IRC has existed for ages, and it will for ages more - there's something you just can't replace about it.
Lord knows people have tried - for instance the introduction of IM software. We've seen internal ways of doing the messagin,m which has led to 'i don't use MSN, so I can't message you'.
Then we saw trillian step in to reunite all of the major protocols, whcih was a good step - but this IM craze is evil in my opinion.
The fact that microsoft shut down chat servers so that you can't meet strangers is something whcih makes me kinda sick. Theres nothing better than talking to someone half a world a way - but moves like that are trying to force us to talk to people we already know.
I don't like talking absolute crap to people that will remember me for it and talk to me the next time i see them in the flesh.
WE've seen HTTP and XML models of client and server communication start to take off - i personally like it, the idea of web services rather than web pages.
I myself am trying new and experimental things, like an RSS (Really simple syndication) client and server for mirc - RSS was made for websites to trade news around, whereas this would be aimed at IRC - whats the most talkative channel, etc etc.
HTTP was a good stepping stone for XML services, we just need to find new applications for IRC protocol - innovation will come from the chat server developers if from anywhere at all.
The mirc developers are the other big innovators, I believe it was mirc that started the idea of DCC, which is now a well established and effective method of file/chat commmunication.
The mirc scripters play a small but effective role in the shape of IRC - we now see lots of spam bots, but we also see anti spam tools, and a high degree of automation. Scripts are something which are unique(ok not really) to mirc compared to many other protocol clients - MSNscript isn't something I'd care to tangle with.
The reason we don't see much innovation is because mirc is closed source - the development team can't stick in every good idea, tehy don't have the time. An opensource community might be able to service it better, but you'd also start getting diversification - lord knows its bad enough with scripts as it is now - no communication, no coding practices, no anything.
I think that the future lies in developing other services which use IRC as a middleman... what these may be we'll never know.
Adding com support to mirc was a great step imho, opening up prewritten code to developers in a great way.
If we saw this support fully completed, we'd see an amazing rise in what we use IRC for i think.
Also, we need more scripters who aren't 14 and barely can spell. Ones that are actual programmers that can carry good habits back to mirc scripting. I know there are practically NO php + mirc integrations about, but imagine the potential...
One simple idea is a profile management system - you fill out your user information, and have a standard blog style site. In addition, you have a mirc script which simply returns a url for a basic CTCP request, then a small client that lets you read it online.
One of the limiting factors in IRC server development is that not many people buy the stupid software, and if they do its exspensive.
It costs a lot in terms of netwrok traffic routing everything through a central server - we need to start escaping the bindings of this network model.
You can maybe register a nickname and read some short messages on most networks today.
Some have news feeds.
Why do they need this? Its really all very primitive, when they could externalize alot of these services.
Also I think we need to see some kind of unification between all of the larger networks - use of common EXTERNAL services would be one i would lvoe to see.
There's more in my head but I'm sleepy
