[mythtv-users] External channel change not passing channel

[f-myth-users at media.mit.edu]

    > Date: Mon, 02 Apr 2007 15:44:49 -0400
    > From: "Michael T. Dean" <mtdean at thirdcontact.com>

[This might be moot, based on the "new interface" below; if not,
I might write a patch, but only if it's not a dumb idea.  See below.]

    > This leaves only Video Source and TV Format.  I guess I just hope 
    > they're scary enough to most users that they don't touch them.  ;)

...or they belong on your "page 2" :)

    > Probably the "rest of the story" behind page 2 being a separate page is 
    > that the items on page 2 only apply to V4L capture cards.

Aha!

    >     						       With digital 
    > capture cards, there is no page 2.

It seems to me that it should be possible to have channels that might
be simultanously capturable by both a V4L and a non-V4L card, which
means the existence of this page 2 is in a superposition of states,
but that's a detail (and I'm probably wrong about this).

    >     				Therefore, moving Video Source 
    > and/or TV Format to page 2 would create a page 2 for digital users where 
    > one didn't previously exist.  However, in my world, I like to think of 
    > page 2 as "not for general use" items.

Hmm.

[Note that I'm not necessarily advocating moving VS or TVF to page 2,
since I've just ignored them completely.  Your "no user serviceable
parts inside" stance would argue that they be moved, but OTOH your
"only-page-2-for-V4L-cards" would argue that they stay on page 1.
It does appear---in my case, anyway---that moving them in order to
make -room- for freqid on page 1 is not necessary, 'cause there's
all that blank space at the bottom.  I'm not sure if that's why you
brought it up or it was just because we're discussing grouping.]

    > OK.  I stand corrected.  The only reason (I know of right now) to change 
    > that value is if you're getting garbage data for your listings.  ;)

Yeah.  I wish DD was better about stirring my listings.   See below. :)

    > I wasn't trying to imply that what you had done was wrong (I didn't even 
    > remember what you had done or why), only trying to discourage users 
    > reading this thread from editing the freqid.  I apologize for the 
    > overly-definitive blanket statement.

No problem.  And your advice in that case was exactly correct.  -My-
problem with this particular bit of UI is that it's the first and only
time I've had to use the channel editor---and putting the freqid on a
separate page from all the other boxes-with-numbers-in-them made it a
total PITA to use.(*)  (After all, I was doing a big cyclic permutation
of a bunch of channel numbers, which meant that I was constantly
flipping back and forth to see "okay, what -was- the channel I'm now
mapping over -here-?" and putting them on the same page would have
made this trivial.  As it was, this seemingly-trivial task was more
like a game of concentration, given that I was also looking a pair of
web pages on another display to see which channels had to be remapped
to what.  It would have been easier if I'd made up a big checklist on
paper and carefully checked off each thing as I did it, but that's
what good UIs are supposed to obviate... :)

I'm -still- not sure it makes sense -to users w/V4L cards- (it sure
wasn't clear to me why things were on which pages), but at least it
seems there's a rationale there, maybe.

(*) And this total PITA was partially responsible for my use of the
phrase "extraordinarily poor" when talking about this part of the UI.
I'm sure a more temperate description, such as "seemingly more
confusing than necessary" would have been a better choice to avoid
riling those who are easily offended, but I can't fix that now.

    > > Of course, the -real- question is, "what does an inexperienced user
    > > think about the layout?" since any experienced user knows exactly
    > > where everything is and how to find it and whether there's anything
    > > missing.  This is why domain experts are not the best people to
    > > evaluate whether information is being presented clearly to non-experts.

    > In SVN trunk, there's a new "live" channel editor.  You press 'e' while 
    > watching a channel and can edit the channel info right there (making it 
    > much easier to identify which channel is which when putting in 
    > XMLTVID's).  In theory, this is the channel editor users will use in the 
    > future.  If so, maybe the whole mythtv-setup channel editor will go away.

That would be most excellent, and moots the whole discussion.  Of
course, the question then is, "when does the m-s CE get dropped?"

(If it's in fact -not- moot and it's likely to stick around, I could
just write a patch to move freqid to page 1, since that scratches my
particular itch and seems much simpler to test than some of the other
functionality I've proposed---but if the devos think as you do,
there's no point in my bothering, since if nobody wants it moved
the patch won't be accepted.  Do you have any sense at all of how
likely it is that someone would veto such a move?)

Random question:  If your channel mapping is messed up such that you
can't even -tune- the channel you want to change, will "e" still work?
I haven't thought this through carefully, but it seems like there
might be circumstances in which you could get stuck w/o something
other than "e" to tweak things with.

    > For me, the /only/ thing I do/have ever done in the channel editor is 
    > set an XMLTVID for digital channels.  Setting other user options (as 
    > enumerated above) is perfectly fine for those who want to do so.

    > I'll agree there's room for improvement, but finding time for 
    > improvement (especially with a whole new UI for channel editing coming 
    > into play) can be challenging.  I think this is much of the reason that 
    > the "one-time-use" mythtv-setup "niceties" are not high priority.  (I 
    > know, "one-time-use" is an exaggeration, but it's still much less-often 
    > used than other parts of MythTV.)

I agree.  Unfortunately, of course, "setup" tasks are (a) the first
thing a new user sees---and hence often doesn't know enough to do the
task well at all, and (b) are done so rarely by, uh, "most/normal/typical"
users (take your pick) that their intricacies are easily forgotten.
That unfortunately makes the -need- for their UI to be done well even
greater, while tending to kick their perceived -priority- through the
floor ('cause they're so rarely needed---and because those best-suited
to fixing them are least likely to see a problem).