[mythtv-users] [mythtv] BrowserBased setup

[Robert McNamara]

On Wed, Nov 24, 2010 at 12:37 PM, Cecil Watson <knoppmyth at gmail.com> wrote:s
>
> Having read the -dev archives it seems, most of the discussion is in regards
> to using Silverlight.  I'd ask another question.  Why move to a browser
> based setup?  Seems the only argument is "there are quite a few of us that
> have headless backends".

Note that while we will be offering a browser based setup experience,
we have been quite clear that we are building a setup *API*, upon
which we will offer a Browser based setup.  There will be nothing
preventing one from writing a MythUI based setup, a silverlight based
setup, a flash based setup, a perl CLI setup, or anything else, so
long as it speaks our standards-compliant setup mechanism.

> Since it's inception, after an install and reboot, KnoppMyth/LinHES has
> automatically started mythtv-setup.  Now, it has been a while since I've
> looked at other distros and I seem to recall reading MythDora has gone to a
> browser based configuration.  My opinion, is that a MythTV setup should be
> looked at like an appliance.  Most everything that needs to be done should
> be doable with a remote via a "10' UI experience".

I have more than a passing interest in writing a MythUI interface to
the new setup API.  AFAIK Stuart Morgan does too.  Moreover, any setup
interface can offer as much or as little of the setup options as it
likes, and organize it however it likes.  For me, this would be a 5-6
step extremely simplified setup.  We've never made any respresentation
that the browser based setup would be the only acceptable one, just
that that's where we're choosing to focus our efforts for the initial
run at it.

> Now, I realize, that BrowserBasedSetup is in it's infancy.  However what
> immediately comes to mind is instead of running mythtv-setup which I can
> fully control with a just a remote, I'll have to launch a browser.  Will
> MythBrowser work?  Will it work well with just a remote?  If I have to go to
> another system to connect to the web server running on the backend, instead
> of making it simpler, another layer of complexity has been added to setup
> process.

The browser setup probably wouldn't work well with the remote.  A
remote-based stup would be easily writeable, however.  I put this
functioning setup together in a weekend last year, and it didn't even
have a setup API to plug in to.  I'd likely pick up the work again
once the API is done:

http://www.fecitfacta.com/simplesetupexample.ogg

> Granted it is a rarity that one should have to run mythtv-setup, moving to
> an all browser based setup seems short sighted.  Whatever issues exist in
> setting up MythTV (distros aside), should be resolved in mythtv-setup.  If
> desired, a broswer based setup should complement mythtv-setup not, replace
> it.

mythtv-setup is, for my money, a total write-off.  It is so confused,
hackneyed, and broken that retrofitting it for the new API is a non
starter.  However, if someone decides they love mythtv-setup in its
current for so much that they're prefer to hook it into the new API,
they can do so and host the code themselves.  I just don't see us
doing it.

On the list of things that I don't see ever, ever happening is us
writing a Silverlight interface ;)  But the OP seems quite into it,
maybe he can scratch his own itch for the tens of Windows MythTV users
out there ;)

Robert