http://getkaiwa.com/ is another Slack alternative that uses an XMPP backend, which IMO is much better than a custom backend. So far the only open source Slack clone I've seen that uses an existing standard for the backend

This is great. Also see: https://zulip.org

And the blog on why Dropbox decided to OSS it:

https://blogs.dropbox.com/tech/2015/09/open-sourcing-zulip-a...

Props for open-sourcing, but I'm putting my money on http://matrix.org/

It's good to have options.

The takeaway I'm getting from this story, and Mattermost, is: 1. Export your critical data from SaaS services if you're business cannot exists without them. 2. Test that this works before putting years of data into a service.

There's nothing wrong with SaaS services, they just mean users must do their due diligence in any business partnership. I can't see how a game company can put their resources into delivering this as an open source project with no future plans for monetization. Frankly, without monetization, open source projects generally wither up and disappear. Then you're no further ahead.

http://www.mattermost.org/why-we-made-mattermost-an-open-sou...

Why do I have to go to slack.com to learn what this is?

I clicked on this link, no explanation. I checked mattermost.org, no explanation. I went to slack.com, no explanation. Then I clicked on a "Product" link on top of the webpage. Finally some information what this actually is.

Even open source projects could benefit from a little bit of marketing.

If what you really want is a pretty web-based way to access IRC, then you might want to check out Glowing Bear -- it connects to your WeeChat IRC client via websockets and works nicely on the Desktop and on mobile. It doesn't have a voice recorder, but it gives you the infinite possibilities of a mature IRC client. It's a project I've been contributing to for a while now and I still absolutely love using it.

https://github.com/glowing-bear/glowing-bear https://latest.glowing-bear.org/

There was a time when I thought something like this was a good idea. But after using Slack for about a week, there's no way I would give up all the benefits of a well-integrated centrally controlled service. All the clients work together perfectly. We have Slack channels with customers. It just all works much better than I can imagine any self-hosted, decentralized service would.

I think Slack serves its stated purpose very well (smaller, business oriented teams), but many groups have started using it for larger communities, mostly because it has unlimited users for free. But it isn't made for that and there is no way that most of these groups would ever be able to pay for a premium subscription due to the per-user costs. 10K messages across all channels is surprisingly easy to hit, need the ability to ignore users, etc. I think this project has great potential to fill that niche if it is marketed properly. Slack is so close to working well in that area but really needs to pivot to be able to serve it well and make money doing it.

I'm trying to decide if this is better than Zulip. They're both open source, backed by someone trusted, and I can run it on my own server.

Interesting that it will be a default feature of Gitlab.

That's a move that seems like it may push Gitlab ahead of GitHub in some ways (well, to me at least).

There are many OSS alternatives to Slack. Some are clones and some are different approaches and many of them are quite good.

The thing these and all other similar efforts miss is the importance of network effects. Everyone uses Slack because everyone uses Slack.

The real problem that needs to be tackled is one layer down: providing an open, distributed alternative for authentication, identity management, and data interchange that is secure and robust enough to provide a backplane for things like this and that is easy enough for anyone to use that it can be pushed out to the mass market. I can't stress the last point enough. It must be stupid simple easy to use or it will fail. It also must offer a good and simple developer experience (DX) or it will fail. DX is part of UX. Things like XMPP are nightmares for devs and sysadmins and fail badly here.

This is a huge missing piece of the web.

Well.... this is depressing -

Mattermost server is made available under two separate licensing options:

- Free Software Foundation’s GNU AGPL v.3.0, subject to the exceptions outlined in this policy; or - Commercial licenses available from Mattermost, Inc. by contacting commercial@mattermost.com

"To simplify licensing, we’ve responded to community feedback and the compiled version of Mattermost v1.0 is now under the MIT open source license" (Emphasis mine)

Why just the compiled version?

Since we are talking about open source software, maybe the guys that own the Mattermost account on Github could create a placeholder repo for Android (I wonder if this idea would work for iPhone, too) and accept Pull Requests until there is at least a beta native app.

Would be happy to move over to an open source alternative, but at the moment they don't seem to support mobile apps.

It would be very difficult for us to move because of this, we talk a lot on the move.

I wonder how this compares to Rocket.Chat, another open-source alternative: https://rocket.chat/

Doesn't this have some heavy hardware requirements? Three machines with at least 2 GB of RAM? Is that really necessary if I'm going to chat with five people?

This looks very nice. Is there any plans for an API/client protocol? Web client is all well and good, but I'd want to have a solid console client, as well as some command line tools (eg: "echo "Some message" | xmpp user@host -- where the equivalent for Mattermost would allow to set the topic, or message a group via a bot etc.)?

Overall I like it because they closely follow Slack's UI. However I question the choice of fully supporting Markdown. A comment isn't supposed to be documentation. Supporting things like bold, italic makes sense for emphasis or making code easier to read. But headings? When would one ever want really large text in a comment?

In my past job, I was desperately looking to get an open source Slack alternative. The ones I tested then (few months back) didnt hold up nicely against Slack. I'm happy to see that finally there's some good competition.

If it supports SSL XMPP it can be a drop-in replacement for a lot of companies.

Feature idea: a canvas where people could draw instead of typing a text.

The blog post is dated October 2nd; Is HN just learning of this announcement late, or is their blog displaying the draft date rather than the publication date?

So, what does this offer over Slack, besides being open-source? I see no mention of any actual features, besides basic chat features.

I've spent a couple of hours trying to get Docker running on my linode Ubuntu server to no avail.

A non-docker implementation would be nice.

Besides being an excellent application, this is a valuable resource for anyone studying Golang.

Does anyone know who the old provider was that locked in their data?

Why not just use IRC ?

Can it send push notifications or other alerts to my phone?

Congratulations, Mattermost team! Huge accomplishment :)

Just what we need, another solution to a problem that was solved two decades ago!