Quote:
- It would steal developer time and energy from the core PVR and media server functionality.
- It would add extra documentation and support overhead and complexity.
- It would increase the complexity and thus the fragility of the system.
- It could adversely affect the primary function of the system as a media server and PVR.
Ditto to this.
Stable, Available, Secure, Updateable multipurpose media-centric networked appliances, with Backup on inexpensive and often recycled hardware and open source software.
Expandable functionality is not as important as availability due to stability and security updates with only periodic core functionality updates for new features. In other words, sacrifice multipurpose for maximum uptime and great useability. Sounds like a marketing thing but I think that its for geeks to build but regular people to use and even geeks don't want to spend to much time keeping it up to date. Adding a destop to this seems like asking for trouble for little gain.
Even though I love knoppmyth, keeping my ubuntu dektop working well is far easier than keeping knoppmyth running. I could only imagine that adding a destop to knoppmyth would make it even tougher even if the multimedia stuff is more a beta thing than most desktop software.
However, part of availability, means being compatible with other desktops. Therefore, I could see a use for a package that an ubuntu desktop user might install that could allow their computer to be a part time backend-slave recorder for a knoppmyth master-backend server. As long as it had the option of being set to be available for recording or transcoding only overnight or when the computer is not being used.
Figuring out how this mash up of programs, drivers, scripts all fit together nicely with almost any hardware is why knoppmyth is so great. I don't want to have to reinvent the wheel and install all of my own packages and configure everything. But that is OK since knoppmyth already includes a good combination of software that makes a seamless presentation of functionality that an average person not familiar with mythtv can easily use is important ot me. There is still plenty of stuff for us to tweak and fine tune and upgrade with out reinventing the wheel. Being able to upgrade between versions more easily would be nice. Security updates would be key as well so we can open this up to other networked and mobile devices.
I do consider whether I should move the master backend to my main debian server that does other stuff and is on raid5 and hasn't had any downtime in years. However, I am happy to use knoppmyth since each frontend is a backend as well and the distributed nature helps reduce cpu/disk waste since each machine can help out.
I just have to think that there is a lot of room for improvement of knoppmyth and it will still be the best distribution for a standalone frontend/backend solution to connect to TVs for the most part.
It is also probably still the best for a master backend in a lan and with the new distributed storage infrastructure all the slave backends will become more valuable as a place to store and transcode stuff.
Anyone who wants to watch/listen to mythtv media from their desktop can do so easily now with samba pretty exported and/or the web interface. They can use Remote Desktop or VNC as well to configure some settings and maybe even watch stuff in a RDP window someday since windows2003 RDP can do a single program in window.
But today, the web interface with full streaming from samba shares is slicker in a lot of ways anyway and should get better wit ajax or flash in the future. This way is better than a knoppmyth desktop version could ever be in my opinion. No offense, but ubuntu, windows, and mac are so far ahead of what you may come up with in packaging together good desktops (they are easily installed on a wide range of hardware with ubuntu at least)
I would concentrate on what ubuntu doesn't do well first which is getting things that aren't already packaged packaged and preconfigured for the most part to work together in a good mash up. I think installing tested packages is far superior method than a livecd reinstall, so hopefully much of the scripts and things could be installed the debian way one day. But untill then it is such a huge advantage to have knoppmyth.
Thanks Cecil and others. Things are changing so fast these days that it is hard to imagine knoppmyth being here forever but I hope so. I'm glad that you are open to listening to what the community thinks are good features to add.