User talk:Atlasowa

From Meta, a Wikimedia project coordination wiki
Jump to: navigation, search

Afrikaans | العربية | azərbaycanca | Boarisch | беларуская | беларуская (тарашкевіца)‎ | български | বাংলা | བོད་ཡིག | bosanski | català | کوردیی ناوەندی | čeština | Cymraeg | dansk | Deutsch | Deutsch (Sie-Form)‎ | Zazaki | ދިވެހިބަސް | Ελληνικά | English | Esperanto | español | eesti | فارسی | suomi | français | Nordfriisk | Frysk | galego | Alemannisch | ગુજરાતી | עברית | हिन्दी | Fiji Hindi | hrvatski | magyar | Հայերեն | interlingua | Bahasa Indonesia | Ido | íslenska | italiano | 日本語 | ქართული | ភាសាខ្មែរ | 한국어 | Limburgs | lietuvių | Baso Minangkabau | македонски | മലയാളം | молдовеняскэ | Bahasa Melayu | မြန်မာဘာသာ | مازِرونی | नेपाली | Nederlands | norsk nynorsk | norsk bokmål | occitan | Kapampangan | polski | português | Runa Simi | română | русский | සිංහල | slovenčina | slovenščina | Soomaaliga | shqip | српски / srpski | svenska | ślůnski | தமிழ் | ไทย | Türkmençe | Tagalog | Türkçe | татарча/tatarça | ⵜⴰⵎⴰⵣⵉⵖⵜ  | українська | اردو | Tiếng Việt | 吴语 | 中文(简体)‎ | 中文(繁體)‎ | +/-

Welcome to Meta![edit]

Hello Atlasowa, and welcome to the Wikimedia Meta-Wiki! This website is for coordinating and discussing all Wikimedia projects. You may find it useful to read our policy page. If you are interested in doing translations, visit Meta:Babylon. You can also leave a note on Meta:Babel or Wikimedia Forum (please read the instructions at the top of the page before posting there). Happy editing!

-- 21:35, 29 October 2011 (UTC)

Wikimedia Travel Guide: Naming poll open[edit]

Hi there,

You are receiving this message because you edited the initial naming straw poll for the Wikimedia Travel Guide.

The proposed naming poll is now open and you can vote for as many of the proposed names as you wish, if you are eligible. Please see Travel Guide/Naming Process for full details on voting eligibility and how the final name will be selected. Voting will last for 14 days, and will terminate on 16 October at 06:59:59 UTC.

Thanks, Thehelpfulone 23:08, 2 October 2012 (UTC)

Talk:Flagged Revisions[edit]

Hello, I've copied your message on the designated talk page and replied there. --Nemo 07:36, 19 March 2013 (UTC)

Also, I see that you participated in a discussion on the right to fork; you may be interested in Talk:Wikimedia Trust where it's linked. --Nemo 07:40, 19 March 2013 (UTC)

Study:Flagged Revisions[edit]

Hi, I have provided some feedback on my user talk page. Cheers.--GlimmerPhoenix (talk) 13:55, 2 October 2013 (UTC)

Publicity Ideas for the Replay Edit tool[edit]

Would you have ideas on getting the word out about the tool ? Some of the ideas I have are:

  • Another video like the Umlaut edit visualisation.
  • Using the slider in some other simple tools ( has the code to the slider as a separate component, the code is old and has not yet been updated with the code being used in the replay edits tool currently).
  • Seeding it with other users who might be interested ?
  • By Copying "importScript( 'User:Jeph_paul/common.js' );importStylesheet( 'User:Jeph_Paul/common.css' );" into your common.js it can be used as a userscript.

Could you send me a mail so that we can start a conversation on this :-)--Jeph paul (talk) 19:14, 8 January 2014 (UTC)

Screenshot of "Replay edits" with user script
Hi Jeph paul!
  • I think another video with the new replay version is a good idea, videos are probably the easiest way to present the tool to other users.
  • I tested the user script, but the play button didn't show up. See screenshot (i made you a new commons:Category:Replay Edits). The problem could be hard to locate, HTTPS? java script? Should i try the mwloader instead of importScript?
  • Asking other users to test is a good idea. I can "seed" on german Wikipedia, if you like ;-) Do you have some ideas for specific questions? How open are you for the inevitable additional feature requests (flying pink unicorn that shits rainbows)?
I think the replayer is a great tool to show people that Wikipedia editing is not a black box operation, and actually not that difficult. All the best --Atlasowa (talk) 18:09, 13 January 2014 (UTC)

Hi, Atlasowa

  • Thank you for the offer to seed it on german wikipedia & sorry for getting back this late. The tool has support for multiple languages now & also has il8n support. I have added some german phrases I got from google translate (Have only added german translations, wasn't sure if google translate would give me accurate results, but the user can access articles from all the listed language wikipedia's).Also added a suggestion dropdown as you type the article name (This should be useful I guess).
  • I'll take a look at the userscript & fix it.
  • Th idea behind asking q's was to know if there would be a usecase for the cases like last 20 edits etc you mentioned. (My UX friend is busy with her work, but we'll get the ball rolling on this.)
  • Made a few UI fixes like adding the arrows between the steps in the homepage. Shall we go ahead and sees it on the german wiki while making the UI/UX other changes paralelly ?

--Jeph paul (talk) 19:18, 3 February 2014 (UTC)

Revamping the View History tab[edit]

What would it take to revamp the View History tab, in terms of community consensus? If we came up with a plan with layout & feature set would it be accepted? Are there plans already on doing this? Is someone working on them? The replay edits tool could be a part of it, we could probably use the graph or other pieces from it. --Jeph paul (talk) 19:25, 3 February 2014 (UTC)

Upcoming IdeaLab Events: IEG Proposal Clinics[edit]

Idea Lab
Idea Lab

Hello, Atlasowa! We've added Events to IdeaLab, and you're invited :)

Upcoming events focus on turning ideas into Individual Engagement Grant proposals before the March 31 deadline. Need help or have questions about IEG? Join us at a Hangout:

  • Thursday, 13 March 2014, 1600 UTC
  • Wednesday, 19 March 2014, 1700 UTC
  • Saturday, 29 March 2014, 1700 UTC

Hope to see you there!

This message was delivered automatically to IEG and IdeaLab participants. To unsubscribe from any future IEG reminders, remove your name from this list

Fwd: $wgMFAnonymousEditing = true is sometimes not respected: cache?[edit]

Hi, I filed your observation there. Thanks for looking into the matter! Please login on phabricator. --Nemo 08:12, 3 March 2015 (UTC)

I won't participate at phabricator because it demands my emailaddress. BTW, i am disappointed in you, Nemo. You're whipping up a crowd for forcing mobile anonymous editing based on bad information and anecdotal tales. I think you know perfectly well that this will open a firehose of crap on hundreds of wikis. It will severely burn small wikis, it will burn out patrollers, it will burn new editors (they will remember their first revert forever), it will burn trust between communities and WMF, and it will burn subsequent trials to allow mobile anonymous editing (when we'll actually have usable UI for mobile editing). If you care to explain what you're really trying to accomplish: i'm listening. --Atlasowa (talk) 10:51, 3 March 2015 (UTC)
Right. [1] Can't you just use a temporary email address?
I'm very sorry that you see it that way. I definitely don't "know" all those things, it's more likely that the effect will be rather minimal. I also see a contradiction between saying at the same time that FlaggedRevs works so well on and that unregistered edits are an unstoppable flood.
What I'm trying to accomplish is simply reason, as always in my life. I think you overestimate my ability to to "whip up a crowd". My job is merely to push people to think (and speak) about some things I feel important. --Nemo 15:13, 4 March 2015 (UTC)
For instance, I can't help thinking that bringing (back) new/old people to edit Meta for the first time in 8 years, to discuss something important, is a good thing. --Nemo 12:03, 9 March 2015 (UTC)
Yes, that is a good thing :-) It's also a good thing if this discussion makes some wikis volunteer to opt-in to mobile anonymous editing, so that we get real experience and more feedback. And a third good thing would be to influence WMF development for mobile web with a reality check.
I'm still traumatized by the mobile-crap-magnet/selfie-pocalypse affair (en 2013, de 2014), did you ever read up on it and what do you think the lessons were? --Atlasowa (talk) 10:38, 11 March 2015 (UTC)
Sure, I followed the matter around Commons back then, though I didn't inspect every detail of it. My understanding is what I told Be..anyone (I think many repeated his same argument without noticing he changed his mind; but I don't want to tell them and risk being too pushy). is not disabled, after all. --Nemo 18:00, 11 March 2015 (UTC)
Hi Nemo: Regarding the mobile anonymous editing drop to nearly nothing after 2015-01-09 until 2015-01-25, i see that there was Research:WikiGrok/Test3 (Test began: 2015-01-14 Test ended: 2015-01-21; Target: 16% of all users, stable mobile site, enwiki - so NOT logged in users only). May be related, maybe not. Thus the bug could come back eventually with another WikiGrok testing(?) --Atlasowa (talk) 08:48, 17 March 2015 (UTC)

Please fill out our Inspire campaign survey[edit]

Inspire astrocat aqua.png

Thank you for participating in the Wikimedia Inspire campaign during March 2015!

Please take our short survey and share your experience during the campaign.

Many thanks,

Jmorgan (WMF) (talk), on behalf of the IdeaLab team.

23:34, 13 April 2015 (UTC)

This message was delivered automatically to Inspire campaign participants. To unsubscribe from any future IdeaLab reminders, remove your name from this list

Edit Activity Graphs[edit]

Hi Atlasowa,

I've made some new graphs visualizing the entire edit activity of a wiki - I'm documenting work at

I've also put down some of the preliminary inferences regarding editor activity on en. Each graph has a selector on top of the page that allows you to filter it. On taking the cursor to the left end of the selector you'll get a resize cursor, you should then be able to redraw/drag the selection.

The thread I started on the research mailing list

I've posted this on your en talk page too.--Jeph paul (talk) 18:00, 15 July 2015 (UTC)

I put together a presentation for the research team at the foundation - It also has some of the preliminary results. It has links to the graphs & says how to interpret & play with them. Let me know if you need any help in interpreting them or if you have other metrics you'd like to see graphed. I've added the data for de too.--Jeph paul (talk) 04:49, 18 September 2015 (UTC)
Hi Jeph paul, this is a really good presentation! I like your analysing "Article Edit Activity" :-) I agree with these observations:
  • Only the articles created till 2007 see continued editing.
  • The articles in the subsequent cohorts see very little edit activity.
  • Newly created articles rapidly lose all editing activity.
I think there are 2 important aspects to this,
While I think these are the "wrong" questions:
  • Do we have a problem in surfacing editable articles?
  • Are we losing editors because they can't find content they can edit? Content they can contribute to?
This sounds a lot like the attitude primarily amongst WMF and developers, that there are those masses of volunteers that just need to be told what to do, by a bot or a software that assigns tasks. This is not how things work, and it's patronizing and also this targeting is really bad. And the answer is not "better targeting" of task assignments by extensive user profiling (again an attitude primarily amongst WMF and developers). Volunteers choose their own tasks. Volunteers need to be given tools to choose their tasks (i.e. feed of new edits by category) and tools to effectively getting done their work (i.e. easier refs by Citoid).
The real question is: How do we find (1), motivate (2) and keep (3) productive editors for Wikipedia?
(1) Productive, knowledgable editors mostly find Wikipedia themselves, not the other way around ("organic" editor acquisition), our best (but still uneffective) shot at finding editors for Wikipedia are University/student assignment programs to edit wikipedia with mentors.
(2) "Motivation" is also self-motivation of editors. But Wikipedia provides a lot of negative feedback (reverts, rules) and should provide more positive feedback (i.e. thanks-notifications for IPs, newbies)
(3) "Keeping" productive editor by not making Wikipedia super-complex (learn wikitext markup, learn editing Templates, learn Wikidata, learn Lua, learn Flow, learn operating a bot, learn to install multiple user scripts, learn Phabricator, etc. etc.) Also, this trend of growing user profiling makes me think about leaving. I am here to volunteer knowledge, not to provide extensive personal data about myself. WMF is collecting more and more data for the sake of maybe-this-could-be-useful-sometime-for-someone and will likely put those data dumps online for anyone interested (including marketing companies, "big data") with a CC-0 licence!
I digress ;-) BTW, I think 2 aspects of Wikipedia editor activity that are under-analysed are:
(A) the impact of flagged-revisions.
(B) the "contagious" effect of edits ([2] Group size and incentives to contribute: A natural experiment at Chinese Wikipedia by Xiaoquan (Michael) Zhang, Feng Zhu. American Economic Review 101 (June 2011): 1601–1615 [3]), (Michael Kummer's paper that looks at a similar topic ("contagion" in pageviews among linked articles) from an econometrics perspective: Spillovers in Networks of User Generated Content – Evidence from 23 Natural Experiments on Wikipedia"), including bot edits (Robin D. Fink and Tobias Liboschik (2010) Bots Nicht-menschliche Mitglieder der Wikipedia-Gemeinschaft. Soziologisches Arbeitspapier Nr. 28, TU Dortmund [4] [5] [6])
Gotta stop  :-) --Atlasowa (talk) 11:23, 21 September 2015 (UTC)

Hi Atlasowa

The edit activity on article cohorts filtered to show only 5% levels.

This shows the article cohort edit activity for de - filtered at atleast 5% levels. (Atleast 5% of the articles created in that cohort are continuing to get atleast 5 edits a month). What we see here is that there is a very steep fall from Dec 05 onward. In fact the fall is very dramatic in a month or two's time the editing pattern has suddenly changed. You might ask me why I'm looking at the 5% levels and not say 30% or 3%, this dramatic change is seen only at 5% level onward. Assuming our assumptions of 5 edits a month being a useful metric of activity etc are correct. We are seeing something that cannot be explained by the low hanging fruit theory, that will show a fall in activity but nothing as steep as this over a few months. I find it difficult to understand that by Nov - Dec 05, all interesting/ important articles were over and suddenly sustained edit activity on the newly created articles nose dived. Having said all this, I have not read the paper fully, I'm still going through it. I think we should dig further. Maybe after say Dec 05, the way people found new articles to edit changed? Wild Guess, show me a random article changed its algorithm or weight ?

Again having said all this, because the articles created after Dec 2005 receive lesser edits mean that they are of a lower quality  ? Is the baseline quality of articles created these days higher?

When I ask the question 'Are we losing editors because they can't find c ....' I'm thinking on the lines of building tools to help the user find new content where the editor can contribute to and less of lets give the editor a recommendation of articles he can edit. Also with my limited knowledge of recommendation systems I'm not sure if they can be accurate for a new comer. Also here a recommendation system not only has to match interests it also has to match skills to. For ex, I'm really interested in say DHT's but I'm probably not going to be able to add any value to the articles. I think there are probably simpler tools we can build that'll go a long way in surfacing editable content.

I'm just getting started with the questions & the graphs. We keep talking on and off. Why don't we collaborate more closely. I'll try to get a few others who might be interested like Shyamal. We can have a common set of questions we can investigate. We can develop investigative tools like the graphs or metric others could borrow. I can bring in my data viz skills. You guys can bring in your skills. We can do more working together.--Jeph paul (talk) 16:08, 23 September 2015 (UTC)

These are some of the questions I have collated to date --Jeph paul (talk) 16:13, 23 September 2015 (UTC)

Enwp retention vs active editors: the famous "holy shit" slide from 2011.

Hi Jeph paul, What happened to Wikipedia in 2005-2007 is very well known, no real need for further analysis. Wikipedia was catapulted from an obscure website by some encyclopdia-geeks, that nobody knew and nobody cared about, to a top10 website that everyone found in search results and read and that was reported about. The "sexy new thing". Mass influx of new editors, Wikipedia flooded in articles. See the stats at Wikipedia.org_is_more_popular_than...#Wikipedia.27s Alexa ranking milestones_.283_month_average.29 and User:Stu/comScore_data_on_Wikimedia. What happened in 2005-2007 has nothing to do with 2015, it's a completely different game now. I see no use in digging into prehistory. Trends from the last 5 years, that is useful to learn from, yes. --Atlasowa (talk) 19:12, 23 September 2015 (UTC)

+en:User:Brews ohare/What happened to Wikipedia in 2006-2007? --Atlasowa (talk) 20:43, 23 September 2015 (UTC)

Hi Atlasowa, Looking at the past may or may not be useful. My larger point is, if we join forces & get like minded people to join us we could do a lot of analysis and produce very interesting insights. We can access most of the edit data publicly through the toollabs db slaves. We can decide on a common set of questions we want to investigate. I can build the graphs and we can get the ball rolling.--Jeph paul (talk) 19:31, 23 September 2015 (UTC)

Yes! Yes, absolutely, Jeph paul. The thing is, this stuff is extremely complex. So many factors at play, incredibly difficult. It's best to identify simple, limited questions that can be analysed and that have real impact for Wikipedia. For example, captchas: Which language wikipedias use captchas at which step (new article creation? user account creation? every IP edit? ...) and from which date on, so we can analyse impact across time and across languages. Or mobile edits. --Atlasowa (talk) 20:27, 23 September 2015 (UTC)

BTW, have i already forced you to read my collection at de:Benutzer:Atlasowa/editor_motivation (mostly english)? Have a look at the graphs. You are not at all an exception with: "I'm really interested in say DHT's but I'm probably not going to be able to add any value to the articles." --Atlasowa (talk) 20:33, 23 September 2015 (UTC)
Hi Atlasowa, You had asked me to look at it once before but I've only skimmed through them :-). I've collated some of my questions/ideas on, can you add the stuff you'd like to investigate. Usage of VE & mobile editing are already on my list. I'll move them over to a separate page/project over the weekend. I'm planning to use the edit activity style graphs to look at the usage patterns of VE/mobile editing. We could also look at the activity of just the new comers on VE/mobile etc. Let me know if there are other graphing/ visualizations techniques you'd like me to use. I looked at the mediaiwki db schema and couldn't find any tables related to captchas. I'm probably missing something.
Also are there other ways to talk to you, email etc?--Jeph paul (talk) 13:23, 24 September 2015 (UTC)
Hi Jeph paul!
Re captchas: I wouldn't expect tables related to captchas in this db. We would need to digg deep in each Wikipedia configuration history to find the diffs that activate some captchas. And also abuse filter configurations of every WP... and ... well, it's complex. There was some analysis for portuguese WP which had a brutal captcha requirement (for any and every edit?), see CAPTCHA at the Portuguese Wikipedia, 22 Mar 2011, deactivation request and results.
Re mobile edits: Let's start with that! How would you visualize Research:Mobile_anonymous_apocalypse? See data sources on Research talk:Mobile anonymous apocalypse. Mobile anonymous edits were allowed in early 2015. How many mobile edits over time, in different languages, IP vs. registered user? How big (bytes?) are mobile edits, over time, IP vs. registered user? etc. Have a look at quarry query stats Should we copy this thread to Research_talk:Editor_Behaviour_Analysis_&_Graphs? --Atlasowa (talk) 14:33, 24 September 2015 (UTC)
The articles edited by new comers in the pt wiki.
Atlasowa Speaking of pt, this is the plot for the edit activity of new comers on pt. It shows the cohorts of articles edited by editors in their first month. I found this interesting because, what this plot shows is that there are a set of articles/many articles that have never been edited by logged in registered editors in their first month. I've not seen this in any other language I have looked at. I haven't yet put this plot online and hence the screen grab.
Lets start with the mobile edits to start with. We can plot similar graphs for registered & logged in mobile editors. We can look at what article cohorts they are editing. Are they creating articles etc? We can also probably do size of edits on article cohorts but its going to be really tricky. I'm adding up all the bytes added by editors in a cohort. Also there were cases where some cohorts had a net negative addition in some months & I didn't know how to sensibly plot it. But lets try to figure this one out. I'm not sure how to plot ip's since there is no notion of a joining date for an ip, again something we'll have to figure out. I'll try to get the plots for registered/logged in mobile editors for the smaller wiki's by tomorrow. Lets move the thread to, its gotten way to long here already. --Jeph paul (talk) 18:36, 24 September 2015 (UTC)

Hi Atlasowa, I've applied for an IEG for the graphs Grants:IEG/Editor_Behaviour_Analysis. I'm working on the mobile edits graphs. I have a graph showing the articles edited by mobile users, first look no surprises there. Also the anon users seem to be reinforcing the logged in users or the vice versa. Next I'll be looking at the editors doing edits on mobile(Are they new comers or the older editors etc). I'll ping you once I put them online.--Jeph paul (talk) 18:45, 29 September 2015 (UTC)

How can we improve Wikimedia grants to support you better?[edit]

Hi! The Wikimedia Foundation would like your input on how we can reimagine Wikimedia Foundation grants to better support people and ideas in your Wikimedia project.

After reading the Reimagining WMF grants idea, we ask you to complete this survey to help us improve the idea and learn more about your experience. When you complete the survey, you can enter to win one of five Wikimedia globe sweatshirts!

In addition to taking the the survey, you are welcome to participate in these ways:

This survey is in English, but feedback on the discussion page is welcome in any language.

With thanks,

I JethroBT (WMF), Community Resources, Wikimedia Foundation.

(Opt-out Instructions) This message was sent by I JethroBT (WMF) (talk · contribs) through MediaWiki message delivery. 01:23, 19 August 2015 (UTC)

Last call for WMF grants feedback![edit]

Hi, this is a reminder that the consultation about Reimagining WMF grants is closing on 8 September (0:00 UTC). We encourage you to complete the survey now, if you haven't yet done so, so that we can include your ideas.

With thanks,

I JethroBT (WMF), Community Resources, Wikimedia Foundation.

(Opt-out Instructions) This message was sent by I JethroBT (WMF) (talk · contribs) through MediaWiki message delivery. 19:08, 4 September 2015 (UTC)