Perhaps I'm the only one who finds this a little concerning in my part, but lately I've been feeling that too many users are trying to watch too much of Wikipedia at one time.
Let me elaborate a little. It isn't uncommon for the normally active user to have hundreds, if not thousands, of pages on their watchlist. Then, when somebody makes an edit that a certain user doesn't agree with, it gets changed or outright reverted. It's like, at the least, a form of a bunch of "Big Brothers" looking over an article and, at the worst, an outright form of page ownership.
I've been on the low end on watchlisting pages myself, but a couple of months ago I decided to "unload" my watchlist, removing most articles that I have extensively worked on since I came onboard -- going from about 50 pages watched to about fewer than 10 pages watched, only keeping those I'm monitoring in the short-term.
Personally, I found unloading my watchlist liberating, and I would hope that more would do the same. There's always that steady stream of vandal-fighters to stomp out any clear vandalism that pops up. It's hard to explain, but I think it's a good exercise in assuming good faith that others will make constructive edits in efforts to improve pages.
-MuZemike
2009/12/10 Mike Pruden mikepruden@yahoo.com:
Personally, I found unloading my watchlist liberating, and I would hope that more would do the same. There's always that steady stream of vandal-fighters to stomp out any clear vandalism that pops up. It's hard to explain, but I think it's a good exercise in assuming good faith that others will make constructive edits in efforts to improve pages.
I gave up using my watchlist in late 2004. Haven't missed it.
- d.
On Thu, Dec 10, 2009 at 11:30 AM, Mike Pruden mikepruden@yahoo.com wrote:
Perhaps I'm the only one who finds this a little concerning in my part, but lately I've been feeling that too many users are trying to watch too much of Wikipedia at one time.
Let me elaborate a little. It isn't uncommon for the normally active user to have hundreds, if not thousands, of pages on their watchlist. Then, when somebody makes an edit that a
People still use watchlists? Mine quickly exploded out of control (I was briefly in the habit of reading the article of the day each day and making a couple of edits). Instead, I use: 1) The 'related changes' view of a list of stubs that I've written. 2) "My contributions".
It's imperfect, but it sort of captures what I'm interested in: changes made to any article I've created (I like watching the little things grow, flower, die) , and changes made to any changes I've made recently (to see how other people react etc). I've requested a single view that captures all this, a couple of times.
The depressing thing about the first one is seeing just how much time and effort is expended on apparently trivial stuff: stub sorting, recategorising, interwikis, very minor style stuff. Probably 2/3 of the edits to my stubs don't affect the actual content of the article at all.
I agree with the central point though: it's very easy to get into the habit of pouncing on any change. And inefficient, too. It would barely take more time for me to assess and react to a month's (or six month's!) worth of changes, than a day's worth.
Steve
Mike Pruden wrote:
It isn't uncommon for the normally active user to have hundreds, if not thousands, of pages on their watchlist. Then, when somebody makes an edit that a certain user doesn't agree with, it gets changed or outright reverted. It's like, at the least, a form of a bunch of "Big Brothers" looking over an article and, at the worst, an outright form of page ownership.
I have around 8000 pages watchlisted at present. Having a long watchlist is actually an antidote to thinking you have to "curate" each change.
I've been on the low end on watchlisting pages myself, but a couple of months ago I decided to "unload" my watchlist, removing most articles that I have extensively worked on since I came onboard -- going from about 50 pages watched to about fewer than 10 pages watched, only keeping those I'm monitoring in the short-term.
Personally, I found unloading my watchlist liberating, and I would hope that more would do the same. There's always that steady stream of vandal-fighters to stomp out any clear vandalism that pops up. It's hard to explain, but I think it's a good exercise in assuming good faith that others will make constructive edits in efforts to improve pages.
The logic is wrong, in that the "pile-up" factor is not the main issue: coverage on someone's watchlist at all is the issue. Divide the number of articles by the number of active Wikipedians and you find that unless many people have four-figure watchlist lengths there will be plenty not watched at all. Vandal-fighting via Recent Changes doesn't do badly, but it's not an exact science (reverting the last edit doesn't get to clusters of bad edits, and can "cover up" more serious issues), and I doubt it is equally effective at all times of day. I reset my watchlist when it hit 30,000 pages (that really was too much), but the problems of "ownership" and excess reversion are not actually problems about how much you watch.
Charles
2009/12/10 Charles Matthews charles.r.matthews@ntlworld.com:
The logic is wrong, in that the "pile-up" factor is not the main issue: coverage on someone's watchlist at all is the issue. Divide the number of articles by the number of active Wikipedians and you find that unless many people have four-figure watchlist lengths there will be plenty not watched at all.
http://en.wikipedia.org/wiki/Special:UnwatchedPages (visible to admins only, for obvious reasons) has nothing in the list at all, so someone bothers putting stuff on at least one watchlist.
How well it's *actually* watched is, of course, another matter ...
- d.
Strangely enough, the flaggedrevisions feature seems to provide a lot of what we need: 1) People don't have to watch changes as they happen, they can stumble on them when they go to save a new change 2) Changes are marked as patrolled, so far more efficient than 10 people all noticing the same change on their watchlist and deciding no action needs to be taken.
Steve
Steve Bennett wrote:
Strangely enough, the flaggedrevisions feature seems to provide a lot of what we need:
- People don't have to watch changes as they happen, they can stumble
on them when they go to save a new change 2) Changes are marked as patrolled, so far more efficient than 10 people all noticing the same change on their watchlist and deciding no action needs to be taken.
This upgrade was due ... about now?
Charles
2009/12/10 Steve Bennett stevagewp@gmail.com:
Strangely enough, the flaggedrevisions feature seems to provide a lot of what we need:
- People don't have to watch changes as they happen, they can stumble on
them when they go to save a new change 2) Changes are marked as patrolled, so far more efficient than 10 people all noticing the same change on their watchlist and deciding no action needs to be taken.
3)The massive backlog in patrolled edits will kill the instant feedback wikipedia currently gives and reduce editing to a level where watching everything is no longer a problem.
On Fri, Dec 11, 2009 at 12:49 AM, geni geniice@gmail.com wrote:
3)The massive backlog in patrolled edits will kill the instant feedback wikipedia currently gives and reduce editing to a level where watching everything is no longer a problem.
Only if all pages are set to show only the patrolled version to newbies. Which is not going to happen. It may happen to a few hundred pages.
Steve
2009/12/10 geni geniice@gmail.com:
2009/12/10 Steve Bennett stevagewp@gmail.com:
Strangely enough, the flaggedrevisions feature seems to provide a lot of what we need:
- People don't have to watch changes as they happen, they can stumble on
them when they go to save a new change 2) Changes are marked as patrolled, so far more efficient than 10 people all noticing the same change on their watchlist and deciding no action needs to be taken.
3)The massive backlog in patrolled edits will kill the instant feedback wikipedia currently gives and reduce editing to a level where watching everything is no longer a problem.
...except both patrolled and unpatrolled edits are intended to be visible immediately.
http://en.wikipedia.org/wiki/Wikipedia:Patrolled_revisions
I doubt that "your edit has, if you look closely, not yet been ticked off by a system that you don't see unless you look at it" is going to kill the instant-editing culture, somehow.
On Thu, Dec 10, 2009 at 7:52 AM, Steve Bennett stevagewp@gmail.com wrote:
Strangely enough, the flaggedrevisions feature seems to provide a lot of what we need:
- People don't have to watch changes as they happen, they can stumble on
them when they go to save a new change 2) Changes are marked as patrolled, so far more efficient than 10 people all noticing the same change on their watchlist and deciding no action needs to be taken.
I do not think we are planning to implement that sort of flagged revisions on enwiki any time soon. The plan is just to enable flagged revisions on problematic articles, as a sort of "semiprotection-light".
I agree that it would be helpful to know which edits have already been reviewed, to save myself the effort of reviewing them again. But this leads to all sorts of problems, such as whether I really trust the other reviewers enough not to look at the diffs myself.
- Carl
David Gerard wrote:
2009/12/10 Charles Matthews charles.r.matthews@ntlworld.com:
The logic is wrong, in that the "pile-up" factor is not the main issue: coverage on someone's watchlist at all is the issue. Divide the number of articles by the number of active Wikipedians and you find that unless many people have four-figure watchlist lengths there will be plenty not watched at all.
http://en.wikipedia.org/wiki/Special:UnwatchedPages (visible to admins only, for obvious reasons) has nothing in the list at all, so someone bothers putting stuff on at least one watchlist.
How well it's *actually* watched is, of course, another matter ...
Perhaps one of our wizards could check how many pages are not watched by anybody who has edited in 2009.
Charles
On Fri, Dec 11, 2009 at 2:38 AM, Charles Matthews charles.r.matthews@ntlworld.com wrote:
Perhaps one of our wizards could check how many pages are not watched by anybody who has edited in 2009.
More useful and precise would be collecting page views of diffs. I don't know if we record them or not though.
Steve
On Thu, Dec 10, 2009 at 2:27 AM, Charles Matthews < charles.r.matthews@ntlworld.com> wrote:
Mike Pruden wrote:
It isn't uncommon for the normally active user to have hundreds, if not
thousands, of pages on their watchlist. Then, when somebody makes an edit that a certain user doesn't agree with, it gets changed or outright reverted. It's like, at the least, a form of a bunch of "Big Brothers" looking over an article and, at the worst, an outright form of page ownership.
I have around 8000 pages watchlisted at present. Having a long watchlist is actually an antidote to thinking you have to "curate" each change.
I agree with Charles (my watchlist is in the mid-7000s) - with a large watchlist you lose that sense of ownership. It's just something you scan from time to time. You pick up the anon edits on backwater pages. But you can't control what's going on, so you just accept...
On Wed, Dec 9, 2009 at 7:30 PM, Mike Pruden mikepruden@yahoo.com wrote:
Personally, I found unloading my watchlist liberating, and I would hope that more would do the same. There's always that steady stream of vandal-fighters to stomp out any clear vandalism that pops up.
What about edits that are not clear vandalism, but are simply erroneous, or demonstrate a complete misunderstanding of the topic? Or edits that are generally correct, but require significant copyediting? I don't believe the vandal-fighters are going to handle those things. I feel much better knowing that many articles have people watching them who are familiar with the subject area.
- Carl