[mythtv] MythWeb

[Gary Buhrmaster]

On Wed, Jan 29, 2014 at 4:59 PM, Steve Campbell <orcusmaximus at gmail.com> wrote:


Hmm, lets look at these points.

> Hmm. Lets look at those points
> a) You should always use the best programming language for the job.
> PHP is much easier to maintain/develop in than C++ (at least for PHP
> devs!), and any good webapp will be making as much use of Javascript
> as possible.

I will agree that if you have a language hammer, everything
looks like a nail.  But if you have no manufactures of that
hammer, the entire issue is irrelevant.  We have someone
who is investing (significant) effort to make this transition.
The existing MythWeb application has been lagging due
to lack of committed development efforts for years.  The
patches (and we should thank those for those efforts) have
just repainted the old house.  It needs significant work.

> b) The integral web interface and frontend seem way outdated compared to Mythweb

If you mean what you see now in 0.27 (or less), that is
true.  Have you tried master?  Is it perfect?  No.  But
it is moving quickly, and is already better in some ways
(and clearly still needs work in others).

> c) Is handled by the distribution package creators. There are good
> arguments for keeping things modular.

Modularity has advantages, and disadvantages.  One
need only look at the Linux kernel to see examples
of various choices.  Perhaps you want to transition
to Hurd.

What is important to most (except for the people who
are maintaining the actual code) is the result.  I believe
the result is going to quite good.

If one had a large team of devs, each doing their
what they do best (some doing the infrastructure,
and some doing php dev, and some doing perl
and some doing python), one could perhaps choose
the best solution for every viewpoint.  Given the
(real) limitations of the project, improving the
capabilities of the MythTV FE today requires doing
things (at least) twice, once in the (usually) FE in C++,
and then (usually) again in MythWeb in PHP.  That
is not entirely scalable, and leads to missed updates
and functionality.

Now, if you are a good PHP programmer, Michael
Dean has previously stated that there is a need for
a (modular/separable) web front end to allow
"Internet" access with security (authentication
and authorization).  Willing to volunteer?