Title of session
Google Meet, Slack, Zoom, YouTube… and other Bugs in Wikimedia Movement - getting Ready for Emergency Exits 🌈
Date and time and Where
- 4th May, 2024 at 4PM (#MayTheFourth)
- Ballroom 1
Cowsay of the session (optional)
Session description:
- Understanding concerns about specific massively-popular collaboration tools (Google Meet, Slack, Zoom, YouTube…)
- Quick overview about possible solutions (and pitfalls) thanks to the Free Software movement
- Call for Action: documenting the current notable software adopted in Wikimedia Organizations (Success cases, migrations, etc.)
- What has been done / success cases / failures (e.g. Wikimedia Jitsi)
- Next steps, including:
- Username for contact: Valerio_Bozzolan
- Session duration (25 or 50 min): 25 minutes
- Session type (presentation, workshop, discussion, etc.): presentation + discussion
- Language of session (English, Arabic, etc.): English
- Prerequisites (some Python, etc.): Some little interest about Free Software / Open Source
- Any other details to share?:
- Interested? Add your username below:
- ...
- ---
Etherpad:
https://etherpad.wikimedia.org/p/wmh2024-Google_Slack_Zoom_YouTube_bugs_in_Movement
Notes from session:
Session Name
Date and time:
4th May, 2024 at 4PM (#MayTheFourth)
Room:
Ballroom 1 \o/
Relevant links
- Phabricator task:
https://phabricator.wikimedia.org/T364081
- Session slides: We test in production.
live stream / video conference
https://meet.jit.si/GoogleMeetSlackZoomYouTubeWikimediaAAAAAAAAAAH
Presenter
[[User:Valerio Bozzolan]]
Participants
- Andrew Lih
- debt
- Jelto
- Laurentius
- Nemo_bis
- Samwilson
- Aklapper
- Tom
- Brennen
- Marado
- Waldir
Notes
MAY THE FOURTH!!!!
Valerio's introduction
We're all here because we're WMF staff and volunteers, we have different roles and have our own decisions. but if we're all here talking about this - the impact is slow
Valerio has been invoved in online wiki events and identified various problems.
3 years ago, talked about Present and Future (2021) - also submited session at 2024 Wikimania: https://wikimania.wikimedia.org/wiki/2021:Present_and_future
We can mention a few goals:
- Respect copyright including copyleft licenses' requirements on attribution and derivatives
- Avoid DRM - digital rights managment
- Creative Commons licenses may require that material be downloadable, but Google/YouTube may discourage or disallow the download of videos.
- Respect privacy: minimise the collection of personal data. When sensitive personal data like gender or sexual orientation are asked from participants, clarify for what purpose. The software used for Wikimania 2022 even used Google Analytics while the Italian privacy authority had just ruled it illegal.
New user group: https://meta.wikimedia.org/wiki/Wikimedians_for_software_freedom . We already have some rules such as not loading third-party proprietary software in Wikimedia projects; we should not force people to become Google customers in order to participate in Wikimedia activities. Ideally Wikimedia Foundation would be proactive in defending user freedom; currently individuals feel they constantly need to lobby and educate staff.
https://meta.wikimedia.org/wiki/FLOSS-Exchange has been augmented with a FLOSS-exchange matrix to summarise the most common choices.
- noteable software that can be adopted. and a list of software used by WMF
- google documents is easy to share internal documents - issue is everybody uses gdocs. can they be extended to the community easily asking for feedback, etc?
- being a google subscriber is a needful thing if the community wants to view the documents - this is wrong
- we shouldn't expect that volunteers need to use google
- can we invest $ in using something else?
- organisation-wide changes are difficult but individual action also helps, for example copy-and-pasting documents into Cryptpad and sharing a link to that rather than
- why can't we use LimeSurvey to pass documents back and forth (from WMF to community and back)
- nextCloud/ownCloud - a bit more expensive, helps avoid legacy Microsoft code
- can also not be self hosted - libre SaaS offered by various suppliers
- Twitter - has only a few Wiki organizations
- mastodon has 8
- we should ask why all our info is in Google Documents
- ask for budget for finding free solutions, rely on volunteers
- wikimedia Meet
- now archived
- didn't have enough knowledge to keep it up, didn't contact company behind the software.
- didn't give WMF / community the support needed to make it success
- WMF went to Google because of issues with Wikimedia Meet
A lot of software isn't free - and we aren't investing in free / open software. Can't compare the level of service and software quality of a paid-for fully managed software service (especially if proprietary) with gratis-hosted software.
Wikimedians for software exchange
- can we ask our organization how much we're paying for all the various software
- how much did Wikimedia spend on the entire Google Suite?
- how much are we paying for windows licenses, zoom, slack?
change is hard, there is always a lot going on, a lot of friction from some people.
direction is clear. but we're stuck in this situation; will we forever require people to use Google?
Wikimedia Italia volunteer sysadmin compiled a list of all services and software in use in the organisation to facilitate transparency and to plan a complete replacement of any surviving dependency on proprietary software.
What should we do? :)
Questions and discussion
Q: using limesurvey (rather than gforms) because of earlier conversations. some of these are drop-in pieces of software. should we use jitsi instead of zoom/meet? zoom/meet is very well supported in the backend - mobile, geography, infrastructure. which ones are easy/ hard to drop in and replace with open software?
- A: The company behind Jitsi... just contacting their help - they should be happy to help us.
Q: the software used is determined by upper management and their background (FLOSS vs corp).- what they're used to using. WMF has guiding principals to use and be in open space. certain tools are to be more convenient - google meet - calendar integration, notes documents created automatically, etc.
A: toolforge and other companies have strict rules about only having free software in your repos, nothing proprietary. It's one thing to use proprietary services like YouTube to reach people who are not yet part of our community, another to force the people inside to use proprietary software.
There was a mention about Wikimedia Europe being involved in procuring free software. Nothing happened yet but it would make sense to pool resources. It's often expensive and difficult even just to find out what the needs are.
What's a possible outcome of the discussion? Do we need to collect information on the situation in different ways? Perhaps a different structure for the data, using WIkibase or something else?
Do we need more transparency on the procurement contracts? Is it really impossible to share information on what's the spending on proprietary software?
Often WMF needs more support than a basic setup of the libre alternatives appear to provide. For example, using a public instance of Cryptpad is not enough; SSO integration would be needed (as supported in https://cryptpad.org/pricing/ ).
Getting the WMF to use open software for business needs. Would the WMF really give up Google Suite? it seems that the Org doesn't care about free software for internal things, it doesn't seem to be on their radar, and out of the expeirence that management has. maybe we can make a list of why free software is better and here's why. to be realistic about the challenges and payoffs. Trying to not be pessimistic, but that is reality.
Seems that we need a lot of resources to run services on our side. to create new instances, it takes a lot of time (1-3 weeks to upgrade the Wikimedia Etherpad instance). I fear that if we buy software for free - we'd need to reduce expectations and our comfort quite a bit.
The cost of managing the software goes beyond any bills Google or others may present to WMF... Proprietary software is probably cheaper overall because it's easier, so cost is not a good motivation for progress. It would help to think of possible migrations with the highest value for effort, for example LimeSurvey for sensitive data collection. LimeSurvey can be configured to safer defaults, e.g. to protect survey participants' data from accidental sharing. Such efforts will not lead to direct cost savings.
WMF IT services team nowadays manages a software catalog which has hundreds of entries and only less than 10 % are FLOSS. Maybe it can be published if we ask nicely?