A brilliant ad

Today I saw a new Belgacom ad for their isp 'skynet': two romantic souls are communicating with email, and 'he' tries to find romantic music. So off he goes to the 'skynet itunes music shop' and searches the list of songs. Then he replies that he cannot select a fitting song from the long, long list. She (I presume) replies that she can, and sends a romantic song.

Let me repeat that last point: she attached an song to the email. Besides the fact that the itunes shop doesn't give you an mp3 (what most people expect), this is also blatant illegal sharing of music unless they went through the whole 'authorise that computer to play my songs' song and dance, which is unlikely given the 'becoming an item' setting.

Just brilliant: seeding the population with expectations that the DRM will never allow them to actually do.

misrouted: goto 1

I've had a little setback with misrouted: it turns out that my original idea of how the universe of CP objects work (ie we have item and containers) turns out to be deeply wrong.

While parsing some complex rulebase with the old implementation it turns out that the objects are divided in 3 classes: items, containers and container-of-listable-items (ideas on naming are welcome). This is because you can have a network, which is a collection, that can contain a group, also previously a collection, in a different way then that the group contains the network: all members of the group are in the network versus a custom method looking if all possible ip's in the network are in the group. Also the trivial method I used the last method kind of broke down when it tried to expand a /8 network into ip's. :-S

So this means surgery at a pretty deep level in the class hierarchy and the old program could just about handle it, but the scars run deep. So misrouted is being rewritten to use the new setup. The good news it that I've thought of a more complex, but not O((N_rules**4)**N_firewalls**2) method of simulating the effects of a rulebase on traffic.

on blobs

I notice that the binary blob discussion is again on lwns front page. Recently I got a new portable from work and this gave me some new insight. In short: binary-only drivers are so bad people will buy other hardware.
The machine (a Lenovo Z61m) uses a X1400 ATI chip, so I have to use the ATI binary-only driver. Well. I'm trying to use it.
The amount of grief I'm getting from it would convince even billg himself that opensource drivers are mandatory in an opensource system. Strange errors pop up and you're just lost, googling only gives you forums where other poor users are trying to make it all work. Due to the dynamic nature of xorg and kernel development the driver will never 'catch up' and problems will keep popping up. And there is little I can do except complain to the responsible person and urge him to buy the Intel only portable next time.
If this keeps on Intel will just sweep the Linux hacker's market, and with it the heart and minds of a lot of people. My next machine will be an Intel, the first one in years, because I cannot get an AMD with a open graphics card, and I think I won't be alone in this.

A deja-vu moment

Here I was reading up on how to install software on the PVR box we bought (a dream 7025). I find the developers information corner and read the ipkg documentation. Talk about a deja-vu. I feel I'm going to get along just fine with ipkg creation.

In other news I've uploaded a new version of slime to main. Yeah. Many thanks to Luca, Marco and the other slime developers!

on GR's

While I battle FTBFS problems with my packages I barely notice the -tank and GR emails flying past, so when I see GR I always think it means GR. Which is something positive, healthy and nice to do, unlike following debian-vote/debian-legal. It must be the effect of my 'think positive' project :-).

Oaxaca teachers strike

During the post-debconf6 trip we stayed in Oaxaca and in my flickr images you will have noticed the striking teachers. People that visited the place may be interested in the fact that Indymedia reports that the government removed the sit-in with force.

stupid CL errors

• Do not assume :always does a check on an empty list:

  [1]> (loop :for x :in '() :always (oddp x))
  T

  is the correct result. But deep inside logic involving "have all entries that property?" it can be surprising.
• 0 is not NIL. Stop laughing, it took me 2 days to find that one part of the program thought it was talking T/NIL while the other talked 0/1 or 2. One (declare (type boolean result)) and a few runtime errors later the program gives sane results again.

Pictures sort-of-up

I've uploaded the Debconf6 and Post Debconf6 trip pictures to flickr.

There is still a time-shift problem with almost all the pictures from the trip, I'll have to explore the flickr API to shift the timestamps forward 41 hours. I already corrected my local copies, which I now need to upload to the debconf gallery :-).

Then I'll write a report on the trip on LJ.

s3cr3t blog message

Tonight we meet some others in front of the Santo Domingo church/museum at 22:00 in Oaxaca.

The formal dinner

I talked about the dinner in my non Debian/lisp blog and put the pictures up on flickr but I think some things need to be said even if there should be no need to say the obvious.

(I'll try to use Enrico's guidelines here so I've deleted 2 pages or rant )

Being part of Debian is more then packaging and flamewars on Debian-devel, it is about working together with respect. In general Debian people (not only maintainers) are intelligent, strong willed people. This makes for some interesting (and strongly worded at times) debates, but we should never lose the respect for the other person and to act when the respect of others is violated.

clc v6 and debconf6

I released common-lisp-controller 6 which has one major change: if you now compile a lisp file not under clc (or slime) control then the resulting fasl file will end up in /var/cache/common-lisp/<uid>/<implementation>/local/<full path of the source file> so that on upgrade of the implementation all cached fasl files will be removed. To give an example compiling /home/luser/cl-world-control/mind-beam.lisp will result (on sbcl) in /var/cache/common-lisp/1001/sbcl/home/luser/cl-world-control/mind-beam.fasl

The only problem with this is that is you remove the source-tree the cached fasl files will not get removed, but I guess people do upgrades with regularity (for sbcl every month or so) and sysadmins know that anything in /var/cache may be nuked at any moment.

For debconf6 I'm blogging about it on my non-debian and lisp blog (so that my non-debian friends read it) and uploading pictures to a special debconf6 flickr set.

ECLM2006: better late then never

As others have already commented here, here, here, here and here ECLM 2006 was great. The level of the talks seems to be improving (not that I want to imply that previous talks were bad) and having companies asking for lispers was a novel experience. I must also agree that as an ex-physicist I found SigLab to be very impressive.

The bad points were the 6hour drive back and the 'reception' at work. I can only now start to catch up with non-emergency stuff. Such is life :-).

There is still a washlist of things to do before going to debconf6, not in the least getting medical coverage for .mx and power-plug adaptors. I've noticed that some other hackers are going with Iberia flight IB6403 (leaving on 13/5/2006 at 12:30). I'll try to dress 'debianista' to attract them.

die fuzzy fonts die!

root@sharrow:/etc/fonts/conf.d# ln -s kill-anti.conf 10-antialias-die-die-die.conf
And still pterm looks like shit. Damn.

a stunning crash

Slime tells me "Lisp connection closed unexpectedly: exited abnormally with code 256". I go and see: (sorry for the long line)

* Argh! gc_find_free_space failed (first_page), nbytes=8.
 Gen StaPg UbSta LaSta LUbSt Boxed Unboxed LB   LUB  !move  Alloc  Waste   Trig    WP  GCs Mem-age
   0:     0     0     0     0     0     0     0     0        0     0        0 2000000   0  0,0000
   1: 121288 15638     0     0 30022     1     0     0        0 122956040    18168 74632320   0 22771058045587471891756344047460264512912328495970408565885262128721370095226093339014016268704097497686156189724993758756525399616019155487351045354729920444619442123787213803893991126769353775124417090005956563018545473316412289883581293782691217408,0000
   2: 131071  6160     0     0 34634     1     0     0       69 141822136    42824 21427888 18640  0,0000
   3: 87088  6169     0     0 19167     1     0     0       23 78496608    15520 2000000 19144  0,0000
   4: 83996  6219     0     0 30404   915   124   109      106 128215256  1021736 2000000 30427 -0,0000
   5:     0     0     0     0  4267  3287  1461   552      283 38188928   997504 40188928 5250  0,0000
   6:     0     0     0     0  6127     0     0     0        0 25096192        0 2000000 5774  0,0000
   Total bytes allocated=534775160
fatal error encountered in SBCL pid 23015(tid 3076041632):

No, this sbcl session was not started before the big bang :-).

Airport Passenger Screening

Another stunning article from Schneier: Airport Passenger Screening. It seems that the success-rate for smuggling things through airport security is 70% for knives and an incredible 100% for bomb-parts.
To print out when flying to Oaxtepec :-).

Swiss

Normally I keep this for on-topic Debian or lisp stuff, but I could not resist commenting on the swiss comment.
One word: Sabena. Most if no all Belgians are still bitter about the whole tragic story and I feel sad that Swiss will claim another nice airline as its (hopeful) last victim.

New common-lisp-controller version

We (I and René van Bevern) have been working on a new version of clc. This should add support for 'create a core from these files' implementations, like ecl. It also uses the UID of the user to create the cache directory, andcustomizationn of the systems loaded by default in the standard images.

It works for us, but we would be grateful if people could try it out and suggest improvements. Get it at:
http://people.debian.org/~pvaneynd/cl-packages/common-lisp-controller/common-lisp-controller_5.10_all.deb

No ecl in debian yet, as I've run up against a strange compilation failure.

Baen books

Thanks to a hint on the Baen CD site I ordered the electronic versions of The Far Side of the Stars by David Drake and War of Honor by David Weber. Both these electronic versions contain a 'bonus CD' image, so I now got a copy of the Slammers CD and of the Baen's At All Costs CD.

I know it is possible to just download the iso's from several sources, but after spending many hours with "Lt. Leary", "Belisarius", "Honor Harrington", "Roger Ramius Sergei Alexander Chiang MacClintock" and the always entertaining Posleen I felt that paying a few dollars was in order.

clisp 2.37 seems unable to append to files

I had already noted clisp bug 1399709 a few days ago. As nobody reacted I started to think something was wrong with my setup. Then I saw Gary King has the same problem on Mac OS-X. So I think maybe the problem does exist after all :-(.

Note to debian people: this problem is delaying the release of a new clisp version with a fix for the xlibs-dev transition.