[mythtv-users] MythTV core structure.

[Steele Price]

<snip>

> The MythMusic tree & playlist layout does just this.  It's acknowledged
that
> the interface is somewhat sub-optimal, so if you have specific about UI
> improvements that a 'non-geek can use and be happy with', please suggest
> them.  Your comments so far are too general and seem to indicate a lack of
> familiarity with how Myth works & what it can really do.

Indeed - I am a n00b to MythTV, out of shear frustration with everything
else I have tried (mostly windows stuff and freevo) I may come off somewhat
annoyed.  Also I am wholly frustrated with the state of my current hardware
and am somewhat(read completely) overworked.

> > When I have more collected architectural concepts I will present them.
> > Resistance to change and defense of the current system is fine,
> > but I really
> > do think I can help make this a better system for everyone, I just need
to
> > present them here in a better manner.

> Agreed; I get the sense that your goals are somewhat in line with goals
for
> future development that have already been expressed.  This project is not
> resistent to change, it's just that you need to understand how we got
where
> we are and what the priorities are for the road ahead.  The general
> direction of development is for Myth to become a media hub much like you
> seem to suggest; it's just that these things take time and the primary
> developers' priorities may not be the same as yours.  After all, it was
only
> around the beginning of this year that the current frontend/backend
> architecture was put in place.  The two biggest things you can do to help
> out are specific (or at least, not overly general) suggestions for
> features/improvements, and more importantly, contribution of code.  Most
of
> the primary active developers spend their time & talents on the core
widgets
> and TV functionality.  Most of the other modules have only a single
primary
> developer, and some don't have anyone currently responsible.  There are a
> lot of great ideas circulating, and only a limited talent pool to
implement
> them.

I do indeed planning on helping in every way I can as I intend to make this
my primary system at home.  I AM a professional developer, albeit not in
C++.  I can read and follow the code, but I am certainly NOT a C++
afecionado.  I tend to work with other languages, but I have a desire to
latch onto projects in others to expand my horizons.  I can contribute alot
in the form of XML and SQL as those are 2 of my primary languages.  I have
been developing client/server applications for many years.

I seem to have totally misunderstood the front/backend scenario currently
being implemented.  I see now this applies strictly to TV an not to the
other modules or the databases as a whole.  I just assumed that
(incorrectly) since most of my client/server implementations are that way.
While most of my database work is either done on other platforms, or in php,
I can certainly apply the same principle to any language given time and
desire.

> > Don't misunderstand me, I think MythTV is great, I only want to make it
> > better, not rewrite the whole damn thing.

> Please don't interpret this as a flame.  I would like to see the same
things
> as you: transparently sharing media files through Myth without having to
> configure NFS or Samba on all the frontends.  It's just that we need to
> collectively come up with a scheme for doing so that fits in with the Myth
> architecture and the talents & time of the developers.

> -JAC

Actually I think that I just need to look at some more core linux management
capabilities such as afs and yp/nis to see if there is a way to implement
those on the file server and just apply some config file info from the
database server into them.  I wasn't actually implying that Myth should
replace any of the already available file systems, just assist in their
configuration for people who could care less (like my mother) about how this
stuff works.  There is also the problem of applying this to various flavors
of linux in transparent ways, that is indeed the whole point I was trying to
make, so some guy posts up an excellent article about how to install xyz in
RH amd Gentoo, but I use debian and it just doesn't apply very well.  Some
of my comments which may have seemed directed at Myth should probably be
applied to linux as a whole.  After all linux is still the choice of hard
core users, not joe average.  So I will just drop this filesystem track as a
whole for now and start using the system as it is.  When I have something
more constructive you will hear it then.

Steele Price
CTO
Digital Dreamshop
http://xtcp.net