Iris update

Bas Wijnen wijnen at debian.org
Sun May 9 19:36:55 EDT 2010


Hello again,

It's been some time since my last update on Iris.  A lot has happened
since then.

For FOSDEM, I worked towards getting a demonstration ready.  That helped
a lot, but of course after FOSDEM I needed a new short-term goal. :-)  I
chose to work towards a standalone alarm clock.

I now have much of that working.  I rewrote the boot system, so that
it's easier to boot from other sources than usb.  That's what took most
of the time; I encountered some very hard to find bugs.  I also wrote
the alarm clock itself (which is pretty trivial), and a gui for it
(which isn't very fancy).

One feature I very much like, is the split between ui and application.
The reason is that a user is likely to want to change things in the ui,
but an average user doesn't know how to change a program.  For example,
in the game Pioneers, we have added an option to change the screen
layout (have a chat panel below or on the side of the main panel), to
support 4:3 and 16:9 screens in a useful way.  As a programmer, I don't
really want to be bothered with this sort of task.  As a user, I don't
want to depend on the programmer to have it available.  So my idea is to
split the program and the ui, and use a (perhaps graphically) scriptable
language for defining the ui.  Any user should be able to make changes
to it, and thus design their own interface for a program.

In the larger picture, I expect programs to be shipped with an interface
which is recommended by the developers, but which can be changed by
anyone who likes it better differently.  An added benefit of this
approach is that it becomes trivial to write a program with a graphical
interface.  All you have to do is mark certain variables as "possibly
viewable by the interface", or "given a value by the interface".

Anyway, back to Iris: the alarm clock is running, but it's not
standalone yet, because it needs to boot over usb.  So the next step is
to write a NAND driver, so it can boot from there.  Wolfgang suggested
that for others to be able to test it, I should make it boot from SD
card as well.  That makes sense, so after the NAND, I'll try to make an
SD driver.  When all that works, I'll publish some binary images which
can be tried without setting up a cross-compiler.

Comments and questions are welcome as always.

Thanks,
Bas
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: Digital signature
URL: <http://lists.en.qi-hardware.com/pipermail/discussion/attachments/20100510/8187bd28/attachment.pgp>


More information about the discussion mailing list


interactive