About this board

Previous discussion was archived at User talk:PAC2/Archive 1 on 2015-11-13.

VIGNERON (talkcontribs)

Salut,

Je n'ai toujours pas testé ton dernier tutoriel mais cela m'a motivé à re-tester Observable : https://observablehq.com/d/7893924aeec79613

J'en ai profité pour tester aussi GraphViz, si tu connais cet outil ou si tu en connais d'autres, je suis preneur de tes remarques et conseils.

Cdlt,

PAC2 (talkcontribs)

GraphViz a l'air très intéressant mais je ne pense pas qu'on puisse y représenter des données. J'irai plutôt regarder du côté des représentations hiérarchiques de type dendogram ou tidy tree : https://observablehq.com/@d3/tree/2?intent=fork

Si ça t'intéresse, on peut tenter de collaborer sur un notebook

VIGNERON (talkcontribs)

Pourquoi pas tester D3 (ce serait logique sur Observable ;) ), par contre fondamentalement les données ne sont pas hiérarchiques donc je partirais plutôt sur une representation sous forme de graphe de force, plutôt comme ceci : https://observablehq.com/@d3/disjoint-force-directed-graph/2 ou https://observablehq.com/@d3/force-directed-graph-component

Je suis fort occupé mais cela m'intéresse beaucoup (je sens bien que j'aurai du mal seul donc autant profiter de ton expérience). Comment veux-tu procéder ?

Reply to "Graphe et Observable"
Xavier Dengra (talkcontribs)

Hi, @PAC2! Happy New Year. I was amazed that someone created a tool like WICA, but it seems that even after your update on 31 Dec, it still does not work (at least for me).

I checked the URL:

But I get the following error:

chart = RuntimeError: NetworkError when attempting to fetch resource.

RuntimeError: NetworkError when attempting to fetch resource.

Xavier Dengra (talkcontribs)

I just realised that in my Safari mobile browser it worked, but only when I entered my username as Xavier_Dengra. In Firefox (desktop) it does not work even if I try with the _.

Maybe is it a good idea to look at it? Have you talked or thought to incorporate this tool into Humanwiki or as a tool in Toolforge? I've discussed with some fellow wikipedians that our biggest struggle is to find the right tool in such a vast ocean of external repositories. Best and thank you in advance, I hope that my feedback is useful!

PAC2 (talkcontribs)

Thanks for your feedback.

I don't know why in some cases I've got network errors. It's kind of erratic and I'm not a specialist of that kind of problem.

I've tried your profile with the English Wikipedia and it seems to work : https://observablehq.com/@pac02/look-at-your-list-of-created-articles-through-wikidata?username=Xavier_Dengra&wikipedia=en.wikipedia.org&lang=en&claim=P31.

I like developing using Observable because it's easy and I can do it from my mobile phone. Maybe one day I'll find the time and motivation to switch to Quarto instead of Observable and push the tool to Toolforge.

Xavier Dengra (talkcontribs)

Thank you very much for your quick reply! Do not worry, it's still good to know that it works in some cases and in some search engines. It seems to work very well in the mobile, indeed.However, the link you sent me it does not work either in Chrome or Firefox, only in Safari via my iPhone.

Best of luck with future additions in the tool. We really need tech users that provide us tools like this.

If you want some suggestions or ideas, one example is the fact that Humanwiki does not allow to combine language+country filters, therefore finding the % of women articles of Andorra written in Catalan language is still not an option. It would be great even to be able to refine data searches to see how many women biographies (or scientific biographies, for example) an exact user has written from the total of biographies of a specific municipality. This kind of searches would be great for some purposes such as organizing local edit-a-thons or promoting better targeted GLAM events and partnerships.

Cheers!

PAC2 (talkcontribs)

"However, the link you sent me it does not work either in Chrome or Firefox, only in Safari via my iPhone."

Are you sure you don't have an ad blocker extension which could stop the scripts?

"combine language+country filters, therefore finding the % of women articles of Andorra written in Catalan language"

I guess it's possible to compute it using a SPARQL query for "small wikipedia versions". I would try to write the query and run it using qlever.cs.uni-freiburg.de.

If you're interested in gender statistics, you may have a look at https://observablehq.com/@pac02/user-level-gender-statistics-for-wikipedia?username=PAC2&wikipedia=fr.wikipedia.org and https://observablehq.com/@pac02/explore-gender-diversity-in-a-single-wikipedia-article

Xavier Dengra (talkcontribs)

Both links are so great! They will be amazing when teaching my students within the wikiproject about food science topics, in case they want to include some gender perspective when editing broader, conceptual articles.

Reply to "WICA is broken"
Framawiki (talkcontribs)
PAC2 (talkcontribs)

OK pour la fusion

Reply to "Fusion"

Automated report of empty item: Q117166187

1
Dexbot (talkcontribs)

Hello, an item that you have edited (and you are the only non-bot editor) is considered empty and will be deleted in 72 hours if it doesn't improve. Your automated cleaner.

Reply to "Automated report of empty item: Q117166187"

Great job on the analytics notebooks!

1
Wil540 art (talkcontribs)

Hello @PAC2, great job on the analytics notebooks here. I'd been curious about the gender breakdown of the articles I've started. Great to see visual representations of the editing! Very useful and interesting tools you've created, thank you!


One small issue I have is that my username has a space so only half of the notebooks worked for me and I would receive "RuntimeError: Failed to fetch". (I wish I never put a space in my username). I tried it with an underscore and I could not get anything to work. The first one and the last three that used the xTools API did not work for me. Is there a search trick I can use to get it to work for me? Thank you again! -

Reply to "Great job on the analytics notebooks!"

una pregunta sobre el inspector gap

1
Tiputini (talkcontribs)

Estoy preparándome para un congreso d'economía feminista y estoy revisando la entrada en inglés: Feminist economics con el Gender Diversity Inspector y me da el resultado siguiente: https://observablehq.com/@pac02/gender-diversity-inspector?wikipedia=en.wikipedia.org&article=feminist%20economics&startyear=1700

Hay alguna cosa que no entiendo, porque cuando hago una lectura atenta del articulo no sé encontrar los 122 nombres masculinos. ¿Qué estoy haciendo mal? ¿Qué es lo que no entiendo?

NOTA: Te escribo en español poque he visto que tienes un nivel avanzado de este idioma, si no es así, por favor avísame y te envio la pregunta en inglés.

Reply to "una pregunta sobre el inspector gap"
MargaretRDonald (talkcontribs)

Fails to work for user:MargaretRDonald, language: en. I look forward to its working. MargaretRDonald (talk) 21:28, 22 August 2022 (UTC)

PAC2 (talkcontribs)

Thanks a lot for your feedback.

It seems that you have created thousands of items. My tool is not robust enough for so many items. I'm sorry for that.

MargaretRDonald (talkcontribs)

Yes, I have. But so have many others: It's the nature of wikidata... (Good luck with your tool) MargaretRDonald (talk) 21:49, 6 September 2022 (UTC)

Reply to "Wikidata's list of created items"

Adaptation des noms des occupations

3
Jérémy-Günther-Heinz Jähnick (talkcontribs)

C'est une belle avancée. Il y a juste un point à régler : c'est au niveau de Wikidata Infobox pour les personnes sur Wikimedia Commons qui reprend le libellé de l'élément au lieu de choisir la forme masculine ou féminine.

PAC2 (talkcontribs)

Oui merci pour le message. Je vais me pencher sur l'infobox Commons. Le fait qu'elle utilise le libellé implique qu'avant la nouvelle prise de décision, les occupations des femmes apparaissaient sous leur forme masculine ? Si c'est le cas, l'utilisation de libellés neutres serait un progrès. Mais on peut faire mieux en utilisant P2521 et P3321. PAC2 (talk) 20:36, 30 August 2022 (UTC)

Jérémy-Günther-Heinz Jähnick (talkcontribs)

Ça vient d'être mis à jour, cf [[:c:Category:Louise Michel]]. Oui, depuis quatre ans et demi que ce programme (désormais utilisé quatre millions de fois) existe, les seuls libellés étaient masculins. Sur la page de discussion de la Rfc, je signale le cas des fonctions (P39) qui avec Wikidata Infobox devraient suivre le même chemin, voir avec [[:c:Category:Sophie Granato-Bricout]] où ses deux fonctions apparaissent ce jour au masculin (puisque le libellé est pris, alors que nous avons à notre disposition une forme féminine du libellé). Cordialement.

Reply to "Adaptation des noms des occupations"

Wikidata_talk:Tools#New_tool_needed

4
Richard Arthur Norton (1958- ) (talkcontribs)

I moved the conversation to Tools, I think the correct place.

PAC2 (talkcontribs)
Richard Arthur Norton (1958- ) (talkcontribs)

I moved it back.

PAC2 (talkcontribs)

Thank you

Reply to "Wikidata_talk:Tools#New_tool_needed"

Gentilé et élément sur les peuples

2
VIGNERON (talkcontribs)

Salut,

Je réfléchis à la remarque sur la RfC par rapport à Allemands(42884) et je ne suis pas sûr que ce soit vraiment un cas similaire où la RfC devrait s'appliquer. Ces éléments ne sont pas des gentilés formellement mais le peuple. On n'est pas sur un individu (comme pour les métiers/occupations/fonctions) mais sur un groupe de personne. « Allemands ou Allemandes » serait vraiment étrange ici et plutôt incorrect, à la limite ou pourrait écrire « Allemands et Allemandes » qui serait plus juste mais toujours bizarre...

PAC2 (talkcontribs)

Oui je me pose la même question. L'enjeu est un peu différent.

Reply to "Gentilé et élément sur les peuples"