User talk:OVasileva (WMF)

RfC on blockers

Thanks for taking the descriptions from Wikidata down from displays of en-WP on mobile. If you would like any input framing the RfC about "blockers" I would be glad to provide it. Jytdog (talk) 17:57, 30 March 2017 (UTC)[reply]

Wrong survey link

I saw a box at the top right of Weightlifting at the Summer Olympics. It contained this text (boxes and some formatting omitted):

Take a short survey and help us
improve Wikipedia

<ext-quicksurveys-affinity-survey-
description>

Visit survey No thanks

Survey data handled by a third party. Privacy

I haven't heard of the survey and post to you because you created MediaWiki:Ext-quicksurveys-affinity-survey-description. "Visit survey" goes to a wrong page. I guess the page should have been [1] where I see a survey. "Privacy" was linked but I didn't test the link and no longer have the page open. I was logged in and have many users scripts but don't know of any that should break the text or link. I have the default "en" interface language. PrimeHunter (talk) 12:43, 9 July 2020 (UTC)[reply]

Hi @PrimeHunter: thanks for pointing this out. We're aware - there were some issues with the initial configuration of the survey that we're currently fixing. The fix will be out in the next deployment (4 hours from now) OVasileva (WMF) (talk) 12:52, 9 July 2020 (UTC)[reply]

Being cautious about changes

I appreciate that you considered possible consequences before implementing Wikipedia:Village pump (proposals)/Archive 185#Next steps for enabling talk links for mobile anon users. While I think the first proposal was overly cautious, I think it's a good thing to at least consider the possibility of disruption before deploying a big change. So strangely..

I see a lot of disruption caused by the current fundraiser, mostly in the form of rants and questions on Wikipedia:Help desk and Wikipedia:Teahouse where helpless Wikipedians get chastised for something they have no control over. I'm not sure if it's worse than previous fundraisers, but subjectively it feels like it is. Where was the caution before launching the fundraiser in its current form?
Cross-wiki uploads have bombarded Commons with copyvios for years and continue to do so. Where was the caution before introducing that?
When structured data was introduced, the implementation was very poorly thought out and the result was a bunch of incorrectly licensed captions (to this day unfixed) for which no use case existed anyway and an ongoing creation of duplicated data that has to be externally synchronized. Where was the caution before enabling that?

I assume you don't have absolute control over everything, but it's a stark contrast between the extreme caution when the community asks for something and the complete lack of caution when the WMF or developers initiate something. (I have more examples but this is already too long) What I'm saying is, please, don't limit your caution to community initiatives. — Alexis Jazz (talk or ping me) 23:02, 11 December 2021 (UTC)[reply]

Metadata presentation for readers

Hi Olga! Picking up from your comment here at VPT, I think there's potential in presenting some metadata to help readers better understand the degree to which they can trust an article. It'll have to be done well, though, since article quality is a tricky thing to measure, and not all pieces of metadata speak to it, and even if they do, not all readers may know how to interpret it.

First, to clarify the goal, it should be what I said above — helping readers better understand the degree to which they can trust an article — not "getting readers to trust us," which is what I saw in the consultant report, and reflects some unfortunate bleedover from the corporate world.

On the pieces of metadata to use, let's go through some of the featured criteria and see which things can speak to them:

well-written – An article that's had more contributors is more likely to have had its grammar refined.
comprehensive – This is correlated with article length. However, it's not equivalent to it: very niche topics like Yugoslav submarine Nebojša can be FAs without being very long. Perhaps you'd want to consider factors like incoming links or pageviews to estimate how long an article ought to be and then compare its actual length against that.
well-researched – This is basically references. With very few exceptions (e.g. leads, plots), everything should be referenced, and any segment of an article lacking references should ding its quality rating.
neutral – This is tricky to measure: more talk page activity (relative to the popularity of a topic) often means it's more controversial, but doesn't necessarily mean it's unbalanced (and in fact often means the opposite). The only really reliable indicator of neutrality issues is the presence of a maintenance tag indicating such.
stable – This can be measured algorithmically, but you have to go beyond just "most recent edit" (which could just be someone archiving a reference). The percentage of article text that has been modified recently would be a much better indicator. Reverts in the recent edit history would be a strong indicator of instability, as would be an open RfC on the talk page.

Brainstorming on some other factors: Protection status could certainly be a relevant piece of info to present here, and per the VPT convo we may be looking to move that anyways. The date an article was created is relevant, as are deletion nominations. The XTools article info page has some things that might be relevant. In particular, the authorship section (readers will likely want to know if one editor has authored 90% of a page) and the year counts (that graph is very useful to see when an article has gone through a growth spurt). Who Wrote That? could be useful if it was improved to the point where it didn't have to be a separate browser extension (something there's no intrinsic need for). And I'm curious about ORES scores, which in my experience tend to be fairly accurate.

Hope all that is helpful, and curious to hear your thoughts! Cheers, {{u|Sdkb}} ^talk 03:14, 3 August 2022 (UTC)[reply]

You are seen.

	The Working Wikipedian's Barnstar
	Thank you, OVasileva. Not only in appreciation for great work on the years-long Vector 2022, but for suffering slings, arrows, incivility, and personal attacks because of it. — Fourthords \| =Λ= \| 23:14, 18 January 2023 (UTC)[reply]

Very bad and corrupt

Hi. You're enabling a very bad and corrupt organization and you should be ashamed of yourself. It's truly despicable that Wikimedia Foundation Inc. is taking in donations from readers by lying to them about the state of its finances and then using that money to hire folks like yourself to push through horrible software changes. A majority of respondents, when asked whether to deploy this worse skin, said no. And yet you and your team have wasted millions of dollars of donor money and barreled ahead despite this, all the while making false claims such as "if the community decides against deploying the skin, no deployment will be made" and writing misleading summaries about what the discussion "really" shows.

Will there be any repercussions or consequences for your team? Of course not, Wikimedia Foundation Inc. management and its board of trustees provide no actual accountability or oversight. But please don't think long-time community members haven't noticed you enabling this trash behavior. --MZMcBride (talk) 17:49, 20 January 2023 (UTC)[reply]

Request for Arbitration notification

You are involved in a recently filed request for arbitration. Please review the request at Wikipedia:Arbitration/Requests/Case#WikiMedia Foundation involvement in software deployment on the English Wikipedia and, if you wish to do so, enter your statement and any other material you wish to submit to the Arbitration Committee. As threaded discussion is not permitted on most arbitration pages, please ensure that you make all comments in your own section only. Additionally, the guide to arbitration and the Arbitration Committee's procedures may be of use.

Thanks, TomStar81 (Talk) 01:34, 20 February 2023 (UTC)[reply]

WikiMedia Foundation involvement in software deployment on the English Wikipedia: Arbitration request declined

Hello OVasileva (WMF),

The case request about WikiMedia Foundation involvement in software deployment on the English Wikipedia has been declined by a majority of Wikipedia's arbitrators.

The request had been explicitly created as an "Ignore All Rules" request, but the Arbitration Committee disagrees about a need for ignoring its policies, which exclude "official actions of the Wikimedia Foundation or its staff" from its jurisdiction and define it as "as a final binding decision-maker primarily for serious conduct disputes the community has been unable to resolve" in the "Scope and responsibilities" section. Its procedures describe an "expectation of prior dispute resolution" that hasn't been fulfilled yet.

For the Arbitration Committee,
~ ToBeFree (talk) 18:12, 23 February 2023 (UTC)[reply]

Update on the user survey

On the 14th of February, you said the WMF were looking into publishing the results of user survey in response to our request and would update us when you know more. Do you have an update for us? BilledMammal (talk) 15:47, 3 March 2023 (UTC)[reply]

User research on GA/FA awareness

Hi Olga! I'm considering reviving the proposal to move GA/FA topicons next to the article title as a way to better highlight a piece of information important for helping readers understand the extent to which they can trust an article. The main stumbling block last time was the fact that a significant contingent of editors seemed to believe that readers already noticed and understood the GA/FA icons. Personally, I find that highly implausible, but at the prior discussion we had only anecdotal evidence. I know you and I previously discussed (here, continuing from here) the possibility of doing some user research testing to get some empirical data, so I wanted to follow up to see if that'd be possible.

I'm not an expert at designing UX research, but I imagine the setup might be something like this. Two different groups would be presented with a prototype page — one with the status quo position for the GA/FA icons, and one with them next to the article title. After the user has had the chance to look over the page, we'd ask questions like "Has this page undergone a peer review? [Yes] [No] [Not sure]," "Which designation does this page have? [Featured article] [Quality article] [Good article] [None] [Not sure]," and "How much do you trust the content on this page? [A lot] [A medium amount] [A little] [Not at all]" The responses to those questions, and how they differ between the designs, would be extremely useful information.

Cheers, {{u|Sdkb}} ^talk 22:53, 4 April 2023 (UTC)[reply]

Pinging SGrabarczuk (WMF) as well in case you have thoughts. {{u|Sdkb}} ^talk 16:11, 8 April 2023 (UTC)[reply]

Hey @Sdkb - thanks for the question! We are actually closing out some somewhat similar user research right now. This research was focused on identifying whether any additional metadata would help with understanding how the wikis work, where the content is coming from, and how it's created. We're still going over the results but hope to publish soon. Generally, we were surprised to find that the baseline of understanding how the wikis work (for example, knowing that Wikipedia articles can be edited) has gone up quite significantly over the past few years. We're currently looking at the data to investigate whether this is somewhat connected to the specific group that we tested with. In terms of the metadata we tested - we looked at number of talk page comments, number of watchers, total edits, last edit time and date, etc. Initial results show that almost all of these (with the exception of the number of total discussions) had a positive effect on understanding the system. I hope we'll be able to publish within the next couple of weeks.

We haven't looked specifically at GA/FA tags though but I think we probably can add them to future research. My only concern with the approach of giving indicators would be that it can be interpreted as a scoring system that can potentially be biased when compared to more neutral metadata around the article that allows the reader to create their own idea of quality. What are your thoughts around this? OVasileva (WMF) (talk) 14:46, 10 April 2023 (UTC)[reply]

@OVasileva (WMF), that sounds like interesting research; I'll look forward to reading it (and feel free to ping me when it's published)!

I think getting readers to understand how wikis work is a good goal, albeit slightly different than the also-important goal of helping them to understand the extent to which they can trust a given article, which is more what the GA/FA icons try to do.

When it comes to that latter goal, my view (and I think the view of most experienced editors) is that the GA/FA icons are a far stronger indicator of quality than any of the pieces of metadata listed above. Talk pages comments just means a subject is popular or controversial, or even that there's one specific part of it that's controversial (as at e.g. Guy Standing (economist)). Watchers I'd guess is basically directly correlated with popularity. Total edits is just how old and how popular a page is. Last edit time could just be a bot correcting an apostrophe with zero human oversight.

We really have two separate questions here, and it's important not to conflate them. The first is, "what pieces of metadata reflect article quality?" and the second is "what pieces of metadata influence readers' perception of article quality?" If we find a piece of metadata that reflects article quality (in the eyes of researchers/experienced editors/etc.) but that readers don't see as such, then the issue is likely with how we present it.

On subjectivity, in a sense every editorial decision made on an article has a degree of subjectivity. One way to see it is that the fact that an article underwent a peer review for a GA/FA nomination is an objective piece of information, even if the contents of that review might be considered subjective. GA/FA icons are certainly fallible to bias just as everything else on Wikipedia is, and in particular they have the issue that older evaluations were a lot less strict than modern ones (which is what WP:URFA/2020 is slowly addressing). Still, despite their flaws, they're the best indicator of quality we have, so I think it's important that we make them noticeable to readers and educate readers about what they mean.

Let me know if you have questions/follow ups about any of that. I would certainly love to see them included in future user research. Cheers, {{u|Sdkb}} ^talk 02:22, 16 April 2023 (UTC)[reply]

A solstice greeting

❄️ Happy holidays! ❄️

Hi Olga! I'd like to wish you a splendid solstice season as we wrap up the year. Here is an artwork, made individually for you, to celebrate. Looking forward to collaborating on New Vector improvements in the coming year! Take care, and thanks for all you do to make Wikipedia better!

Cheers,
{{u|Sdkb}} ^talk

Solstice Celebration for OVasileva (WMF), 2023, DALL·E 3.

(View full series)

Note: The vibes are winter solsticey. If you're in the southern hemisphere, oops, apologies.

{{u|Sdkb}} ^talk 06:52, 24 December 2023 (UTC)[reply]

Reminder to vote now to select members of the first U4C

You can find this message translated into additional languages on Meta-wiki. Please help translate to other languages.

Dear Wikimedian,

You are receiving this message because you previously participated in the UCoC process.

This is a reminder that the voting period for the Universal Code of Conduct Coordinating Committee (U4C) ends on May 9, 2024. Read the information on the voting page on Meta-wiki to learn more about voting and voter eligibility.

The Universal Code of Conduct Coordinating Committee (U4C) is a global group dedicated to providing an equitable and consistent implementation of the UCoC. Community members were invited to submit their applications for the U4C. For more information and the responsibilities of the U4C, please review the U4C Charter.

Please share this message with members of your community so they can participate as well.

On behalf of the UCoC project team,

RamzyM (WMF) 23:09, 2 May 2024 (UTC)[reply]