I believe that Aaron and I discussed the theory that injecting some energy into WikiProjects might be a productive avenue for editor retention and productivity.
Is there a dashboard somewhere that shows community health statistics for WikiProjects, such as:
1. Number of recent edits to articles that have been templated with that project's template 2. Number of active editors in those articles 3. Number of active editors in those articles who are also members of the project 4. Number of editors who have recently edited in the WikiProject's project space and talk pages 5. Whether the project has a newsletter, and if so, readership statistics for it.
Thanks!
Pine
1) plus the number of edits to un-templated articles in the same category, I suppose 2) of those active editors, have they ever posted on Wikiproject talk pages, and if so, which ones? (we should be able to create Wikiproject editor profiles based on # edits to WP talk pages) 3) How to measure participation? Signup page or posts on WP talk pages? 4) it would be nice to know this for overall, e.g of the active editor group of +/- 3500 on enwiki, what percentage has ever written on WP talk pages and who are the top posters? 5) Can only think of one (military history)
On Thu, Jan 7, 2016 at 1:42 AM, Pine W wiki.pine@gmail.com wrote:
I believe that Aaron and I discussed the theory that injecting some energy into WikiProjects might be a productive avenue for editor retention and productivity.
Is there a dashboard somewhere that shows community health statistics for WikiProjects, such as:
- Number of recent edits to articles that have been templated with that
project's template 2. Number of active editors in those articles 3. Number of active editors in those articles who are also members of the project 4. Number of editors who have recently edited in the WikiProject's project space and talk pages 5. Whether the project has a newsletter, and if so, readership statistics for it.
Thanks!
Pine
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
On the topic of Wikiprojects and retention, have there been any attempts at directing newcomers to Wikiprojects?
Sent from my iPhone
On Jan 6, 2016, at 19:42, Pine W wiki.pine@gmail.com wrote:
I believe that Aaron and I discussed the theory that injecting some energy into WikiProjects might be a productive avenue for editor retention and productivity.
Is there a dashboard somewhere that shows community health statistics for WikiProjects, such as:
- Number of recent edits to articles that have been templated with that project's template
- Number of active editors in those articles
- Number of active editors in those articles who are also members of the project
- Number of editors who have recently edited in the WikiProject's project space and talk pages
- Whether the project has a newsletter, and if so, readership statistics for it.
Thanks!
Pine _______________________________________________ Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
*Gabe/Nemo:* There is at least one piece of research that indicates it does, under certain circumstances: http://research.microsoft.com/en-us/um/redmond/groups/connect/CSCW_10/docs/p...
At one point, I started building a WikiProject-matching workflow on the Teahouse (with Nettrom, using SuggestBot). But we never finished or tested it, and life moved on. I believe there have been other plans to do this at other points, inside and outside WMF. To my knowledge, none of these plans came to fruition, but I'd love to hear otherwise.
With the advent of ORES and the recommendation API, I bet a better workflow for matching newbies and WikiProjects could be developed, if there is will enough, and time.
*Pine/Jane:* There's no dashboard, but it's possible to gather these data from the replica DBs (through Quarry, for instance). I may even have some datasets lying around that contain some of the data you're asking for in historical form (or at least pointers to research based on said data). I'll look around and post what I find.
Jonathan
On Thu, Jan 7, 2016 at 9:43 AM, Federico Leva (Nemo) nemowiki@gmail.com wrote:
Gabriel Mugar, 07/01/2016 18:35:
On the topic of Wikiprojects and retention, have there been any attempts at directing newcomers to Wikiprojects?
Many attempts across years and languages. Sending people to places is easy, the issue is whether that does any good.
Nemo
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Thanks Jonathan and Jane. The reason that I ask is that I'm planning to encourage newbies to participate in the Teahouse and in WikiProjects via my grant-funded video project https://meta.wikimedia.org/wiki/Grants:IEG/Motivational_and_educational_video_to_introduce_Wikimedia. It would be nice to know which WikiProjects are currently the healthiest, and more importantly if it's possible, how newcomers can determine for themselves which projects look like a good fit for them at any given point in time.
Pine
On Thu, Jan 7, 2016 at 2:08 PM, Jonathan Morgan jmorgan@wikimedia.org wrote:
*Gabe/Nemo:* There is at least one piece of research that indicates it does, under certain circumstances: http://research.microsoft.com/en-us/um/redmond/groups/connect/CSCW_10/docs/p...
At one point, I started building a WikiProject-matching workflow on the Teahouse (with Nettrom, using SuggestBot). But we never finished or tested it, and life moved on. I believe there have been other plans to do this at other points, inside and outside WMF. To my knowledge, none of these plans came to fruition, but I'd love to hear otherwise.
With the advent of ORES and the recommendation API, I bet a better workflow for matching newbies and WikiProjects could be developed, if there is will enough, and time.
*Pine/Jane:* There's no dashboard, but it's possible to gather these data from the replica DBs (through Quarry, for instance). I may even have some datasets lying around that contain some of the data you're asking for in historical form (or at least pointers to research based on said data). I'll look around and post what I find.
Jonathan
On Thu, Jan 7, 2016 at 9:43 AM, Federico Leva (Nemo) nemowiki@gmail.com wrote:
Gabriel Mugar, 07/01/2016 18:35:
On the topic of Wikiprojects and retention, have there been any attempts at directing newcomers to Wikiprojects?
Many attempts across years and languages. Sending people to places is easy, the issue is whether that does any good.
Nemo
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
-- Jonathan T. Morgan Senior Design Researcher Wikimedia Foundation User:Jmorgan (WMF) https://meta.wikimedia.org/wiki/User:Jmorgan_(WMF)
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
I'm not convinced that the first three of those tell you much about the health of a wikiproject. For example when I first reviewed the word staring I replaced most of them in Bollywood related articles with "starring", or I felt jaundiced "appearing". That would have boosted the first two stats, but I'd dispute that it said anything about the health of the wikiproject.
A few years ago during the cleanup exercise of unreferenced BLPs, I was involved in an exercise where we used bots to notify about 700 wiki projects of unreferenced BLPs in their areas. It was an interesting test of the health of wikiprojects, and hopefully some day an Internet historian will find it a suitable topic for a thesis. I've since dabbled in the idea of testing a welcome to newbies that steers them to a relevant wikiproject based on the area of their first edits, but I haven't done enough or in a sufficiently organised way to generate useful data.
If I were trying to judge the health of a wikiproject in terms of whether they are a good thing to direct newbies to I would be more interested in questions such as:
How many active editors are watchlisting that wikiproject?
Is there a collaborative exercise going on there that newbies can join?
If a newbie asks a question there how likely are they to get a worthwhile reply and how quickly do they get that reply?
Does the wikiproject have some useful information in terms of sources that one can trust or that are outdated or otherwise flawed?
More broadly it would be good to know if wikiprojects are good for editor recruitment and retention. My hypothesis is that if someone if someone tries out editing Wikipedia and is steered to an active and relevant wikiproject then they will be more likely to continue after that first trial edit. I simply don't know whether introducing people to inactive wikiprojects is worthwhile or what the cutoff is on activity.
As for retention I am pretty sure that being involved in an active wikiproject is a positive for editor retention. I suspect that being involved in multiple wikiprojects is also a positive, I would be fascinated to know whether being involved in smaller wikiprojects and especially being the sole champion of a wikiproject makes active editors more or less likely to stay on the project. If I'm at least partially right in that editors in active wikiprojects are more likely to stay active longer than other editors then we could have a phenomenon here that will over time exacerbate wikipedia's problem of patchy coverage with the better covered topics improving faster than the gaps. Conversely if each topic has a founder effect then over time Wikipedia will become less uneven as more and more topics go through the phase of having an active editor or editors making their mark on the topic by radically improving articles.
Regards
Jonathan
On 7 Jan 2016, at 04:42, Pine W wiki.pine@gmail.com wrote:
I believe that Aaron and I discussed the theory that injecting some energy into WikiProjects might be a productive avenue for editor retention and productivity.
Is there a dashboard somewhere that shows community health statistics for WikiProjects, such as:
- Number of recent edits to articles that have been templated with that project's template
- Number of active editors in those articles
- Number of active editors in those articles who are also members of the project
- Number of editors who have recently edited in the WikiProject's project space and talk pages
- Whether the project has a newsletter, and if so, readership statistics for it.
Thanks!
Pine _______________________________________________ Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
On Fri, Jan 8, 2016 at 12:45 AM, Jonathan Cardy <werespielchequers@gmail.com
wrote:
More broadly it would be good to know if wikiprojects are good for editor recruitment and retention. My hypothesis is that if someone if someone tries out editing Wikipedia and is steered to an active and relevant wikiproject then they will be more likely to continue after that first trial edit. I simply don't know whether introducing people to inactive wikiprojects is worthwhile or what the cutoff is on activity.
That's probably right. I think a nice cutoff on activity would be: ask all wikiprojects to come up with a banner to show to a subset of newbies, to indicate how many newbies or impressions they want (what they think they can handle), and to create a page/section with an intro and projects for newbies, if they don't already have one. Any project that can manage this is welcome to get a few newbies to work with if they want, in my book.
we could have a phenomenon here that will over time exacerbate wikipedia's problem of patchy coverage with the better covered topics improving faster than the gaps. Conversely if each topic has a founder effect then over time Wikipedia will become less uneven as more and more topics go through the phase of having an active editor or editors making their mark on the topic by radically improving articles.
Isn't that how the projects have worked so far? the above happens, but also when a topic is fully covered it becomes boring to all but the completionists, so they look for other things to do. So patchwork hyperfocus flutters across fields and topics and ends up covering quite a lot. That type of individual focus is probably less biased towards 'the popular stuff' than the diffuse tidbit updates that add recent links and current events: the unevenness of the latter is more noticeable, since it is steady over time.
Cheers, Sam
On 2016-01-08 07:27, Samuel Klein wrote:
On Fri, Jan 8, 2016 at 12:45 AM, Jonathan Cardy werespielchequers@gmail.com wrote:
More broadly it would be good to know if wikiprojects are good for editor recruitment and retention. My hypothesis is that if someone if someone tries out editing Wikipedia and is steered to an active and relevant wikiproject then they will be more likely to continue after that first trial edit. I simply don't know whether introducing people to inactive wikiprojects is worthwhile or what the cutoff is on activity.
That's probably right. I think a nice cutoff on activity would be: ask all wikiprojects to come up with a banner to show to a subset of newbies, to indicate how many newbies or impressions they want (what they think they can handle), and to create a page/section with an intro and projects for newbies, if they don't already have one. Any project that can manage this is welcome to get a few newbies to work with if they want, in my book.
Actually, already knowing how many WikiProjects are alive (for example, I watch several, and most of them are dead) would be already valuable. May be even posting a question at the talk page of every WikiProject whether the project is alive and able to set up smth would give the answer. (Number of watchers certainly does not - many projects are watched by a lot of inactive users).
Cheers Yaroslav
I would say that projects have a number of levels of activity:
1. dead 2. someone is running around tagging articles with the Project banner 3. there is genuine conversation (not just spam) on their Project talk 4. there is some kind of To-Do list that gets added to 5. items actually come off the To-Do list because they've been done
In my own editing, I've never seen level 5. I know of a few at levels 3 and 4. There's a lot of level 2 and many are dead. I think you'd need a project at least at level 3 to make it worthwhile to point a newbie at it, but that's no guarantee that the conversation taking place will be encouraging or welcoming.
While I say I have never seen level 5, I am nonetheless aware of very small groups of editors that act like they have a mission but seem to coordinate via User Talk than a project page. I must say I tend to operate in that mode because I find the formalised projects attract too many people who want to "lay down the rules to everyone else" rather than get on and do the job.
Kerry
-----Original Message----- From: Wiki-research-l [mailto:wiki-research-l-bounces@lists.wikimedia.org] On Behalf Of Yaroslav M. Blanter Sent: Saturday, 9 January 2016 2:34 AM To: Research into Wikimedia content and communities wiki-research-l@lists.wikimedia.org Subject: Re: [Wiki-research-l] Community health statistics of Wikiprojects
On 2016-01-08 07:27, Samuel Klein wrote:
On Fri, Jan 8, 2016 at 12:45 AM, Jonathan Cardy werespielchequers@gmail.com wrote:
More broadly it would be good to know if wikiprojects are good for editor recruitment and retention. My hypothesis is that if someone if someone tries out editing Wikipedia and is steered to an active and relevant wikiproject then they will be more likely to continue after that first trial edit. I simply don't know whether introducing people to inactive wikiprojects is worthwhile or what the cutoff is on activity.
That's probably right. I think a nice cutoff on activity would be: ask all wikiprojects to come up with a banner to show to a subset of newbies, to indicate how many newbies or impressions they want (what they think they can handle), and to create a page/section with an intro and projects for newbies, if they don't already have one. Any project that can manage this is welcome to get a few newbies to work with if they want, in my book.
Actually, already knowing how many WikiProjects are alive (for example, I watch several, and most of them are dead) would be already valuable. May be even posting a question at the talk page of every WikiProject whether the project is alive and able to set up smth would give the answer. (Number of watchers certainly does not - many projects are watched by a lot of inactive users).
Cheers Yaroslav
_______________________________________________ Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
I believe that Operation Majestic Titan, a subproject within Wikiproject Military History, was operating at level 5 for awhile, largely thanks to the work of a small number of high-frequency contributors. Perhaps there were and are other projects active in this manner. Also, the Signpost, when it is going well -- it has ups and downs -- functions at level 5.
J-Mo, is there a chance that I can set up a meeting with you in a month or two to discuss using Quarry to extract Wikiproject activity data on a semi-automated basis, if that's possible?
Pine
On Fri, Jan 8, 2016 at 1:07 PM, Kerry Raymond kerry.raymond@gmail.com wrote:
I would say that projects have a number of levels of activity:
- dead
- someone is running around tagging articles with the Project banner
- there is genuine conversation (not just spam) on their Project talk
- there is some kind of To-Do list that gets added to
- items actually come off the To-Do list because they've been done
In my own editing, I've never seen level 5. I know of a few at levels 3 and 4. There's a lot of level 2 and many are dead. I think you'd need a project at least at level 3 to make it worthwhile to point a newbie at it, but that's no guarantee that the conversation taking place will be encouraging or welcoming.
While I say I have never seen level 5, I am nonetheless aware of very small groups of editors that act like they have a mission but seem to coordinate via User Talk than a project page. I must say I tend to operate in that mode because I find the formalised projects attract too many people who want to "lay down the rules to everyone else" rather than get on and do the job.
Kerry
-----Original Message----- From: Wiki-research-l [mailto:wiki-research-l-bounces@lists.wikimedia.org] On Behalf Of Yaroslav M. Blanter Sent: Saturday, 9 January 2016 2:34 AM To: Research into Wikimedia content and communities < wiki-research-l@lists.wikimedia.org> Subject: Re: [Wiki-research-l] Community health statistics of Wikiprojects
On 2016-01-08 07:27, Samuel Klein wrote:
On Fri, Jan 8, 2016 at 12:45 AM, Jonathan Cardy werespielchequers@gmail.com wrote:
More broadly it would be good to know if wikiprojects are good for editor recruitment and retention. My hypothesis is that if someone if someone tries out editing Wikipedia and is steered to an active and relevant wikiproject then they will be more likely to continue after that first trial edit. I simply don't know whether introducing people to inactive wikiprojects is worthwhile or what the cutoff is on activity.
That's probably right. I think a nice cutoff on activity would be: ask all wikiprojects to come up with a banner to show to a subset of newbies, to indicate how many newbies or impressions they want (what they think they can handle), and to create a page/section with an intro and projects for newbies, if they don't already have one. Any project that can manage this is welcome to get a few newbies to work with if they want, in my book.
Actually, already knowing how many WikiProjects are alive (for example, I watch several, and most of them are dead) would be already valuable. May be even posting a question at the talk page of every WikiProject whether the project is alive and able to set up smth would give the answer. (Number of watchers certainly does not - many projects are watched by a lot of inactive users).
Cheers Yaroslav
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
That's fair, Kerry. The structure of a wikiProject as often conceived is so high-maintenance that it just barely worked when there was 2x-10x more wiki activity in focused areas. At present, they're no longer really viable; and our historical memory of what wikiProjects can be or do isn't that relevant today.
There has always been lighter-weight coordination through scripts (and those who use them) and talk pages. Thoughtful tracking of coordination would look across all of those types of efforts; find the very active people reaching out to others and let them template & script their work; and perhaps help create wikiProject-like dashboards for every such group.
That said, the largest initiatives - from typo-fixing to quality-tagging to tree of life - did have wikiProjects associated with them. Whether those came first, or second, after there was a group actively doing the work, is a different question.
SJ
On Fri, Jan 8, 2016 at 4:07 PM, Kerry Raymond kerry.raymond@gmail.com wrote:
I would say that projects have a number of levels of activity:
- dead
- someone is running around tagging articles with the Project banner
- there is genuine conversation (not just spam) on their Project talk
- there is some kind of To-Do list that gets added to
- items actually come off the To-Do list because they've been done
In my own editing, I've never seen level 5. I know of a few at levels 3 and 4. There's a lot of level 2 and many are dead. I think you'd need a project at least at level 3 to make it worthwhile to point a newbie at it, but that's no guarantee that the conversation taking place will be encouraging or welcoming.
While I say I have never seen level 5, I am nonetheless aware of very small groups of editors that act like they have a mission but seem to coordinate via User Talk than a project page. I must say I tend to operate in that mode because I find the formalised projects attract too many people who want to "lay down the rules to everyone else" rather than get on and do the job.
Kerry
-----Original Message----- From: Wiki-research-l [mailto:wiki-research-l-bounces@lists.wikimedia.org] On Behalf Of Yaroslav M. Blanter Sent: Saturday, 9 January 2016 2:34 AM To: Research into Wikimedia content and communities < wiki-research-l@lists.wikimedia.org> Subject: Re: [Wiki-research-l] Community health statistics of Wikiprojects
On 2016-01-08 07:27, Samuel Klein wrote:
On Fri, Jan 8, 2016 at 12:45 AM, Jonathan Cardy werespielchequers@gmail.com wrote:
More broadly it would be good to know if wikiprojects are good for editor recruitment and retention. My hypothesis is that if someone if someone tries out editing Wikipedia and is steered to an active and relevant wikiproject then they will be more likely to continue after that first trial edit. I simply don't know whether introducing people to inactive wikiprojects is worthwhile or what the cutoff is on activity.
That's probably right. I think a nice cutoff on activity would be: ask all wikiprojects to come up with a banner to show to a subset of newbies, to indicate how many newbies or impressions they want (what they think they can handle), and to create a page/section with an intro and projects for newbies, if they don't already have one. Any project that can manage this is welcome to get a few newbies to work with if they want, in my book.
Actually, already knowing how many WikiProjects are alive (for example, I watch several, and most of them are dead) would be already valuable. May be even posting a question at the talk page of every WikiProject whether the project is alive and able to set up smth would give the answer. (Number of watchers certainly does not - many projects are watched by a lot of inactive users).
Cheers Yaroslav
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Jonathan Cardy, 08/01/2016 06:45:
If I were trying to judge the health of a wikiproject in terms of whether they are a good thing to direct newbies to I would be more interested in questions such as:
How many active editors are watchlisting that wikiproject?
action=info now gives a better information than this: https://meta.wikimedia.org/wiki/Wikimedia_Forum?action=info https://phabricator.wikimedia.org/T105392 still open but not hard.
Nemo
wiki-research-l@lists.wikimedia.org