From: Paul Louden <paulthenerd_at_gmail.com>
Date: Sat, 21 Nov 2009 15:05:19 -0600

Sam Pattuzzi wrote:
>
> If you open up a playlist, then use the menu button it shows a context
> menu, it doesn't bring you back to the main menu.
>
>

The playlist viewer happens to actually technically be a plugin, so this
isn't really fixable as the situation currently stands if I understand
correctly. It has to do with how the viewer is handled.

>
> It sounds fair to me that hold shouldn't really do anything but as you
> said, some plug-ins use all the keys. Maybe as a plugin starts, if it
> has some unusual exit pattern it should say so. I'm interested in the
> current attempts at consistency, have you got any links?
>

Most of this was done a considerable time ago. If you search back in the
changelog far enough you'll find efforts to make sure many plugins exit
the same way. There's no concentrated effort though; plugins are often a
secondary concern.


> I find that an unnecessary insulting statement. I don't think that a new
> user (given that Rockbox is installed for them) should have to read the
> manual to figure out how to use it, a process of trial and error should
> be enough really.

If Rockbox is installed for them, whoever did the installing is
responsible for letting them know the controls are complex. Rockbox
doesn't currently come on any players, and we strongly encourage people
to read the manual. Trial and error is enough to use the basic
functionality of the player. There's very, very little functionality
that can't be found through trial and error, honestly, but the manual
simply streamlines the period of time it takes to learn about things. I
don't think "the manual is unnecessary" is ever going to be a goal. It's
simply unrealistic, and your definition of usability doesn't scale to
software as complex as Rockbox.


> I take the point that more key mappings are needed than actual keys are
> available. I think at the moment Rockbox has it right in that only power
> user stuff is on these long press keys. Maybe some visual cue could be
> useful in letting the user know they activated the long key press.
>

I don't understand. Is the expectation that the user doesn't know
they're holding down a button?

> Sorry about that, I realise that looking back at it. I will try and make
> future posts more concrete.
>

Thanks. I know there are usability concerns, but the effort should be on
where things don't work, or where things work in a way different than
other parts of the software have trained you to expect. I would strongly
suggest dropping the idea that "a user should always be able to guess"
though. Most single, short presses should be guessable and expected
based on button labels and/or behaviour taught in screens leading to the
screen the function is in, but long presses and combinations will always
either require the user to guess or read the manual.
> Sorry, don't know what OF stands for.
>

Original Firmware. For discussion you may wish to reference
http://www.rockbox.org/wiki/ProjectGlossary for looking up things.

>
> Maybe not all of them, but even at the moment it is not too hard to
> guess most of them.
>
If it's not too hard to guess most of them, what's the problem?
Received on 2009-11-21