GNU Xnee can replay on Maemo / N900

Last week I got a N900 from Pelagicore to make sure GNU Xnee executes on the device. And as soon as Xnee is verified to work on the N900, we can start testing the work done in Xnee to support X Input Extension input devices.

Last Friday I tested Xnee on the N900, by logging in to my development computer (x86) from the N900, using ssh -X. It didn’t work. Since I didn’t know where the problem lied, Xnee or the X server or perhaps the ssh client/server,  I decided to natively build Xnee for N900…. but when to do that?

Today, I finally had some spare time to hack on Xnee and N900. First thing, set up scratchbox. It was great getting to know scratchbox. After scratchbox was installed there was nothing hindering me … or was it?

It took an hour to play around with some new ./configure options, change the autotools config files, add some #define and to add some libs to the linker in order to get an executable cnee for the Maemo device. The compilation worked well (after a while), so I continued with testing cnee on the device.

* Record: does not work

* Replay: works

* Retype: works

so I guess it’s time to dive down into the X server shipped with the device to figure out what’s wrong with RECORD extension in the N900 X server. More info to come….

Thanks Pelagicore for the N900

Andrew Gerrand and GO / FSCONS extra event

June 8, I opened the doors of IT University for an extra FSCONS event arranged by FSCONS and GTUG.  Andrew Gerrand from Google was here in Gothenburg to talk us through the language Go.

I liked Andrew’s talk btw. Clear. Easy to understand. He seemed to be interested in the questions asked. During the talk theere was a discussion on IDEs and according to Andrew there’s no need for an IDE when writing Go (which is the case (IMHO) when you’re programming for example Java). That sounded promising. Anyhow, let’s talk about Go. I am usually skeptical to new things (a clear sign of a geezer) and to some extent I’ll stay that way regarding Go. At least for a while more. Without any deeper knowledge I present my reflections on the language (as if anyone cares):

When it comes to memory management I feel pretty ok. As a developer it’s easier to let some one else do memory allocation for you (i.e not alloc and free), but at the same time when doing C/C++ I am in control and I can predict a bit better when things happen.

[In a previous version of this blog post I by mistake used the word concurrency to mean the distribution mechanism in Erlang. This is such a big mistake I’ve decided to rewrite history and edit my blog post – 2010-06-11]

Looking at “cross machine distribution mechanism” (think Erlang) I miss it. I asked Andrew about this and he said that there is none, but I interpreted him as if the door is not closed, but until then we will have to rely on a library (think pthreads in C).  However, if concurrency comes with a virtual machine or run time system (or what ever you want to call it) I am not sure I think it’s worth it. At least not when it comes to the kind of programs I normally write.

I asked Andrew about shared libraries and there is no such thing in Go. I surely understand why, but it would be neat to have. This could be done I guess with the C bindings mechanism in Go.

Enough of my complaints and ramblings. What do I think?…. what is my gut feeling. I am not a big fan of leaning on arguments such as “my inner feel says”, but that’s more or less what I will do.

Why Go?: It’s free software. It seems to be a language with which I can quickly get my things done.

Why not Go?: Not that many targets supported, no shared libs, non concurrency (why not use C instead of Go then?)

I guess I will give it a try…. and that’s not something I say that often. Perhaps write some Xnee bindings for Go.

Makes me wonder, should we invite some one to talk about Go at “the real” FSCONS 2010?

Pedagogical prize to Software teacher Carl Magnus Olsson

My colleague and friend Carl Magnus Olsson just received a prize,  “University of Gothenburg awards teachers for excellent pedagogical contributions to education”. I’ve had the pleasure of working tightly with Carl Magnus for a couple of years and can only say he’s worth it. The amount of hours spent preparing the course, reading and commenting hand ins, reading student papers …..   and his usual verbose email replies.

It is inspiring ….. CMO, You rock