Basic rationale is in Bug 538396 (Comments 0, 4 and 9). Net::IRC is unmaintained and we need to move away from it.
In 538396, Wolf asked: >What areas have you seen that would be improved for us? You mentioned the >amount of code that could be eliminated, trying to get an idea where, as i'm >not familar with what is different between the two modules. These are some of the things you are implementing on top of Net::IRC which POE::Component::IRC would handle for you: * Joining channels automatically (on connect, rejoining after kick, etc) * Reconnecting to servers (optionally withserver list to cycle through) * Automatic CTCP replies (VERSION, TIME, etc) * Flood protection / message queueing * Parsing user and channel mode changes * Changing to a temporary nickname when needed (e.g. user -> user_) and reclaiming the desired nickname when possible It also sports a sophisticated plugin system, which you may or may not want to use for the bot modules.
#(In reply to comment #1) > It also sports a sophisticated plugin system, which you may or may not want to > use for the bot modules. I'd like to see the existing module system staying or at least provide legacy support for them.
Wolf: How will this impact Bug 531785? Should finish the work in there and then move to this, or scrap that work and redo it after we've made this transition?
(In reply to comment #3) > Wolf: How will this impact Bug 531785? Should finish the work in there and > then move to this, or scrap that work and redo it after we've made this > transition? If possible, finish that work first, particularly if its reasonably close to being able to land. This is a priority for 2.8, but not critical enough to trample over other work in progress.