| Authors: | Andrew Wilcox |
|---|---|
| Version: | 1.0 |
| Status: | Draft |
| Copyright: | © 2015 Andrew Wilcox. NCSA open source licence. |
The current IRC daemons available for production use today are lacking in many areas: features, compatibility, reliability. A new, modern IRC daemon is not only desired, but needed.
As the IRCv3 protocol moves forward on standards track[1], having an easily extensible server will allow us to capture more market share. Implementing new features of the protocol first will be a defining trait. It will also enable the protocol details to be hashed out and made better, making a lasting impact on the larger Internet community.
[1]: http://ircv3.net/
- Implement the entire IRCv3 protocol.
- Be production-quality by Q4 2015.
For peer-to-peer chat systems, GlorIRCd is an IRC daemon that will scale well, be fault tolerant, and compliant with bleeding-edge standards. Unlike other IRC daemons, GlorIRCd is designed to be easy to extend and maintain. GlorIRCd is also designed to be highly fault tolerant for mission-critical chat systems. It is also designed to operate stand-alone, without an extra "services" package.
- Modular core allowing for new features to be written and tested easily, and allowing for on-the-fly patching for high availability.
- Standards-compliance with IRCv3.
- Fault tolerance via a different form of server linking.
- Nickname and channel registration is handled inside the daemon.
The initial release will focus on the core and linking protocol. Therefore the initial release may not be immediately production-ready. It will be a testbed for ideas and innovations in the IRC ecosystem.
The second release will focus on extremely high availability and production scenarios.
Continued evolution in the design and linking protocol are anticipated. Additionally, more IRCv3 extensions will likely be introduced.
Linking to other IRC daemons is not a goal.