31 May 2002
(updated 31 May 2002 at 15:25 UTC) »
Freecell Solver Lecture
I finished it, gave it some final touches and uploaded it to vipe.
You can find it here.
Now I have to review the slides before the presentation on Monday.
The 3-D impression of a BSD Daemon we created now has moving parts,
thanks to a lot of effort I invested in it. Writing Inventor Scenes
in C++ sure is tedious. I still have to inform my partner about it.
An exercise that we were given in the "Internet" course, turned out
to contain a bottom note saying that we should better use the book
"Routing in the Internet". I left it at my apartment, so now I'm
stuck without it. But who is John Galt? The way things are
progressing it seems Roy and I will have to spend some time off the
summer vacation working on the project. We just can't seem to find
enough time during the semester proper.
I eventually finished the SICP 2 exercise. I implemented the
Minimum Spanning Tree Algorithm with awful complexity because I was
too lazy to implement a binary heap from scratch. Whatever: the
lecturer said that complexity was not the issue here.
Yesterday, I could not drive through my usual trail in the Yarkon
Park because there was a Food Festival there. Today I saw that
people started dismantling all the equipment and stages that was
left there. So it's probably back on track. There's some kind of
activity going on in the middle part of the park, but I don't know
what it is. There are just several large heaps of stones or dirt.
Today, a girl jogger sort of accompanied my trail for a large part
of my ride through the University St. I then lost her as she
jay-walked (or rather jay-ran) across the red-lights in Qeren
Qayemet. It was no the first jay walk of her that I witnessed.
I placed the source on the CVS repository on BerliOS and since then
I was able to hack on it. So far, I did some cleanups of the code,
added an unlimited tests order, and implemented compact allocation
for some resources. In the compact allocation scheme, I allocate
contigous blocks from a buffer, rather than individuallly malloc
The CVS is by far more convenient that keeping tar.gz archives. The
problem is that it takes some time to commit it, because the
connection is slow.
I downloaded the TuxRacer 1.1 Demo by SunSpire Studios. The
graphics are better than the old Open-Source version. It took me
some time to get used to it, but now I like it a lot. I'll probably
talk my dad into buying the full game, because it looks very
I've learned a bit of STL yesterday, and covered some of the basics
of vectors, lists and iterators. Programming one of the exercises
the tutorial gave me turned out to be quite tedious, and it took me
a long time to get it right. The problematic part was the data
structure logic more than the API.
I don't think I will invest much more extra time in it, because I'm
not using it now, and until one needs to use something, he has
little motivation to learn it.