Talk:Community Notes/GA1

GA Review

Article (edit | visual edit | history) · Article talk (edit | history) · Watch

Reviewer: Czarking0 (talk · contribs) 21:43, 14 January 2024 (UTC)

Thanks for detailed review, I will work on making those changes as advised, all very useful. I was using inline citations as per WP:ATTRIBUTEPOV, but you think they should all be removed if not quoting the source? Happy to remove them if you think not necessary, maybe the information is more factual than biased opinion? My only other query here is ADL not being a reputable source, despite being listed as reliable WP:RSPADL? Hence the inline attribution there, as advised for what appears to be an opinion piece. Thanks again. CommunityNotesContributor (talk) 10:41, 15 January 2024 (UTC)

Great response, I am working on these comments here. As far as the ADL point goes I think you have the right of it. WP:ATTRIBUTEPOV should only be used for statements you see as biased. If you think statememts are biased give it a second thought as to if they are sufficiently notable as to be included in the article or if there is a less biased source/way to phrase the point that should be made. I won't harp the on inlines any more Czarking0 (talk) 20:25, 20 January 2024 (UTC)

Have made changes based on suggestions up to the criticism section, as waiting on confirmation of the above, so just WP:INTEXT to remove from there as far as I can tell. Have removed Le Monde opinion, as indeed isn't very notable. It was leftover text from original article prior to more sources becoming available for context sake. CommunityNotesContributor (talk) 11:57, 15 January 2024 (UTC)

Working on getting through all your changes Czarking0 (talk) 20:26, 20 January 2024 (UTC)

I added comments below Czarking0 (talk) 00:45, 21 January 2024 (UTC)

Thanks, will sort out the rest of the sources, thanks for expanding. Have had a few computer problems in last few times so have been unable to attend to the rest of this GAR but will sort it out soon. CommunityNotesContributor (talk) 16:56, 24 January 2024 (UTC)

Hello and thank you for your contribution. I believe this article is close to GA and I want to add some comments to be addressed before approving. I was skeptical of this being sufficiently noteworthy but your sources demonstrate notability. This section of the review is just based on your writing. I will review sources if you implement these changes. Also the sections are just history, operation, and criticism. Are there potentially other sections that should be added? "

Introduction

• "Community Notes, formerly Birdwatch, is a feature on X (formerly Twitter) where contributors can "add context" in order to provide fact-checks, under a post, image or video." My understanding is that fact checking is not the sole purpose here I think think this would be better "Community Notes (formerly Birdwatch) is a feature on X (formerly Twitter) where contributors can "add context" such as fact-checks to posts, images, videos, or comments.  Done

• Multiple times throughout the article you say "As of [month] " put the year in afterwords otherwise in 5 years it becomes confusing to read.  Done

• Try to keep your coverage of events in the last 90 days brief since wikipedia is not for current events

• "It has been considered as an attempt to debunk propaganda and misinformation[12] and as a replacement for Trust and Safety staff,[13] although former head of the department, Yoel Roth, states this was never the intention.[9]" The grammar here is not great. Try to never user "It has been considered" has this naturally raises the question: by whom? Alternatively try, "Its goal is to debunk propaganda and misinformation, but its scope was greatly expanded when it replaced the Trust and Safety staff".  Done

• "Posts receiving notes on X are no longer eligible for ad revenue[20][21] and users are no longer able to report misleading content.[22]" This sentence does not really read as intended. It sounds like user's are only no longer able to report misleading content on posts with notes, but I think the sources indicate that no user reports for misleading content are still supported.  Done

• "The program has also been inconsistent in its application of notes and combating of misinformation, especially from the 2023 Israel-Hamas war.[a]" This falls squarely into current events. I like the note itself but I don't think including it in the introduction to the article is appropriate at this time.  Done

New feedback:

• " It has been described as a community-driven content moderation program" why not: "It is a community-driven content moderation program"?

History

• "but was considered to be a very small portion" by whom?  Done

The reason I ask by whom here was meant to be rhetorical sorry for being unclear there. This is not a biased claim so it does need in next citation. The better presentation to the user would be "This then increased to 156 on the day of the invasion, a small fraction of the propaganda by the Internet Research Agency. You can either use a ref from that page or

Still slightly confused here. I assume you mean "does not need in text citation" as opposed to "does need"? The reference you provided is from 2019, so also don't understand how this could be used to reference community notes in 2022. I've however removed the in-text, and changed "considered" to "estimated" as per the source "probably a tiny sliver", as opposed to definitively a very small amount. Hope that's better. Otherwise, this sentence might be better removed, as is based on speculation rather than definitive numbers. CommunityNotesContributor (talk) 14:13, 21 January 2024 (UTC)

I am using that source as an estimate of how much propaganda the Internet Research Agency produces. Correct I meant to say does not need. Czarking0 (talk) 03:03, 24 January 2024 (UTC)

• "In October 2022 The Verge found that the most commonly published notes were related to COVID-19 misinformation based on historical usage." This is not encyclopedic. The encyclopedic way to write this is "The most commonly published notes in 2022 were related to COVID-19 misinformation". The citation will show The Verge. You should fix this anywhere you see the same issue.

This has been done by the way. CommunityNotesContributor (talk) 14:16, 21 January 2024 (UTC)

 Done Czarking0 (talk) 02:58, 24 January 2024 (UTC)

Operation

• "The program prioritises notes that receives ratings from a "diverse range of perspectives",[7] rather than be based on majority rule,[18] by an open-source algorithm described as "insanely complicated". The structure of this sentence is i,d,d where i is independent clause and d is dependent. This is not a valid sentence structure in the English language. Also prioritises is spelled prioritizes at least in American English. Try "Rather than majority rule publishing, the open-source Community Notes algorithm prioritizes notes that receive positive ratings from a "diverse range of perspectives" The fact that the algorithm is described as complicated is notable but again I ask "by whom?".  Done

You addressed this and now the grammar is valid (I think) but it is still a bad sentence. Avoid the present perfect continuous tense. Try something more like "The Community Notes algorithm publishes notes with agreement from contributors who have a history of disagreeing. Vitalik Buterin described it as 'insanely complicated' after reviewing the open-source implementation."

Apologies, my use of tenses has never been good. Hopefully this is better now? CommunityNotesContributor (talk) 14:51, 21 January 2024 (UTC)

• "The contributor gets points if their note is validated,[38][17] known as "Rating Impact", that reflects how helpful a contributors' ratings have been. A contributor unlocks the ability to write notes once they have a "Rating Impact" of at least 5.[10][39][40] Any registered X user with an account older than 6 months can apply to become a contributor, provided they supply a mobile number, the user agrees to abide by the Community Notes guidelines,[7][41] and the user hasn't broken any X rules recently.[32] Users on the platform can additionally vote on whether they find the note helpful or not.[2]" I am not convinced the specific guidelines for how one becomes a contributor are notable. Also they could more easily be changed while the notable part can stay the same: Users must apply to becomes contributors and are restricted based on their "Rating Impact".

"X users are able to vote on whether they find notes helpful or not,[2] but must apply to become contributors, that is restricted based on "Rating Impact" as well as the Community Notes guidelines" It is not clear what the that in this sentence refers to.

Think this is clearer now. CommunityNotesContributor (talk) 14:51, 21 January 2024 (UTC)

Agreed  Done Czarking0 (talk) 03:04, 24 January 2024 (UTC)

Criticism

• "Le Monde concluded that Community Notes were useful, but were not a substitute for conventional moderation." This seems like an opinion piece from a newspaper. Is this really notable?  Done

• "The fact-checking website Snopes discovered three posts from verified users, who had shared a video of a hospitalized man from Gaza with false captions claiming it showed "crisis actors", had failed to receive any Community Notes after 24 hours.[61]" This is another example of non-encyclopedic writing. The content is fine just reword it  Done

• " The ADL documented the possibility of conflicting notes appearing, after Jackson Hinkle falsely claimed a graphic image of the 2023 Hamas attack on Israel was AI-generated, and notes initially appeared in disagreement due to unreliable information from AI detector software.[62] " The ADL is not a reputable source.  Done

• " Wired has documented that Community Notes is susceptible to disinformation, after a graphic Hamas video shared by Donald Trump Jr. was falsely flagged as being a year old, but was instead found to be part of the recent conflict.[25] The original note was later replaced with another citing the report from Wired.[9]" I just wanted to point out that this might feel like another case of non-encyclopedic writing it is actually an exception to the rule since Wired becomes part of the notable factual content.  Done

• "In November 2023, the Atlantic Council conducted an interactive study of Community Notes, with analysis from Bloomberg, " non-encyclopedic

Sources

• Mashable: you have sufficient other sources this should not be included and places where it is cited should be removed.  Done

• https://uk.finance.yahoo.com/news/bird-watching-going-x-twitter-111442959.html?guccounter=1&guce_referrer=aHR0cHM6Ly9lbi53aWtpcGVkaWEub3JnLw&guce_referrer_sig=AQAAAAfXT-IYiwYEHwJxWFpLzr6a4ZL-Br1g6SY3KdFkOadZ8A0e7etL1uu4G6vU8hRA8HdBoTyEiZTJk_XvRrbFqxYjZp8ZGVvmNBJYPiKbPJevcehPkQ-GzhxporLngv3cmDVTScMVKOZtP_vWEUzQGGhiyJGCXiScMwV99HCjnciU This is an opinion piece. Reconsider using it.

• https://www.mediaite.com/news/elon-musk-claims-state-actors-are-manipulating-community-notes-after-his-post-gets-fact-checked/ Not a reliable source  Done

More to come — Preceding unsigned comment added by Czarking0 (talk • contribs) 00:54, 21 January 2024 (UTC)

• Bump, Philip (December 11, 2023). "A new benchmark on the right's descent into surreality". The Washington Post. ISSN 0190-8286. Archived from the original on November 12, 2023. Retrieved January 22, 2024. This is an opinion piece and the claim is already sourced. I recommend it is removed.  Done

• "In October 2023 the program was updated for notes to appear faster on the platform, described as a "massive speed improvement" by X, as well as scaled to send notifications to users who had interacted with the posts.[31]" Do you have another source for this? I am not sure that this is notable. Seems like it could be a corporate fluff piece.  Done
  • No other source I could fine, agreed it's likely fluff in hindsight, have removed. CommunityNotesContributor (talk) 11:04, 5 February 2024 (UTC)

• X community notes: Weapon against fake news or free speech?". The Japan Times. September 3, 2023. Archived from the original on October 26, 2023. Retrieved December 5, 2023. This source does not provide the info it is cited for. Also very little in it that is not covered by other sources  Done
  • I think this cite was misplaced, I don't remember it, it's gone. CommunityNotesContributor (talk) 11:04, 5 February 2024 (UTC)

• "Novak, Matt. "Ron DeSantis Ends Campaign With Fake Quote From Winston Churchill". Forbes. Retrieved January 22, 2024." This is the most blatant violation of WP: NO TNP I have ever seen. Please remove  Done
  • It's part of a list of "notable" examples of CN uses. Either the whole list of examples is irrelevant and considered WP:NOTNP, or it's not. I don't understand why this single example is being singled out. It is not elaborated why DeSantis received a note, like the other examples, only that he has received one has a prominent politician. CommunityNotesContributor (talk) 11:04, 5 February 2024 (UTC)

• Remove newsweek sources  Done
  • WP:NEWSWEEK is marginally reliable I'm aware. Moreover, the claims referenced are not "extraordinary" claims, they are regurgitating relatively public information, in order to avoid using WP:TWITTER as a source. I think in this case WP:RSCONTEXT matters. CommunityNotesContributor (talk) 11:04, 5 February 2024 (UTC)

The reliability of a source depends on context. Each source must be carefully weighed to judge whether it is reliable for the statement being made in the Wikipedia article and is an appropriate source for that content.

• "They can also note when an image is digitally altered or AI-generated,[5][52] in order to crack down on such content." I would end this sentence at AI-generated and not site the Evening Standard  Done

• Morrissey, Brian (May 2, 2023). "Elon Musk: Web 2.0 man". therebooting.com. Archived from the original on October 22, 2023. Retrieved October 21, 2023. This is an opinion piece from an unreliable source that adds little to content of the article. Remove.  Done

• Rand, David; Martel, Cameron; Lee Allen, Jennifer Nancy (December 2, 2021). "Birds of a feather don't fact-check each other: Partisanship and the evaluation of news in Twitter's Birdwatch crowdsourced fact-checking program" (PDF). osf.io. Archived from the original on November 17, 2023. Retrieved November 24, 2023. This is an open access publication which is unreliable. Anyone can upload this here. Remove. I would remove the claim as well unless you have other sources.  Done
  • Have removed the last sentence referencing this. It's the study referenced by WIRED, so thought it'd be useful to reference as the RS did, but indeed is open-source published so have removed. CommunityNotesContributor (talk) 11:04, 5 February 2024 (UTC)

• "Blue-Checked, 'Verified' Users on X Produce 74 Percent of the Platform's Most Viral False or Unsubstantiated Claims Relating to the Israel-Hamas War - Misinformation Monitor: October 2023". NewsGuard. Archived from the original on October 20, 2023. Retrieved November 18, 2023. This is not a reliable source. I would remove related claims or demonstrate that it is reliable.  Done
  • Another case of citing the study that RS references, have removed. CommunityNotesContributor (talk) 11:04, 5 February 2024 (UTC)

• Weatherbed, Jess (October 20, 2023). "Blue checkmarks on X are 'superspreaders of misinformation' about Israel-Hamas war". The Verge. Archived from the original on October 20, 2023. Retrieved November 18, 2023.
• Spangler, Todd (October 20, 2023). "X/Twitter Verified Blue Check-Mark Users Are 'Superspreaders' of Disinformation * About Israel-Hamas War, Study Says". Variety. Archived from the original on October 20, 2023. Retrieved November 18, 2023. These two sources are suspiciously similar. Please investigate.  Done
  • Yes, because they reference the NewsGuard study, so there is nothing suspicious here. Two reliable sources are referencing a source considered not reliable, because they believe the study is reliable. If I had to guess, the RS were able to analyse and verify the credibility of the study, which would explain why they consider it reliable. CommunityNotesContributor (talk) 11:04, 5 February 2024 (UTC)

  Nice work on most of the sections. I am getting to it all. In this particular case I have to disagree with you. The Verge and Variety are reputable in some regards but they are not top tier. WP:VARIETY is reliable for entertainment which this is not strictly entertainment. Overall this seems more like a case of two reliable sources that decided to publish an article that is not reliable. We can take them as a reliable source for their reporting but a reliable source choosing to publish an unreliable one does not make the unreliable one reliable. In either case, the Community Notes article goes on to directly quote the NewsGuard study which really can only be attributed to the NewsGuard source. Czarking0 (talk) 02:27, 14 February 2024 (UTC)

  On this note, that last paragraph is just Mashable and NewsGuard. It also presents a bunch of statistics that are not very well contextualized. I think this whole paragraph should be removed. Czarking0 (talk) 02:44, 14 February 2024 (UTC)

• "The study also included Jackson Hinkle." Is this notable?  Done
  • Yes, he's a notable participant, have amended line to reflect that. CommunityNotesContributor (talk) 11:04, 5 February 2024 (UTC)

• Syme, Pete (November 21, 2023). "Elon Musk's pet feature on X took as long as 3 days to flag fake news about Israel and Gaza, analysis shows". Yahoo Finance. Archived from the original on November 24, 2023. Retrieved November 21, 2023.

Syme, Pete (November 21, 2023). "Elon Musk's pet feature on X took as long as 3 days to flag fake news about Israel and Gaza, analysis shows". Business Insider. Archived from the original on November 24, 2023. Retrieved November 24, 2023. Two references created for the same article  Done

Czarking0 (talk) 03:53, 24 January 2024 (UTC)

Updated this section to indicate what is done and what is not. Czarking0 (talk) 03:18, 5 February 2024 (UTC)

I have updated all that was not done. Can you respond saying why? @CommunityNotesContributor

Czarking0 (talk) 03:20, 5 February 2024 (UTC)

• @Czarking0 Apologies for delay, have removed unreliable sources and responded to others with indented bullets. Where I haven't responded, the source should be removed. Willing to remove whatever else is left if necessary, but first thought I'd try to explain why I believe the sources deserve to stay. CommunityNotesContributor (talk) 11:04, 5 February 2024 (UTC)

Positive Remarks

GA Reviews can get negative so I wanted to leave a few things that I think really stand out as good. Maybe future editors can be inspired.

• This source is really good and the paragraph that summarizes it is as well: https://web.archive.org/web/20231017224528/https://www.wired.com/story/x-community-notes-disinformation/ — Preceding unsigned comment added by Czarking0 (talk • contribs) 02:35, 14 February 2024 (UTC)

Stability

• One of the sources is titled "Twitter's Community Notes is the fiercest 2024 primary battleground".

• Community Notes for images was launched in the 90 days.

• By my count 16 out of the 53 sources were published in the last 90 days.

These three points seems to me to indicate reviewers should not expect this article to demonstrate a level of stability sufficient for GA. However @CommunityNotesContributor has demonstrated a high level of understanding of the norms of wikipedia so i will wait for them to respond before passing a judgement.

• This is not what stability means: it does not change significantly from day to day because of an ongoing edit war or content dispute.. The note clearly states: Stability is based on the article's current state, not any potential for instability in the future. CommunityNotesContributor (talk) 10:49, 21 February 2024 (UTC)