Antony Stone wrote:
> I guess I'm still working
> from the premise that the more people want a feature, the more effort the
> developers might like to put into implementing it, though.
First I'd like to clarify that there is no offense intended in my
This is a silly assumption. Firstly, even if users were the target
audience, they don't have the technical knowledge to understand what is
feasible, how time intensive an idea is, and so forth. Even if an idea
is worthwhile, the idea most interesting to the largest number of users
(or rather, the idea the largest number of users make noise about) is
often one that will either take ridiculous amounts of developer time, or
one that will have negative impacts causing many developers to say "let
some other dev code it."
But Rockbox is open source. Not only that, but it is entirely
independent from users. I mean this in the most literal sense: it does
not depend on them. The "essentials" of the project are donated, almost
entirely by the developers themselves. There is a fund of donations from
the users, yes, but without it the project would still be able to
continue. Why then should the interest of the majority be a major
concern? Yes, we like to make things better for users. But why should a
developer, when given the option, say "I want to code this feature that
70 people who aren't willing to donate their time want, rather than this
features that I personally think is a better fit, when I'm the one
donating my time?" Most of the people working on Rockbox are doing it
because they personally think that Rockbox is a great thing, and they
have interest in making it a greater thing while either entertaining
themselves or learning. That is, they have a personal itch to scratch.
In some cases that itch could be public recognition, but I think with
our project there are few or no people who are in it for that. Instead
the majority leans toward "make Rockbox turn my MP3 player into the best
one possible" and of course every dev has his own vision on this.
Usually this includes a degree of user friendliness, but there's also,
somewhere in probably any dev, a feeling that "users don't always know
where time should be spent, because they rarely look at the whole picture."
Vetted feature requests, then, would help filter ideas because you've
had at least one developer agree that "I think this is good for the
project." As it stands, the current tracker is full of "public opinion"
but in the views of many developers, not full of "good sense."
Received on 2008-06-22