[mythtv] [mythtv-commits] mythtv commit: r13275 by stuartm

[Dan Wilga]

At 2:14 PM -0700 4/21/07, Bruce Markey wrote:
>I look at the what's new list almost every day which is why I
>wrote it. Even if I hadn't, I'd be familiar with what is or isn't
>included in the list by now. And even if I didn't notice, if the
>person who wrote it told me that there is one entry per title,
>I'd pay attention to that rather than continuing to misrepresent
>the facts to someone who knows better =).

Yes, you are correct that, in the case of New Titles, there is only 
one entry per showing. I had typed my original statement without the 
benefit of actually having a copy of Myth to run here to check with. 
However, I have since looked at the various search screens and found 
several that I believe many users would most often prefer to sort by 
title--most notably Categories. My original assertion still stands, 
despite the fact that one of the examples I gave was incorrect.

>The point is that
>any user, regardless of habits, should understand and trust
>what they are seeing every time they enter the page regardless
>of what they are searching for, how they got there or what they
>did last time.

[and later on you also say]

>It may not be obvious what the sort order is and that there may
>be more showing available.

I agree with both of these statements. I have often thought that it 
was not as easy as it could be to distinguish which sort order was in 
use, even when the default is always Time.

Would it not, therefore, be better to try to come up with some sort 
of user interface change that would improve the visual feedback to 
users, so they can see what sort method (time/title) and direction 
(ascending/descending) is being used? This would have benefits even 
without my proposed change to store the sort type.

Perhaps the relevant section of the string "(1) time (2) title" could 
be highlighted. Or perhaps an even more radical change would be 
better, such as changing the background appearance of the entire list 
depending on its sort mode. These are, of course, just wild 
suggestions as my knowledge of the possibilities supported by the UI 
code is far from complete.

>you've not yet acknowledged that my point is that even
>though the code is trivial, it is a bad idea.

I am acknowledging that it is a bad idea based on the current UI of 
Myth, for the reasons you gave. That does not necessarily mean that 
improvements cannot be made to the UI that would accommodate the 
change I'm proposing.

Perhaps the needed UI changes are too radical, and simply not 
feasible. But I do think it's worth discussing.
-- 
Dan Wilga                                                        "Ook."