I think the TL;DR is:
Zoom is proprietary and many members of the community would ideally like to find a way to avoid it. JitSi and BigBlueButton are possible alternatives but we need to ensure that they are stable for large calls where not all participants can guarantee fast internet. For community calls there are very important discussions to be had, and the pragmatic approach is to use a tool we know is stable. Maybe we can have some less important calls on BigBlueButton to try it out.
Maybe we can use this thread to talk about both open meeting tools, but also other alternatives to proprietary services for collaboration. I know @hpy already maintains a list of such services. The one thing that I think we need to consider carefully is how viable these alternatives are. I feel some level of pragmatism is required, open tools will always be preferred, but if we can’t use them effectively to work towards the core GOSH goals we are shooting ourselves in the foot.
Hi Jenny, and everybody around,
Could you consider having the next call over Jitsi Meet instead of Zoom? I know you said you chose Zoom based on the number of participants. May be we can start thinking about running our own instance of Jitsi.
We have in the past tried running JitSi off our own instance. We has more problems than on JitSi meet. 8x8 is open source, it is the paid instance of jitsi made by the developers of JitSi. I think long term it may be worth investigating if we can use that as an open but supported solution.
I think using zoom for these calls is the pragmatic approach. We would rather be using open tools but we need to take things one step at a time. For these important meetings to function we need the calls to go smoothly, and zoom is the only platform we have found that does this. Sadly it is not open, when we have the time we can experiment with something like 8x8.
The main thing is that Jitsi requires more bandwidth out of every participants connection for a stable call making it not usable for large calls with many participants and varying quality of internet connections. This may improve in the future. Extinction Rebellion faced similar issues apparently (there’s a whole talk about their infrastructure if you want to geek out).
Wow! I know about BigBlueButton as an open source replacement for Zoom and friends, but didn’t know there’s a hosted service, not to mention a co-operative, that one could sign up for. +1 indeed. @bhaugen: Have you tried it? How well does it work? Any eyewitness accounts?
Even if we can’t make this work in time for the meeting on 14 July, I suggest we look into @bhaugen’s suggestion “very strongly…” Given the resources now available via the new grant, surely we can implement a more ethical solution than Zoom!
@jarancio and I discussed the possibility of a GOSH Open Hour. For an informal way for the community to catch-up and video chat, but without the fixed agenda. Maybe this would be a good way to trial some other video platforms and see how they perform. But I think we need to do a lot of testing before we run a critical meeting on another platform.
I’m a member of a group that is talking to them and will most likely start to use their hosted version, but I have tried a couple of other hosted BigBlueButton sites and they worked pretty well.
I mean, any of the hosted video meeting apps can have bandwidth and other difficulties, and I am sure BBB will not be immune. I don’t have enuf experience to say where it ranks among the contenders.
I think the TL;DR is:
Zoom is proprietary and many members of the community would ideally like to find a way to avoid it. JitSi and BigBlueButton are possible alternatives but we need to ensure that they are stable for large calls where not all participants can guarantee fast internet. For community calls there are very important discussions to be had, and the pragmatic approach is to use a tool we know is stable. Maybe we can have some less important calls on BigBlueButton to try it out.
Maybe we can use this thread to talk about both open meeting tools, but also other alternatives to proprietary services for collaboration. I know @hpy already maintains a list of such services. The one thing that I think we need to consider carefully is how viable these alternatives are. I feel some level of pragmatism is required, open tools will always be preferred, but if we can’t use them effectively to work towards the core GOSH goals we are shooting ourselves in the foot.
We’ve evaluated Jitsi in the past (public instances and self-hosted) and we can say that it didn’t hold up for large calls. I don’t think we need to spend time evaluating it further unless we hear some news that they have released significant improvements.
Happy to give BigBlueButton a go, but it’s very hard to evaluate anything for large calls without taking up a lot of people’s time, by definition.
I get the sense there is broad agreement (from any corners of the open universe) around Jitsi, and I’m wondering if we can’t hop on that bandwagon in suggesting or building use cases around.
Does any have insights into the Jitsi dev community that we could connect to? We’re a tiny fish in that big sea, but just curious.
If you have a look on their forum there are threads about performance and the discussion seems to hinge on upgrading from VP8 to VP9. There is also a GitHub issue about VP8/9. I don’t understand any of this well enough to do more than sign up to their forum and announce we also have seen issues.
I mentioned meet.coop and their hosted Big Blue Button service in that other thread, and said that Open2020 had been able to handle a fairly large number of people.
If you can’t access that, ask for admission and I or somebody will let you in.
But here’s the TL;DR version:
meet.coop is working hard on formulating its service offering in the light of hands-on piloting, and is constrained as a start-up by (human and financial) resource shortages, small-coop economics, practicalities of coop-to-coop federation and shortcomings in the available software (eg the front-end for administering bundles of rooms). Thus, the service at first will be of an early-adopter Alpha-release nature, and users will need to be ready to go through a bit of a learning process in how the coop membership works, and how the service evolves and becomes richer.
…the current server provision (Koumbit coop, Montreal) is good, so long as lots of users don’t go crazy with big rooms and lots of cameras on. As revenue comes in, load balancing across multiple servers is part of the roadmap. All good.
The BBB tech itself works, no problem, the challenges are more about the large-scale administration of the front end, as a user-facing platform and a paid-for service. A BBB room works fine (depending on browsers being used and of course broadband connections) and quite large numbers can participate (with cameras off, preferably). If an organisation just wants a virtual space to meet in - especially, in smallish teams - a BBB room will do the job (in a slightly different way than Zoom/jitsi but the learning curve isn’t too great). At this stage, with regard to meet.coop (there are other providers of BBB access too) the issue is more about wanting to contribute to the coop project per se, and engaging the challenge of putting core infrastructure in the commons, rather than in corporate hands. Being a commons-cooperative federated venture is a big part of the proposition.