Older blog entries for jameson (starting at number 32)

19 Oct 2002 (updated 19 Oct 2002 at 04:00 UTC) »

Studying in the US is... different from Germany. Then again, part of what confuses me could be a particular property of this university; I haven't had classes that lasted only 50 minutes but were scheduled three times a week since, well, the rough equivalent to high school (although, from what I've been told, high school doesn't quite cover the German Oberstufe, but that's only slightly related).

In many ways it could be argued that the United States are weird. Then again, I'm pretty sure that many people from here think the same of Europe, so let's change that to 'different' (think "bread", "butter", "coffee", "washing machines" and "stoves"; then again, they do have things like peanut butter and bottomless cups, so there are upsides as well as downsides). Getting used to people running around with weapons and the government being entitled to kill people is a bit strange, too, but at least they have recycling, bike lanes and a working bus system here (which doesn't really make up for that strange feeling, as I'll admit, but turns out to be more relevant to everyday life).

I'm still not sure what I'll do next, though. There are a couple of Ph.D. positions open here, and students from Darmstadt are rumored to have a pretty good chance at getting them. There are also one or two places I could go to back at home to do a similar program-- although what I'd really like to do would be to go to Edinburgh (to the semantics department), but I guess my chances there would be too close to zero to even deserve logarithmizing. Still, one can try...

pizza did some more work on the sound system recently; unfortunately, my laptop doesn't really do much in the way of sound (except for beeping... hmm... is there an ALSA driver for the PC beeper or something like that?), so I cannot comment on how much the situation has improved.
Meanwhile, I got two projects approved at university-- one to document and bugfix the FreeSCI parser, and one to get extended VM support up and running, for SCI01/SCI1/SCI1.1 and perhaps even SCI32 support (thanks to Prof. Gary Nutt). This should be fun-- working on my favorite FS project and getting graded for it...

People have argued that the academic world is a major contributor to the Free Software world, which I believe there is little doubt about. It's good to experience this first-hand.

Sister's 40th birthday today, and I can't be there (one ocean away). I guess I'll miss a lot of birthdays while I'm here, though.

Programming Languages
A course in programming languages reminded me of how horrible the favorite languages of choice of our oh-so-advanced Free Software community are. Correct me if I'm wrong, but I believe that C, Perl, Python, and perhaps also C++ (and, to a lesser extent, Java) are the most popular picks out there; I've recommended some of them in the past and have come to the conclusion that the quality of these recommendations is debatable.

Sure, it can be (reasonably) argued that the size of the set of modules/libraries/classes available for a particular language (and the maturity of its elements) gives some indication regarding its practical usefulness, but ignoring key weaknesses in a language (such as a lack of type parameters aka parametric polymorphism (which is closely related to "generics" and "templates), static typing, limitation of side effects ("pure" programming style), data hiding or modularity) means that, even though lots of libraries may be available, they may be of poor inherent quality-- a great example for this is Java's lack of type parameters, which means that using container classes requires typecasts (most of the languages given above don't even have static typechecking), or the limited amount of optimizations possible in impure programming languages (since side effects can rarely be estimated in their entirety). Now I'm well aware that many of these things are difficult problems, and that (which is probably a much worse problem) UNIX and Linux are based around C, a C core set of libraries, a C-style dynamic linking mechanism etc., requiring other languages either to scale down (and thus raise the question of whether they do anything but add overhead in the first place) or to re-implement major parts of the library (which people will, in general, be even less enthusiastic about, since it would add even more overhead, unless these other people stop using C programs, which won't happen).

I guess the only alternative would be to make C the language using the wrappers. Although it would be preferrable to have a decent language to build on top of first...

Sorry for the rant, but maybe someone will consider this to be interesting in some way.

Have been certified as a Master. While I'll admit that I'm flattered, I have to point out that this rating is silly (please refer to the certification guidelines-- No matter how much I'd like to, I do not work full-time on Free Software; the other points are more vague but don't apply to me in general, either).

On an unrelated note, I'm now in the lucky position of participating in an exchange program with the University of Colorado at Boulder. I won't go into detail about the various "accidents" and "voluntary cancellations of applications" that occured to the other applicants here...

So I'll be spending 8 months in the US of A now, starting in mid-August. I've been told that I might have to pay an additional $275 in order to allow the university to monitor my movements more precisely (as if this would be of any relevance to my plans for World Domination), thanks to some weird new law.
Suffice it to say that observing the US from within is probably going to be a major change of perspective...

Anyway, is anyone from UCB an (active) Advogato member?

Very little in the way of news, unfortunately, except for some bug fixes, cleanups, Adlib emulation, and the beginnings of a Dreamcast port.

Hey, looks like someone's actually reading my diary entries ;-)

Still closer to the release. IMHO, the UNIX part is done now (thanks to the remains of the Compaq TestDrive compile farm, I could test this one on a few new setups this time. 'Test' as in 'compile and run ./freesci -h', because they don't have X11 forwarding. Oh well.). However, support for Win32 seems to be broken; since I don't have the equipment to test or the knowledge to work on that, the release will have to wait for the Win32 porters.

Meanwhile, anyone sufficiently bored to try what is likely to already be 0.3.3 as far as non-Win32 is concerned can try this here.

cactus: Glad you like it! I can't seem to reproduce the problems you're having here (FreeSCI halting for a short while and CPU usage increasing dramatically) and haven't received any similar bug reports yet, but this sounds interesting (at least from the point of view of someone who's been fixing collision bugs for the last week). I'll contact you per e-mail, since advogato probably isn't the best medium to use for bug hunting...

Well, it's been quite a while since the last update once again. I am certain millions of advogato readers have been anxiously waiting for a new one... oh well... who am I fooling... if not even I bother to proof-read it, why should someone else care about what I write?

Anyway, many things have changed, so this might get a little lengthy. Here it comes...

New computer
Yes, I actually bothered to buy a new system recently (named it Asuka), to do stuff my "old" box didn't really handle well. It has built-in graphics accelleration, hardware DSP mixing etc., but it came shipped with a proprietary OS which I still have to replace (or, at least, partially replace, for dual-booting) with Debian. It's probably one of the fastest of its kind available, an Amiga 1200 with a Blizzard IV turbo card. sic. Now I have an Amiga and an Alpha. I guess I'd start buying Betamax hardware if I had a TV.
Anyway, it'd probably be a nice little system if I'd take a day or two to get it into shape, and I'm quite looking forward to doing this once the semester is over. It's a bit hard to use without manuals, though; I haven't even found an easy way to enter the shell yet (short of starting to boot, then removing the disk I'm booting from).

FreeSCI: New Release
FreeSCI 0.3.3 is actually going to be released shortly. This is one free software project with a highly erratic development cycle...

hasn't seen much of me lately, which, despite probably being better for it, is rather unfortunate, IMHO- I'm still hoping on "fixing" it for building on DEC cxx this weekend, but it's just a vague hope.

New currency
Here it comes, the Euro. Shiny new coins, looking a bit like the newer 10 Franc pieces. The notes are a bit too colorful for my personal taste, but I thought the same of the new DM notes when we introduced them a few years ago; I'll get used to this just as I'll get used to having an EUR 0.2 coin.
Of course the Euro isn't just a new currency, it's a political event as well. It's up to TPTB now to use the momentum they've generated to push any pro-Europe agendas they might have in store; let's just hope those agendas will be something better than certain "Anti-terrorism" laws introduced by opportunistic politicians recently.
Bright and shiny they are, these Euro coins, not as scratched and dirty as the other coins I'm still carrying around. I wonder how they'll look in a few years.

So I've actually seen it now. I've always wondered whether it's possible to take a book of the narratory depth of one of the LOTR's parts and turn it into a movie- a long one, of course- whithout missing a lot and still having to rush everything.
Now that I've seen one of the best directors around try it with a very large amount of money at his disposal, I'm pretty certain it's not.

Just returned from it. Southern France is a great place for people who enjoy beach-style holidays, with sunny weather, refreshing breezes, great food and mostly acceptable prices (It's not Poland, but it isn't Switzerland, either). Unfortunately, being naturally receptive to sunburns, I'm not particularly into lying on the beach all day, but it was still quite entertaining- while the area may be lacking the inns that tend to mark places well-suited for hiking and no major (> 250m) height differences can be found there, it's still a fun place to just wander around and explore. Still, next time I go on a vacation I should take the time to learn the basics of the native language spoken in the country I'm going to- while not being able to understand even some of the most basic signs and phrases is sufficiently embarassing already, having to fall back to a third language (usually english) is downright impolite. At least IMO. Although I'd be the first to admit that the necessity to visit a bathroom wins over politeness most of the time.

10 days without net access. And while I'm away, HP buys the big Q, the Andromeda TV series starts, and an utterly insane amount of e-mails piles up in my inbox. Ten days sure are more than they used to be... I blame time dilation. Time has been slowing down perpetually ever since we started collecting information in information networks. The conclusion is obvious- information has a negative mass, thus we're building a white anti-hole now.
And now for something completely different. claudio: You do qualify for a 'master' ranking- you've worked on a wide variety of Free Software projects, both commercial and non-commercial, in user- and in kernel space, both contributing and leading development, on a large number of platforms, in war and peace, above and beyond the call of duty. Besides, you like Pokey the Penguin. Anyway, you have a much better idea of what you're doing than I have [of what I'm doing], and this ought to be reflected in some way.

Time to start focussing on a new release. I've been told that the Win32 port has been improved in some major ways, but I'd prefer it if there were more updates to the UNIX side than just bugfixes. Well, it's not as if there was nothing left on the TODO list, but starting something grand and new doesn't fit particularly well with "focussing on a new release"...

School project
Working on a communication protocol for this thing (page in German) here. It's a, well, bike of sorts. And it's stuffed with an insane amount of technology (No, you won't be able to buy this thing any time soon). We were originally planning to use one of Compaq's iPAQ systems (running Linux) to act as the speed indicator and general configuration management system (allowing centralized parametrization of the devices connected over the bus), but Compaq's reluctance to support us and their recent troubles may force us to fall back to a different device.

Exams are next week, so please bear with me if I take a little longer to reply to non-urgent e-mails. Unfortunately, a friend convinced me to visit LinuxTag on saturday, so I loose yet another day.


    I'm going down to Stuttgart, gonna have myself a time
    Friendly hackers everywhere, making code and presentations
    going down to Stuttgart, gonna leave my code behind.
I'm wondering if I'm going to meet anyone I know- a few people from the local LUG appear to be going there, but I haven't met many of them physically, so this isn't really going to help. Maybe there'll be a few names I recognize at the Debian booth...

Graphical web browser
Wo-hoo! I finally have one of those for myself! The thing I have is called 'Konqueror'. It appears to be a bit unstable- segfaults every third URL or so with a warning that the received header has a size of -1- but, otherwise, it's nice, particularly for a first version. Wonder why it says 2.2something, though...
Kudos to Chris Chimelis and RevKrusty for getting it to work, BTW!

Requiem for the Alpha?
This doesn't look good, particularly if they guessed the buyer correctly. I believe that IA64's an interesting addition to the 64 bit marketplace, but it is not a replacement for the Alpha. Intel will think differently, of course...

Spending most of my time on university stuff lately, so there's not much going on development-wise. With some luck, a university project might turn out to become another addition to the Free Software world, though. OK, to be honest, it's the other way 'round- I'm trying to get a piece of code I'm working on to be accepted as a replacement for a much more boring project I'm supposed to be working on.

GCC 3.0's out, and I'm supposed to be happy. Unfortunately, Compaq's (soon Intel's?) cxx is faster by an order of magnitude. I haven't compared the code quality yet- neither compiler manages to build Mozilla or the kdelibs on my box, and I'm not really in the mood to do benchmarking.

Since the last update, I went to the theaters exactly four times: Twice to see Mononoke Hime, which I highly recommend- the German synchronized version is surprisingly good-, once for Perfect Blue, which wasn't bad either, and once again for The Mummy 2 (Or whatever the official title is), which I didn't enjoy quite as much as most of the others I talked to. It had a few nice ideas, but it was just too cheesy ("Hollywood-ish") in places.
Anyway, I think I'm going to skip Perl Harbour- according to two reviews and one oral report, it's apparently not worth watching. Not quite on par with Saving Private Ryan, Das Boot or Hotaru no Haka('Grave of the Fireflies').

Frightening news from claudio: It's back! With a vengenance! It's got the Star Generator, and it's probably trashing Mega-Tokyo (no, not Megatokyo) right now!
It's definitely great to see this project (one of the original motivations for starting FreeSCI) resurrected. However, the URL would appear to be slightly different than indicated in the original post- the other project referenced looks interesting, too, though...
I still think the new FreeSCI gfx subsystem (which desparately needs a name!) can be strapped onto Sarien. It'd probably add some overhead, but it would allow Sarien to take advantage of all of FreeSCI's graphics drivers, such as GGI, and... uh... the Xlib driver... and... the partially implemented SDL driver.... OK, still, there might be some benefits, such as trilinear filtering, or maybe some performance improvements on system with more memory. Or maybe not. But it'd be something I should try eventually...

/dev/sequencer doesn't like me. Somehow, we run out of partials (voices) much too soon, even though a sufficient number of note-offs (MIDI 8x... or was that 9x?) are sent. Also, far too many gfx optimizations appear to be hitting worst-case situations (especially noticeable in the HQ character setup screen). Don't like that at all.
A few interesting suggestions regarding extending SCI have been suggested on the IRC channel, mostly regarding BSD socket support. Personally, I think a more interesting challenge would be to extend the interpreter to have 32 bit support without breaking the existing 16 bit code. Anyway, general interest in using SCI for new stuff bodes well regarding Brian Provinciano's SCI development thingy. Don't think I'd want to use a 16 bit interpreter for a new project, though...

Interpreters and virtual machines
I got an interesting mail from someone who slightly mis-judged the order of magnitude of FreeSCI and wanted to pit it against Java and .NET. He was rather persistant with this, especially since, as he pointed out, a proprietary VM called ICVM was much faster than Java on his system. From what I can tell, ICVM looks like a more register-based approach (like Dis) with a highly CISC instruction set. Due to the CISCness, the interpreter overhead was supposed to be rather small, making a JIT unneccessary.
However, the design of ICVM looks rather PC-centric- most of its registers are 32 bit, and it only has 6 general purpose integer and 3 general purpose floating point registers (On most architectures, you could play Space War in the remaning registers without this noticeably affecting performance...).
Anyway, the main point he was making was that there ought to be a free VM design around. Creating a good real-life VM would certainly be an interesting challenge, but I'm not sure whether it'd help Free Software in general- after all, it would just encourage people to keep their stuff closed again. Then again, it might help making non-mainstream platforms more popular, which might turn out to affect the BSDs and GNU/Linux positively... Of course, the amount of work needed to create something in the order of manitude of Java would be immense. A more sane starting point would probably be to start off an existing project (such as Python) and use its libraries, tweak its VM for performance, and write a gcc backend for it...
Well, I guess I'm spending too much time thinking about this- a project of this kind couldn't happen without massive interest from a group of powerful hackers, and I don't think we'll see that.

Looks like Exult will be going into Debian's contrib. IMHO that's pretty good news, but I'm not certain whether the auto-built Alpha port will work well (since they're probably building it with gcc rather than cxx).

Watched the first ten episodes of Cowboy Bebop for the third time (with a constantly increasing audience). I do have the third DVD lying around here, but, not having a DVD player myself, I'll have to wait for other people to have some spare time in order to finally learn what happens after Ganymede Elegy...
Mononoke Hime is going to be shown in our local theaters RSN. Only two more weeks or so...

Eventful 1.5 weeks lie behind us now. The most interesting part for me was the new bug tracking system- this was my first real-code encounter with Python.
One of the other interesting problems was how to deal with badly behaving single-user code. SCI games sometimes loop over the GetEvent() system call without explicitly waiting (they didn't have select() or anything like this). It's not unlike the "repeat until kepressed;" thing I used to do back in my Turbo Pascal days, and it sucks about as much. FreeSCI tries to trap it now by using the most obvious route- it keeps track of calls to functions which wait explicitly or retreive time values, and, if none of these was called after two subsequent GetEvent() calls, it executes a "penaly sleep" period. CPU usage in some sequences dropped from 1.25 to 0.22, so this appears to be a useful optimization.
Traffic in the IRC channel has increased significantly. It's fun, of course, but IRC can easily distract from doing real work (although this doesn't appear to be a problem for most developers yet).

This appears to be pretty much the most convenient language I've used so far. From what I've heard, its integration into C appears to be pretty good, too, making it a good choice for scripting languages. Of course, it still has a few things I personally consider problems:

  • Indentation rules. Changing control structures sometimes becomes a battle between me and EMACS' tabbing rules. Also, they doesn't help with automatically generated code (this may not be what Python was designed in mind with for, though).
  • list comprehensions and static lambda expression evaluation are missing. I know, they're in Python 2.1, but that doesn't help much until the licensing issues have been resolved (alternatively, those could be counted as the problem).
I still have to check out some of the other functional languages around to be able to judge it more appropriately. However, I agree with most of the other opinions regarding it that this should make an excellent teaching language, and, thanks to its extensive libraries and C integration, a good language for general purpose programming and prototyping, too.

Rumors about Samsung ditching API are about- don't know what to make of that yet... OTOH, Samsung has information about the upcoming UP1500 board on their page, whereas I can't find anything about it on API's page...
Anyway, if we disregard politics for a second and examine the specs, this looks like just the board everybody and their 400W power supply have been waiting for: Compared with the UP1000, memory bandwith was doubled everywhere (more than doubled in some places, IIRC), including AGP, and it comes with SRM rather than AlphaBIOS by default. Seems to require an EV68, though, so I'm left to drool...
But back to programming: For some reason, Compaq's ccc appears to have problems if people pass more than 6 parameters into inline assembly blocks. This breaks the current alpha blending code, of course... Any ideas?

23 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!