Talk:Community Resources and Partnerships/India Rapid Project/Lingua Libre SignIt 2026

Timeline

Timeline

Contributor	Workdays (est.)	Github issue	Task description
Kabir Singh	2 days	#104	Add other sign languages (Armenian, Indonesian) [critical for project internationalization and growth]
Kabir Singh	3 days	#87	Popup>Settings with persistent memory on all browsers [critical bug]
Kabir Singh	2 days	#66	Change SPARQL query to Commons APIs for long term resilience
Kabir Singh	4 days	#1,#9	Migrate away from oojs and css bug (deep js and CSS revamp to avoid occasional UI collision)
Kabir Singh	3 days		Adapt code so it passes Chrome web store security requirements
Kabir Singh	1 days		Other smaller tickets
Hugo Lopez	3 days	#85	Web store registration, management and publication on Chrome and Firefox web extension stores
Hugo Lopez	1 day	#46	Update github documentation
Hugo Lopez	2 days		Printable e-flyers design and call for translation
Hugo Lopez	1 day		Mailing announcement to ~20 Wikimedia Chapters and few identified deaf community organisations.
Hugo Lopez	1 day		Project administration and final report
	15+8 workdays

Yug (talk) 12:00, 30 October 2025 (UTC)

Follow up on your project proposal

Dear @GonFreeaks,

Hope you are well. Thank you for applying for India rapid project. Since, your project was worked on during GSoC 2024 (T361550) and you've proposed it again for GSoC 2026 (Phase 1 proposal). Please provide detailed responses to the following questions and clarify:

1. Since, Wikimedia France (WMFR), which already maintains Lingua Libre and has an existing budget for this project? Please provide a clear explanation of why WMFR cannot or is not supporting this project?
2. What exactly was delivered in GSoC 2024 vs what remains unfinished? Can you provide specific GitHub issues/PRs?
3. Why are you requesting India Rapid project funding now when you have already proposed this same work for GSoC 2026?
4. Have you announced this project on Wikimedia mailing lists asking for volunteer contributors? Why or why not?
5. The components like updating GitHub documentation, creating e-flyers, calling for translations, and sending emails. Why can't these promotional/administrative tasks be done by volunteers as part of normal project outreach?
6. What is the current status of the codebase, and what specific technical deliverables remain to reach stability?

Looking forward to your response by 30th Nov, 2025. Thanks PDas (WMF) (talk) 09:31, 25 November 2025 (UTC)

Dear @PDas (WMF),

Hope you are doing fine as well. Here are the responses :

1. WMFR focuses its resources on Lingua Libre and have provided no funding for SignIt according to Yug, who lead the project since 2025.
2. You may scroll down this list of PRs that I made during GSoC 2024. Here is what remains unfinished :
   • Change SPARQL query to Commons APIs for long term resilience : SPARQL queries are slow when compared to Wikimedia Commons API. Not to mention they are hard to grasp for new contributors.
   • Add other sign languages : We have around 1211 Armenian sign language videos and 26 Indonesian sign language videos that need to be integrated in the extension for i18n. Also , they can only be accessed via commons API and not SPARQL. We can say that this and the above issue go hand in hand.
   • Popup>Settings with persistent memory on all browsers : As mentioned above, it is a bug that prevents users' settings to persist across tabs or page reload.
   • Migrate away from oojs and css bug (deep js and CSS revamp to avoid occasional UI collision) : huge revamp to prevent vanilla CSS and library conflicts. At the initial time of development it was not an issue but later the OOjs library itself went under maintenance and preferred the new code written in Codex as mentioned here in the README. While there are other issues as well , they are not high priority that question the stability and working of the extension. I hope this answers the last question as well.
3. I had no awareness about the GSoC 2026 application. I manage the technical side, while application-related matters are handled by the project owner, Yug. Same goes for question 4 and 5.

GonFreeaks (talk) 05:30, 29 November 2025 (UTC)

Hello @PDas (WMF), as the former single project lead from 2020 to 2024 I can confirm what GonFreeaks shared with you. Following your numbering, there are additional information or confirmation :

1. SignIt is not funded by WMFR. WMFR kick started the project in 2019. I took over SignIt in 2020, looking for support from WMFR/Lingua Libre and occasional funding when necessary, applying the mixed approach similar to Lingua Libre. But after 5 years with 600hours invested and the largest bottlenecks stalling due to lack of support, WMFR's director confirmed at the Wikimania 2025 that « available funding by the French Ministry of Culture and dedicated to languages are forwarded to the recording system itself (Lingua Libre) alone ». SignIt has no WM sponsor. Acknowledging this policy we recently separated SignIt from WMFR and Lingua Libre and we are looking for WM sponsor(s) and grants necessary to unlock SignIt.
2. In 2024, Google phased out web extensions v2. Our GSOC24 therefor focused on recoding and upgrading SignIt into v3. We (mostly GonFreeaks but me as well) successfully upgraded the web extension code and security methods. But we kept an eyes on additional tasks already listed by GonFreeaks above (2).
3. GSOC26 is not yet open, we haven't applied to it. So I'm not sure what you are referring to.
4. I/We haven't announced SignIt on that particular Wikimedia mailing lists, we could, yes. It would add one more vector. We also think we most need to ship to "market" with 2 sign languages to demonstrate SignIt as a multilingual toolchain.
5. Outreach efforts: GitHub documentation is done by the developers during the sprint. Creating e-flyers, calling for translations, and sending emails could be done as volunteer over a longer period of time, yes. It's slower but it's ok yes. I think you may adjust a grant offer accordingly since the proposal is itemized.
6. See point (2) by GonFreeaks above : the plugin works with manifest v3 code... but it needs consolidation (API, Memory, Move away from buggy OOJS) for stability, and expansion (add Armenian) for a more meaningful demonstration.

Ok, this should give you the larger and more detailed picture. Globally, those remaining issues are the heavy issues which we were unable to solve via volunteer approach alone, and it's why we here level up to grant request.

Thank for reviewing this grant application. Yug (talk) 01:29, 3 December 2025 (UTC)

Recommendation for non-funding

Hello @GonFreeaks, thank you for submitting this India Rapid Project application. Based on the overall technical review of the project plan and proposed implementation, as well as the reasons outlined below, we do not recommend this proposal for funding:

• The India Rapid Project supports tech projects that demonstrate clear community demand and alignment with community needs. In this case, we were not able to identify sufficient project demand, even after explicitly requesting that it be added. The overall usability of the project also remains unclear.
• The roles and responsibilities within the project team are unclear.

Best regards, DSaroyan (WMF) (talk) 09:51, 8 December 2025 (UTC)