From: Paul Louden <paulthenerd_at_gmail.com>
Date: Sun, 07 Oct 2007 13:25:07 -0500

> But I don't believe that the question is about support, coding effort,
> documentation problems, or anything else, but the lack of interest from the
> devs. This is pretty acceptable, since this is not a feature which all users
> demand.

It's not merely a "lack of interest." People have coded patches that
allow customizable keys. If it were merely a lack of interest, then were
someone else to do it, it would get accepted.

It is a specific desire on the part of the key developers, specifically
those who choose the direction of the project, NOT to have customizable
keymaps in Rockbox.

It is also a feeling among many (possibly the majority, but I can make
no guarantee) but not all of the developers that configurable keymaps
generate more trouble than they're worth.

They make code complex when it is not necessary to do so, make support
more frustrating (people already have enough trouble with the concept of
"unsupported builds", it adds an extra support question to verify if
they're using default keymaps with an official build), and simply
increase code size when the ideal is still to keep the Rockbox binary
slim where possible.

It's really a situation of "It's more trouble than it's worth, and it
creates trouble for us, but worth for you." When you think that it's the
"us" that are working on it, you can probably see why they don't feel
it's good to generate trouble for themselves, especially if they've
decided the worth isn't very much.

And arguing "but we say it's a lot of worth" isn't going to change
things. Not really. If you feel there are problems with the keymap,
please recommend how the main keymap can be dramatically improved. If
it's worth "a lot" then surely there must be dramatic problems with the
existing keymap.
Received on 2007-10-07