User talk:Philippe

Archives: 1, 2, 3, 4, 5, 6, 7, 8, 9

Archiving

Did it again. :-) -Philippe (talk) 18:20, 13 June 2015 (UTC)

Please delete this edit and edit summary

https://en.wikipedia.org/w/index.php?title=Karl_Klug&curid=31658515&diff=672163636&oldid=670779637--TM 19:18, 19 July 2015 (UTC)

Yeah, that seems reasonable. Certainly not done to improve the encyclopedia. Thanks for pointing it out. -Philippe (talk) 19:51, 19 July 2015 (UTC)

comment

With respect Philippe, I suggest that we should not be speculating on an editor's mental health; even when tempered with what may or may not be best for it. — Ched :  ?  07:10, 7 October 2015 (UTC)

Ched, you're quite right. I will remove that portion. Not sure what I was thinking there. Thank you. -Philippe (talk) 07:12, 7 October 2015 (UTC)

UTRS account

Hi Philippe! Just to let you we're deactivating the all-access UTRS account associated with Philippe (WMF). If you want, you're absolutely welcome to request a new account from Philippe and we'll activate you right away, as well as flag you as a CU. Thanks! ☺ · Salvidrim! · ✉ 02:21, 4 November 2015 (UTC)

File:Philippe.JPG listed for discussion

A file that you uploaded or altered, File:Philippe.JPG, has been listed at Wikipedia:Files for discussion. Please see the discussion to see why it has been listed (you may have to search for the title of the image to find its entry). Feel free to add your opinion on the matter below the nomination. Thank you. Stefan2 (talk) 00:07, 5 November 2015 (UTC)

ArbCom elections are now open!

Hi,
You appear to be eligible to vote in the current Arbitration Committee election. The Arbitration Committee is the panel of editors responsible for conducting the Wikipedia arbitration process. It has the authority to enact binding solutions for disputes between editors, primarily related to serious behavioural issues that the community has been unable to resolve. This includes the ability to impose site bans, topic bans, editing restrictions, and other measures needed to maintain our editing environment. The arbitration policy describes the Committee's roles and responsibilities in greater detail. If you wish to participate, you are welcome to review the candidates' statements and submit your choices on the voting page. For the Election committee, MediaWiki message delivery (talk) 13:57, 23 November 2015 (UTC)

Precious anniversary

word cloud
Thank you, Philippe, for serving WMF, giving galaxies of barnstars, for fighting vandalism and nonsense ("We could really use your help to create new content, but ..."), for reminding us of the colourful word cloud and for clear words of reaching out, - repeating: you are an awesome Wikipedian (4 March 2010)!

--Gerda Arendt (talk) 08:55, 15 December 2014 (UTC)

A year ago, you were recipient no. 1063 of Precious, a prize of QAI! --Gerda Arendt (talk) 06:54, 15 December 2015 (UTC)

Requesting rollback

The user Helmoony is mass-removing "Tunisian arabic" in articles that was in place since several months. after the debate in Tunisia portal in september, an admin finally decided to grant the mention of Tunisian arabic for the names of people and places. Things were so settled since then, but the user Helmoony is recently removing all the mentions, could you please do something about it ? Thank you.Zangouang (talk) 16:23, 5 January 2016 (UTC)

I will be glad to take a look, but honestly it will likely not happen until this evening at the earliest. If you need quicker action, you might want to leave a note on WP:ANI for an administrator there to have a look. -Philippe (talk) 19:21, 5 January 2016 (UTC)

UTRS Account Request

I confirm that I have requested an account on the UTRS tool. -Philippe (talk)

• I've just activated your account, and asked our developper to flag you as CU. Glad to have you back! :) ☺ · Salvidrim! · ✉ 15:37, 15 February 2016 (UTC)

Question

As a former employee of WMF, you could perhaps give me some insight. You know the results of the harassment survey that were published (I saw your reaction on the mailing list). You may or may not know that those results were hugely inflated because they counted all responses, including those where people gave a value of zero (for the question: how many times did you experience the following types of harassment).

Quick overview:
The first two days of the survey, people couldn't skip items that didn't apply (Kalliope (WMF) reported on nov 4 that it had been fixed). This will have resulted in a lot of zero answers present.
The raw data of the results confirm that zero values were included: the average and standard deviation for revenge porn and hacking are mathematically impossible with only values between 1 and 100. The revenge porn results contain at least 381 zero's. It's a weak lower bound, the actual number cannot be determined from the average and standard deviation (doxxing for example could contain even more zero's, or none at all, the lower average/sdev ratio makes it impossible to conclude anything )
The revised report simply dropped the percentages (respondents who reported each type of harassment) and instead gives the averages for each type. This is useless information, because people gave inflated numbers. The average and sdev for hacking for example are only possible if the 14 highest responses add up to more than 1000 (49% of total). People having their account hacked more than 70 times?? The most likely explanation imo is that with 10 sliders from 0 to 100, people will have indicated the importance, rather than the number of incidents (or the sliders were difficult to move accurately, or a few trolls..).
There has been no straight answer to the claim, it seems obvious to me that they don't want to admit the error and give the correct results. The qualtrics application may or may not provide a way to count non-zero results (the help page for slider controls mentions limited analytic tools as a possible drawback ), I don't know, but the site explicitly mentions the option to download all data if users want to perform additional processing.
The most likely reason why they won't address the issue, imo, is that the real figures are much lower, and releasing them would raise the question why they didn't double-check the very unlikely results.

That's the short summary. Few people (outside the WMF) know the exact details I think, my first (erroneous) hypothesis (a technical error, forcing users to enter non-zero values) was based on a feedback post from nov 3 (saying he could only proceed to the next question after entering non-zero values for all the ten types listed), and it was this version that people posted on reddit and on Jimbo's talk page. People reading the survey talk page could also easily get the wrong idea. The two versions would give the same results btw, in both cases all surveys from 2 to 4 november would be affected. The big difference (which I should have realized much sooner): it's pretty unlikely that hundreds of participants would submit wrong answers because they couldn't finish the survey otherwise, and only one of them would report it.

My question: if you accept this version of events, do you think there is a chance that the WMF will publish the results eventually, or will they ignore such request because the impression of a "botched survey" is preferable to the alternative? I've seen little interest from the community, Patrick Earley's reactions (on the survey page and on his talk page) don't answer my questions, so I've given up trying to get answers on-wiki. I'm currently considering which science journalists or statistics blogs might be interested in doing a story about it.

BTW: Do you think any of the resignations last week (and the comments from Siko) could be related, or am I simply too focused on this case? Prevalence 12:45, 15 February 2016 (UTC)

Response

I should begin by saying that while this report was issued by members of my former team, I had nothing to do with its execution, and was not around for most of the time period in question, so I am not privy to any details about how this was constructed or what design choices were made.

You ask

• if you accept this version of events, do you think there is a chance that the WMF will publish the results eventually, or will they ignore such request because the impression of a "botched survey" is preferable to the alternative? - I do not yet know enough to know if I accept this version of events. However, my experience with this team suggests that if errors were made, they will be corrected.
• Do you think any of the resignations last week (and the comments from Siko) could be related, or am I simply too focused on this case? - No, my understanding is that none of them related in any way to this case. Rather, they were exactly what they appear to be... a mix of people moving on for various reasons. I would place this case's importance in those decisions at around 0%. -Philippe (talk) 12:52, 15 February 2016 (UTC)

In my scenario the decision would have come from the top, not from inside Support and Safety or CE. It hasn't exactly been a good news show lately, maybe they wanted to avoid yet another WMF controversy on top of it. Thank you for your answer. Prevalence 13:34, 15 February 2016 (UTC)

I think that in the fullness of time, we will realize that this is exactly what it appears to be: errors that were made at an unfortunate time. Sometimes with hoofbeats you chase zebras... but sometimes, they're just horses. This is all horse, I think. -Philippe (talk) 13:49, 15 February 2016 (UTC)

When Patrick Earley gives "the data and graphs weren't as informative as they should have been" as motivation for removing the percentages and displaying the averages instead, what else can one conclude? I changed my opinion about the team not responsible for the decision btw, I just noticed that even though these averages are useless as averages, they do correspond fairly well to the number of users who would have reported each type of harassment (if you take 30 as corresponding to 100%, most values seem reasonable). So the team could easily justify it for themselves, no harm done...

Doesn't change much as far as I'm concerned, for someone who understands averages, the results for the rare types are still absurd. And they introduced another error: saying that the statistics are based on 1215 respondents, instead of on the specific number of responses received for each of the ten questions. For hacking there were 760 responses, which gives a total of 2044 (760*2.69) reported incidents, not 3268 (2.69*1215). Prevalence 15:13, 15 February 2016 (UTC)

The Signpost: 09 March 2016

• News and notes: Katherine Maher named interim head of WMF; Wales email re-sparks Heilman controversy; draft WMF strategy posted

• In the media: Wikipedian is break-out star of International Women's Day; dinosaur art; Wikipedia's new iOS app and its fight for market share

• Op-ed: A modest proposal for Wikimedia’s future

• Featured content: Five articles, four lists, a topic, and five images were promoted this week.

• Technology report: Wikimedia wikis will temporarily go into read-only mode on several occasions in the coming weeks

• WikiCup report: First round of the WikiCup finishes

• Blog: The new alchemy: turning online harassment into Wikipedia articles on women scientists

• Systemic bias: Revenge of "I can’t believe we didn’t have an article on ..."

• Traffic report: All business like show business

• Read this Signpost in full
• Single-page
• Unsubscribe
• MediaWiki message delivery (talk) 19:54, 13 March 2016 (UTC)

A cup of coffee for you!

Just wanted to give you thanks for conducting that gigantic CU at Isla Riordan, and some coffee to help you stay awake through it. Much appreciated! GABHello! 22:48, 13 March 2016 (UTC)

Aww, thanks! Much appreciated.  :) -Philippe (talk) 22:50, 13 March 2016 (UTC)

Ping me if you need a hand. Risker (talk) 22:52, 13 March 2016 (UTC)

The Signpost: 16 March 2016

• News and notes: Wikipedia Zero: Orange mobile partnership in Africa ends; the evolution of privacy loss in Wikipedia

• In the media: Wales at SXSW; lawsuit over Wikipedia PR editing

• Op-ed: Hard work needed to address Wikimedia’s leadership challenges

• Discussion report: Is an interim WMF executive director inherently notable?

• Featured content: This week's featured content

• Technology report: Watchlists, watchlists, watchlists!

• Traffic report: Donald Trump, the 45th President of the United States

• Wikipedia Weekly: Podcast #119: The Foundation and the departure of Lila Tretikov

• Blog: It “revolutionized the way German-speaking people inform themselves about the world”: Fifteen years of the German Wikipedia

* Read this Signpost in full * Single-page * Unsubscribe * MediaWiki message delivery (talk) 21:12, 19 March 2016 (UTC)

Egaplaicesp and Tirgil34 are the same person

I salute your decisiveness when recommending a duckblock for Gushtasp as a Egaplaicesp sock. Egaplaicesp is however definately the same person as the prominent sockmaster Tirgil34. Having this fact established by an administrator would be of great help to future SPIs. Aside from their same behaviour of promoting Turkic at the expense of Indo-European history in Central Asia through edit warring and extensive socking, their connection can be proven with solid evidence:

• Proto-Turkic language/Urheimat: Tirgil34 (IP sock)[1]/Egaplaicesp[2]
• Haplogroup R1b: Tirgil34[3]/Egaplaicesp[4]
• Identical way of marking edits: Tigil34 (IP sock)[5]/Egaplaicesp[6]
• Quoting "starling.rinet.ru" source: Tirgil34[7][8][9][10][11][12]/Egaplaicesp[13]
• Promoting the image "Eickstedt Europe Turanid.jpg": Tirgil34 (image created by Su4kin)[14]/Egaplaicesp[15]
• Pazyryk culture/Pazyryk rug: Tirgil34[16]/Egaplaicesp[17]
• Karasuk culture/Karasuk languages: Tirgil34[18]/Egaplaicesp[19]
• Quoting "ncbi" source: Tirgil34[20][21]/Egaplaicesp[22][23][24][25][26]

Regards. Krakkos (talk) 21:30, 21 March 2016 (UTC)

While I agree that it would be useful for future purposes, I have a fairly strong belief that checkuser is a tool of last resort - when we can demonstrate account relatedness in other ways (as we very clearly can here), there is no need to use Checkuser. Using it probes into the activities of anyone using the same IP, and any innocent parties deserve to have their privacy protected as much as possible. Therefore, where we have behavioral evidence that can clearly connect users, I always consider that a better way to do so. I agree with you that this evidence is damning, and would consider it to make the connections. -Philippe (talk) 04:33, 22 March 2016 (UTC)

The Signpost: 23 March 2016

• Interview: Exclusive: interview with interim ED Katherine Maher

• News and notes: Lila Tretikov a Young Global Leader; Wikipediocracy blog post sparks indefinite blocks

• In the media: Angolan file sharers cause trouble for Wikipedia Zero; the 3D printer edit war; a culture based on change and turmoil

• Traffic report: Be weary on the Ides of March

• Editorial: "God damn it, you've got to be kind."

• Featured content: Watch out! A slave trader, a live mascot and a crested serpent awaits!

• Arbitration report: Palestine-Israel article 3 case amended

• Wikipedia Weekly: Podcast #120: Status of Wikimania 2016

* Read this Signpost in full * Single-page * Unsubscribe * MediaWiki message delivery (talk) 03:24, 27 March 2016 (UTC)

The Signpost: 1 April 2016

• News and notes: Trump/Wales 2016

• In the media: Saskatoon police delete Wikipedia content about police brutality

• WikiProject report: Why should the Devil have all the good music? An interview with WikiProject Christian music

• Traffic report: Donald v Daredevil

• Featured content: A slow, slow week

• Technology report: Browse Wikipedia in safety? Use Telnet!

• Recent research: "Employing Wikipedia for good not evil" in education; using eyetracking to find out how readers read articles

• Wikipedia Weekly: Podcast #121: How April Fools went down

• Blog: Growing hashtags: Expanding outreach on Wikipedia

* Read this Signpost in full * Single-page * Unsubscribe * MediaWiki message delivery (talk) 20:12, 1 April 2016 (UTC)

User:Gmartin1122/Wikia

Philippe, can you explain what exactly is going on at User:Gmartin1122/Wikia? I was just dating the draft and it looks like you created a WP:UP#COPIES of Wikia with incomplete attribution which is fine but why is it in the userspace of another editor who's only edit was a COI notice and why are you working on it there? -- Ricky81682 (talk) 02:52, 9 April 2016 (UTC)

Gmartin is a coworker of mine at Wikia. We are collaborating on a new draft of the article to present on the talk page as a beginning point for an update to the article, in hopes that we can convince some folks to make the necessary updates. That is also why both of us placed COI notices up pertaining to Wikia. We won't be directly editing the article, of course, just providing a draft as a starting point, on the talk page. -Philippe (talk) 02:57, 9 April 2016 (UTC)

Ok, that makes more sense. Thanks. The userspace choice was a bit unusual to me. -- Ricky81682 (talk) 03:05, 9 April 2016 (UTC)

Sure, I absolutely understand that. Obviously, we're working very hard to be sure that we are within policy here - when we take the suggestions to the article talk, we will again disclose our COI, and use the appropriate templates. At the same time, because this article is a derivative work of the one that's already there, the attribution issue will also be cured. If there's anything we can do to make this more transparent, I welcome the suggestion. (Obviously, the intention was to lay all of this out when we go to the article talk, you just beat us to it - and beat me to getting the article cleaned up and ready as well, btw.) -Philippe (talk) 03:07, 9 April 2016 (UTC)

Bay Area WikiSalon series kickoff, April 27

Please join us in San Francisco!

Panel discussion at a recent Wikipedia & Journalism event.

The last Wednesday evening of every month, wiki enthusiasts in the San Francisco Bay Area will gather to collaborate, mingle, and learn about new projects and ideas. We have two brief presentations lined up for our kickoff event in downtown San Francisco:

• The Nueva Upper School recently hosted the first ever high school Wikipedia edit-a-thon. We will hear what interests them about Wikipedia, what they have learned so far, and what they hope to achieve.
• Photojournalist Kris Schreier Lyseggen, author of The Women of San Quentin: The Soul Murder of Transgender Women in Male Prisons, will tell us about her work and how she researched the topic.

We allow time for informal conversation and working on articles. Newcomers and experienced wiki users are encouraged to attend. We will have beverages and light snacks.

Please note: You must register here, and bring a photo ID that matches your registration name. The building policy is strict on this point.

For further details, see here: Wikipedia:Bay Area WikiSalon, April 2016

We hope to see you -- and until then, happy editing! - Pete, Ben & Wayne

The Signpost: 14 April 2016

• Op-ed: Should prison inmates be permitted to edit Wikipedia?

• News and notes: Denny Vrandečić resigns from Wikimedia Foundation board

• In the media: Wikimedia Sweden loses copyright case; Tex Watson; AI assistants; David Jolly biography

• Featured content: This week's featured content

• Traffic report: A welcome return to pop culture and death

• Arbitration report: The first case of 2016—Wikicology

• Gallery: A history lesson

* Read this Signpost in full * Single-page * Unsubscribe * MediaWiki message delivery (talk) 00:45, 14 April 2016 (UTC)

Re: Your edit to Wikia

Hello!

Sorry about that! I should have checked the link. Reverted it. Thanks for the heads up! Fgh (Talkpage)