I'm thinking of launching a MythTV distro website

[jarod-ajLrJawYSntWk0Htik3J/w at public.gmane.org (Jarod Wilson)]

On Dec 18, 2009, at 2:29 PM, Rich Braun wrote:

> Jarod Wilson pointed out:
>> But wasn't your core issue that mythtv wasn't stable enough? That
>> sounds like a major technical issue, so your technical approach seems
>> rather key to me...
> 
> The basics are that I start with a fixed set of hardware,

*THAT* is the key piece I was looking for. If you do that, sure, you can have a much more stable and consistent experience. Until your vendors stop carrying the parts you were using. Or switch the silicon shipped under the exact same part number. Or the parts are simply discontinued. :)

But at that point, you could of course move to functional replacement parts. Its still an issue to keep in mind though.

> that I have a great
> deal of problem/solution notes, that I set up a knowledge-base based on
> Atlassian Software, that the underlying base system is OpenSUSE 11.2, and that
> I push out the configs from a central server using Capistrano, puppet and
> other tools.

Not that I have anything against OpenSUSE, but if I were really serious about making this as bullet-proof as possible for as long as possible, I'd use one of the Red Hat Enterprise Linux clones or an Ubuntu LTS release or something like that. Not sure if anyone has ever tried a SLES clone, or if Novell even makes it possible...

> So the technical approach is /not/ the interesting part of the project.

No, not really. I'd wager that any of the MythTV distros could decide to only target specific hardware and achieve the same stability.

> The
> fact that none of the other distros take this approach isn't interesting
> either.

There actually have been attempts at creating a "certified supported/functional" hardware package by, iirc, the guys at mythic.tv, which was designed for knoppmyth (again, iirc).

> Gathering a devoted group of volunteers to keep the distro current as
> time goes by, and to build more enthusiasm among novice users, is the
> interesting part.  At least to me.

That is interesting to me as well. Unfortunately, my experience with MythDora is that very few technically competent people actually have a significant desire to spend their free time improving software that is already working for them for people who are... Well, not technically competent. It takes a LOT of dedication, time and effort, to fully flesh things out -- at least if you're building a general-purpose, runs-on-all-hardware distro. Might not be so bad if you limited things to say, only a specific ION platform.

> Being able to just plug the thing in and
> have *everything* work (remote control, HDMI audio, S3 suspend, non-jittery
> 1080p, the Schedules Direct stuff--*all* of it!) is what will interest others.

Yeah, that definitely appeals to a fair number of people. Myself and a few other MythTV guys, including Chris Peterson when he was working at Silicon Mechanics, actually talked about trying to have a pre-built package made available by SiMech. However, there were two major obstacles:

1) very low margin on the hardware
2) legal hurdles in the form of the MPEG-LA

Are you prepared to pay codec patent licensing fees to MPEG-LA, and/or have legal counsel on retainer for when they come after you for patent violations? This almost immediately sucks up any margin from the hardware, so a support subscription fee model might be the only way could could keep things afloat. (That, and/or trying to resell SchedulesDirect data via some sort of bulk purchasing deal?). Or are you just going to sell the hardware, and leave it up to the user to go download the software and install it? (And also note that technically, you may well be violating the codec patents by simply distributing the software in the US, so you'd need an offshore hosting provider...)

>> From your comments it doesn't seem to me that you're in the target audience of
> my appeal:  seeking motivated people who either want to volunteer to handle
> some of the technical stuff, or perhaps to lead a new non-profit (but
> money-collecting, a la Schedules Direct) organization.

I don't have time for all the things I'm already working on anyway, but no, I'm not part of your target audience.

-- 
Jarod Wilson
jarod-ajLrJawYSntWk0Htik3J/w at public.gmane.org