Mh, I have the fear if we switch now, we might loose some community members. Additionally, you need a GitHub account, which would somehow lock out non-developers (actually a clear negative side for me).
I’d rather first ensure we have a healthy community on discourse; and then we from my side can switch to a different channel. But until discourse is fully up and running, it is too disrupting to switch I think…
So I would not switch now, that is!
Greets, Sebastian
It’s not urgent. My suggestion would be to test Gitter (or other solutions) thoroughly and then, when discuss.neos.io is launched) run a poll to see what our current Slack users would like to use.
Glitter as a public stream, so you don’t need to login to see the discussions. For me this is an important point, as we don’t have to handle the “public” archive somewhere else.
This switch can wait to after the github switch, the argument that github is only for dev, is less and less important, see http://danwin.com/2014/04/a-guide-to-using-github-for-non-developers/ so many project (author, publishing, …) start to use Github for other tasks than pure coding.
But I still lost the point, why we need to leave Slack?
Not having a history would encourage people to use it the right way: for quick chat and not decision making.
Mjello Not sure the Free model for Slack is not preferable actually. That messages older than 10000 isn’t searchable is not a big issue. Personally I rarely use search and I don’t expect to be able to find old messages. Embracing discourse more is a good thing in any case, to keep information that’s valuable for longer and make it searchable (indexable). One of the reasons why Slack works well is because it’s made very well, which I don’t get the feeling for with Gitter. Additionally I’m fairly certain I personally won’t participate much if there’s only one channel, it will simply be too much irrelevant information to channel through and likely just become a support + random channel. As reference there’s already way too much in our current Slack to keep up with, but having separate channels helps a lot in terms of prioritization.
If the question is really about the history being searchable, then we can easily export it or fetch it automatically using the outcoming webhook and Jenkins. Actually I doubt Gitter is any more searchable than slack, at least the search is simply not working from my experience trying it out just now. Couldn’t find “bacon” in our Neos chat even though it was written multiple times, but hey I found Kevin so I guess that’s better anyway? Additionally the search is cross project and you find users totally irrelevant from the project in question. Additionally it doesn’t solve the problem of being indexable AFAIK, so to achieve that we would still need to export and expose the history ourselves. I also second the argument about being developer focussed, which it does seem to be build around. It might get better with time of course, but for me Gitter is not a feasible alternative to Slack as it is now.
Regarding the limited integrations, that’s really not a problem. The incoming webhook can be used for everything you need. We just have to prioritize a little for the other 4 integrations. Writing integrations is dead simple and Jenkins is more than happy to do the work required. Also gives a lot more flexibility since many of the integrations aren’t that well made IMO.
Also agree with the timing being bad to make such a switch. Since I sound like an old grumpy man I leave you with a
TL;DR: -1 for using Gitter, +1 keep slack as free plan, +1 embrace discourse more
Kim from Slack just replied to my second email. He (or she ) wrote that they can’t provide the resources at the moment for communities as large as ours, but that they are reevaluating their nonprofit program and will get in touch with us if they make changes in our favour.
So, let’s continue using Slack as it is, but I suggest that we still go through the channels we have at some point and think about how to move more discussions to Discourse. We should simply collect examples of conversations which we think belong to Slack and other which we think should have been had in Discourse and then prepare an FAQ item where we can point to.
Gitter seems to require a github account, right? That’s a big NOT like for me in terms of building a diverse community with non coders as well. Sure you can argue that creating one account at slack or one account at github is the same, but it feels different. You create an account for code sharing (which you might never intent to do) to get into a chat. Sounds like a chat for coders only to me…
I don’t really consider this an issue because we can create accounts more or less automatically based on your neos.io profile and in the end, almost all chat solutions will require an account. No matter if developer or designer, there will always be people who don’t like having to sign up at Slack, Github, Google, Skype, Freenode, …
Well, there’s a client. And the server component doesn’t seem “done done” yet
Also it needs more than just a package of course. But wouldn’t it be cool to have a neos.io oAuth provider that one could use in custom solutions?