Realize that this all sounds great, but Khaled makes a living off of maintaining mIRC. This is how he feeds himself. While this fact may not be important to you, it is important for the author of the program to be able to benefit from his work as well.
There are many open source clients that you can contribute to, if you want. You'll find that the bureaucracy of actually contributing to projects like those is generally pretty high, and it's not as idealistic a situation as you make it out to be. Simply open sourcing a product doesn't automatically mean everything you hate will be fixed (or can be). There are many great open source examples of this; they all have problems, they all take a long time to address certain issues. Some of them never fix certain things. Will an open source mIRC mean you will get better error handling? Unless you're willing to implement it yourself (and are offering to commit to maintaining the code you wrote for a while longer), there's really no guarantee.
More importantly, open sourcing a 15 year old product isn't necessarily going to birth any new (interesting) development. You need a community of people willing, but more importantly, able to contribute. I'm sure you want things fixed in mIRC. I'm not sure you would even know where to begin implementing them. I'm sure there are tens of thousands of lines of code in mIRC (if not hundreds of kLOC). It would take a while for people to figure out the entirety of the existing code base before they could start adding new features. It would be a long, messy transition. Eventually, if people still cared, there might be something good that comes out of it.. between that period you probably won't get much in terms of quality. And then you won't be guaranteed that the maintainers would be any more productive than K is now. I'd rather not make such a gamble.
