As I men­tioned, I saw Kill Bill Vol. 2.

While I en­joyed Vol­ume 1's child­ish no­tion of grandeur by ac­cu­mu­la­tion, I must con­fess I was wait­ing for that emo­tion­al mo­men­t, like S. Jack­son's mono­logue at the Din­er in Pulp Fic­tion.

Or the whathe­heck mo­ment like the death of Robert de Niro's char­ac­ter in Jack­ie Brown.

Or even the goofy quar­ter-­pounder-with­-cheese mo­men­t.

But all I got in Vol­ume 1 was Uma Thur­man in the slick Bruce Lee out­fit, blood, Yakuza­s, and the brass of RZA­'s ver­sion of "Bat­tle with­out hon­our".

Which is not ex­act­ly a small amount of things to get, don't get me wrong, but not quite what I wished for.

And much lat­er I start­ed read­ing the re­views for Vol­ume 2, and it seemed a more QT flick, and I like those, and I saw it.

And it's good. And it has the quater-­pounder-with­-cheese mo­ment (Su­per­man), and the whatthe­heck mo­ment (Elle's), and it has the emo­tion­al mo­ment (a lot of them), even if they are dark­er, and let's face it, ba­si­cal­ly evil.

So, there's a lot I liked. I was even hap­py about David Car­radine's part (although his lisp drove me nuts (and I don't mean LISP (I mean he says yeth in­stead of yes))).

BTW: I read on a news­pa­per an ex­cerpt from an in­ter­view he did for Un­cut mag­a­zine... he's one crazy guy.

The Pai Mei stuff is hi­lar­i­ous. Michael Mad­sen's Budd is awe­some. The very pic­ture of moral deca­dence wrapped around the nas­ti­est lit­tle-broth­er syn­drome ev­er.

So, I ac­tu­al­ly quite loved this movie. It's not my favourite QT flick ev­er (that's Pulp Fic­tion, then Jack­ie Brown), but it's a hell of a lot bet­ter than al­most ev­ery­thing else I saw in the last year or so.

All I want to know is what hap­pened to Sofie Fa­tale (cheesi­est name ev­er for a french char­ac­ter), and I would be hap­py as a clam.

I looked at my old sto­ries, and found some rather new com­ments :-)

In my Squid au­then­ti­ca­tion via POP or IMAP ar­ti­cle, Ed­win Groothuis asked for it to be up­dat­ed for Squid 2.5. Which is pret­ty rea­son­able, so I did it ;-)

And plovs, at 04:00:27 AM on April 05, 2004 post­ed the first com­ment on the very first ar­ti­cle I ev­er post­ed here, Sim­ple KDE Trick #1 (first and last of the se­ries ;-).

That first feed­back on­ly took... 14 months and 4 days, since the ar­ti­cle was post­ed on March 1st 2003.

Is­n't it amaz­ing how noth­ing is ev­er lost?

I am try­ing to do some­thing which is, I think, pret­ty cool, in python.

How­ev­er, I like show­ing things work­ing, and I am hav­ing trou­bles with the last fi­nal step on what I am try­ing to achieve.

Since I know a few bet­ter python pro­gram­mers read this...

Sup­pose I have this:

def fun(self,x):
        pass

class C:
        pass

C.a=fun
C.b=fun

What code should be in fun() so that it fig­ures out if it has been called as C.a or as C.b?

I am think­ing some­thing like read­ing the high­er step in a back­trace, but I don't know enough python to fig­ure out how to do that.

I have been us­ing my Palm m100 for ap­point­ments and read­ing etexts for two months al­ready and it has been re­al­ly re­al­ly nice.

How­ev­er, af­ter read­ing a slash­dot dis­cus­sion, I am up­grad­ing, to a al­most-new Sony Clie SJ-30

Rea­son­s:

• Col­or vs Black­­&White (not grayscale)

• High­­er res­o­lu­­tion (320x320 vs 160x160)

• Smal­l­­er

• Recharge­able bat­tery vs AAA bat­ter­ies. This means that as long as I re­mem­ber to plug it ev­ery once in a while, I won't lose da­­ta.

• I can sell the m100 and pay part of the Clie that way, so it's not all that ex­pen­­sive (about US$100 in the end)

• More mem­o­ry (8MB vs 2M­B)

• Mem­o­ry stick slot

• Al­­most ev­ery­one likes the Clies as far as I can see. Al­­most ev­ery­one hates the m100. Why? ;-)

The on­ly pos­si­ble con is that with a nice set of Du­ra­cell­s, the m100 could go on for two weeks of read­ing about 4 hours a day, and the bet­ter screen in the Clie will eat bat­tery much faster (I've read about 2-­day bat­tery lifes from some user­s).

A stu­dent asked me how to au­tolo­gin on a lin­ux ter­mi­nal. I thought... that can't be hard... I just don't know how it's done.

So, I googled, and there's a patch for minget­ty that al­lows au­tolo­gin... but it's for an old­er ver­sion (0.9.4) and I can't even find it.

Af­ter a while, I check the SRPM from RH9... and it has the patch in the source tar­ball, but does­n't ap­ply it. No won­der, since it does­n't work.

So, I patched the patch, and here it is: a au­tolo­gin patch for minget­ty 1.0.1... I have so clear in my mind why I don't pro­gram in C :-) (although it was rather easy).

The patch

I even have a patched SPEC file (no big­gie there), and even sr­c.rpms that should work on any Red Hat like thing.

Af­ter you in­stall the patched minget­ty, you can put a line like this in your /etc/init­tab

1:2345:respawn:/sbin/mingetty tty1 --autologin=ralsina

And a ses­sion for us­er ralsi­na will open in tty1. You can't even quite lo­gout, since it logs you back in.