3 ms·
explain it like I'm 5 please: why can't.we just build an actually user friendly IRC client (including non technical people) that doesn't look like a dos app, an
by jdc0589 9y ago
explain it like I'm 5 please: why can't.we just build an actually user friendly IRC client (including non technical people) that doesn't look like a dos app, and add client specific extensions for expanding image urls inline, etc... ?
- ge0rg 9y agoIn theory, we can. However, building a well-working multi-platform chat client is a significant task, imagine a 5-10 person team working over a year, plus the long-term maintenance support. IRC might not be the right protocol for that, because you'll be adding quirks on top of quirks, but you could easily do that with XMPP. However, you won't find anyone willing to sponsor that development. Old protocols like IRC and XMPP just don't cut it in Silicon Valley. Examples for creating a new IM solution out of nothing, provided funding, include Signal and Matrix/Riot.
- jrochkind1 9y agoThe centralized nature of Slack actually makes it a lot _easier_/cheaper to develop, than developing for true federation. See for instance: https://signal.org/blog/the-ecosystem-is-moving/ https://signal.org/blog/the-ecosystem-is-moving/ and https://lwn.net/Articles/687294/ https://lwn.net/Articles/687294/ Okay, could you implement some IRC extensions to provide some of the features Slack has, deploy them on your own central server cluster, and provide a client that has access to those slack-like features with a UI as good or better than Slacks, so long as you're using channels on that server? Sure, you totally could. You could even take the extra time and energy to make those extensions open standards instead of just something you did (and it definitely is lots of extra time and energy to do that). A lot of work, but not a lot of unsolved problems, it's pretty straightforward. You could do it with IRC or XMPP. Now who's gonna pay for all that?