From: Neon John <johngd_at_bellsouth.net>
Date: Mon, 22 Mar 2004 17:54:49 -0500

On Sun, 21 Mar 2004 11:49:59 -0800, "David H. Straayer" <mail_at_dhstraayer.com>
wrote:

>Ok, several of you have taken to the idea of soliciting Asian
>manufacturers to make players for Rockbox.
>
>My idea is this: let's work on a magazine article for placement
>in design magazines in Asia. The article is likely to welcomed
>by the editors of these magazines, and it will get to the target
>audience we'd like to reach.
>
>I've done a (very crude) first draft of such an article, below.
>Let's pass it around and comment on it. I'll dredge through my
>old notes, and Google around to try to find placement for it.
>
>What do you think?

Several thoughts.

* advising potential manufacturers to seek design information from open source
authors is bad advice for two reasons. One, The authors probably don't know
all the hardware interface details. Rockbox and the MAS chip, for example.
Second, few mfrs would want to clone an old design. More likely is a mfr
wanting to consult with open source authors on designs they (the mfr) have
developed with regard to ease of porting existing open source code.

* Advising mfrs to seek out people to write open source code (presumably from
scratch) is not very good. In effect you'd be advising the mfrs to look for
people who will take on a programming job for no pay. Not likely. Open
source people tend to write code for what interests them. Only when that
interest happens to intersect with a hardware mfr will there be joy.

This is bad from another perspective. If the mfr simply announces that it
will provide free hardware to programmers willing to code, the mfr will end up
giving away a lot of hardware to people who do nothing but talk a big story.
BTDT. Leaves a very bad taste regarding open source.

What the mfr should be advised to do is seek out existing open source projects
that come closest to what they need and then contact the principle players to
see if they'd be interested in doing a port. These are people with proven
track records.

* I suggest advising mfrs that one way to benefit from open source is to have
their paid programmers get involved with an open source project, port it to
their hardware and release the changes back to the community, working with the
principle players all along the way. That's the way a lot of Linux
development happens.

* I would be very careful about promising more than can be delivered. In my
experience, porting an open source product to a new hardware platform is
almost as expensive as writing it from scratch, at least going in. Over the
long term it is both cheaper and better to use open source but the startup
cost is high. I'd rather manage expectations so that the mfr is surprised at
how much MORE he benefited from the project than he expected, rather than him
being disappointed due to overzealous expectations.

John

---
John De Armond
johngdDONTYOUDARE_at_bellsouth.net
http://bellsouthpwp.net/j/o/johngd/
Cleveland, Occupied TN
_______________________________________________
http://cool.haxx.se/mailman/listinfo/rockbox

Received on 2004-03-22