:[[User:Dragons flight|Dragons flight]], I think the graphs can answer some of your questions.
:#[[File:Editor Activity Of New Comers In Every Month - (en).png|thumb|Editor Activity Of New Comers In Every Month - (en)]] Using the filter on the top of the page we can filter [https://cosmiclattes.github.io/wikigraphs/data/editors/html/en/monthly_activity_cohort_value.html Monthly Editor Activity Split By Cohort - Values] to show only the edit activity of editors in their first month as part of the entire activity in a month. Here editors are grouped by the month of their first edit and the graph plots active edit sessions(edits >=5/month). So we can look at (Total edit sessions in 1st month of editing) over the years. If you want to look at the same in terms of percentage you could look at https://cosmiclattes.github.io/wikigraphs/data/editors/html/en/monthly_activity_cohort_percentage.html. If you want to look at the entire edit activity of editors joining in a given month( 1st month, 2nd month, 1st year etc) you could use https://cosmiclattes.github.io/wikigraphs/data/editors/html/en/cohort_activity_value.html and filter as above. These graphs should help you answer (1 & 2)
:#[[File:Editor cohort longevity en 5- levels.png|thumb|Editor cohort longevity en filtered at 5% levels.]] To answer (3) you could use https://cosmiclattes.github.io/wikigraphs/data/editors/html/en/cohort_longevity.html and filter it to say 5% to see how long atleast 5% of the editors who joined in a month remain actively editing. There is a filter on top of the page which you can use to filter the graph. [[User:Jeph paul|jeph]] ([[User talk:Jeph paul|talk]]) 18:21, 5 October 2015 (UTC)
:# (4) could be answered by the graphs mentioned above(Monthly Editor Activity Split By Cohort - Values, Monthly Editor Activity Split By Cohort - Percentage). Let me know if it doesn't.
:# I haven't yet looked at (5) & (6).
:# I'm working on similar lines currently, I'll put them up when they are done.
The idea lab section of the village pump is a place where new ideas or suggestions on general Wikipedia issues can be incubated, for later submission for consensus discussion at Village pump (proposals). Try to be creative and positive when commenting on ideas. Before creating a new section, please note:
If you're ready to make a concrete proposal and determine whether it has consensus, go to the Wikipedia:Village pump (proposals). Proposals worked out here can be brought there.
Before commenting, note:
This page is not for consensuspolling. Stalwart "Oppose" and "Support" comments generally have no place here. Instead, discuss ideas and suggest variations on them.
Wondering whether someone already had this idea? Search the archives below, and look through Wikipedia:Perennial proposals.
It would draw more attention to those article deletion discussions that otherwise would have gotten so little attention, some of which didn't draw attention to patrollers because they weren't proposed for deletion first. Blackbombchu (talk) 18:40, 20 August 2015 (UTC)[reply]
So essentially a drive to point out discussions which aren't receiving enough input? There's definitely an interesting idea here, perhaps similar to the way editors are messaged to vote in RfCs, but with a focus on AfDs with low participation? Sam Walton (talk) 20:44, 20 August 2015 (UTC)[reply]
I guess this could work like this; users opt-in to a messaging service (in the same way that Suggestbot or the RfC bot work), and receive some amount of notifications per period of time regarding AfDs which have reached their 4th or 5th day with less than one or two votes. Sam Walton (talk) 21:08, 21 August 2015 (UTC)[reply]
An "opt-in notification service" of some kind to solicit input in AFD discussions lacking participation would be fine; @Esquivalience: I would be opposed to a bot moving transclusions up and down a page, unless it was given a separate page of its own or implemented as a "sort function" to allow users to affect only their view of the page. I quite often check discussions (listed by date and time added) that I have commented in or have interest in, or I'll get through a few on the list and come back later; the order plays some role in my memory of them. If the discussions shift positions on occasion, while it's an interesting idea, I think it would have undesirable effects. Whether or not the benefits would outweigh them I'm not sure, but I would definitely have to adjust my methods considerably. I would support a page that lists discussions lacking input in the manner you suggest separate from the current system (perhaps broken down by day as well or simply the last seven days combined).—Godsy(TALKCONT)08:08, 23 August 2015 (UTC)[reply]
I think that we ought to scrap AFD entirely, and re-create it as a purpose-built system that actually handles the whole workflow from start to finish.
Imagine an AFD tool that has simple forms to fill out for the nominator, that never sees nominations get "lost" due to transclusion problems, and that automatically counts !votes and tracks how many separate individuals participated. Imagine one that notices when a page is ready for closing (i.e., because it meets our standard criteria, such as having ≥3 participants and being 7 days old, or whatever we decide), and that puts the page into a list or category for action. Imagine one that could sort or filter by any criteria that you care about: the most attention (maybe it's SNOWing?), the least attention, only BLPs, only articles tagged by my favorite WikiProject, etc. Imagine one that can be withdrawn or closed by clicking a few buttons with a built-in script (including direct access to page deletion for admins and maybe a scripted blank-and-redirect button for everyone), rather than having to type special codes into a template and separately processing the page.
Yeah, that would be an improvement. Currently, there are regular complaints about the process being overly complicated. Maybe a gadget or a gadget-bot combination might serve to create such a system. Jo-Jo Eumerus (talk, contributions) 18:39, 28 August 2015 (UTC)[reply]
I think the process of nominating articles for deletion and having well researched deletion discussion pages can be made better by Wikipedians collaborating in a complex way, and I think the only way that's going to happen is if Wikipedia accepts help from Harvard Catalyst which will do a lot of participating in deletion discussions. Blackbombchu (talk) 20:49, 2 September 2015 (UTC)[reply]
No, WhatamIdoing, it would be a whole lot worse. Perhaps you have forgotten over the years of not particpating on that area that AfC is a discussion not a vote. It should be kept as complicated as possible to deter newbie NAC from experimenting with it. We also already have the fully automatic AfD tool, but that's also embedded in NPP which you told me years ago is a superfluous process.Kudpung กุดผึ้ง (talk) 11:57, 6 September 2015 (UTC)[reply]
I believe that you meant that it's supposed to be a consensus-finding process rather a numeric vote, but (a) both admins and NACs are actually closing these as votes, and (b) keeping a tally of the votes doesn't force an admin to delete on that basis. My bigger concern is about getting people to nominate suitable articles, and to let interested users know about the nomination. If you still want to close manually, then that's okay with me. WhatamIdoing (talk) 22:28, 6 September 2015 (UTC)[reply]
I think the problem of some deletion discussions getting so little attention can be solved by having a way to patrol the relisting of deletion discussions. However if that change gets made, Wikipedia should also make another change of having a policy to ban editors for relisting a deletion discussion that's less than a week old so that there won't be greedy people trying to get a certain deletion discussion more than its fair share of attention. In addition to that, for each deletion sorting Wikiproject, there should be another deletion sorting Wikiproject only for relisted deletion discussions for that category. That way, experienced editors can easily decide to only participate in relisted deletion discussions, giving relisted deletion discussions quite a lot of attention by experienced editors. Blackbombchu (talk) 17:56, 18 September 2015 (UTC)[reply]
The WikiProject for a major topic area seems to be dead
It's been a long time since any post to WT:WikiProject Engineering has received a meaningful reply. Imho there are certain projects that are "too big to fail". For a major topic area such as engineering to not have a functional WikiProject is a serious problem. I think we could convert the main WikiProject Engineering page into a type of "disambiguation" page that lists active projects that cover various sub-topics of engineering - chemical engineering, electrical engineering, etc. Roger (Dodger67) (talk) 08:08, 16 September 2015 (UTC)[reply]
Sounds like your the new de facto head of Wikipedia:WikiProject Engineering. Some comments: Just because nobody is using the talk page, doesn't mean that nobody is using the project's main page. You can try changing the main page. That might bring people out of the woodwork. I don't know if I'd change the main page too much though. Maybe adding a header notice using {{mbox}} or something would be sufficient. Jason Quinn (talk) 15:37, 22 September 2015 (UTC)[reply]
What would people think about implementing some kind of software that allows users who can't edit a semi-protected article to edit it similar to pending changes, however the edits wouldn't "go live" until any auto-confirmed user accepted the change? This would make semi-protection much less forceful and encourage editing, rather then dealing with the wiki-markup, talk pages, and templates that go along with semi-protected edit requests? I feel like there's some kind of objection to this, or it would have been implemented in the past, so, what are those? Kharkiv07 (T)01:32, 23 September 2015 (UTC)[reply]
I would not be in favor of making these tweaks to semi-protection. We often get articles hit with sustained vandalism (e.g., when a celeb does something stupid) and editors should not have to spend hours hitting revert. --NeilNtalk to me14:18, 24 September 2015 (UTC)[reply]
As Sam Walton said, this would basically be implementing a "pending changes level 0". I don't think this is necessary or even desirable. If there are no beneficial IP/new account edits a page can be semi-protected, and if there are then PC1 can be used. The backlog at Special:PendingChanges is never egregious, so PC0 would not be a workload benefit. Also by not screening the editors who accept the change the system could be gamed. Having yet another type of protection would just mean more bureaucracy and squabbles about what level should be used when. BethNaught (talk) 14:26, 24 September 2015 (UTC)[reply]
I generally oppose making anything harder for those who are willing to edit as auto-confirmed users in order to better accommodate those who are not. The way to "encourage editing" is to encourage registration. Semi-protection is only one of the many good reasons to do so. ―Mandruss☎14:30, 24 September 2015 (UTC)[reply]
Pageview statistics for all articles created by a user
I was thinking about how to personally mark the 5 millionth ENWP milestone, and this idea occurred to me. What have been the most popular articles I've created, by all-time pageviews? It's a sort of long list so I'd rather not gather the data by hand. Is there a tool that can do this? — Brianhe (talk) 17:43, 23 September 2015 (UTC)[reply]
I think the tool is just really slow, Panyd? You had 47.610 Total views in 2015-08! Since i don't create new articles on enwiki (but redirects that i already forgot about) it works faster for me. --Atlasowa (talk) 14:01, 25 September 2015 (UTC)[reply]
Took tens of minutes to run on my account. Apparently after creating a few hundred articles / redirects this kind of check is maybe not a great idea. Only just barely missed hitting 500k views though (495,064 in 2015-08). Dragons flight (talk) 14:39, 25 September 2015 (UTC)[reply]
I knew most of the articles I've actually created were way out in the long tail of reader interest, but wow, this is embarrassing! I'd also really like to know who linked to my top-viewed article to make the traffic look like this: who are all these people suddenly interested in flying ice cubes? Opabinia regalis (talk) 21:08, 26 September 2015 (UTC)[reply]
new namespace: Chronicle, for recording events comprehensively
I have an idea for a new namespace to be entitled "Chronicle." It would be a place to note or record all events or items within a particular area. doing so would allow us to create a common space and resource where historical events could be noted and referenced, without requiring us to change the regular historical articles to record new events before their eventual significance is fully understood.
currently, there is no centralized place to create a central narrative of events as they occur.
one major potential of wikipedia is to serve as an ongoing and evolving record of events as they happen. a shared central space for such information would make it much easier for editors to be able to have a central resource to review recent events and to see if they warrant inclusion in various higher-level articles, such as history articles, science articles, technology, etc. --Sm8900 (talk) 01:32, 27 September 2015 (UTC)[reply]
thanks for your reply. but no, that's not what I meant.rather, I meant exhaustive and inclusive articles which would chronicle and record every occurrence in a particular topic or area.so, for instance, the US CONgress currently addresses a huge number of items which never get reflected here. a Chronicle article on American politics could reflect many items addressed by Congress without, for example, bloating the regular encyclopedic articles on Congress. --Sm8900 (talk) 20:50, 27 September 2015 (UTC)[reply]
hi Jim.henderson. no, not for that reason, but rather because an exhaustive compendium of everything which happens in Congress, the UN, or the Chicago City Council is not necessarily warranted for a regular Wikipedia article. however, it would be a good informational resource which should still be available somewhere at this site.
additionally, it might not just be a chronicle of official proceedings, but could also be a comprehensive record of all historical events and current events of interest. it could include links to existing articles in order to do so.
So a chronicle which collates links to every article here on US current events, or alternately a chronicle which records every political news item in the State of Colorado, or in all major US cities during the year 2015, might not be warranted for a regular article. but it still might be useful in the long term as an informational archive and resource. ---- Sm8900 (talk) 00:11, 30 September 2015 (UTC)[reply]
My gut reaction to this proposal is: no. would allow us to create a common space and resource where historical events could be noted and referenced, without requiring us to change the regular historical articles to record new events before their eventual significance is fully understood That's original research and we can't decide what to include/not include without a thesis for the page. If we have a thesis which explains what qualifies to be included/excluded, then it should be a standard History of X article. We shouldn't be in the busniess of create a central narrative of events as they occur, we have to go based on what Reliable sources report (either as new coverage or as historians writing and drawing the inferences). Reporting on recent events is either the perview of WikiNews or the perview of a article that is written to support a In the News point on the frontpage. Anything else pushes the Recent-isim factor of day in the sun coverage. Hasteur (talk) 18:12, 30 September 2015 (UTC)[reply]
My gut reaction is likewise no. There are some instances where "long list of everything" articles are warranted—List of Statutory Instruments of the United Kingdom, 1994 and the like—but those are adequately covered in the existing list format. "A comprehensive record of all historical events and current events of interest" would be unworkable (are you aware of how many events take place in even a small city in any given year?) since they would either be unworkably large, or require the invention of arbitrary inclusion criteria. ‑ iridescent18:17, 30 September 2015 (UTC)[reply]
hm, okay.
well, firstly, i only meant to include only those events and items for which reliable sources do exist. in other words, each event would need to have been reported in some other reliable source or citations, just like regular articles. so I agree, no original research should be involved at all for these.
as far as your other point, I hear ya. however, the problem is that currently, there is no central article to note current events of notability, no matter how notable they are, unless they are already accepted as being of genuine historical significance. and even then, there is no central place to chronicle them. so an item of medium but genuine significance might never get recorded anywhere here.
we could still retain the same standards of notability as would currently govern any encyclopedia article. the difference would be that this would consciously be a place, if you like, for timeline articles on current events of some genuine significance,but which otherwise might not get their own entries until a year or two after they occurred.
so I am willing to modify this idea almost infinitely to retain adherence to our basic notability rules, yet still to provide a different type of article which would provide some degree of a new approach.
I recognize that this may still not allay all your concerns or reservations on this, but does that at least improve the idea somewhat? thanks for all your input here. --Sm8900 (talk) 18:24, 30 September 2015 (UTC)[reply]
Perhaps you could write a "sample" in userspace of what you think a Chronicle would look like? It doesn't have to be long, or even based on real events, but it might help to make clearer exactly what you are proposing. QuiteUnusual (talk) 11:17, 1 October 2015 (UTC)[reply]
hm, that's a good idea. thanks for your helpful idea on that. I will start working on that. anyone, feel free to keep adding comments if you want. thanks! --Sm8900 (talk) 13:15, 1 October 2015 (UTC)[reply]
I've been working on visualizing editor behaviour on wikipedia over the past few months. I've put together graphs that visualize editor activity, retention etc. I made a presentation for the research team at the foundation - http://slides.com/cosmiclattes/edit-activity-graphs-analysis/. It also has some of the preliminary results. It has links to the graphs & says how to interpret & play with them. Please let me know if you guys have other metrics or ideas you'd like to see graphed. I'd love to hear what you guys think of the graphs. I have proposed an IEG to continue working on the graphs.
I feel like we've known how this works for years, but nobody wants to confront it - instead, we'll just keep compiling numbers on it to make ourselves feel like we're doing something about it. Samsara14:40, 4 October 2015 (UTC)[reply]
Besides insights on editor retention/longevity etc which are well known already the analyses have interesting inferences about how long an article stays actively edited (I don't yet have the data for en, every other big wiki, after about 2007, most of the articles only see activity for a month or 2 after creation, then they hardly see any activity). Also I have noticed cyclical editing patterns in wikis like de etc. Jan-Dec sees a spike in edit activity. Retention rates amongst very active editors(100+ edits/month) sometimes show more active editors than who joined, meaning to say some people with lesser edit activity become more productive in their second month (I haven't put up theses graphs online yet). These analyses are not solutions, but could help us get more insights into macro editor behavior.--jeph (talk) 05:28, 5 October 2015 (UTC)[reply]
In the last 2 years we've seen a slow but important uptick in editor activity on the English Wikipedia, after several years of declines. Personally, I'd like to understand that shift with an aim towards identifying the origin of the improvement (and perhaps gain insight into how we might be able to do even better). A simple first step is probably to pull out the changes in new editor registration, changes in editor activity of existing editors, and changes in editor retention. My personal suspicion is that most of the uptick is due primarily to growth in new editors, but I'm not entirely sure. Dragons flight (talk) 10:44, 5 October 2015 (UTC)[reply]
Hi Dragons flight, Some of the graphs that are up already are:
Editor Cohort Longevity - Shows the longevity/retention of a given editor cohort in a month.
I haven't looked at user registrations yet. Some of the other ideas I'm working on are here and here. Would you have any specific ideas for me or directions you would like me to explore?jeph (talk) 12:47, 5 October 2015 (UTC)[reply]
Those displays are visually attractive, but in that format I have trouble getting at the information I would want. For example:
What is the trend in editing for new editors? (Total edits in 1st month of editing, 1st year of editing, etc.)
What is the trend in editing for established editors?
How has short-term editor retention changed, e.g. the trend in new editor percentage that keep editing after 6 months? 1 year? 2 years?
How has the percentage of edits from new vs. old accounts changed through time?
How has the number of newly registered accounts changed through time?
How has the 1st edit conversion rate changed (i.e. fraction of registered accounts that actually edit)?
How has the fraction of editors going from 1 edit to 10 edits changed? From 10 edits to 100 edits? From 100 edits to 1000 edits?
(4) could be answered by the graphs mentioned above(Monthly Editor Activity Split By Cohort - Values, Monthly Editor Activity Split By Cohort - Percentage). Let me know if it doesn't.
I haven't yet looked at (5) & (6).
I'm working on similar lines currently, I'll put them up when they are done.
Done. DMacks (talk) 20:33, 27 September 2015 (UTC)
.png)
