I was a huge proponent of IRC but the lack of developments the past... decades has killed it for me. Discord is straight up better. Yes it's proprietary, but no, IRC is just not a good alternative at this point, it hasn't kept up.
And I love to death what the IRCCloud guys are doing but it's just not good enough to use as a proper communication tool especially in a company.
Admitting this is the first step to fixing it. If you want open source alternatives to win, you can't get stuck telling people the things they like aren't worth supporting. I like the fact that I can do text, voice and video in the same tool; that I have a searchable message history; that I don't have to manage the server myself; that I can make interesting bots with a webhook system and a websocket API; that I can interact with people programmatically over an OAuth2 API; that I can use markdown in messages, embed files and youtube videos, pin messages; that I can manage large communities using a thorough group and permission system (I'm managing a Discord server that has 20k+ people on it, this is stuff that can't be done over IRC quite simply).
And you know what, I like the custom emojis too.
Quassel isn't good enough (speaking as someone who loves quassel). IRCCloud isn't good enough (speaking as someone who loves IRCCloud and heartily recommend everyone here to support them by buying a subscription).
I used to say: Our best chance is that Hangouts is open sourced. Nowadays, I'm rooting for Matrix but I think our best chance is that Discord is open sourced. These protocols, they get developed with very little awareness of what people actually want -- they copy features left and right, try to either support everything and end up a bloated or unusable plugin mess (XMPP) or support nothing and end up unpopular. Most of them are toys. In the end, we need serious players, passionate about creating not just protocols but good interfaces to them. This is hard to find in open source.
None of the features you mention are useful for actually communicating and getting work done.
Your concept of "better" is distinctly different from mine. In my world, software without completely useless and unnecessary features is better than bloatware. irssi + tmux + logging + grep works great for 100% of actual not-embedding-useless-cat-videos-to-avoid-having-to-actually-get-work-done use cases.
In the interest of adopting your own borderline-uncivil tone: Stop assuming objectivity in your notion of evolution and improvement being the same as everyone else. It's not. Just stop. Please.
> None of the features you mention are useful for actually communicating and getting work done.
I respectfully disagree. It's incredibly helpful to be able to video/voice call a coworker instead of typing back and forth. It's also incredibly helpful to be able to quickly past a screenshot of an issue you're seeing, or to paste a quick snippet of code, or log, with nice formatting.
Yes, I could use my phone or another app, and I could paste my image to imgur and share a link, or point them to github - but that's another step, another roadblock on the road to communication and understanding.
> It's incredibly helpful to be able to video/voice call a coworker [...]
In my experience it is pretty rare in software development to have the need for a voice/video call. There are use-cases but they are rare and better handled by a specific software.
> It's also incredibly helpful to be able to quickly past a screenshot of an issue you're seeing [...]
And also painful. Sometimes I get screenshots from terminal screens showing error messages and have to retype them when e.g. grepping for keywords in source.
> [...] quick snippet of code, or log, with nice formatting [...]
If it's short just blob it into the chat window. If it's long then please use some paste service, as screenshots of logs are a nightmare (see previous point).
> [...] but that's another step, another roadblock on the road to communication and understanding.
Why is there another step? You can have multiple programs, tabs, terminals, shells, sessions, etc. open at the same time.
For developers that might be true that voice/video call is not happening often but I don't think that other business functions would share the same perspective. The ability to call, video conference and easily share images and docs is an important need for collaboration in Design, Marketing, Product Management and as a PM I often had to share screenshots back with engineers to illustrate what I'm talking about. Then, from my experience large orgs do not want to have multiple chat systems that compete with each other internally, so they'll take the one that offer more versatility while being also easy to use.
Disclaimer: I'm an ex-Atlassian (Product Manager) and my feedback is genuine, coming from experience. I'm also super happy to see my ex-co-workers getting Stride out but my opinion is not a Marketing plot. Software now requires the collaboration of many different roles, many of them who would prefer Slack/Stride over IRC.
> In my experience it is pretty rare in software development to have the need for a voice/video call
And in mine, it's very common! I lead a distributed team, and frequently it's easier and quicker to move from text chat to voice in order to explain something.
Why not XMPP though? Why hang onto IRC? I mean sure this is cool but it's client specific features with no standardisation of the protocol (correct me if i'm wrong).
If I have to install a special client (quassel) to access these features how is that any better than installing a XMPP client?
Only thing IRC has going for it is the existing userbase, most of who run clients that won't support these new features.
Actually, all these features are being standardized into the IRC protocol. That's the good part.
And we've got all the developers of all the major IRCds and cloents together, working on these things. In contrast to XMPP, where support for an XEP between clients is spotty, and in contrast to Matrix, where a single dev team runs the server with ~50% of global users, builds the clients, and servers.
I agree that XMPP is a better protocol, but with the people we have, we can, long term, change the IRC protocol, too.
And when it gets here, then IRCCloud and Quassel would be a good answer to my original comment. But now, those clients don't provide those features, while Slack does.
I’m writing on these things right now. They’re going to be in Quassel at least before the end of the year, on desktop and mobile. In developer builds they’ll already be in in a few weeks or a month.
And almost all functionality already is there anyway, as said.
Except it's not going to be part of the standard, and there's no way that I can guarantee that the other people in the chat room are using the same thing.
At the end of 2014, I was in a solid ~40 irc channels and active in maybe 15 of them.
Today, I'm in 10 and active in none. All the channels I left, even some of the ones I'm still in now, have switched to Discord, Slack or Matrix. The two only channels I care about are actually both mirrored on Discord using the excellent Matterbridge (https://github.com/42wim/matterbridge/).
I was a huge proponent of IRC but the lack of developments the past... decades has killed it for me. Discord is straight up better. Yes it's proprietary, but no, IRC is just not a good alternative at this point, it hasn't kept up.
And I love to death what the IRCCloud guys are doing but it's just not good enough to use as a proper communication tool especially in a company.
Admitting this is the first step to fixing it. If you want open source alternatives to win, you can't get stuck telling people the things they like aren't worth supporting. I like the fact that I can do text, voice and video in the same tool; that I have a searchable message history; that I don't have to manage the server myself; that I can make interesting bots with a webhook system and a websocket API; that I can interact with people programmatically over an OAuth2 API; that I can use markdown in messages, embed files and youtube videos, pin messages; that I can manage large communities using a thorough group and permission system (I'm managing a Discord server that has 20k+ people on it, this is stuff that can't be done over IRC quite simply).
And you know what, I like the custom emojis too.
Quassel isn't good enough (speaking as someone who loves quassel). IRCCloud isn't good enough (speaking as someone who loves IRCCloud and heartily recommend everyone here to support them by buying a subscription).
I used to say: Our best chance is that Hangouts is open sourced. Nowadays, I'm rooting for Matrix but I think our best chance is that Discord is open sourced. These protocols, they get developed with very little awareness of what people actually want -- they copy features left and right, try to either support everything and end up a bloated or unusable plugin mess (XMPP) or support nothing and end up unpopular. Most of them are toys. In the end, we need serious players, passionate about creating not just protocols but good interfaces to them. This is hard to find in open source.