(I'm quoting from http://mail.wikipedia.org/pipermail/wikien-l/2005-December/033880.html)
Jimbo wrote once: "Today, as an experiment, we will be turning off new pages creation for anonymous users in the English Wikipedia."
This was way back in 5 Dec 2005. Has the experiment run long enough? What sort of experiment varies the independent variable only one way?
Let's turn page creation back on for anons. We turned it off, so let's see what happens when we turn it back on; otherwise we're simply running on sheer blind inertia and unthinking myopia. Every month since Dec. 2005 we should have been asking whether the costs have been worth the benefits.
"1. Annoying anons may simply decide to create accounts and make annoying nonsense pages anyway. This will certainly be true in some cases, but it is an empirical question as to how many."
Quite a few. I haven't seen much of a reduction in PROD or AFD or speedies ([[User:Dragons flight/Category tracker]] shows that Speedy has at times reached 349 entries, and it tracks back to late July 2006).
"2. We will lose good new pages created by anons of good will. This may cause the growth of English Wikipedia (in terms of the number of articles) to slow a little bit. With 800,000+ articles, and ever-increasing traffic to the website, this seems to be a worthwhile cost."
PR-wise, turning off page creation wasn't good, to say the least; it has forced all sorts of ugly hacks to pages and annoyed many many people (such as persons like me; one cannot even create a simple redirect when not staging out of one's computer/account). Take a look at the monumental failure that is AFC sometime, which turning off page creation has forced on us. Valid, good articles are being entombed there.
There's no time like the present. It's summer now, so dedicate Wikipedians have plenty of time, and the September deluge is not yet upon us; there are even more Wikipedians than ever, and more articles, and better tools. How much have the anti-vandal tools (too many to name now, even excluding bots) proliferated and improved since that long-ago December?
"But preveneting [sic] anons from creating new pages is a different matter, and it seems a worthy time to make an experiment of it."
Yes, let's. Experiments go both ways, remember...
~maru
On 8/20/06, maru dubshinki marudubshinki@gmail.com wrote:
(I'm quoting from http://mail.wikipedia.org/pipermail/wikien-l/2005-December/033880.html)
Jimbo wrote once: "Today, as an experiment, we will be turning off new pages creation for anonymous users in the English Wikipedia."
This was way back in 5 Dec 2005. Has the experiment run long enough? What sort of experiment varies the independent variable only one way?
Let's wait a bit before we do that. I understand there's a study of the effect that turning off anon creation had, but it's not yet ready for publication. I think we should wait to see that study until we act.
Kelly
On 8/21/06, maru dubshinki marudubshinki@gmail.com wrote:
This was way back in 5 Dec 2005. Has the experiment run long enough? What sort of experiment varies the independent variable only one way?
Let's turn page creation back on for anons. We turned it off, so let's see what happens when we turn it back on; otherwise we're simply running on sheer blind inertia and unthinking myopia.
We don't need to. We saw what happened in the 4 years before it was turned off.
Take a look at the monumental failure that is AFC sometime, which turning
off page creation has forced on us. Valid, good articles are being entombed there.
And look at how many nonsense creations are being stopped there too.
MacGyverMagic/Mgm wrote:
On 8/21/06, maru dubshinki marudubshinki@gmail.com wrote:
This was way back in 5 Dec 2005. Has the experiment run long enough? What sort of experiment varies the independent variable only one way?
Let's turn page creation back on for anons. We turned it off, so let's see what happens when we turn it back on; otherwise we're simply running on sheer blind inertia and unthinking myopia.
We don't need to. We saw what happened in the 4 years before it was turned off.
But how can we tell what changed in the year after it was turned off relative to those four years? Other than purely anecdotal reports, which are effectively useless for a system the size and complexity of Wikipedia, what analysis has been done?
Take a look at the monumental failure that is AFC sometime, which turning
off page creation has forced on us. Valid, good articles are being entombed there.
And look at how many nonsense creations are being stopped there too.
The question isn't simply whether nonsense creations have been stopped. If that was the only goal, then the perfect solution is obvious; disable new article creation entirely.
We need to see some sort of rigorous statistical analysis of article creation patterns before and after the anon-disabling, producing objective results rather than subjective opinions. I've long been annoyed that this experiment was apparently set off without any plans for this. It's fortunate (though less than ideal) that Wikipedia's database retains enough information to still do some analysis after the fact.
On 8/21/06, Bryan Derksen bryan.derksen@shaw.ca wrote: [snip]
We need to see some sort of rigorous statistical analysis of article creation patterns before and after the anon-disabling, producing objective results rather than subjective opinions. I've long been annoyed that this experiment was apparently set off without any plans for this. It's fortunate (though less than ideal) that Wikipedia's database retains enough information to still do some analysis after the fact.
Analysis has been done, results will be published.
In any case, I'm curious about what you believe we lost that we could have had with additional planning?
On 8/21/06, Gregory Maxwell gmaxwell@gmail.com wrote:
Analysis has been done, results will be published.
In any case, I'm curious about what you believe we lost that we could have had with additional planning?
Well traditionaly it is a good idea to lay out your method for stastical anaysis in advance.
Gregory Maxwell wrote:
In any case, I'm curious about what you believe we lost that we could have had with additional planning?
My support for the whole endeavour, for one. The appearance of professionalism, for another. This thing looks like it was just a knee-jerk reaction to bad publicity.
We've lost a lot of time as well. It's been almost nine months now, during some of the most widespread publicity Wikipedia's ever had. If when analysis is finally published it turns out that this experiment had a major negative impact on new user participation it would have been nice to know that more quickly to limit the damage.
From a purely data standpoint probably not a whole lot was lost thanks to the thoroughness of the database's records, though having to rely on the deletion database for long-term information retention does make me a bit nervous. But that's not the only consideration here.
On 8/21/06, Bryan Derksen bryan.derksen@shaw.ca wrote:
My support for the whole endeavour, for one.
[snip]
I hope you understand that getting the permission of each and every user before making a change is a solution which doesn't scale.
We've lost a lot of time as well. It's been almost nine months now, during some of the most widespread publicity Wikipedia's ever had. If when analysis is finally published it turns out that this experiment had a major negative impact on new user participation it would have been nice to know that more quickly to limit the damage.
I don't believe we could have gotten useful data about the possible positive impacts of this change any sooner than now. The most useful metrics take time (i.e. life expectancy of an article), and there is a lot of noise.. so to make confident statements we need lots of data.
For your reference, there have been people watching this for clearly negative impacts all along. Had there been substantial evidence that it was doing a significant amount of harm, it would have been aborted.
Here is some data, http://myrandomnode.dyndns.org/new_articles.png The graph shows the rate of article creation, excluding articles which were eventually deleted, based on the timestamp of the oldest visible revision over the life of the project. X is time in julian day numbers, the ticks are about 200 days apart, and the graph is about a month old.
This doesn't tell us that the policy *helped* but it does indicate that it isn't causing serious harm. A more comprehensive statistical study of the impact is forthcoming.
From a purely data standpoint probably not a whole lot was lost thanks to the thoroughness of the database's records, though having to rely on the deletion database for long-term information retention does make me a bit nervous. But that's not the only consideration here.
Our retention of deleted metadata is just as robust as anything else...
Gregory Maxwell wrote:
On 8/21/06, Bryan Derksen bryan.derksen@shaw.ca wrote:
My support for the whole endeavour, for one.
[snip]
I hope you understand that getting the permission of each and every user before making a change is a solution which doesn't scale.
Who said anything about asking anyone's permission? I would have supported the experiment if its design had simply been made clear from the outset (unless it had been a stupid design, of course, in which case I would have spontaneously voiced objections and suggestions to improve it. No need to actually _ask_ me about it).
For your reference, there have been people watching this for clearly negative impacts all along. Had there been substantial evidence that it was doing a significant amount of harm, it would have been aborted.
Back on 6/20/2006 there was this exchange on the mailing list in the thread ""the experiment" - did it work?":http://mail.wikipedia.org/pipermail/wikien-l/2006-June/049320.html
Jimmy Wales wrote:
Guy Chapman aka JzG wrote:
It worked, though.
It is not clear to me that it did. I would love for us to have some serious analysis of that.
Apparently even Jimbo was unaware that there was any analysis going on, and he's the one who initiated the experiment in the first place. How is anyone else supposed to know what's going on?
Our retention of deleted metadata is just as robust as anything else...
No it isn't. from http://en.wikipedia.org/wiki/Wikipedia:Deletion_review: "The archive of deleted page revisions may be periodically cleared. Pages deleted prior to the database crash on 8 June 2004 are not present in the current archive because the archive tables were not backed up. This means pages cannot be restored by a sysop. If there is great desire for them it may be possible to retrieve them from the old database files. Prior to this, the archive was cleared out on 3 December 2003."
Deleted versions can't be relied on, who knows when the deletion database will be cleared again?
All I ask is a little transparency. There's no reason for any of this to be kept secret.
On 22/08/06, Bryan Derksen bryan.derksen@shaw.ca wrote:
All I ask is a little transparency. There's no reason for any of this to be kept secret.
Seconded.
- d.
On 8/22/06, Bryan Derksen bryan.derksen@shaw.ca wrote: [snip]
Deleted versions can't be relied on, who knows when the deletion database will be cleared again?
That doesn't reflect our current practices.
All I ask is a little transparency. There's no reason for any of this to be kept secret.
What do you think is being kept secret.
Don't mistake Jimbo's inability to be simultaneously aware of the work of every single participant as an indicator of secrecy.
On 22/08/06, Gregory Maxwell gmaxwell@gmail.com wrote:
On 8/22/06, Bryan Derksen bryan.derksen@shaw.ca wrote:
All I ask is a little transparency. There's no reason for any of this to be kept secret.
What do you think is being kept secret.
It's that your messages in this thread are the first I've heard that anyone was actually looking at the situation rather than just treating no anon page creations as the way things are henceforth. And it is reassuring that it's being looked at, and particularly that you are :-)
- d.
Gregory Maxwell wrote:
On 8/22/06, Bryan Derksen bryan.derksen@shaw.ca wrote: [snip]
Deleted versions can't be relied on, who knows when the deletion database will be cleared again?
That doesn't reflect our current practices.
Do you know of anyplace I could find out what our current practices are? I hunted around a bit and deletion review was the only policy page I could find that mentioned the issue one way or another. If it's inaccurate I should correct it.
All I ask is a little transparency. There's no reason for any of this to be kept secret.
What do you think is being kept secret.
In your previous post all you said was "Analysis has been done, results will be published." Who's doing the analysis? Is there any indication of where and when will the results be published? Finding out anything about what's been going on with this "experiment" has been like pulling teeth, it's nine months in now and this is the first time I've heard even this much.
Don't mistake Jimbo's inability to be simultaneously aware of the work of every single participant as an indicator of secrecy.
It's my own inability to find out any of these details myself that I'm more worried about.
On 8/22/06, Bryan Derksen bryan.derksen@shaw.ca wrote:
Gregory Maxwell wrote:
On 8/22/06, Bryan Derksen bryan.derksen@shaw.ca wrote:
All I ask is a little transparency. There's no reason for any of this to be kept secret.
What do you think is being kept secret.
In your previous post all you said was "Analysis has been done, results will be published." Who's doing the analysis? Is there any indication of where and when will the results be published? Finding out anything about what's been going on with this "experiment" has been like pulling teeth, it's nine months in now and this is the first time I've heard even this much.
C'mon now, that's the great thing with retrospective analysis. You get to find out which data support your point *first*, then your results are sure to be "correct".
Anthony
Bryan Derksen wrote:
Gregory Maxwell wrote:
On 8/22/06, Bryan Derksen bryan.derksen@shaw.ca wrote: [snip]
Deleted versions can't be relied on, who knows when the deletion database will be cleared again?
That doesn't reflect our current practices.
Do you know of anyplace I could find out what our current practices are? I hunted around a bit and deletion review was the only policy page I could find that mentioned the issue one way or another. If it's inaccurate I should correct it.
All I ask is a little transparency. There's no reason for any of this to be kept secret.
What do you think is being kept secret.
In your previous post all you said was "Analysis has been done, results will be published." Who's doing the analysis? Is there any indication of where and when will the results be published? Finding out anything about what's been going on with this "experiment" has been like pulling teeth, it's nine months in now and this is the first time I've heard even this much.
Don't mistake Jimbo's inability to be simultaneously aware of the work of every single participant as an indicator of secrecy.
It's my own inability to find out any of these details myself that I'm more worried about.
I still haven't seen any answers to the questions I posted here and the thread seems to have moved on from discussing them. Should I assume that these answers don't actually exist?
On 8/21/06, Bryan Derksen bryan.derksen@shaw.ca wrote:
My support for the whole endeavour, for one. The appearance of professionalism, for another. This thing looks like it was just a knee-jerk reaction to bad publicity.
There's no doubt, it *was* a knee-jerk reaction to bad publicity, but not a bad one. Calling it an experiment was a mistake, since everyone now expects introduction, method, results, analysis, conclusion, in that order. However, we haven't had any more Seigenthalers, and there's certainly no sign that anything has become dramatically worse. Maybe Wikipedia could have been "even better" than it is now, for the same time period? We could have been up to 1.6 million articles? Who knows.
Steve
On 22/08/06, Steve Bennett stevagewp@gmail.com wrote:
that order. However, we haven't had any more Seigenthalers, and
What this means is that we haven't had anyone with ready wide media access hate their bio. We've had lots of people who hate their bio. I'd say that if anything's changed in that regard, it's a hardarsed attitude to living bios. I'd REALLY like some more solid cause and effect stuff before you attribute "no more Seigenthalers" to blocking anon page creation.
- d.
On 8/22/06, David Gerard dgerard@gmail.com wrote:
What this means is that we haven't had anyone with ready wide media access hate their bio. We've had lots of people who hate their bio. I'd say that if anything's changed in that regard, it's a hardarsed attitude to living bios. I'd REALLY like some more solid cause and effect stuff before you attribute "no more Seigenthalers" to blocking anon page creation.
Agreed, drawing any conclusions other than "nothing terrible happened" is pretty difficult - we didn't set out to collect any data, and we don't really have any.
Steve
On 8/22/06, Steve Bennett stevagewp@gmail.com wrote:
Agreed, drawing any conclusions other than "nothing terrible happened" is pretty difficult - we didn't set out to collect any data, and we don't really have any.
I asked what data we are missing, and no one has named any.
Please don't cloud a complicated issue further with uninformed conjecture.
On 8/22/06, Gregory Maxwell gmaxwell@gmail.com wrote:
On 8/22/06, Steve Bennett stevagewp@gmail.com wrote:
Agreed, drawing any conclusions other than "nothing terrible happened" is pretty difficult - we didn't set out to collect any data, and we don't really have any.
I asked what data we are missing, and no one has named any.
Please don't cloud a complicated issue further with uninformed conjecture.
Ok: - How many new pages were created per day by anons prior to the change? - What proportion of those new pages survived 1 month? - How many new entries at AfC were created per day since the change (preferably, month by month stats to see the long term trend) - What proportion of those entries became articles which survived 1 month? - What proportion of attempts to create a new page by an anon result in an AfC entry? - What proportion of attempts to create a new page by an anon result in the anon registering an account? - What proportion of attempts to create a new page by an anon result in the anon registering an account then coming back 4 days later and doing the job?
Steve
On 8/22/06, Steve Bennett stevagewp@gmail.com wrote:
We could have been up to 1.6 million articles? Who knows.
Is it a bad thing that we're not?
We have a *lot* of articles, and I think we're well past the point where we can say that more is better without careful consideration.
It is clear that it's generally easier to monitor one low traffic article for vandalism than 10 very low traffic articles. I don't know exactly where those breakpoints are... but it does seem intuitive that past some point additional articles will increase our workload more than they increase the informative value of the project.
On 8/22/06, Gregory Maxwell gmaxwell@gmail.com wrote:
On 8/22/06, Steve Bennett stevagewp@gmail.com wrote:
We could have been up to 1.6 million articles? Who knows.
Is it a bad thing that we're not?
......
Yes, it is. We've got a really long way to go before we reach anything like comprehensiveness. We still have articles to write that are treated in other general encyclopedias. Go see the Missing Encyclopedia Articles Project, for example. Or take a look at this thing Piotrus wrote, wherein he estimates en will be largely complete in terms of article entries at 400 million entries: [[User:Piotrus/Wikipedia interwiki and specialized knowledge test]].
~maru
On 8/22/06, maru dubshinki marudubshinki@gmail.com wrote:
Yes, it is. We've got a really long way to go before we reach anything like comprehensiveness. We still have articles to write that are treated in other general encyclopedias. Go see the Missing Encyclopedia Articles Project, for example. Or take a look at this thing Piotrus wrote, wherein he estimates en will be largely complete in terms of article entries at 400 million entries: [[User:Piotrus/Wikipedia interwiki and specialized knowledge test]].
Yes, I talked to Piotrus about his research. ... I also contributed one of the longest lists of missing articles (the missing music subjects, 46k subjects).
I'm still skeptical. I too can generate numbers to support a postion, for example Wikipedia's word count is what.. 10x Britannica? Article acount... 18x?
Lets start with a simple question: What percentage of new articles created today fulfil requests from the Missing Encyclopedia Articles Project? Of the ones that do, what percentage of those are created by 'new' users?
To the first part, we know the answer must be vanishingly small because there are a huge number of new articles created every day (>5k). If more than a tiny fraction were created every day we would be done already. I can't guess at who created them, but it would be easy enough to check the lists and see who created them... Anyone want to try?
The argument I think you're presenting here is counter productive: that we are still working on making an encyclopedia. We're not. We've made an encyclopedia. Now need need to move on and work on making an encyclopedia that doesn't suck.
New articles are still important, as you've pointed out... But I've seen no evidence that new article creation belongs even on the top 100 task list for making our encyclopedia not suck.
Just because we know a lot more about new article creation than things like stability, verifiability, and consistency is not a reason to give new article creation more importance than it is due.
On 22/08/06, Gregory Maxwell gmaxwell@gmail.com wrote:
New articles are still important, as you've pointed out... But I've seen no evidence that new article creation belongs even on the top 100 task list for making our encyclopedia not suck.
True. OTOH, writing new articles is easy. OTOOH, it behooves those of us who find something we wanted to look up in Wikipedia isn't there and promptly write an article for it to write a *good* article. I try to. At least a clear description and a good reference. Something that will be *of use* to a reader who wants to know about the topic. (Last example created by me: [[lxrun]]).
I don't want to add something saying this to http://en.wikipedia.org/wiki/MediaWiki:Newarticletext unless we can sensibly remove something else (per the wisdom of [[m:instruction creep]]. Ideas to talk page thereof please.
- d.
Gregory Maxwell wrote:
New articles are still important, as you've pointed out... But I've seen no evidence that new article creation belongs even on the top 100 task list for making our encyclopedia not suck.
I'd certainly put it in the top 100, even the top 10, although probably not the top 3. When a user looks up an article, the worst possible result would be to find a misleading or otherwise incorrect article. The second-worst result would be to find no article. The third-worst would be to find a somewhat shoddy but not misleading or wrong article. So I'd rank "we don't cover that at all" pretty high on the list of ways our coverage of a topic could suck.
I know I'm still frequently annoyed that I have to look elsewhere for information in seemingly basic subjects because we don't yet cover them. For example, there are 20+ former prime ministers of Greece on which we have no article, and thousands of basic computer science topics.
-Mark
On 8/22/06, Delirium delirium@hackish.org wrote:
I'd certainly put it in the top 100, even the top 10, although probably not the top 3. When a user looks up an article, the worst possible result would be to find a misleading or otherwise incorrect article. The second-worst result would be to find no article. The third-worst would be to find a somewhat shoddy but not misleading or wrong article. So I'd rank "we don't cover that at all" pretty high on the list of ways our coverage of a topic could suck.
[snip]
I think you need to interject some [[bayesian]] style reasoning into the above...
Even if it were REALLY terrible to get no article, would it matter much if it were very rare? If it were only mildly bad to get inaccurate information, but it was fairly common how bad would that be?
With 18x the article count of our competition... I *hope* that 'no result' found isn't really due to a significant coverage problem... If it is, then what reason do we have to believe that continuing the same old process of creating articles will *ever* solve that problem?
Although our lack of logging makes it currently impossible to substantiate with data, I strongly believe that we could get substantially more improvement on 'no article found' front for our *readers* by improving our search ([[Double Metaphone]] enabled title search, for example), creating redirects, better navigation tools, improvements to main page layouts... etc.. than simply by writing more articles at this point.
Unfortunately I can't even tell people what to fix until we have better data.
Recently, by using some third party search data we determined that there were no pages for a large number of fairly common search strings. A couple of coverage gaps were found... but their numbers were dwarfed by the number of alternate representations and typeohs. Many redirects and a few new pages have been created as a result ... I believe many of these will be useful to readers, but we won't be able to tell without page view data for them.
On 8/23/06, Delirium delirium@hackish.org wrote:
I'd certainly put it in the top 100, even the top 10, although probably not the top 3. When a user looks up an article, the worst possible result would be to find a misleading or otherwise incorrect article. The second-worst result would be to find no article. The third-worst would be to find a somewhat shoddy but not misleading or wrong article. So I'd rank "we don't cover that at all" pretty high on the list of ways our coverage of a topic could suck.
I agree. I personally feel that a one sentence description with a URL link to a page that covers the subject thoroughly is a *very* valuable addition to our encyclopaedia. I wonder if others agree?
Steve
On 23/08/06, Steve Bennett stevagewp@gmail.com wrote:
On 8/23/06, Delirium delirium@hackish.org wrote:
I'd certainly put it in the top 100, even the top 10, although probably not the top 3. When a user looks up an article, the worst possible result would be to find a misleading or otherwise incorrect article. The second-worst result would be to find no article. The third-worst would be to find a somewhat shoddy but not misleading or wrong article. So I'd rank "we don't cover that at all" pretty high on the list of ways our coverage of a topic could suck.
I agree. I personally feel that a one sentence description with a URL link to a page that covers the subject thoroughly is a *very* valuable addition to our encyclopaedia. I wonder if others agree?
Yep. aka a good stub. (e.g. [[lxrun]], which tells an interested person not a whole lot, but at least what it is and a reference to look further.)
- d.
On 8/22/06, Delirium delirium@hackish.org wrote:
Gregory Maxwell wrote:
New articles are still important, as you've pointed out... But I've seen no evidence that new article creation belongs even on the top 100 task list for making our encyclopedia not suck.
I'd certainly put it in the top 100, even the top 10, although probably not the top 3. When a user looks up an article, the worst possible result would be to find a misleading or otherwise incorrect article.
Why's that so bad? They can just edit it to fix it.
The second-worst result would be to find no article.
Why's that so bad? They can just create it.
Oh wait, they can't. So that IS bad.
The third-worst would be to find a somewhat shoddy but not misleading or wrong article.
Again, fixable. I'd think the obviously wrong article might be more likely to motivate someone to fix it so maybe shoddy is a bad thing.
So I'd rank "we don't cover that at all" pretty high on the list of ways our coverage of a topic could suck.
I know I'm still frequently annoyed that I have to look elsewhere for information in seemingly basic subjects because we don't yet cover them. For example, there are 20+ former prime ministers of Greece on which we have no article, and thousands of basic computer science topics.
On 8/23/06, The Cunctator cunctator@gmail.com wrote:
On 8/22/06, Delirium delirium@hackish.org wrote:
Gregory Maxwell wrote:
New articles are still important, as you've pointed out... But I've seen no evidence that new article creation belongs even on the top 100 task list for making our encyclopedia not suck.
I'd certainly put it in the top 100, even the top 10, although probably not the top 3. When a user looks up an article, the worst possible result would be to find a misleading or otherwise incorrect article.
Why's that so bad? They can just edit it to fix it.
how do they know it is misleading or otherwise incorrect?
On 8/23/06, The Cunctator cunctator@gmail.com wrote:
The second-worst result would be to find no article.
Why's that so bad? They can just create it.
Heh, "I went to this book looking for some information which I was lacking. It didn't have it either, so I just added it."
Nothing wrong with this picture? :)
Steve
On 23/08/06, Steve Bennett stevagewp@gmail.com wrote:
On 8/23/06, The Cunctator cunctator@gmail.com wrote:
The second-worst result would be to find no article.
Why's that so bad? They can just create it.
Heh, "I went to this book looking for some information which I was lacking. It didn't have it either, so I just added it." Nothing wrong with this picture? :)
"looked it up and added it."
Don't you find yourself searching in vain for the "edit this page" link on other websites?
- d.
On 8/23/06, David Gerard dgerard@gmail.com wrote:
Don't you find yourself searching in vain for the "edit this page" link on other websites?
Yep. Well, no - I don't use other websites much any more, for that reason ;)
Steve
On 8/22/06, Steve Bennett stevagewp@gmail.com wrote:
On 8/21/06, Bryan Derksen bryan.derksen@shaw.ca wrote:
My support for the whole endeavour, for one. The appearance of professionalism, for another. This thing looks like it was just a knee-jerk reaction to bad publicity.
There's no doubt, it *was* a knee-jerk reaction to bad publicity, but not a bad one. Calling it an experiment was a mistake, since everyone now expects introduction, method, results, analysis, conclusion, in that order. However, we haven't had any more Seigenthalers, and there's certainly no sign that anything has become dramatically worse. Maybe Wikipedia could have been "even better" than it is now, for the same time period? We could have been up to 1.6 million articles? Who knows.
Steve
Have you ever considered the possibility that there haven't been any more Seigenthalers is because the MSM is jaded with that sort of thing?
And that's a rather cavalier attitude to take about the possibility that we've shot ourselves in the foot to the tune of thousands and thousands of articles and possibly editors.
~maru
On 8/21/06, MacGyverMagic/Mgm macgyvermagic@gmail.com wrote:
On 8/21/06, maru dubshinki marudubshinki@gmail.com wrote:
This was way back in 5 Dec 2005. Has the experiment run long enough? What sort of experiment varies the independent variable only one way?
Let's turn page creation back on for anons. We turned it off, so let's see what happens when we turn it back on; otherwise we're simply running on sheer blind inertia and unthinking myopia.
We don't need to. We saw what happened in the 4 years before it was turned off.
Yes. We saw Wikipedia take off and far surpass Nupedia (which had anon page creation turned off, incidentally).
Take a look at the monumental failure that is AFC sometime, which turning
off page creation has forced on us. Valid, good articles are being entombed there.
And look at how many nonsense creations are being stopped there too.
It's a lot easier to destroy bad articles than it is to write good ones.
~maru
On 8/21/06, maru dubshinki marudubshinki@gmail.com wrote:
Jimbo wrote once: "Today, as an experiment, we will be turning off new pages creation for anonymous users in the English Wikipedia."
That statement is much better understood if you substitute "trial" for "experiment".
Personally I don't think anonymous page creation serves our goals much. *shrug*
Steve
On 8/20/06, maru dubshinki marudubshinki@gmail.com wrote:
(I'm quoting from http://mail.wikipedia.org/pipermail/wikien-l/2005-December/033880.html)
Jimbo wrote once: "Today, as an experiment, we will be turning off new pages creation for anonymous users in the English Wikipedia."
This was way back in 5 Dec 2005. Has the experiment run long enough? What sort of experiment varies the independent variable only one way?
Let's turn page creation back on for anons. We turned it off, so let's see what happens when we turn it back on; otherwise we're simply running on sheer blind inertia and unthinking myopia. Every month since Dec. 2005 we should have been asking whether the costs have been worth the benefits.
I don't think we're both blind and myopic, but otherwise I wholeheartedly agree. Turn on anon page creation, maybe in Dec 2006 so we have a year's worth of data to compare to.
On 8/20/06, maru dubshinki marudubshinki@gmail.com wrote:
(I'm quoting from http://mail.wikipedia.org/pipermail/wikien-l/2005-December/033880.html)
Jimbo wrote once: "Today, as an experiment, we will be turning off new pages creation for anonymous users in the English Wikipedia."
This was way back in 5 Dec 2005. Has the experiment run long enough? What sort of experiment varies the independent variable only one way?
Thanks for asking this, I was wondering too.
Yeah, I would have much preferred a statement more like
"Today we are starting an experiment-- we are turning off new page creation. This experiment is to last until (t). Analysis of (typeA), (typeB), and (typeC) will be done by (userA), (userB), (userC). If you have other ideas for analysis, please let us know. Also we are looking for someone to do specialized (analysisZ). Please see the discussion at (http://en.wikipedia.org/wiki/Wikipedia:No_Anonymous_Page_Creation) if you would like to comment on or discuss it."
(or something like that...)
Instead, it's an experiment of indefinite duration with analysis (of an unknown kind? where is this analysis described?) presumably being done by some person or persons, probably, and the results to be revealed at some later unspecified date.
Most distressingly, as far as my (admittedly quick) search turned up, there's no particular place that this experiment is being discussed by the community. The archive page at http://en.wikipedia.org/wiki/Wikipedia_talk:Articles_for_creation/Archive-20... is the closest thing to a discussion I could find in the WP: namespace, but it's no longer a discussion. It's an archive.
Am I missing a discussion of this kind somewhere obvious?
Thanks,
-Stefan user:kiaparowits
Let's turn page creation back on for anons. We turned it off, so let's see what happens when we turn it back on; otherwise we're simply running on sheer blind inertia and unthinking myopia. Every month since Dec. 2005 we should have been asking whether the costs have been worth the benefits.
"1. Annoying anons may simply decide to create accounts and make annoying nonsense pages anyway. This will certainly be true in some cases, but it is an empirical question as to how many."
Quite a few. I haven't seen much of a reduction in PROD or AFD or speedies ([[User:Dragons flight/Category tracker]] shows that Speedy has at times reached 349 entries, and it tracks back to late July 2006).
"2. We will lose good new pages created by anons of good will. This may cause the growth of English Wikipedia (in terms of the number of articles) to slow a little bit. With 800,000+ articles, and ever-increasing traffic to the website, this seems to be a worthwhile cost."
PR-wise, turning off page creation wasn't good, to say the least; it has forced all sorts of ugly hacks to pages and annoyed many many people (such as persons like me; one cannot even create a simple redirect when not staging out of one's computer/account). Take a look at the monumental failure that is AFC sometime, which turning off page creation has forced on us. Valid, good articles are being entombed there.
There's no time like the present. It's summer now, so dedicate Wikipedians have plenty of time, and the September deluge is not yet upon us; there are even more Wikipedians than ever, and more articles, and better tools. How much have the anti-vandal tools (too many to name now, even excluding bots) proliferated and improved since that long-ago December?
"But preveneting [sic] anons from creating new pages is a different matter, and it seems a worthy time to make an experiment of it."
Yes, let's. Experiments go both ways, remember...
~maru _______________________________________________ WikiEN-l mailing list WikiEN-l@Wikipedia.org To unsubscribe from this mailing list, visit: http://mail.wikipedia.org/mailman/listinfo/wikien-l
On 8/25/06, Stefan Sittler kiaparowits@gmail.com wrote:
Instead, it's an experiment of indefinite duration with analysis (of an unknown kind? where is this analysis described?) presumably being done by some person or persons, probably, and the results to be revealed at some later unspecified date.
Think of it as a "trial" or a "change with unspecified duration", rather than an "experiment", and the pain will go away.
Steve
I don't like trials of indefinite length either :)
On 8/25/06, Steve Bennett stevagewp@gmail.com wrote:
On 8/25/06, Stefan Sittler kiaparowits@gmail.com wrote:
Instead, it's an experiment of indefinite duration with analysis (of an unknown kind? where is this analysis described?) presumably being done by some person or persons, probably, and the results to be revealed at some later unspecified date.
Think of it as a "trial" or a "change with unspecified duration", rather than an "experiment", and the pain will go away.
Steve _______________________________________________ WikiEN-l mailing list WikiEN-l@Wikipedia.org To unsubscribe from this mailing list, visit: http://mail.wikipedia.org/mailman/listinfo/wikien-l
Steve Bennett wrote:
On 8/25/06, Stefan Sittler kiaparowits@gmail.com wrote:
Instead, it's an experiment of indefinite duration with analysis (of an unknown kind? where is this analysis described?) presumably being done by some person or persons, probably, and the results to be revealed at some later unspecified date.
Think of it as a "trial" or a "change with unspecified duration", rather than an "experiment", and the pain will go away.
It doesn't, because "trial" is still just as misleading as "experiment" without some plan for evaluating the trial's results and "change with unspecified duration" is just a euphemism for "change with indefinite duration."
I don't have any fundamental problem with making this a permanent change provided a rigorous analysis actually shows that it helped the situation. I'm told that an analysis has been done. So there's an easy and obvious way to make me quit complaining about this issue.
On 8/25/06, Bryan Derksen bryan.derksen@shaw.ca wrote:
Steve Bennett wrote:
On 8/25/06, Stefan Sittler kiaparowits@gmail.com wrote:
Instead, it's an experiment of indefinite duration with analysis (of an unknown kind? where is this analysis described?) presumably being done by some person or persons, probably, and the results to be revealed at some later unspecified date.
Think of it as a "trial" or a "change with unspecified duration", rather than an "experiment", and the pain will go away.
It doesn't, because "trial" is still just as misleading as "experiment" without some plan for evaluating the trial's results and "change with unspecified duration" is just a euphemism for "change with indefinite duration."
Or to quote Jimmy wales, "we should proceed experimentally to see if this helps or not. If not, we try something else. Calvinball."
On 26/08/06, Anthony wikilegal@inbox.org wrote:
Or to quote Jimmy wales, "we should proceed experimentally to see if this helps or not. If not, we try something else. Calvinball."
I think that that's true atleast.
I'm coming at the new article issue from the opposite direction.
Isn't the question how we can leverage the editors so that they can easily remove new, inappropriate articles, rather than trying to stop people from creating new articles?
And that ties back to visibility. Right now, new articles have low visibility. That's really bad.
Perhaps we should insist that all new articles be classified in some way, like the dewey decimal system or something, and allow editors to see new articles in their favourite classification in their watchlists or something.
That way, the new articles will appear and immediately get jumped on; and the pain of policing new articles is distributed across all the editors as much as possible.
Ian Woollard ian.woollard@gmail.com wrote:
On 26/08/06, Anthony wrote:
Or to quote Jimmy wales, "we should proceed experimentally to see if this helps or not. If not, we try something else. Calvinball."
I think that that's true atleast.
I think it's smart too and will turn out to be VERY beneficial.
I'm coming at the new article issue from the opposite direction.
Isn't the question how we can leverage the editors so that they can easily remove new, inappropriate articles, rather than trying to stop people from creating new articles?
If you stop or significantly slow the creation of new, inappropriate articles, editors have more time to work on existing articles. I think you'll also find that the majority of new articles that are kept or don't need complete rewriting are by registered users. In fact, I'd bet if you pulled the stats that the same is true of all edits, not just creation.
I'd like to see the experiment applied to all edits. I.e., no anon edits period. Registration is so simple and requires so little info (name and password only) that it serves less as removing anonymity and more as simply forcing vandals to do more in order to vandalize. If anything, it protects naive users from publicly broadcasting their IP to those looking for easy targets.
And that ties back to visibility. Right now, new articles have low visibility. That's really bad.
Perhaps we should insist that all new articles be classified in some way, like the dewey decimal system or something, and allow editors to see new articles in their favourite classification in their watchlists or something.
With the exception of any actual dewey devices, that seems like a useful idea regardless. There appears, however, to already be a page that can be bookmarked and reviewed: http://en.wikipedia.org/wiki/Special:Newpages If you go to http://en.wikipedia.org/wiki/Special:Recentchanges you can see a variety of other utilities already in place.
~~Pro-Lick http://en.wiktionary.org/wiki/User:Halliburton_Shill http://en.wikipedia.org/wiki/User:Pro-Lick http://www.wikiality.net/index.php?title=User:Pro-Lick --------------------------------- Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls. Great rates starting at 1¢/min.
On 26/08/06, Cheney Shill halliburton_shill@yahoo.com wrote:
If you stop or significantly slow the creation of new, inappropriate articles, editors have more time to work on existing articles.
No, volunteers want to do what they want to do. If you stop them doing what they want to do, that doesn't mean they'll do what they don't want to do.
- d.
On 26 Aug 2006, at 11:41, David Gerard wrote:
On 26/08/06, Cheney Shill halliburton_shill@yahoo.com wrote:
If you stop or significantly slow the creation of new, inappropriate articles, editors have more time to work on existing articles.
No, volunteers want to do what they want to do. If you stop them doing what they want to do, that doesn't mean they'll do what they don't want to do.
I agree. But one of the core aims of Wikipedia is to promote free content. By stopping people adding non-free content, some proportion of people who don't care whether their content is free or not will add free content.
It's a conflict between building an encyclopaedia and building a respository of free knowledge.
On 8/26/06, David Gerard dgerard@gmail.com wrote:
No, volunteers want to do what they want to do. If you stop them doing what they want to do, that doesn't mean they'll do what they don't want to do.
I think it's /possible/ he meant that if users have to spend less time deleting articles that shouldn't have been created, they could spend time editing rather than deleting.
I'd disagree with that, too; the kind of people who like going through recent changes and doing janitorial work won't start writing; they'll start doing other kinds of janitorial work. However, such work could be more useful, granted.
-Matt
On 8/26/06, Cheney Shill halliburton_shill@yahoo.com wrote:
Ian Woollard ian.woollard@gmail.com wrote:
On 26/08/06, Anthony wrote:
Or to quote Jimmy wales, "we should proceed experimentally to see if this helps or not. If not, we try something else. Calvinball."
I think that that's true atleast.
I think it's smart too and will turn out to be VERY beneficial.
I'm coming at the new article issue from the opposite direction.
Isn't the question how we can leverage the editors so that they can easily remove new, inappropriate articles, rather than trying to stop people from creating new articles?
If you stop or significantly slow the creation of new, inappropriate articles, editors have more time to work on existing articles. I think you'll also find that the majority of new articles that are kept or don't need complete rewriting are by registered users. In fact, I'd bet if you pulled the stats that the same is true of all edits, not just creation.
I agree with David that volunteers are going to do what they want. However, I also agree with Matt. *IF* you can significantly slow the creation of new, inappropriate articles, (and all other things remain equal), the people who enjoy dealing with new articles will have more time, time which at least some of them will likely spend productively.
This would be a good thing, for sure. But having seen no study conducted, I'm not convinced that this can be achieved. And also, it's pretty much obvious that all other things *don't* remain equal. At least some people have switched their attention to Articles for Creation, for instance, an activity which wasn't necessary before this change.
I think Ian explained it well. "how we can leverage the editors so that they can easily remove new, inappropriate articles". Moreover, how can avoid excessive duplication of effort when identifying new, appropriate articles? It would be nice if we could flip a switch and turn off bad articles, but short of requiring adminship to create new articles I don't see it happening. Improving the collaboration process, on the other hand, that's much easier, and more the basis of Wikipedia - trust that most people are trying to do the right thing, and give them the tools to do it.
I'd like to see the experiment applied to all edits. I.e., no anon edits period. Registration is so simple and requires so little info (name and password only) that it serves less as removing anonymity and more as simply forcing vandals to do more in order to vandalize. If anything, it protects naive users from publicly broadcasting their IP to those looking for easy targets.
I wouldn't mind having the experiment applied to all edits, but only if it was for a very short period of time (definitely less than a month), and only if there were clear criteria for measuring the effectiveness.
That said, I don't really see the point. It makes it harder for vandals to edit, but it also makes it harder for non-vandals to edit. Other than making it harder for anyone without an account to edit (as well as anyone with an account who isn't always logged in), it doesn't really accomplish anything (protecting naive users could be accomplished by simply not publishing the IP address of "anons" but instead assigning them a unique number which can only be tied to their IP address through CheckUser). In fact, I'd say vandals tend to be more persistent than casual copyeditors, and would be *less* likely to give up when they find they have to register.
One last point to consider - you don't want to have a four day wait before you can edit, right? Assuming not, the barrier to vandals (and to non-vandals) would be very low.
Anthony
On 8/26/06, Anthony wikilegal@inbox.org wrote:
This would be a good thing, for sure. But having seen no study conducted, I'm not convinced that this can be achieved. And also, it's pretty much obvious that all other things *don't* remain equal. At least some people have switched their attention to Articles for Creation, for instance, an activity which wasn't necessary before this change.
The quality of articles coming out of AfC is higher than the quality of articles written directly by anons. Or at least it is when I do it ;)
Steve
On 26 Aug 2006, at 01:01, Bryan Derksen wrote:
Steve Bennett wrote:
On 8/25/06, Stefan Sittler kiaparowits@gmail.com wrote:
Instead, it's an experiment of indefinite duration with analysis (of an unknown kind? where is this analysis described?) presumably being done by some person or persons, probably, and the results to be revealed at some later unspecified date.
Think of it as a "trial" or a "change with unspecified duration", rather than an "experiment", and the pain will go away.
It doesn't, because "trial" is still just as misleading as "experiment" without some plan for evaluating the trial's results and "change with unspecified duration" is just a euphemism for "change with indefinite duration."
I don't have any fundamental problem with making this a permanent change provided a rigorous analysis actually shows that it helped the situation. I'm told that an analysis has been done. So there's an easy and obvious way to make me quit complaining about this issue.
You seem to be asking for NPOV - a bit revolutionary ;-)