Discord has neither.
I don't have any influence over these giant communities but if I were starting them from scratch I would put them on https://spectrum.chat because it explicitly solves these issues. (no affiliation) Now the network effect makes it a lot harder. I would encourage -new- open source communities to check Spectrum out.
https://zulipchat.com/for/open-source has more detail on Zulip for open source projects.
Open source communities that communicate via mailing lists do have threaded discussions that can be found via a Google search. Though the websites they link to have a rather limited interface for viewing the discussion. NNTP gateways like gmane make it much easier to read through the threads (though it would be better if the web pages presented a more HN or reddit type view of the discussions).
So mailing lists then? OSS communities have been using them for years, they even have offline browsing and offline search. There are a number of user interfaces available from online ones like google groups or gmail, desktop readers, terminal apps and even editor plugins.
I had this discussion with Max Stoiber (founder of Spectrum.chat) on Twitter a couple months ago  . I agree that threadedness and searchability are nice features of Spectrum, but I'd say that both a forum-like searchable model and a real-time chatroom model have a place, and that the use cases only partially overlap.
Are you in any way being compensated to say this? I don't expect you to disclose the details here, but I will at least ask. Publishing this blog post I understand that it is important for Discord to win over open source project to their platform. As users, developers are often early adopters and "organic" influencers.
I think you should give https://riot.im a try. It's open source and federated and has IRC bridges.
I had no idea they were even writing that article until I saw it posted here on HN.
Not sure what else I can say to prove that, but it's true :)
Spectrum looks similar to a 2011 text message client, and a niche platform is less welcoming than something that a good portion of people already have an account on.
This just isn't true. Compare  to . Nor does it look like VB, phpBB, etc. from 2011.
> .. a good portion of people already have an account on.
Ah yes, my N accounts for N communities on Slack.
OK, they are flat bubbles now rather than shiny blobby ones, but it's the "bubble for everything someone says" that makes it look dated.
I hated the "bubbles" look back then and always used themes that just presented text as text, with possible unobtrusive delimiters between people's comments. Like Slack and Discord do.
im not concerned with the looks. and this niche argument is pretty weak, every platform incl discord starts somewhere.
Besides, Reactiflux is a totally independent group - started by the community, run by the community. We've had several Facebook employees come in for Q&A chats, but we're not an "official" part of React in that sense.
(Note: I'm a Reactiflux admin.)
I am deeply disappointed by Discord's approach to Free Software in general, although I understand that they deeply desire the same business model as Slack or Hipchat. It'd be nice if we could stop balkanizing chat already.
Discord doesn't have the luxury of being a bought-out company like Curse client which makes money for Amazons other services. Discord is on its own, and as such has to make back their revenue. You can't make revenue unless you charge for something, and if people can build something better than the discord client (They can), then you're going to be hard pressed for features.
Remember it took Discord a year and half to actually implement video calls, when right from the start they advertised heavily that they were a 'replacement' to Skype. Took them two years to just have video embeds. Even still they have no proper administration system for servers. If you want to admin a server, you have to build and host your own bots for it.
And thats just the iceberg. It gets messier the deeper you delve into the personalities and inexperience this company has.
Better to stay away and donate time/money to open source competitors even if they lack network effects.
It's also great for creative writing groups, clubs, hobby groups...great tool for casual organisation of 3-50 people where most of your time is going to be spend casually chatting and linking images and posting jokes. It's also much better for creating ephemeral groups.
But I'm not really sure I think it works for open source projects. "Better than Slack" is setting the bar really low; Slack was a bad solution for this as well. In particular I'm concerned by: 1) Walled garden 2) uncertain corporate future 3) lack of public, searchable logs 4) lack of interoperability and federation 5) lack of deep links into archives 6) barrier to new members.
Reactiflux on Slack was a poor experience; they outgrew Slack and were forced to jump to Discord and...it's still a poor experience. It's like hitting yourself on the hand with a hammer, realising it hurts, so getting a different hammer. This isn't progress. I'm not sure it's worse, so sure, if you're already on Slack and you want more of the same then sure, switch to Discord. But maybe first think about alternatives that aren't "the same thing I already have with a dark theme"?
As for Github, their are absolutely concerns relating to their role in the open source ecosystem, and many people advocate for Gitlab and other solutions as a result. However, the concerns with Github are much more theoretical; the actual experience is quite good, whereas with Slack it's pretty mediocre. Plus, it's a lot easier to migrate a project away from github; because it's built on git you keep all your commit history, etc., and your issues and wiki are easily exportable. If I want to move my open source project from github to a self-hosted gitlab, it's the work of an afternoon.
None of that's true on slack.
>> Slack, like so many others before them, pretend to care about interoperability, opening up just so slightly, so that they can lure in people with the promise of "openness", before eventually closing the gate once they've achieved sufficient size and lock-in.
> Spot on. People are lured in by hype and forget the long-term consequences. Always chose “open” by design, never by charity.
The one I liked the most was WPF, which I would say is web inspired (XAML is like HTML in many ways).
The later apps I did were web-based, the whole app was an HTML page running inside a WebView. This was years before Electron appeared, and much harder. But it was still so much better.
Yes, performance sucks, but you can do in one day what it takes you one week in a classic GUI framework. Layout which is a major pain in old frameworks is trivial in HTML. Same for fancy fonts/colors and using images everywhere.
Slack is designed for trusted teams and does not handle malicious members. There are several won’t-fix issues that drove us out:
1. Anyone can set reminders for the slackbot to message to anyone in the channel. These reminders look official and don’t say who sent them. Great for phishing. You can’t disable this.
2. By default, anyone can download a list of all users, including email addresses and real names. Awesome again for phishing.
3. It’s now super hard to tell imitation accounts from the real thing on Slack. Multiple people can have the same usernames in the channel.
Discord, being forged in the fires of massive gamer communities approaches everything with the expectation that some of your users will be evil, and gives you good tools to deal with it.
I will not stop promoting the open source projects I'm working on on Twitter and Facebook just because they happen to not be open source.
At the time where Reactiflux needed to find a new home, I tried a ton of alternatives, both open source and closed source and the general level of quality for all of them was a huge step down from Slack, except for Discord which actually felt better.
Awesome. What all open source projects you evaluated? Are you really telling that there are no open source projects that are as good as Discord out there?
"Discord is the easiest platform to join. New users can immediately join our conversations without having to create an account."
That is, frankly, the killer feature.
Notwithstanding, I find it odd that developers here don't use E2E chats. It's always something clear text with permanent history, and in the case of Slack, key logging too! There's a JS based xmpp framework out there.
It puzzles me why no one uses such things.
So I would say, they are clearly doing something wrong over there and while I searched for solutions I got the impression that my problem wasn't an unique incident. At least for me it was enough to not recommend Discord.
I wish more communities would use Discord. The single sign-on alone makes it super convenient, but all around I just like the experience better than Slack, from login, to sharing channels, going by the actual UX (I haven't used the app in a while. Back then it needed a bit of love. Dunno if its still the case).
Every time I try to join a Slack community, it starts with "Ok, where' the darn link for it", to having to sign up, and looking at my computer's resources dying under the load and everything lagging if I'm navigating old comments from search (if there even is any, since free communities get truncated pretty hard).
I know there's a bunch of missing features, especially for corporate use, but for the stuff a lot of these communities use it for, Discord just works better.
The only thing that used to be a big issue for me was dealing with servers with a large amount of channels (since you kind of always join all of them by default. Is that still the case?). Channel organization by category is quite nice though.
Annoying, but it's a one-time thing.
That would not be fun in Discord-land.
From my understanding, the free Slack version doesn't offer unlimited history, so I'm wondering how Discord can offer these features.
Based on the description given in that article it's a technical hassle to store that many messages, and it could be Slack just thought it didn't want to deal with that.
Discord might forgo some profits right now for a better future.
But now I'm pretty sure Slack is in panic mode internally. Discord had unbelievable growth in the last months.
With search, I'm really not sure how they pull it off.
Cost isn't actually too bad, given how we've designed our search schema - and how we lazily index.
Maybe my mental math is failing me (or I'm overestimating how many messages there are), but that still seemed expensive.
In the end, as long as you don't go out of business, because I'd be very very sad :)
On GCP, 50TB of Persistent SSD is roughly $9000/month before any discounts.
So, for my own benefit:
Discourse is the open source born-again web forum software from the people behind Stack Overflow: https://github.com/discourse/discourse
Discord is a proprietary Slack clone whose main attactions are bundled TeamSpeak and a dark theme: https://discordapp.com/
Sure there's plenty of people using their service and not paying for it, but it's also great marketing. If I do need chat for my paid organization, I'm much more likely to pick Slack if I'm using it daily for public channels.
Additionally, we run a very slim and optimized infrastructure relative to other chat apps (elixir/BEAM makes this very effective; our current VM footprint for our real time services is ~4 thousand CPUs over a hundred or so VMs). We realized early on we needed to write efficient code that is cheap to run in order to be viable as a business.
I read a blog-post back in 2016 from Slack which mentioned their footprint was in the 5k+ VMs . In 2016, Slack was roughly at 1/8th the scale we are at now.
If you're asking people to upvote this submission, that's against HN rules and that doesn't work.
EDIT: Since this submission got unpenalized (and this comment now obsolete), the Discord page the article links to (http://discordapp.com/open-source) may be a better source.
In this case, there is no discussion about the article, just about the upvote brigade. So the only reason for sharing the link to the HN discussion would be to get people to upvote it.