We’ve started a trial of Slack and that’s getting a lot of momentum in the company. So much so that we’re considering upgrading to a paid account (primarily to get more history).
We would also like to integrate chatops and customer live chat into our production Meteor app.
In terms of ease of integration into our stack, Rocket Chat would seem to be a better fit, but clearly Slack has huge mindshare. Even in the general Meteor community, I frequently see “project x’s Slack channel” being referred to.
Anyway, I installed Rocket Chat on my laptop just to get a feel for what it could offer - and I must say I’m really impressed. It offers everything we’re currently using in Slack and a whole lot more. Sure, there are deficiencies in the docs, but it’s not yet at 1.0. The livechat integration is particularly impressive.
We’re shortly going to do a company-wide of trial of Rocket Chat to see what other people think, but I know it’s going to be a hard sell (Slack mindshare again). Cost is not really the issue - I suspect the in-house hosting/management of Rocket Chat may even make it more expensive over the first year.
Can anyone tell me if there’s some overwhelming reason to choose Slack over Rocket Chat, or even better, if there’s some overwhelming reason to choose Rocket Chat over Slack?
Rocket.Chat is quite easy to manage, but is also resource hungry. You’ll need at least 1 GB of RAM if the database sits on the same server for it to run smoothly. It has its problems, but the development team is very active and happily accepts pull requests, plus the code is very straightforward to modify. Overall for me the only reason to use Slack over Rocket Chat is audio/video calling, which I couldn’t really get to work using Rocket Chat (but probably there is a way to get them to work).
Another advantage of Slack over Rocket Chat is its search capabilities, I find Rocket Chat’s search featur quite slow and unreliable at times, but I don’t use it often so it’s not a big deal.
Still, I think Slack will be much more expensive to use than Rocket Chat, especially if you have more than few users, because you pay per user. So it will be about $80 per month for 10 users if you pay monthly. For Rocket Chat it’s pretty much fixed cost for you (and it can run on a $10 DO droplet just fine) no matter how many people use it (unless it’s hundreds).
We have a completely separate 3-server MongoDB replica set, so I’m assuming that will cope (current activity is quite low).
It will be interesting to see what video chat is like when we install it properly - I was able to start a one-side conversation with myself, but that was clearly never going to work!
Good point about search - I haven’t tried that - thanks for the heads up.
The other thing to consider is how much developer time it will take to setup/manage rocket chat. If you’re paying someone at the company $65/hr and they spend 15 hours setting it up and managing it through out the year, you’re spending $1095/yr (DO for $120 + $975 labor) for rocket chat instead of the $895/yr for slack (plus you can use it right away and you’re not liable for managing it or responsible for when bugs pop up). My guess is setting up and managing rocket chat throughout the year will suck out more than 15 hours (e.g. 2 workdays) in labor per year.
@a.com I do manage a Rocket Chat instance for more than a year now and I’ve only spent few hours at the beginning setting it up (modifying the look to match the style I wanted, replacing the logo, etc.). After that it didn’t require almost any maintenance whatsoever (only if I wanted to update to the newest version, which was as easy as doing a git pull then restarting the server).
I guess even if you save $100 a year using rocket chat, is it really worth worrying about it over just paying slack and moving on to work you actually get paid for… but each to their own.
Of course that depends on who you are and what are your requirements, but if you are a software house and already do manage servers and Meteor apps, deploying yet another one is not really a big hassle. It all boils down to the number of active users, if its only few then I would go with Slack, but if you have dozens of users then suddenly Slack becomes quite expensive.
I think we’re working on about $4k for an annual Slack cost. We have our own servers and devops, so I don’t think it’s likely to be a concern about in-house costs. As I said earlier, I’m expecting broadly similar costs with either.
So, I’m more concerned about features and usability.
In that case I think Slack is still a better solution just because it’s backed by a company that actively works on it and has the capacity to work on the UX. Rocket.Chat can be a bit rough around the edges UX wise, which for me is not a big deal, but for less technical oriented people it can be.
Discord! I just recently tried it (to join the Reactiflux community to hang with fellow React devs), and I love it. Give it a shot. Discord has the advantage of having live voice channels which can be handy at times.
I’ll have to jump in and also back Discord here. Our dev team started with Slack and as voice became more important now moved to Discord. Its voice features is in another league, always clear channels even in regions where internet connection fails on other apps (google hangouts, slack voice). Overall good experience - it beats Slack. (Where the cost doesn’t matter)
Have not tried Rocket.chat but am looking at implementing it as a chat server for app users through api hopefully in the near future. (Where the cost does matter)
We have already decided to go with Rocket Chat for our business. But we are facing issue with the speed. We need some changes to the look and feel so we took github latest version and working on it. Now when we are trying to start using “meteor npm start”, its hanging for hours together and not able to complete the linking step. Read somewhere that Livechat package is the biggest and can be removed for better speed. Can anyone help me suggest the best way to speed up the build step in development? Any suggestion on hardware requirement also will help us.
