Version 3 (modified by anonymous , 15 years ago) (diff)


Possible Future Development Tasks

This is not a general feature request list, please don't turn it into one.

If anyone sees things they'd like to work on, please email the mythtv-dev list for details/questions/etc.

  • A more standard communication protocol between the frontend/backend/etc. Currently uses a custom protocol, and it would be nice to replace with something that's more of a standard. Note: this really needs to be fairly efficient, since this is a common task. Parsing XML queries/responses several hundred times a second really isn't a good thing to be spending CPU time on. dbus a possibility?
  • uPNP support, both server and client. Interoperability with CE devices, as a lot of them are starting to use uPNP for remote video playback. (David Blain)
  • Multicast streaming of video/live tv.
  • Finish up Thor's work on mfd (Myth Frontend Daemon). This would handle video streaming between machines, audio playing (music everywhere, not just in mythmusic), media importing (cd ripping), burning, etc. The frontend process becomes more of just the UI.
  • zeroconf/rendevous/bonjour/avahi/whatevertheycallitnow support for the frontend/backend. Less setup work. mfd already uses this. Move the backend into a mfd process? Could make the scheduler (ie., the master backend) be auto-promoted from existing backends on the network.
  • Live-tv rework - #340.
  • UI rewrite - #12. This actually encompasses more than the UI itself - one of the intents is to sanitize a lot of the code that's grown over the years, contributed by different people. It's not terribly cohesive as it is. Big task, I intend to write a larger doc once I have time for this.
  • Abstract the SQL database somehow? People doing embedded platforms want this, but it gets tricky due to access permissions/etc. Might be better to make the frontend ask the backend for stuff, but it seems rather unwieldy to potentially bounce a query through the backend to the sql server, especially since that might be on a completely separate machine.
  • New website. I hate updating the current one. So, it doesn't get updated often.
  • Port to QT 4.x. Need to wait until the major distros have this new version of Qt included. We will probably no longer be able to support QT 3.x.
  • Possibly get rid of some of the various plugins by incorporating them into the core app? Might allow the user interface to be consolidated more.
  • Improve developer documentation by doxygenifying all the major classes, and the major subsystems of MythTV.
  • Merge DVB and HDTV recorder MPEG stream parsers. (Daniel Kristjansson)
  • Add MPEG stream parsing in the frontend to support three functions:
    • Auto-select of subtitle and audio based on language and number of channels.
    • Auto-detection of useful portion of the video for the aspect ratio and framing.
    • Handling DSM-CC based interactive television.