However, I don't think the future of collaboration will be Slack. It is just too expensive and quite buggy at the moment. I'm hoping there will be competition that will either make Slack better or replace Slack with a cheaper, less buggy alternative.
I was thinking about the impact of taking it away again, if Hangouts supported "channels" that people in the domain could join I think we would have a pretty seamless departure. The channels feature is about all I've remained to love after having slack for 2 years.
Because otherwise I don't really see any inherent performance issues in the problem domain that would require hardware more powerful than what we had in the early 90s. IRC, for all it's UX issues, proves exactly that.
Especially voice/video chat, holy crap Slack is ridiculous. The first time I started a Slack voice chat, I actually had time to disable my extensions and log into my router before I figured that no, it's not a problem on my end, it's just that slow and the damn thing loaded.
Slack is just really bad in terms of performance.
Yet, Slack manages to have lag between pressing Enter and showing a message. It's baffling to me how almost nobody doesn't seem to have a problem with that. Miranda IM worked much faster when connected to 6 networks... on a machine that was 10x slower than my phone.
Like, come on.
Can't blame the entire OS for a single program being badly written.
- Scrolling up to something that happened fairly far back in history can feel really awkward and broken, image loads and ajax fetches can make the scrollbar behaviour really janky (to be fair, I just tried this and it was better than I remember, although still far from perfect)
- Switching rooms takes about a second, even if I was just in that room, which while not horrible, is still a little laggy for something that should be nearly instant if I have the room's content cached.
Overall while it's gotten better, the "native" Slack app still feels very much like a webpage with OS chrome around it (which is what it is, but other apps like VS Code, Atom, and Discord don't give this impression nearly as much)
This is all on the latest Macbook Pro upgraded as much as possible in every respect save the video card.
Granted that it's an older model, but why does an IRC chat program need 10 minutes to run? What is it doing?
I've a high end workstation on windows 10, and a gigabit jnternet connection. There's no excuse for the delays we see. Another major issue for me at least is their editing system. If you send a message, and edit it "too soon", the UI pretends to edit it, and some seconds later comes back and gives you a modal dialog saying it failed to apply the edit, and undoes your edit. The delay is _huge_ - 10-20 seconds.
Don’t people “just not respond” even more with email? Maybe I’m an outlier but I probably reply to less than 50% of things that come into my email.
And if you need a realtime convo, come tap me on the shoulder or schedule a time to talk. That’s the main benefit of slack (and email): async communication. Async is good.
If I had a 50% response rate for emails that clearly need a response, that would severely impact my ability to do my job and would likely result in my manager taking steps to help me improve that.
I meant for my real email.
As for my manager taking steps to help me improve that, he’s even worse. Or just has a stronger things-that-are-important filter than I do for Slack. Email is for papertrailing things that were previously agreed upon in person or on slack.
The problem is not 'Slack' or 'Email' - it's the wasted communications energy at all these companies.
Most companies have not figured this out, and think that tech will solve if for them. Not so much.
I feel like slack (and competitors) have a lot of room to grow still.
Especially when timezones are involved
Given that they steadfastly refuse to implement an ignore user feature while at the same time they've implemented mandatory channels, of course the signal-to-noise ratio falls and people just ignore Slack. I know I do.
If you have to ignore your colleagues you have far worse problems than Slack not having an "ignore user" feature.
Slack's been pretty clear that they don't want their product to be used for this use case, and have even nudged some larger "casual" communities and open-source projects off. So it's not surprising that they plan their feature roadmap according to what (they think) will be useful to businesses.
Try working at a large enough company. Or, try using any number of bots. I'm in a few mandatory channels of zero relevance (ok, mute) and a few of high relevance that are punctuated with lots of bot chatter I'd like to ignore. And, sure, there are a few coworkers I'd prefer to ignore.
Even IRC will let you manage notifications with more granularity than Slack will.
I do consulting for a bunch of startups and they often invite me to their slack. I'm pretty fascinated by the hugely different "slack cultures" at each company.
One company I worked for would militantly tell people to never use @all, @channel, etc, because it was distracting (THANK YOU!!!!) and another client will literally do "@channel, I'm WFH today" in the all company channel. I've had 2-3 hour blocks where I'll have 6-7 notifications that aren't for me.
If you ask me, Slack is the best way to distract your employees from doing work. It's like putting everyone in ~1 hour of meetings a day for no good reason.
I could see something similar to Slack as a free, open source project. It could possibly develop out of something like Signal [https://signal.org/]. Making and supporting an open source solution can be a viable business too (e.g. CoreOS, Docker). I doubt that it would kill Slack either. It should be interesting to watch, and will benefit the end users as a whole through an improved Slack and/or a similar free option.
There are PLENTY of other chat protocols other than IRC. Not sure why people always make the comparison. IRC would be a poor choice for a workplace.
XMPP is a pretty good alternative with stuff like movim that have a friendly modern looking client.
Slack being hacked is generally just not something most business need in their threat model. Same goes for Gmail, GitHub, and most other business services.
I can imagine people lackadaisically sharing things like ssh keys / passwords on it. I certainly have.
Any time something is hacked where the participants thought they had privacy is a potential disaster. Not sure how you can say otherwise.