We abandoned hipchat because the latest app was horrible; constant signouts, lost configurations, sso issues.
Skype is also a mess right now; with chat application bugs, UX issues (why does the freakin' mute button move around) calls that can't be joined or start cyloning, participants are suddenly dropped mid call. Throw in the spying allegations ...
Hangouts is great for video but is a really clunky experience otherwise.
With everbody dropping the ball, there certainly is room for Slack to come in and take control of this space.
Seconding Discord. It rocks for real time communication and works very well in a browser, with rock solid native apps.
And they send you really friendly emails.
edit: heh. Apparently I tried posting it to HN before. Perhaps they're going a little hard on the "for gamers" branding when it could really be for anyone.
Indeed, the React community moved from Slack to Discord. Slack has made it clear they are a tool for small teams. I don't necessarily see this announcement as a change from that.
Oddly enough, Discord started out as a chat/voip tool for smaller gaming groups too, but as a side-effect of awesome engineering we've been able to handle large communities really well. Over the winter break, one community sprung up, going from no users, to about 2400 concurrents at peak, with about 600 of them in voice chat at the same time.
Also, our tech stack for the native/web client is React & our iOS app is React Native, so it was pretty awesome to see Reactiflux migrate to Discord.
We came to Discord from Hipchat... mainly because I'm championing it. The killer feature to me are the voice channels when you're dealing with remote staff. They nailed this.
However, for a company, there are a lot of missing features, so definitely take some time to evaluate it first to make sure it's a fit. For example searching chat history (something we miss a great deal, but can live without). Also, the permissions are horrendously complicated and the fact that you can't remove or lock down the "public" room is super annoying.
That being said, we're still using it and pretty happy. It probably doesn't measure up too well against Slack if you don't value the group voice channels.
Just saw the threaded discussions feature for Zulip from a thread in here. Wow, that would be killer.
Chat history search is something that's coming soon.
Although the permission system may be something that's complicated, it fits the gamer clan niche well. A lot of people actually have been requesting that we add more complex permission hierarchies.
I think your permissions are fine but the weird permission locking of the default room needs to go away. There's no reason that room should be more special than the other ones.
It'd also be nice to make it clearer whether someone can view a specific channel or not (eg see it in the list).
Oh and... IRC gateway? :) Or at least open spec on the protocol, pretty please.
We do not have any retention policy settings right now. Messages are kept until you delete the channel they're in.
Our product focus is on making VOIP the best it can for gamers/gaming communities. Definitely not huge enterprises!
However, 7000 member rooms for announcements? No sweat. We've had @everyone notifications targeted at upwards of 20,000+ users at once in a single server. I actually wrote the code that handles that. Lots of streamers, game developers and stuff actually use discord for announcements. One of the bigger ones in my recent memory was the Star Citizen PTU server, the devs used it for chat + announcements.
You could make a good amount of money if you focus on the few and relatively easy enterprise features of retention policies, sso & message export. Many large corps are sticking with hip chat but hating it because it's the only thing that kind of works.
I exaggerated a bit - they have 7656 at the moment (after being 'shut down'). It was the mostly-official place to discuss react, flux, webpack, other tech and libraries in the react ecosystem etc.
In it's peak, there were lots of channels for all the different parts and facets, including geo-oriented channels as well.
> 8000 people? Is it like a forum with many threads and topics going on or is it really like an IRC chat room with 8000 people in it?
I'm a member of the UX Slack (https://www.designerhangout.co/) who have ~6k members, and it's pretty active. It's a cross between a forum and IRC, and (from a user perspective) fails pretty badly because of it :)
A major reason is Slack's limit of 10000 messages per free Slack. So conversations get lost within 2-3 days, there's no way to find them unless you've starred them. With no threading discussions don't happen well - there's a lot of noise (like IRC) which can't be filtered out.
Summary: No way to organise/archive knowledge (same q's again and again); noise of IRC interspersed w discussions, Slack's hard limit of visible messages.
If you're not paying they won't be too keen. Slack just doesn't work very well with that many people and channels - the app just grind to a halt on my Mac.
If you're paying 35k/year however, it would be interesting to see how they react to that...
+1. To my knowledge Discord is and will always be intended for gamers for the forseeable future. They don't have the infrastructure to support enterprise expectations like LDAP/SSO, dedicated account managers, support hotlines, guaranteed SLAs (although I know their tech team is highly competent), etc.
I just took a look at Discord and it's really funny how it seems more of a Slack/Skype alternative than a TeamSpeak/Mumble alternative, even though they try to target the gaming market.
For example, there is no way to self-host a Discord server. While that might be OK for Slack/Skype users, some TeamSpeak/Mumble users might not like that.
It hasn't stopped it from becoming widely popular in the gaming community though. All gaming-related projects I follow are on Discord. This includes streamers and guilds.
My thoughts exactly. Since adopting Slack, the only time we use anything else is Hangouts for videoconferencing. Even using the attached chat in the hangout feels clunkier than using Slack chat during a videoconference.
When you really feel it is when Slack goes down and you have to go back to using something else. There are just so many little details in the presentation that make a text chat in Slack feel right. Adding videoconferencing to their platform is the logical next step.
In an ideal world they'll find a way to improve on video chat similarly! I'd be plenty happy though with an experience comparable to hangouts, but without leaving Slack.
Zoom (zoom.us) is really good for video conferences. It's got a slightly clunky interface, but the excellent audio quality and graceful fallback are clutch.
Skype is absolutely a disaster, but it's the only video chat solution I've seen that handles audio well at all (things like echo detection). The super-expensive fancy solutions like Cisco's offering are probably better, but I haven't tried. All the web-based tools seem to be different UIs on top of the same fundamental technology. I think Slack is nice and they generally seem to understand how to make software for humans, so I really hope they pull this off.
You would think that... But at $10k per room for video conferencing, it was a disaster half the time, phantom calls, all kind of problems. I can't tell you how many times we just put my laptop at the end of the table and fired up Skype or a Google Hangout. We've pretty much abandoned Skype for anything and use Hangouts when we need group video, otherwise it's just FaceTime.
I'll be happy to see how Slack integrates voice/video. If the quality is 80% of Hangouts and the experience is better, I'll take it.
I found the usability of Hipchat pretty subpar vs Slack as well. Things like editing or deleting messages, extensibility with integration, alerts and syncing across devices..
Snippets!? Come on! These should be basic, foundation features.
Hangouts is great for video if you don't need things like readable smaller text on presentations or detail in design reviews when casting your screen. Our designers hate it and our customers are always asking things like "what's that button say? I can't read it". We used to use BlueJeans which had much better video quality, but the in-room systems were an order of magnitude more expensive than chromebox for meetings.
That was my first thought as well - in light of recent discussions here on HN concerning just how terrible Skype seems to be these days (I wouldn't know) it feels like an obvious next step for Slack. Good for them.
I wish slack would concentrate on making the app more productive for teams to work together remotely. We can type fast enough, in fact, hackernews tells me I'm typing too fast.
The problem is organizing and collecting the information and ensuring it is useful later, rather than just during the 5 minutes. Let users break a chat message out into a threaded discussion for example. We already have plenty of voice and video chat options, we can just pick up a phone. Those things are ephemeral. Text is forever. Let's keep more of that so we can use it to further the business objectives.
I like to treat Slack like e-mail. I'm also a fan of GTD(Getting Things Done), and a pretty big part of that (and things like Inbox Zero) is that _you don't leave important stuff in your inbox_.
If there's some info in slack that will be valuable for more than 15 minutes, it's likely that it belongs somewhere else!
When dealing with bugs, I copy the chat (well, a link to the chat) into a card in Trello. Same for basically any request that requires me to do something. I do use Slack reminders for calendar-based events.
I treat Slack like I treat talking to a person in real life. If I don't write it down, I'll likely forget it, so I better write it somewhere (not in Slack) quickly. You can probably accomplish some of this with pins inside Slack, but I've never accomplished it.
I think that this is one of those intractable problems where, because people work differently together and in individual workflows, it's hard to solve the "long-term coordination" problem without bogging down the "short-term coordination" problem.
Anyways, text is forever, but so are newspapers on microfilm. But we still make history books.
Exactly! A channel is just like a meeting room, you can expect me to constantly follow more than one or two channels, for the others, @channel or @myusername.
A great point, but saving from Slack isn't the most straightforward. It's not as brainless as "Select text & hit 'Save to Evernote'" which I do in my web browsers. I've resorted to screenshots a few times, to keep the richness of the messages.
> If I don't write it down, I'll likely forget it, so I better write it somewhere (not in Slack) quickly.
My philosophy on this is "don't be afraid." It's all 1s and 0s to the machine when you're digital, so it's OK to dump 10 paragraphs into certain text boxes. If you want to do something, you can usually do it! A lot of programs let you paste images in directly nowadays.
If you're OK with having links to chats instead of the chat text itself, you can click the timestamps in a Slack chat to get a permalink . Sometimes I just copy the entire chat and paste it into Trello.
Which leads to "where". Most things I end up with are small tasks like bugfixes. Actionable tasks for the immediate. As much of that goes into Trello as possible. I'm not afraid of making multi-paragraph cards that copy the chat verbatim either (it's all 1s and 0s, right).
I also do the whole "screenshot" thing too. Take a screenshot and also paste it right into Trello.
Some stuff is important in the immediate but I'll throw away later. Temporary notes for things like user support. I use OS X's Notes app, which is basically notepad with tabs (I would likely use Notepad++ on Windows to accomplish this). It's all temporary notes, but still nice because it doesn't scroll up as other people start talking.
If I have something like a long-term reference, I try really hard to find a good place for it on a case-by-case basis. Usually that involves writing a text file (markdown) and putting it into Dropbox. On OS X I use LightPaper (a great little app), on Android I use Draft[1] (iOS has a thing called Drafts). Both apps are set up to point to the same folder, so I have these references on my phone too.
I also use text files for journaling what I'm trying to accomplish. For example, at work I'm working with a system that will physically mail documents back to me, so I wrote down exactly my operations so that when the docs get back to me I know what I did:
Sent two pdfs (located in Documents) on invoice `weFFEA123231`. One is Letter size instead of A4, another is A3 size.
Also sending Invoice `weFFEA123232` with just the letter-size page(`2015.pdf`).
A "user space unit" in PDF is 1/72nd of an inch.
Sometimes this log dead-ends, most recent entry:
For compressor and source maps, thing that's likely needed is to rewrite the JS compressor.
An easy solution would be
-----
I also use my calendar! I'll also just paste random stuff into my calendar for information I'll need at a certain event (for example reservation confirmation numbers).
This veered slightly off topic but tl;dr is:
Trello for tasks, Notes for temp notes, LightPaper for reference+journaling, Calendar for time-sensitive information.
not sure if they changed course, but slack has been talking about becoming the hub for all business comms and files and eating into ms office/google apps
Threaded discussion support is basically the thing that's been holding Slack back from complete world domination, and for some reason they're taking forever to do it:
The fact that they are ignoring this is a very bad sign. It's way easier to do and hasn't been done well and instead they are doing something really hard that people can already do easily outside the system. They've shifted their focus from their customers needs to beating the competition and it will their downfall.
Threaded discussion support is a major feature their competition touts, so arguably by adding it Slack would be crushing their competitors and serving their customers at the same time.
> Let users break a chat message out into a threaded discussion for example.
Threaded chat would be awesome! In 'noisy' public Slacks the conversations are impossible to follow. Sure IRC had the same problem, but it wasn't used as a general replacement for forums in the way Slack is.
Lots of different instant messengers like skype. Google chat, facetime, just the regular old phone. Video conferencing tools. Even screen sharing systems like gotomeeting. It's a solved problem mostly, but hard to do technically, so that's why it concerns me that Slack is trying to do it when there are lots of things no one does well and slack isn't addressing those. This is a bad sign for slack, IMO.
I do know about Skype, Google Hangouts, and videoconferencing tools like Webex, GotoMeeting, etc. And have used enterprise videoconferencing tools in companies where I worked earlier, and Skype and Google Hangouts. But there are issues with both of the latter at times (e.g the recent thread about Skype issues on Linux), hence my question, which was more about other (new) ones, and only for voice and video chat, not screen sharing or text chat (which work fine with many providers), really.
But thanks for the answer. I've tried out a few of the new WebRTC sites like appear.in and some others, but they seem to be alpha level. Not tried them a lot though.
Has anyone tried Firefox Hello, which was mentioned in the recent thread about Skype issues on Linux? Is it alive? Going to try it with someone soon, if it is alive, but interested in hearing others' experiences.
What are the options for development team discussion boards?
We have GitHub Issues, but they're not a discussion forum... the GitHub wiki, which nobody likes to use for anything except info they don't have any other place for... and Slack, which is too synchronous and spammy to be a place for calm, longer-term discussions...
Any interesting contenders? Perhaps something aimed at remote async teams?
Funnily enough, I've been toying with the idea of building a new service, something like a weird fusion of Github and Reddit, for exactly this use-case.
It feels like there should be a good solution for dev-team (or even OSS community) discussion baked into the project hosting solution. The only thing holding me back is that it looks like a huge undertaking.
Yay for weird fusions. The internet is just a big collaboration machine and existing tools are calcified clichés.
When me and my brother worked on a project, we set up our browser home pages to show a CGI gadget that grepped (well, ag'd) the source code, and commit log, for @-mentions and XXX comments. This way we could talk asynchronously inside the coding process, e.g.
// @dbrock do you know a better way to do this?
That was ridiculously simple to build but very effective. So I'm a fan of this linking between code and communication, and encourage you to keep experimenting!
I'm envisioning a github/gitlab alike source hosting service, but with the UI and UX tuned towards enabling discussion and collaboration. Sort of trying to get to Github's original motto of "Social Coding". Put more emphasis on the social and human aspect than on tweaking the product for enterprise sales.
The basic idea would be to have a reddit-style threaded discussion available from any point in the repository: the repo itself, a commit, a branch, a tag, and to have easy hyper-linking between those objects. Good search would help too.
Random crazy idea that might be useless or contraproductive: have users install a small daemon that notifies the server about their workspace changes in between commits, so you can see what people are working on live.
In a team setting such as where I work, it's somewhat arbitrary that I only learn about work once it's committed and pushed. It would be interesting to see in real time what people are messing around with.
But I also like the idea of asynchronicity, and taking inspiration from open source collaboration into the business world, rather than the other way around...
By the way, it seems to me that the basic functionality of GitHub and GitLab is quite simple, and the kind of thing you could make an MVP of in a weekend or two, if you carefully avoid complications and keep it ruthlessly simple...
I think we already have discussions on commits in GitLab. We used to have discussions on repo's (the wall) but this wasn't used much. I agree with the hyperlinking, if I mention an issue in the chat the issue should get a backlink to the chat archive.
Am I the only one who feels like they should stick with what they're good at? Don't add too many features, bells, and whistles. Just keep building the core product.
I had the opposite thought -- its such a simple next step (at least, basic video chat integration is really simple with webrtc) I was wondering what they were waiting for. Without it, the door is wide open for a competitor to come in and be Slack + (on demand) Video Chat.
its such a simple next step (at least, basic video
chat integration is really simple with webrtc)
Said like someone who's barely used videoconferencing software!
Having tried systems with built in video chat, Skype, Hangouts, Facetime, Radvision and Scopia I can tell you they all suck - and none of them give you the diagnostic information to tell /why/ they're broken.
In the past year I've seen echos that come and go, sound problems that are present in group calls but not individual calls, pages that work in Chrome but not Firefox, temporarily losing video or audio with no helpful diagnostic information, dropped calls with no helpful diagnostic information, people being unable to join group calls with no helpful diagnostic information, browser plugin problems with no helpful diagnostic information, problems reconnecting dropped connections with no helpful diagnostic information, users being unable to hot-plug their camera while a call is in progress, mobile apps only finding the rear-facing camera, users having to log out on android so the call will ring on their computer, nonexistent or out-of-date Linux clients, nonexistent Windows clients, and crap video quality even when users are on high-bandwidth connections that can easily stream 4k netflix.
I wish it was simple to make a decent video conferencing system, then someone might have done it.
He didn't mean it would be simple to implement. He meant the reasons for why Slack should include it are so numerous that it is easy and "simple" to make a case that Slack should implement it right away.
What's that fallacy called which leads us to assume that other peoples' jobs are easy? If there are a few major players in an area and all of their offerings are terrible, is it more reasonable to conclude that they just all suck despite somehow being major players, or to conclude that it's actually a much harder problem than you think?
I only meant its an obvious next step, and until they close that loop the door is wide open for competitors. Once they do close that loop, the door becomes tiny. The only easy part would be doing a basic webrtc implementation (as they are using Chrome via Electron, and hence get the vast majority of that implementation for free). For reference, my last job was working on a web based webrtc video conferencing app in a lead role at one of the major video conferencing companies.
Although simple from a product roadmap pov it's incredibly difficult from a technical pov - no doubt the reason that they acquired a great team a year and we've still seen nothing released. It also requires a completely different set of knowledge and expertise to websockets/rest.
I would like it if they fixed all the bugs and usability issues with the core chat functionality first. They are trying hard to be a good IRC client, but failing at it in fundamental ways.
I have high hopes for it, then. ScreenHero remains my favorite screensharing application. Among my circles, its usage has been growing even though you can now only register an account by invitation.
Agreed. I actually hope it provides the screen sharing capability with the possibility of video also. There are tons of non-developers that I always feel like I wish I could just Screenhero them and show them something, but having a random person take over your computer can be a little jarring.
We have a channel for our team to BS around in and one for more serious discussion which inevitably trends towards BS. Someone has to be an adult and tell people to quit it eventually. It's a problem with our team but not slack.
I'm surprised there hasn't been more of an emphasis on creating better dashboard. One of the biggest hang-ups I experience as a user is it's too difficult to find and keep windows open for every team that I'm on that has it's own slack address. If I could sign in once and see all of the teams I'm on in one place I would be much more inclined to keep a slack tab open in my browser all the time.
I'm still convinced that Hipchat was bought by Slack and the Hipchat horribleness lately is the Slack migration plan. The new client is just terrible. I asked a coworker to zip up their HipChat.app so I could revert back to a state of somewhat stability. Hopefully someone who works at Atlassian is reading this and will read: We just need to chat. I miss whatever happened to the HipChat of 2013/2014.
HipChat definitely is Atlassian's dropped ball,... and after they saw Slack's success, they leant down to pick it up ... and accidentally kicked it away again. It's surprising just how badly they've handled that project.
A pity, because I liked Hipchat's ability to show more than three comments at once, sane display of notifications, and non-rapacious pricing schedule.
This makes me wonder. How well does video-chat over WebRTC work? Are there any open-source video chat applications?
I know there is icecomm.io, but it is closed source. I wonder how they handle video, and if they are doing any video processing and/or compression in javascript.
Too soon to tell yet in my case, but I've been pleasantly surprised by the rapid rate of improvements, so I'd say probably yes.
Skype and Hangouts are still a little better in terms of quality and overall experience, but a standard WebRTC app is definitely "good enough", and with some tweaking, it can be great (see Appear.in).
And WebRTC brings some very important advantages over Skype, Hangouts, etc (mainly reduced friction, not having to login, better embeddability, and overall control of the experience)
I've yet to test the new VP9 codec, which should bring huge gains. And there are other tools being added to the ecosystem which will enable lots of neat use cases.
That graph shows the drop of cca. half a million users around the end of 2015. Is that reporting glitch or they really did loose and regain half a million users in a month?
If they can do this right, they will get a lot more customers.
Right now we use Lync/Skype for communication, which I loathe almost as much as Lotus Notes. This sentiment is shared by many. Slack usage has organically grown (as a reaction to the failings of Lync) and I can envision Slack displacing Lync in many companies.
This. I think Slack is going a new territory with voice/video. It will compete now with MS Lync and Cisco Spark. Both are eating market shares in collaboration and Slack will be affected. The two giants are investing a lot of money in their cloud based collaboration tools and might want to buy slack in the future, if it becomes a real threat.
Screener's video/audio/screen-sharing tech with Slack's UX will be a good marriage, and when I heard about the acquisition I was expecting good things.
Use both products at work, and Screenhero is good enough we can code in pairs remotely across Europe. Bake that right into Slack and improve the UX (because Screenhero's interface is not great in places), and it's going to be a really, really solid stack.
I try to avoid it because its such a tease trying to get it just load up. To put things into perspective it's faster to load up Facebook with a web browser and just chat with that than to load the Mac Slack app.
It's cool they have those funky loading messages but its just too slow.
Might point out im in Australia and latency is a bit higher. I noticed the same thing in a few African countries.
While latency is higher by a hundred or so milliseconds the loading time is significantly slower. This too on a line that is 99.4%+ better than the average.
It is badly engineered. I had a look at the inner workings and first off they use a web app in a native wrapper, second their websocket negotiation process has a lot of going back and forth just to get started.
The reason for the move to dedicated servers (AFAIK) was because without it, they couldn't really do offline messages. Once WhatsApp etc started doing this, the Skype team realised that their greatest strength (no paying for bandwidth or servers etc) was actually their greatest weakness, since they couldn't deliver the features that people had begun to expect from chat apps. There was a great article on this .. but I can't find it.
Skype is also a mess right now; with chat application bugs, UX issues (why does the freakin' mute button move around) calls that can't be joined or start cyloning, participants are suddenly dropped mid call. Throw in the spying allegations ...
Hangouts is great for video but is a really clunky experience otherwise.
With everbody dropping the ball, there certainly is room for Slack to come in and take control of this space.