| ppelberg | |
| Mar 19 2025, 11:37 PM |
This task involves the work of compiling the list of templates and policies the wikis listed below use to signal that a given edit/sentence/paragraph/section/article/etc. contains non-neutral language.
The list this task is asking for will enable@achou – as part ofT388215 – to identify what top-20 wikis contain enough eval data for the Peacock Check model to train on.
More broadly, this task sits within the larger choice we are making inT387925 about what languages v1 of the Peacock Check model will support.
For each of the wikis listed below:
Please fill the table below with information regarding promotional, non-neutral, biased... content.
Please use a comma-separated format.
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Open | None | T265163Create a system to encode best practices into editing experiences | |||
| Open | None | T365300Introduce Edit Checks that encourage behavior aligned with Manual of Style consensus | |||
| Open | None | T365301Peacock Check: Prompt people to revise promotional language | |||
| Open | None | T387918[MILESTONE] Run an A/B test to evaluate impact of Peacock Check | |||
| Open | ppelberg | T388471Invite volunteers to review Peacock language model (v1) | |||
| Open | None | T387925Determine language support for Peacock Check (v1) | |||
| Open | None | T388736[SPIKE] Create a lightweight estimate for proportion of new content edits reverted because of peacock language | |||
| Open | Trizek-WMF | T389445Compile list of templates, jargon and policies relevant to NPOV |
For Romanian, I suggest reaching out to User:Klaul who lead this research project on NPOV:https://meta.wikimedia.org/wiki/Research:Grounding_NPOV_pillar_in_post-censored_information_ecology
InT389445#10669368,@Strainu wrote:For Romanian, I suggest reaching out to User:Klaul who lead this research project on NPOV:https://meta.wikimedia.org/wiki/Research:Grounding_NPOV_pillar_in_post-censored_information_ecology
Thank you, I shared it with the team!
Hi, I added some templates and policies from Spanish Wikipedia. Do we need to add a policy (if it is available) about non-sexist language?. I'll ask some power users about jargon in edit summaries.
InT389445#10675289,@Bachounda wrote:@Superzerocool I see an edit conflict—let me know when you’re done. :)
I'm done. Please, continue :) (I'll copy my links if there is something wrong ;))
I found these templates from this page in arwphttps://ar.wikipedia.org/wiki/%D8%AA%D8%B5%D9%86%D9%8A%D9%81:%D9%82%D9%88%D8%A7%D9%84%D8%A8_%D8%B5%D9%8A%D8%A7%D9%86%D8%A9_%D8%A7%D9%84%D9%85%D8%AD%D8%AA%D9%88%D9%89
InT389445#10675273,@Superzerocool wrote:Hi, I added some templates and policies from Spanish Wikipedia. Do we need to add a policy (if it is available) about non-sexist language?. I'll ask some power users about jargon in edit summaries.
Thank you for your help!
Please add any policy that fits your community's definition of NPOV.
@Msz2001, thank you for your help! Can you add the parameter(s) used for{{Dopracować}} specific to neutrality issues? I identifiedpov,zależne, andreklama, but I prefer to rely on your expertise. :)
Thank you@Amire80! Are there any templates used to mark pages for promotional/tone issues at he.wp?
InT389445#10670032,@Trizek-WMF wrote:InT389445#10669368,@Strainu wrote:For Romanian, I suggest reaching out to User:Klaul who lead this research project on NPOV:https://meta.wikimedia.org/wiki/Research:Grounding_NPOV_pillar_in_post-censored_information_ecology
Thank you, I shared it with the team!
@Strainu, meanwhile, could you start filling the table for Romanian Wikipedia please?
InT389445#10678410,@Trizek-WMF wrote:Thank you,@Wkee4ager! Does no.wp use templates and edit summary jargon?
The two templates listed are the ones I could think of. I'm not very aquainted with the jargon, but "egenreklame, npov, objektivitet, reklame" might have been used. I'll update the description.
Thank you@Wkee4ager!
A suggestion for everyone -- please share this task with your community if you think it is worth the effort: the more precise terms we have listed, the better the model would be.
We really appreciate the time and effort you put into gathering templates and terms for this project. Thank you!
If you are curious about the next steps, please add your names toT388471: Invite volunteers to review Peacock language model (v1): we are looking for testers for the model that will be built using the data you gathered!