[mythtv-users] Where to find Myth Setup Information?

[Simon Hobson]

Mark, I've also been the frustrated newbie so I have some idea where 
you're coming from. Can I ask that you calm down, take a deep breath, 
and try to read this email in a constructive manner ? I can see that 
you've got into a really negative frame of mind over this, but 
really, if you can overcome that you may be surprised.
For reference, I can't remember how long it was, but I think from me 
first dabbling to having a working system took me a couple of years 
(not that I was trying all that time). I'm not a programmeer, so when 
it fails to compile then I'm lost. Things have improved, and still 
are improving - back then I didn't have the option of popping in a 
MythBuntu disk !
But now I hardly watch anything other than via Myth - it **really** 
does have that impact on your viewing habits (even with it's 
deficiencies).

BTW - before I used Myth, I used EyeTV on an old laptop. That was a 
commercial, polished, 'appliance' package - and it was at least as 
frustrating to use as Myth !

mark Deneen wrote:

>Well then, you are the right person for me to address. You pitch the 
>product as an application, which it most assuredly is not. It's a 
>programming toy, and it ought to be advertised as such.

Really ?
The 'headline' text on the project main page says :
>MythTV is a Free Open Source software digital video recorder (DVR) 
>project distributed under the terms of the GNU GPL. It has been 
>under heavy development since 2002, and now contains most features 
>one would expect from a good DVR

I can see that you can interpret that as meaning it's an appliance, 
but it also says it only has 'most' of the features you'd want.

>1. The web site is dishonest, and leads people to believe this is a 
>working application of the type most people would expect today.e.g. 
>Install, Setup, Run. That isn't what this is and you know it.

Actually, it is. Pick the right packaging and you can do just that - 
MythBuntu being one such packaged example.
But as already explained to you, it is (and always will be) a complex 
bit of software to set up. I see where you are coming from - you are 
new to the project and you just want to click a few buttons and it 
will work for you. Sadly a project like this cannot EVER be like that 
for more than a small minority of users for the simple reason that it 
has so many options it has to support.

To see why, you need to look beyond your own requirements. For me 
(I'm in the UK), the setup is fairly simple - I have one DVB-T tuner 
and I needed to upgrade the kernel to get working support for it. Pop 
around Europe and you will find a few more variations of DVB-T, all 
subtly different. Plus you'll find people wanting to record from 
proprietary satellite boxes (Sky in the UK comes to mind) where the 
only option is an analogue grabber to re-digitise the analogue out 
from the receiver box. Once you get to that, then that's difference 
hardware and you also need to integrate a means of channel changing 
on the box.

Read the lists and it's clear that the guys in teh US are having a 
lot of fun. Not only are they having a digital switchover for 
terrestrial, but it seems the cable companies are cashing in on the 
confusion to screw their customers with a similar switchover on cable 
services. So you've got guys with multiple tuners & capture cards. 
Analogue tuners for some remaining analogue broadcasts, digital 
tuners for the new digital transmissions (terrestrial and cable), and 
analogue grabbers for the analogue outputs from the cable tuners they 
need to use to get encrypted/protected channels.

So hopefully you can see that there is a **HUGE** range of inputs 
needed to be supported. Note that Myth doesn't directly 'support' any 
item of input hardware - it supports anything that's supported under 
Linux and so sometimes there is a delay while other people working on 
hardware drivers catch up (often without support from the hardware 
manufacturers).

Then there are all the output options. Again, as far as Myth is 
concerned, if Linux can display a picture on it then it will use it - 
but there are still options to be configured for each display (and 
some of them may well require a knowledge of configuring X).

And then there are options for system topology. At one extreme you 
can have a single machine that runs everything - 
database/capture/storage/display. That is useful, but misses out on a 
lot that Myth can do. At the other extreme, there are users with 
multiple backends (capture & storage), networked storage (eg NAS), 
and multiple frontends (TV in every room). This is something that I 
don't think ANY consumer system will do - allow you to watch any of 
your media on any TV in the house independently of what anyone else 
is watching.

The downside is that such complexity will never be "point and click" 
for installation.

I know this sounds like a sales pitch, but I'm trying to explain 
where this project comes from, and where it's headed. If you 
understand that, then you are a good way to understanding why it's 
not been simplified down to an appliance - and probably never will be 
as a mainstream part of the project. It is not attempting to replace 
the consumer PVRs you can buy off the shelf - that would be a 
pointless project. It is attempting to be what they probably never 
will be - a system that can accept multiple arbitrary inputs, record 
whatever you want with powerful rules and scheduling, and play it all 
back on any device you have.

In project terms, it's still relatively young (note it's still a "0." 
version number).

>2. At a minimum, some effort ought to be put into making grammatical 
>sense out of errors and menus. What's in this program ought to be 
>embarrassing to any educated person claiming to an engineer. It's 
>loaded with unexplained jibberish. The horrendous misuse of language 
>in the program and the docs is simply appalling.

I can't say I've noticed, but I've almost got to the point where I've 
switched off to poor language use - are grammer and spelling still 
taught at school these days, because it sometimes doesn't seem like 
it.
The fact that you notice these things means you are probably someone 
better equipped to help in that area.

>3. The "docs" you are all so proud of are meaningless recitations of 
>the menus, which are unexplained faux techno-jargon. It is not a 
>How-To, or a User Manual. Find a professional technical writer to 
>volunteer. Programmers know nothing about technical writing, and 
>create nothing but hacked rubbish when it comes to user manuals.

There is the rub, programmers like writing code, and as you say often 
are not good at writing docs. "Find a professional technical writer 
to volunteer" sounds great - until you realise that you can't 
"volunteer" someone to do this, they have to do it themselves. That's 
a common problem that few projects ever get round.

Again, it sounds like you might be equipped to help here - and you 
have the advantage of being able to see what experienced people 
can't. It's like putting up road signs, the people that do it know 
where they are going and can't see when the signs are rubbish !

>4. A real engineer ALWAYS wants to know how customers experience 
>their products. And for SW, I don't simply meaning collecting bugs. 
>This Rube Goldberg contraption is a mess from any customer 
>perspective. You ought to be thankful I would waste the time to tell 
>you what is wrong with it. Instead you are defensive, as though the 
>mere act of applying hours of your free time is type of a substitute 
>for competence. It is not. The program is a hacked mess, and it 
>doesn't matter if you have 100 hours into it, or a million. Surely 
>you understand that, right?

As someone who's been following the project for some time I can tell 
you that they are cleaning things up. Something like this doesn't get 
built quickly, or come out 'right' in it's first incarnation. At 
first things are hacked together to see what works, and how things 
can work - user experience is waaay down the list after actually 
getting something to work at all. As time goes on, the basics get 
sorted, and people start to see what others want from it - when 
someone started the project, I doubt they even considered a small 
fraction of the things people are doing with it now.
As I read it, a lot of the development work now does go on tidying 
things up. For example, the current incarnation now uses themes for 
the use interface - thus making it a lot easier for someone to 
customise the UI by writing a theme without having to change any of 
the underlying code. Networking is also a lot easier - previous 
versions had some "interesting" requirements to make multiple 
frontends and backends work together for all the different media 
types.

Things are improving, but no-one is paying for this, and so it goes 
at the pace that people are prepared to put their own time into it. 
That is the nature of such development.

>5. You've spec'd a project that is 10X larger than you could ever 
>competently complete with part time coders, and it shows. All I can 
>see that looks anywhere near finished is the menu appearance.

It's got to the size it is because different people want different 
things. Yes, it could be a lot simpler if those running the project 
simply locked down the requirements and (for example) didn't support 
anything but US users using US TV systems that are open and 
unencrypted. But then the project would probably die off because it 
would exclude so many people.

>Get as haughty and insulted and defensive as you like. It makes me 
>laugh, actually. I may be the dumbest user you've ever heard of, but 
>I have 5 computers and dozens of large applications I've purchased 
>and installed, and please trust me on this - - none of them have 
>this level of McGuyver bailing-wire nonsense associated with their 
>installation or use. Have I ever had to recompile parts of a program 
>to get it to work? Uh, no.

Sorry, then you are clearly in the camp Microsoft aim at - never mind 
the quality if the user can randomly click boxes and get something 
that "sort of works". That's not intended as an insult, it is their 
primary market.
What I can tell you, from shoulder surfing with the guys who do this 
at work, but a once you get past the basics you can do with tick 
boxes, configuring a Microsoft server can be just as arcane and 
unintuitive ! That is what comes with complexity.

>I was happy to sign off a couple hours ago by saying, sincere 
>thanks, and sorry for wasting your time (and mine). But now, this 
>hyper-defensiveness of the "developers" has me amused. I am assuming 
>you developed this thing for yourselves? Because if you ever had 
>intentions of having customers (paying or not) you'd want to hear 
>what they had to say. It's free!

Well actually, most of them do do it for themselves. That is the 
underlying theme of many large projects - the guys doing the meat of 
the work do it to scratch a personal itch. There's never (as far as I 
know) been an intention to make it into something you could "sell" - 
though there are a great many developers who are working on improving 
the user experience that comes from it.

And I'll finish off by adding that people who land on this mailing 
list and ask sensible questions do get a lot of help. Seriously, I've 
seen long threads and some serious support time going into solving 
users problems. But that's for users who come along and ask "how do I 
do this", or "I don't understand what these options mean, can you 
help me" (cf your complaint about the options to be set for capture 
cards).

The surest way to not get help is to arrive at a part built project, 
shout out that it's a pile of manure, and generally go all out to 
p*ss off those who might otherwise have helped. It's like walking 
onto a building site where the roof isn't on yet and complaining that 
the wallpaper isn't up yet.

Trust me, the developers do listen to what goes on, they do listen to 
what user want and what they have trouble with. It just takes time to 
get there.
Yes, it's got areas that can be improved, but it IS usable with a 
little effort.

-- 
Simon Hobson

Visit http://www.magpiesnestpublishing.co.uk/ for books by acclaimed
author Gladys Hobson. Novels - poetry - short stories - ideal as
Christmas stocking fillers. Some available as e-books.