Hacker News new | past | comments | ask | show | jobs | submit login

Since you offered, I'll ask. What's the down low on IRCv3? Is it actually going to ever have any sort of impact? It feels like too little too late. Far too late to ever have support in most clients.

IRCCloud implements it but servers don't, and IRCCloud feels kind of dead development wise. And it doesn't really solve the problems that Discord solves, for example.

The dream of instant messaging being built on top of open protocols, just like email, doesn't feel terribly far out of reach but it also doesn't feel like we're making progress towards it and the efforts spent on keeping IRC alive feel, at least to me, kind of futile compared to say, efforts spent on Matrix.

I don't think it's crazy to leave IRC behind, and most people who mourn it will either do so out of nostalgia, or because of the loss of an open protocol. I'd rather centralize on something that has a future though.




>What's the down low on IRCv3?

Well, it is, but it's slow. IRCCloud and KiwiIRC both use it, and IRC.com will support it as well. There are some exciting new plans surrounding encrypted voice and video on the Kiwi side, for example.

>doesn't really solve the problems that Discord solves

What specifically? Hard to address that one without more granular discussion points.

>it also doesn't feel like we're making progress towards it

It could be argued that IRC.com is going to be a major step toward more rapid progress, between the foundation funding development, and the likelihood that it will be an enormous network that's V3 compatible.


I've heard that Twitch implements some of IRCv3 (or at least uses that format for some of their custom message attributes); however my impression the last time I gave just a glance at IRCv3 was more of a cautionary learning example.

My biggest issue with the protocol is that if I want to develop for an advanced data interchange protocol there shouldn't be any optional extras. Everything should be in the spec and required of real clients. (I can, however, envision a protocol in which 'relay servers' exist that don't need to fully understand a message to pass it's content.)


Protocols which have a minimum implementation are fine. Protocols which prohibit extensions end up with them anyway.

Trust me, I attempted to veto a few things in IRC's early life, they are all in there now, can't stop them.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: