required. (There is really also (5), completism for lists, which gets through to me, but perhaps is a minority interest.)
So what we get is a rate of growth by the article-number metric (not the only interesting measure) where one component is mostly to do with outside 'push', while the others are 'pull', and depend on how Wikipedians self-assign to tasks.
Charles
2009/11/17 Charles Matthews charles.r.matthews@ntlworld.com:
Bod Notbod wrote:
On Tue, Nov 17, 2009 at 12:19 PM, Carcharoth carcharothwp@googlemail.com wrote:
As long as history doesn't come to an end, and new people keep getting born and (annoyingly) becoming notable enough for a Wikipedia article, there will always be a need for new articles.
Not to mention people's irritating and continuing habit of publishing successful books, making notable films (running the risk of creating notable actors and other staff), creating successful companies with successful products, progressing with scientific enquiry, advancing technology, releasing new software...
Yes. To be tedious and pedantic about it all, once we have finished Phase 1 of enWP, where we were playing catch-up with the obviously encyclopedic topics like chemical elements and US Presidents (etc.), we get to Phase 2, where the new articles fall into several distinct classes:
(1) articles about newly notable topics; (2) articles about fairly obviously encyclopedic topics, for which sources were available without too much trouble and which fit into existing coverage, but had been missed for whatever reason; (3) articles which are much like those in (2) to create, but only came to light after someone expanded existing coverage somewhere (new redlinks); (4) articles for redlinks where the supporting sources took a bit of quarrying out.
So we are really saying that (1) is generally speaking the 'reactive' class. To some extent the rate of creation is not under "our" control (these articles will be started in some form anyway). The others are the 'proactive' classes: (2) really just requires people to read the site and notice places where redlinks are or should be, and create good stubs that are not a huge effort (the traditional form of growth). (3) requires upgrading stubs to generate fuller coverage, and then we are back to (2). While (4) takes us back to the "librarian" discussion: deeper-cutting research skills required. (There is really also (5), completism for lists, which gets through to me, but perhaps is a minority interest.)
So what we get is a rate of growth by the article-number metric (not the only interesting measure) where one component is mostly to do with outside 'push', while the others are 'pull', and depend on how Wikipedians self-assign to tasks.
Charles
Indeed. Looking at this:
http://www.floatingsheep.org/2009/11/mapping-wikipedia.html
Gives us some idea where the gaps are but not to the extent you might think (there are simply fewer citable sources referring to things in say the Central African Republic than the UK).
On Tue, Nov 17, 2009 at 1:08 PM, geni geniice@gmail.com wrote:
<snip>
That is fascinating. Thanks for posting that link.
Gives us some idea where the gaps are but not to the extent you might think (there are simply fewer citable sources referring to things in say the Central African Republic than the UK).
"there are more Wikipedia articles written about the fictional places of Middle Earth and Discworld than about many countries in Africa, the Americas and Asia"
Oops! :-/
The closest I've come to writing about things in other countries is here:
http://en.wikipedia.org/wiki/John_Allan_Broun
I had a load of sources on the talk page to follow up, but never found time and left them there for anyone else who happened to chance along (actually linking the articles from other articles would help there).
The history of astronomical research in India during the time of the British Empire is actually quite interesting (if you like that sort of thing). I've just been reading about the research carried out by John Evershed:
http://en.wikipedia.org/wiki/John_Evershed
I've glad to see we have an article on the observatory he did most of his work at:
http://en.wikipedia.org/wiki/Kodaikanal_Solar_Observatory
This article could do with expanding:
http://en.wikipedia.org/wiki/Madras_Observatory
And this one could do with tidying up:
http://en.wikipedia.org/wiki/Colaba_Observatory
Carcharoth
On Wed, Nov 18, 2009 at 12:29 AM, Carcharoth carcharothwp@googlemail.com wrote:
The closest I've come to writing about things in other countries is here:
Aww, I'm a *much* better person than you:
New Zealand: Broken River, New Zealand, Craigieburn Valley, Fox Peak, Invincible Snowfields, Mount Dobson, Mount Lyford, Porters, Rainbow, Snow Park, Mount Cheeseman, Temple Basin, Mount Olympus Ski Area, Mount Potts, Roundhill Ski Area, Hanmer Springs Ski Area, Mount Robert, Manganui
France:Carnac stones, Couesnon, La Trinité-sur-Mer, Sée River, Château de Beauregard, Loire Valley, (Puiseaux River), Canton of Morez, Parc de la Tête d'Or, Pontorson, Phare de la Vieille, Ar Men, Tourelle de la Plate,
Central Europe: Mount Klin, Starý Smokovec, Adršpach-Teplice Rocks, Lomnický štít, Czarny Staw pod Rysami, Tatranská Lomnica, Mount Bystrá, Slavkovský štít, Váci utca, Hrebienok
UK and Ireland: Guinness Storehouse, Calcot Manor, Staxigoe
Bolivia: Villazón
Peru: Castaño Overa Glacier, Alerce Glacier
Other: Kushk River, Kushk, Saint Mary of Valencia Cathedral, Mustafa Centre, Temple of Olympian Zeus (Agrigento), Breckenridge Reservoir, A1 Motorway (Italy), Cirat, Peacock Alley (jazz club)
Don't worry though, I'm sure you'll get there! One day you can wallow in your own multicultural global self-satisfaction too!
Steve
On 17/11/2009, Carcharoth carcharothwp@googlemail.com wrote:
As long as history doesn't come to an end, and new people keep getting born and (annoyingly) becoming notable enough for a Wikipedia article, there will always be a need for new articles.
Maybe, but I don't know how many. That level doesn't seem particularly obvious on the data yet.
It might be interesting for somebody to try sampling the new article lists and try to work out what percentage of new articles are on a topic that couldn't have been written before (say) the previous month.
Anyway, the world isn't ending (at least until 2012 ;-) ) and the article quality is still marching ever upwards.
Carcharoth
----- "Steve Bennett" stevagewp@gmail.com wrote:
My long-time observation is that the people who obsess about FA over the long term want to keep the number of articles with that status approximately constant by making the standards more and more difficult to meet.
Yeah, we see that on FPC (featured pics) - and RfA (admins) for that matter.
Does that mean we should have the same process for "delisting" admins - not really fair that candidates today are held to a higher standard than candidates who got through three years ago!
Andrew
2009/7/14 Andrew Turvey andrewrturvey@googlemail.com:
----- "Steve Bennett" stevagewp@gmail.com wrote:
My long-time observation is that the people who obsess about FA over the long term want to keep the number of articles with that status approximately constant by making the standards more and more difficult to meet.
Yeah, we see that on FPC (featured pics) - and RfA (admins) for that matter.
Does that mean we should have the same process for "delisting" admins - not really fair that candidates today are held to a higher standard than candidates who got through three years ago!
That very question has been discussed to death and no agreement has been reached. Discussing it again here will serve no purpose. There are lots of discussions about governance going on, maybe once one of them reaches a conclusion we'll be able to clear up this question.
On Tue, Jul 14, 2009 at 12:01 PM, Andrew Turveyandrewrturvey@googlemail.com wrote:
Does that mean we should have the same process for "delisting" admins - not really fair that candidates today are held to a higher standard than candidates who got through three years ago!
Actually it is "fair", in a warped way. RfA is supposed to be purely a risk management exercise: we subject prospective admins to a couple of tests to reduce our risk that they go feral. The chance of a 3 year old admin going feral is much much lower than the average prospective admin, sort of by definition.
(So I don't see a benefit of desysopping/retesting, just on that basis.)
Steve
Steve Bennett wrote:
RfA is supposed to be purely a risk management exercise: we subject prospective admins to a couple of tests to reduce our risk that they go feral.
I thought it was mainly an exercise to see if you cared enough to look up the standard acceptable answers to the standard questions. Anyway it mostly selects people who are fine to have the tools. Some small proportion show up as not really suitable on a time scale of three months; and some others do go a bit strange after a longer period in the front line. I doubt these cases are the sort of things RfA as filter can catch, though - too formalistic and not based on "interview" techniques.
Charles
2009/7/16 Charles Matthews charles.r.matthews@ntlworld.com:
Steve Bennett wrote:
RfA is supposed to be purely a risk management exercise: we subject prospective admins to a couple of tests to reduce our risk that they go feral.
I thought it was mainly an exercise to see if you cared enough to look up the standard acceptable answers to the standard questions.
It used to be. These days people go through your contributions with a fine toothed comb to make sure you've never mistagged a CSD, called a spade a spade or spelt "opportunity" with an 'e'.
2009/7/16 Thomas Dalton thomas.dalton@gmail.com:
It used to be. These days people go through your contributions with a fine toothed comb to make sure you've never mistagged a CSD, called a spade a spade or spelt "opportunity" with an 'e'.
Not so. In fact for the most part other than some computerized checking people don't normally check back even 1000 edits. This can result in an interesting situation where RFAs can biol down to luck. Does the person who remembers the mistake you made 3000 edits ago find out about the RFA before it closes?
On Mon, Jul 13, 2009 at 3:29 AM, Steve Bennettstevagewp@gmail.com wrote:
http://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2009-07-06/Feature...
I couldn't help but notice:
- Five articles were promoted to featured status this week
- Four articles were delisted this week.
- Twelve lists were promoted to featured status this week
- Eight lists were delisted this week
What a lot of churn. So the overall rate was merely +1 FA, +4 FL (and also 3 topics and three images).
Is it always this bad?
There are long-term stats somewhere, and they could be updated if you asked. I suggest identifying which of the featured areas you want to see long-term stats for, and asking at the relevant talk pages. An approximation to these stats could be obtained by going through the Signpost summaries.
Carcharoth
On Mon, Jul 13, 2009 at 5:26 AM, Carcharothcarcharothwp@googlemail.com wrote:
There are long-term stats somewhere, and they could be updated if you asked. I suggest identifying which of the featured areas you want to see long-term stats for, and asking at the relevant talk pages. An approximation to these stats could be obtained by going through the Signpost summaries.
If you don't need week-by-week stats, then the best place is: http://en.wikipedia.org/wiki/Wikipedia:Featured_article_statistics
The total number of featured articles has been rising fairly steadily, but the ratio of featured articles to total articles (a little under 0.1%) has remained more or less steady over the years.
In contrast, the percentage of articles that are Good Articles has been rising for some time: http://en.wikipedia.org/wiki/Wikipedia:Good_article_statistics
I think the main reason why this week stands out is the unusually low number of FA promotions; sometimes the process hits a lull, and sometimes a bunch of pending promotions don't get processed until after the Signpost reports its numbers, which can make the week-to-week numbers a fun tea leaf reading exercise but they don't necessarily mean much.
-Sage (User:Ragesoss)
----- "Steve Bennett" stevagewp@gmail.com wrote:
From: "Steve Bennett" stevagewp@gmail.com To: "English Wikipedia" wikien-l@lists.wikimedia.org Sent: Monday, 13 July, 2009 03:29:06 GMT +00:00 GMT Britain, Ireland, Portugal Subject: [WikiEN-l] Featured churn
http://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2009-07-06/Feature...
I couldn't help but notice:
- Five articles were promoted to featured status this week
- Four articles were delisted this week.
- Twelve lists were promoted to featured status this week
- Eight lists were delisted this week
I often wondered - what's the point of delisting? Surely if a previous version of an article was good enough to be Featured, if the current version isn't, you should just restore the one that was?
Or am I missing something?
Andrew
Andrew Turvey wrote:
----- "Steve Bennett" stevagewp@gmail.com wrote:
From: "Steve Bennett" stevagewp@gmail.com To: "English Wikipedia" wikien-l@lists.wikimedia.org Sent: Monday, 13 July, 2009 03:29:06 GMT +00:00 GMT Britain, Ireland, Portugal Subject: [WikiEN-l] Featured churn
http://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2009-07-06/Feature...
I couldn't help but notice:
- Five articles were promoted to featured status this week
- Four articles were delisted this week.
- Twelve lists were promoted to featured status this week
- Eight lists were delisted this week
I often wondered - what's the point of delisting? Surely if a previous version of an article was good enough to be Featured, if the current version isn't, you should just restore the one that was?
Or am I missing something?
Andrew
The problem is that the FA standards have changed over time, becoming more strict. So, it is very possible that if an article reached FA several years ago, then no version of it would pass today's FA criteria.
Firestorm
On Mon, Jul 13, 2009 at 8:09 PM, Andrew Turvey <andrewrturvey@googlemail.com
wrote:
I often wondered - what's the point of delisting? Surely if a previous version of an article was good enough to be Featured, if the current version isn't, you should just restore the one that was?
Or am I missing something?
Andrew
It's the standard for initial listing that has changed - what qualified for FA in 2005 or 2006 may not meet standards today even in its best historical form, so they take a fresh look at articles where problems are identified and delist them if they don't meet current criteria.
Nathan