http://www.guardian.co.uk/technology/2009/aug/12/wikipedia-deletionist-inclu...
Much familiar argument from threads here. Some of the usual suspects commenting, and everyone putting in their two cents. Somewhere in the middle is a debate struggling to get out: is the volume of reversions indicative of good gatekeeping (poor edits to popular and well-developed articles have little chance of sticking), or bad gatekeeping (established editors assert ownership)? Stats from 2007 and 2009 show a step-change of some sort, as we know, but don't really prove that there is a current trend (we could be going sideways).
Charles
----- "Charles Matthews" charles.r.matthews@ntlworld.com wrote:
From: "Charles Matthews" charles.r.matthews@ntlworld.com
is the volume of reversions indicative of good gatekeeping (poor edits to popular and well-developed articles have little chance of sticking), or bad gatekeeping (established editors assert ownership)? Stats from 2007 and 2009 show a step-change of some sort, as we know ...
I personally have seen a considerable change in reversions since I started editing in ~2006 - driven by automated reversions. When I started editing I would often see vandalism on my watchlist which I would go and revert. I haven't done that for ages, because it's rarely there by the time I get to it!
On that basis I would suggest - good gatekeeping!
Andrew
The 'limit' that's being reached is the article count; so reverts aren't the question.
The real question is whether the AFD process is working correctly, particularly for new articles, now that the low-hanging fruit is gone.
I've personally seen several of my referenced articles that in all honesty didn't violate a single policy get AFDd; one was 'merged' in 40 minutes of the review starting by the admin who also voted in the review, and then he unilaterally decided the results of review was something that not even he voted for!?!
That marks a new low point for the AFD process I think;
(FWIW it got overturned at DRV, but then deleted anyway, but not for violating policy that I could point you to...)
Right now the AFD process never looks for potential in articles and never looks at violations of policy, it's simply a popularity contest for articles; articles that haven't been created yet are inevitably less popular topics, so are even more likely to get deleted out of hand.
That's not the way it's supposed to work, but that's the way it does work. ;-)
There is a similar discussion at the en.wiki Village pump on this, and I think it runs parallel to this discussion.
I tend to think that the bigger problems with AFD is the lack of participation on many of them; if you disagree, then look at the bottom of each page of any given day's AFD and observe all the relisted (some twice) discussions that have zero, perhaps one other comment.
I think people are discouraged from participating in AFD discussions or even sounding off what they think the best actions should be taken for two reasons. First, there are those who are afraid to participate in AFDs for the fear of being pigeonholed or labeled (i.e. as an inclusionist or a deletionist when they are actually in the middle). Second, there are those users who are discouraged or perhaps disgusted over the level of wikilawyering and incivility that permeates especially in the more contentious discussions.
I partially fall into both camps myself. When I first came on over a year ago, I took a lot of advice from the Help pages (as the article indicates, taking it in by 'osmosis'). [[Help:Contents]] mentioned "deleting pages", which I kind of took as some sort of a Wiki-responsibility to participate in AFD discussions, or at the very least, chime in. And that's what I did for about the first six months or so in which I have been active. However, that has slowed down greatly since then for the reasons that I have explained above.
Not to say that I don't regret not having participated in AFD discussions frequently. In fact, it was through AFD where I have learned about everything policy and guideline-wise that I know now, how to interact with other users as well as what to do and what not to do in a general sense (not just in AFD discussions but in any WP discussion).
With that said, I don't think the pros or cons of AFD are going to necessarily effect article volume. There are always going to be new topics out there, which will perpetually facilitate new articles. Looking at the big picture with regards to the current information that exists on Wikipedia, it ultimately becomes an exercise on "how" to organize said information as opposed to squeezing every last bit, every last letter, into everything (not that you don't see the latter, as that happens very often, as well).
Speedy deletion is going to happen, PROD is going to happen, AFD is going to happen. The rest, as Ian puts it, depends on the integrity and open-mindedness of the editors out there to make sure they're running with the intention of the encyclopedia's net benefit in mind.
-MuZemike
--- On Thu, 8/13/09, Ian Woollard ian.woollard@gmail.com wrote:
From: Ian Woollard ian.woollard@gmail.com Subject: Re: [WikiEN-l] "Wikipedia approaches its limits" - Technology Guardian To: charles.r.matthews@ntlworld.com, "English Wikipedia" wikien-l@lists.wikimedia.org Date: Thursday, August 13, 2009, 11:08 AM The 'limit' that's being reached is the article count; so reverts aren't the question.
The real question is whether the AFD process is working correctly, particularly for new articles, now that the low-hanging fruit is gone.
I've personally seen several of my referenced articles that in all honesty didn't violate a single policy get AFDd; one was 'merged' in 40 minutes of the review starting by the admin who also voted in the review, and then he unilaterally decided the results of review was something that not even he voted for!?!
That marks a new low point for the AFD process I think;
(FWIW it got overturned at DRV, but then deleted anyway, but not for violating policy that I could point you to...)
Right now the AFD process never looks for potential in articles and never looks at violations of policy, it's simply a popularity contest for articles; articles that haven't been created yet are inevitably less popular topics, so are even more likely to get deleted out of hand.
That's not the way it's supposed to work, but that's the way it does work. ;-)
-- -Ian Woollard
"All the world's a stage... but you'll grow out of it eventually."
WikiEN-l mailing list WikiEN-l@lists.wikimedia.org To unsubscribe from this mailing list, visit: https://lists.wikimedia.org/mailman/listinfo/wikien-l
"Michael Pruden" mikepruden@yahoo.com wrote in message news:515438.44185.qm@web32604.mail.mud.yahoo.com... ...pigeonholed (i.e. as an inclusionist or a deletionist when they are actually in the middle).
Merjists are both, and they do not need to participate in any AfD discussion, because the articles they redirect do not actually get deleted. IOW, any user can undo a merj, because both articles exist: Seeing the history for the deleted articles is only a matter of either writing or finding ?redirect=no. So, in a way, they are also neither, because deleted material should appear at the redirection destination, so I guess they are net inclusionists. This is of course only applicable to notable articles that are longstanding synonyms or close cousins. I think it is also possible to be a pre-emptive, deletionistic merjist and prevent new articles from being created when their content already exists, elsewhere, under a synonym. I tried doing some of that in [[recent changes#requested articles]], and I was chastized for some of it -- did not hit the best mark, I suppose. _______ http://ecn.ab.ca/~brewhaha/WP_CRYSTAL.HTM written from the merjist POV.
On Thu, Aug 13, 2009 at 4:58 AM, Charles Matthewscharles.r.matthews@ntlworld.com wrote:
http://www.guardian.co.uk/technology/2009/aug/12/wikipedia-deletionist-inclu...
Much familiar argument from threads here. Some of the usual suspects commenting, and everyone putting in their two cents. Somewhere in the middle is a debate struggling to get out: is the volume of reversions indicative of good gatekeeping (poor edits to popular and well-developed articles have little chance of sticking), or bad gatekeeping (established editors assert ownership)? Stats from 2007 and 2009 show a step-change of some sort, as we know, but don't really prove that there is a current trend (we could be going sideways).
Charles
Regarding the familiar arguments related to this... should the Signpost be a venue for discussing thing stuff? See: http://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/Newsroom/Suggestio...
-Sage
Sage Ross wrote:
On Thu, Aug 13, 2009 at 4:58 AM, Charles Matthewscharles.r.matthews@ntlworld.com wrote:
http://www.guardian.co.uk/technology/2009/aug/12/wikipedia-deletionist-inclu...
Much familiar argument from threads here. Some of the usual suspects commenting, and everyone putting in their two cents. Somewhere in the middle is a debate struggling to get out: is the volume of reversions indicative of good gatekeeping (poor edits to popular and well-developed articles have little chance of sticking), or bad gatekeeping (established editors assert ownership)? Stats from 2007 and 2009 show a step-change of some sort, as we know, but don't really prove that there is a current trend (we could be going sideways).
Charles
Regarding the familiar arguments related to this... should the Signpost be a venue for discussing thing stuff? See: http://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/Newsroom/Suggestio...
I think you're right to suspect that this would be hard to cover _properly_ in the Signpost's usual and gratefully concise style. Just picking out the different strands of "deletionism" looks like several pages of philosophy tutorial to me. Stats are interesting, but stats on reversions without a proper indication of their distribution (are they largely in the top 1000 articles by readers?) seem fairly inconclusive.
Charles
Maybe we should stop reverting vandalism. It would improve our statistics, after all.
-Luna
Ha, okay, one of those rare lols where I actually laughed out loud.
-- Sent from my Palm Pre Luna wrote:
Maybe we should stop reverting vandalism. It would improve our statistics,
after all.
-Luna
_______________________________________________
WikiEN-l mailing list
WikiEN-l@lists.wikimedia.org
To unsubscribe from this mailing list, visit:
"Luna" lunasantin@gmail.com wrote in message news:1adf38660908131222j772acaa8h544580d05e716d02@mail.gmail.com...
Maybe we should stop reverting vandalism. It would improve our statistics, after all.
What? And let it out that wikipedians hav WebCams on celebrity sex life?
On Thu, Aug 13, 2009 at 6:58 PM, Charles Matthewscharles.r.matthews@ntlworld.com wrote:
http://www.guardian.co.uk/technology/2009/aug/12/wikipedia-deletionist-inclu...
"Meanwhile, for those who did not invest vast amounts of time in editing, the experience was very different. "For editors that make between two and nine edits a month, the percentage of their edits being reverted had gone from 5% in 2004 all the way up to about 15% by October 2008. And the 'onesies' – people who only make one edit a month – their edits are now being reverted at a 25% rate," Chi explains."
My sense is that the reason for this is that the experienced editors are now more certain of what is a good edit and what is a bad one. Back in the day, any edits were welcome. You could write a sentence on [[George W Bush]] and it would be an improvement on the void before it. But now, most random edits are going to be to articles that are quite polished, and don't need random edits. They need thought out edits that conform to policies.
Summary: With the encyclopaedia being bigger and more complete, it's less likely that a "onesie"'s edit is worth keeping.
The 1% reversion rate for experienced editors was also interesting. I doubt my edits get reverted at anything like that high a rate.
Steve
Steve Bennett wrote:
The 1% reversion rate for experienced editors was also interesting. I doubt my edits get reverted at anything like that high a rate.
Yes, the mean here might tell less than the median. (I.e. you'd expect to see very different figures for controversial and non-controversial articles, and lumping all articles and frequent editors together and averaging isn't going to be that helpful.)
As http://asc-parc.blogspot.com/ admits, they (the PARC people) have basically done a press release on a conference paper that won't be produced until WikiSym in October. This would account for the rather sensationalised tone of it all.
Charles
2009/8/17 Steve Bennett stevagewp@gmail.com:
Summary: With the encyclopaedia being bigger and more complete, it's less likely that a "onesie"'s edit is worth keeping. The 1% reversion rate for experienced editors was also interesting. I doubt my edits get reverted at anything like that high a rate.
It can be problematic. I frequently edit as an IP when I'm at another machine and can't be bothered logging in. The unexplained reversion rate is *much* higher than when I edit logged-in, even though the edits are exactly the same sort of thing.
(Usual culprit: overenthusiastic use of Twinkle. When you say "that was me, what was the purpose of this reversion?" the usual response is blustering and "HOW CAN I KEEP UP WITH THE EDITS IF I HAVE TO THINK ABOUT THEM" or similar. I know that's nothing like all Twinkle users, but a lot of this does noticeably come from Twinkle users.)
I urge any editor who's been around a while to try editing as an IP, and see what the reversion rate is. Then ask the reverter what their reasoning was for each reversion. They should be able to justify it, after all, even with a "sorry, slipped up." Which is fine too, just please show evidence of thought.
- d.
On Wed, Aug 19, 2009 at 3:04 AM, David Gerarddgerard@gmail.com wrote:
It can be problematic. I frequently edit as an IP when I'm at another machine and can't be bothered logging in. The unexplained reversion rate is *much* higher than when I edit logged-in, even though the edits are exactly the same sort of thing.
Ah, yes. This was an obvious test I should have thought of.
One of my pet hates: when an IP changes a figure in in infobox or somewhere in article, with no comment, and no source. I've heard reports of people doing this as sport, just to be annoying, but in my experience, they're often right. But it leaves you in a real quandary, if you can't verify it either way.
Steve
On 19/08/2009, Steve Bennett stevagewp@gmail.com wrote:
On Wed, Aug 19, 2009 at 3:04 AM, David Gerarddgerard@gmail.com wrote:
It can be problematic. I frequently edit as an IP when I'm at another machine and can't be bothered logging in. The unexplained reversion rate is *much* higher than when I edit logged-in, even though the edits are exactly the same sort of thing.
Ah, yes. This was an obvious test I should have thought of.
One of my pet hates: when an IP changes a figure in in infobox or somewhere in article, with no comment, and no source. I've heard reports of people doing this as sport, just to be annoying, but in my experience, they're often right. But it leaves you in a real quandary, if you can't verify it either way.
I normally revert those, unless you can verify it it's just an unreferenced change. You can leave a message on their talk page though asking for a ref. Same goes for logged-ins.
Steve
On Wed, Aug 19, 2009 at 6:55 AM, Ian Woollardian.woollard@gmail.com wrote:
One of my pet hates: when an IP changes a figure in in infobox or somewhere in article, with no comment, and no source. I've heard reports of people doing this as sport, just to be annoying, but in my experience, they're often right. But it leaves you in a real quandary, if you can't verify it either way.
I normally revert those, unless you can verify it it's just an unreferenced change. You can leave a message on their talk page though asking for a ref. Same goes for logged-ins.
I see a lot of these patrolling recent changes in Huggle. I look at the user's other contribs and provided I can find just one in the same day where he's blanked the page and written "SUCK MY ASS!!!" I'll revert the numeric change and put "rv numerical change by bad faith editor but editors may wish to double-check" as an edit summary.
Another warning sign is a number of numeric changes, without any other sort of edit, in completely unrelated types of articles. I wouldn't necessarily rv on that basis but I probably would if they've had any sort of warning that day.
On Wed, Aug 19, 2009 at 5:48 PM, Bod Notbodbodnotbod@gmail.com wrote:
On Wed, Aug 19, 2009 at 6:55 AM, Ian Woollardian.woollard@gmail.com wrote:
One of my pet hates: when an IP changes a figure in in infobox or somewhere in article, with no comment, and no source. I've heard reports of people doing this as sport, just to be annoying, but in my experience, they're often right. But it leaves you in a real quandary, if you can't verify it either way.
I normally revert those, unless you can verify it it's just an unreferenced change. You can leave a message on their talk page though asking for a ref. Same goes for logged-ins.
I see a lot of these patrolling recent changes in Huggle. I look at the user's other contribs and provided I can find just one in the same day where he's blanked the page and written "SUCK MY ASS!!!" I'll revert the numeric change and put "rv numerical change by bad faith editor but editors may wish to double-check" as an edit summary.
What you need to check here is that the editor in question isn't reverting vandalism by someone else. In other words, you need to check back further through the edit history to make sure you are reverting to the last clean version. I've seen cases of HUGGLE and TWINKLE users reverting a vandalised page to a still-vandalised state, and no-one else checking, and such vandalised pages (now with the "legitimacy" of a revert from an "approved" user) staying in that state for months.
Another warning sign is a number of numeric changes, without any other sort of edit, in completely unrelated types of articles. I wouldn't necessarily rv on that basis but I probably would if they've had any sort of warning that day.
Same comment as above. Reverting should never be done without checking what you are reverting TO.
Carcharoth
Carcharoth wrote:
...I've seen cases of HUGGLE and TWINKLE users reverting a vandalised page to a still-vandalised state, and no-one else checking, and such vandalised pages (now with the "legitimacy" of a revert from an "approved" user) staying in that state for months.
Indeed. And I've seen canny vandals instigate a deliberate chain of contradictory vandalism (perhaps involving sockpuppets) with the apparent intent of goading well-meaning but careless vandalism patrollers into doing precisely this. (It's an annoyingly effective technique.)
On Wed, Aug 19, 2009 at 6:41 PM, Carcharothcarcharothwp@googlemail.com wrote:
I see a lot of these patrolling recent changes in Huggle. I look at the user's other contribs and provided I can find just one in the same day where he's blanked the page and written "SUCK MY ASS!!!" I'll revert the numeric change and put "rv numerical change by bad faith editor but editors may wish to double-check" as an edit summary.
What you need to check here is that the editor in question isn't reverting vandalism by someone else. In other words, you need to check back further through the edit history to make sure you are reverting to the last clean version.
I do check what I'm reverting back to. The "editor in question" as I say, is the "SUCK MY ASS!!!" guy usually. If I go back through the article history I often see that there's a revert war, but usually the vandal leaves no edit summary and the person reverting in good faith does. So that's easy enough to spot. In short I tend to flick through both the editor's contribs AND the article history and get a rounded picture.
Steve Summit:
Indeed. And I've seen canny vandals instigate a deliberate chain of contradictory vandalism (perhaps involving sockpuppets) with the apparent intent of goading well-meaning but careless vandalism patrollers into doing precisely this. (It's an annoyingly effective technique.)
Well, I confess that would probably fool me. Although would I be correct in saying that in these cases it would either be a case where; the first sock guise is an anon-IP and the other sock a named user; OR both are named users? Which is to say it would HARDLY EVER be the case that the two guises of the sockpuppet are editing using different anon-IP numbers?
If that's the case, virtually everything that reaches the top of Huggle's To Do list is anonymous-IPs... I hardly ever see an edit by a named user. So provided I'm correct to say that two different anon-IP addresses editing in (usually quick) succession are not going to be one man and his sock, I shouldn't be falling into that pit.
I hope I've made that clear to follow. I have tried.