IRC is awesome. The biggest downside is that when you join a channel, you don't get to read all the previous history. In that important feature Twitter/Slack is superior.
By the way, #startups on freenode seems to have the most HN users on it.
By the way, #startups on freenode seems to have the most HN users on it.
I use bip[1], which is an IRC bouncer. I basically connect to networks from a remote machine, and when I hop into IRC, I connect to that machine - which gives me some playback of what's been said beforehand, so I know at least some of the context.
This doesn't solve the 'problem' for joining new channels, but it's nice if you have regular channels you hang out in.
1: https://bip.milkypond.org/ (their cert is broken :/)
erlang vs python .. connected clients vs historical search.. apples and oranges i guess...
In my use of it - primarily social, less technical - this is actually a huge benefit. Whereas things I say on Facebook, on Twitter, even on Skype groups float around forever, this doesn't happen on IRC.
Only the other people in the channel log things, and if someone else turns up, they don't see what I've been talking about - the exact contents of my discussions aren't floating about forever so long as people don't intentionally distribute them.
However, these days there are IRC loggers and 4chan archivers.
I miss when more of the web was like that. It's nice.
How it actually accomplishes that within the framework of the "protocol," I don't know.
The Quassel protocol supports this, unfortunately it looks like the quassel team doesn't care about taking it further than where it is.
Senders – there's absolutely no reason to fake senders, since it was always valid for messages to come from people not in the channel.
In a future release InspIRCd will adopt support for the server-time extension which will allow us to get rid of the timestamp duplication if a user's client supports it.