Difference between revisions of "Lausanne 2024"

Revision as of 10:07, 27 March 2024

This page summarizes the plans for the Spring Hackathon 2024.

Date & Venue

From 28 April to 5 May 2024 in Lausanne.

House is located at Route de Lausanne 107, Bourg-en-Lavaux [1].

Getting there by plane

Landing at Geneva airport, there is a train station next to the arrivals. All trains go to Geneva Central, then most continue onwards to Lausanne (about one every 15').

Getting there by train

From Geneva, Geneva airport, or Zurich: go to Lausanne and then take a local train to Cully or Villette (the house will be 10-15' away along the main road). Train schedules are here. I recommend downloading the most excellent SBB app Android iOS. Trains in Switzerland do not require advance booking.

Logistics

DO NOT FORGET TO BRING AN EXTENSION CORD (and an adapter if you are not joining from mainland Europe).

Goals

The main goal of the hackahton is to focus on

Discussions

Hotspot

Discuss Hotspot permanent (across SD update) storage
Discuss Hotspot maintenance policy: when/under which circumstances do we stop maintaining a version/deployment/HW? Whats our limit on client support (browser version for instance)?
What is/are the perfect(s) hotspot(s) hardware configuration?
Should we better serve companies (vs NGOs / Foundations)?
Should we better serve preppers (what is the outcome of the custom prepper image offering?)?
Should we sell all-in-one hotspot configurations? What is it? hardware + ZIMs? virtual machine? continuous updates?

Catalog management

Discuss relevance of per-scraper Tag in ZIM metadata
Have Content Team present Tag/Category strategy (including i18n issue)
Discuss availability of content: once we've started to provide a content, do we consider we have to do our best to continue to provide it and update it on a regular basis
Challenge the metadata limit: is 30 chars for Title too short?

Scraping

Merge zimit / warc2zim? Merge youtube / ted ? Merge all python scrapers ?
How tolerant are we with item failures in scraper? In other words, is it preferable to produce a ZIM with only 99% of source content because the scraper fails to process 1% of them, or do we want to target 100% or nothing? mwoffliner is very strict, one failing article and the scraper stops. zimit is very permissive, there is no limit in number of failed pages. iFixit has a middle ground, allowing to configure the percentage of failed items.

Cross-projects

Discuss Support Policy: document existing non-written or loosely written: apple, etc and define/document for public services (library.kiwix.org)

Meta/org

Presentation of last boards slidewares and outcomes (redacted from confidential information if any)
Retrospective last year
Roadmap for 2024, for the next 2 years, 5 years ; post-mortem
Discuss possibility to have updates of Christina's UI work on the weekly (via Stephane)

Achievements

Agenda

From Friday 3rd to Sunday 5th there is also the Wikimedia Hackathon in Tallinn.

Attendees

Kiwix

Reg
Kelson
MGautier
Jaifroid
Levon
Travis
Benoît
Stephane

Budget

Hosting:
F&B:
Travel:

@@ Line 33: / Line 33: @@
 *Have Content Team present Tag/Category strategy (including i18n issue)
 *Discuss availability of content: once we've started to provide a content, do we consider we have to do our best to continue to provide it and update it on a regular basis
+*Challenge the metadata limit: is 30 chars for Title too short?
 ==== Scraping ====
@@ Line 46: / Line 47: @@
 *Retrospective last year
 *Roadmap for 2024, for the next 2 years, 5 years ; post-mortem
+*Discuss possibility to have updates of Christina's UI work on the weekly (via Stephane)
 == Achievements ==