[asus-port] wiki

Hi, people!

I found you by looking for Familar Linux, which took me to the
handhelds.org web site, where I discovered the "Asus MyPal" project,
the wiki and the mailing list.

In page http://www.handhelds.org/moin/moin.cgi/MyPal I have noticed a
potential inaccuracy:I own a "Typhoon *MyGuide* 5500 XL", not a MyPal
and I don't think Typhoon ever called their model MyPal. I think you
should change what the subtitle says ("Plus OEMs like...") to reflect
this. If I'm wrong, you can add my model to the OEMs. :-)

It was sold at a quite small price through Lidl, the discount chain,
so many people must have bought it. Chances are, many of them would be
at least curious about trying out Linux on it, maybe some are also
programmers and could help. If the subtitle would mention the model's
correct name, they, too, could find this project through search
engines.

I had good fun today trying to get Familiar working on it, because I
enjoyed the challenge. It was not easy to duplicate the best results
obtained by you and the best results were less than I thought at the
beginning. Well, I probably didn't even manage to get as far as you
got, but that only proves my point that you don't make it easy for a
newcomer to get the best result that you managed to obtain so far (or
I'm dumb). I've waded through the mailing list's archive, which kept
referring newcomers to the wiki, to patches to the latest CVS at the
time of the mesage and to precompiled stuff on web sites.

I don't have a problem with this project having a high entry barrier
to usage and contribution. Given how experimental it still is, you may
not even want clueless users around, wasting your development time
with support requests. I'm only mentioning it in order to make you
aware of it.

Another, perhaps better way to keep away those who aren't willing to
live on the bleeding edge (and this brings me back to my "results were
less" statement above) would be to warn them early that the MyPal port
is not ready for prime time.

On my entry path, the first such point is in
http://www.handhelds.org/projects/mypal.html, where it would be
helpful to find the description of how far you've got. If we are to
believe Eric S Raymond, software will attract contributors if it
"works" in some way. Users who come from the GPE and OPIE screenshots
and try the MyPal port will fail to get to the point shown in the
screenshots and for them the port will "not work".

The project web page should make it clear what the "prize" is if
someone wants to try it out. For example, it could say "we managed to
get Linux to load from under PocketPC and then access it then through
serial console or USB networking." The minuses/dangers could be
mentioned here, too: "BootBlaster doesn't work, your PDA will be reset
to factory settings...", whatever is appropriate to help users decide
if they even want to give it a try and to minimize the surprise in
case something nasty happens.

All right, if editing the non-wiki page is a hassle for its
maintainer, visitors should find more detailed information at the
second point on my entry path, the wiki. The wiki looks useful to
developers, but not to new users. Even knowing the prize from the
previous step, currently they would have a hard time putting together
the pieces of the puzzle in order to get it working (or is it just
me?), so for most of them the MyPal port would still "not work".

On the wiki, newcomers should be provided straight away with a recipe
for winning the "prize". If currently the most spectacular thing that
can be done is to ssh to the PDA while it's running Linux, please
provide the steps for newcomers to reproduce it. Backup this, then
download that, then edit here, then run so and so and watch it work!
It rewards the newcomer for their interest and gives them something
they can build on and watch it grow. Not all paths to the prize must
be given, just the most robust one. Once one way works for someone and
they become interested, they might discover that another way suits
them best, but if zero ways work in the beginning, they might not stay
around to discover the others.

I must have reset my PocketPC about twenty times today. :-) At about
the tenth (including failed copies to the CF) I gave up on the
bootopie-h3900.tar and started trying with linexec. At about the
fifteenth I had the idea which would prevent linexec from freezing the
PDA (see my next e-mail) and the rest I spent fine-tuning the solution
until something finally worked.

I got as far as seeing "asus620_udc_is_connected:
asus620_udc_is_connected" messages on the PDA's screen and getting
network interface 'usb0' up and down when I plugged the PDA into and
out of the cradle. No local prompt, no network connection. I'm sure
those will eventually come, too. :-)

Thanks for the effort, already.
Thomas.
Received on Sat Mar 05 2005 - 22:17:31 EST