The fabulous jwz [EN] links to Gizmodo [EN] ridiculing Apple customers. Penny Arcade, of course, already knew long ago [EN]. Well, it’s true. Apple [EN] seems to be the only corporation that can indulge in all kind of nastry practices, like suing [EN] news sites that announce unreleased products, or patenting all [EN] kinds [EN] of [EN] stuff [EN], without getting the bad reputation in the technophile community that this kind of behaviour usually entails. Somehow, the genius of Apple marketing seems to get the better of our sense of moral consistency.

Of course, there’s also all kinds of cool stuff about Apple – hear Steve Wozniak [EN] talk about the beginnings (or read about it in Hackers [EN]), for example, or read the story of Graphing Calculator [EN].

But who am I to talk… The news of the Mac mini [EN] was hardly out when I was already clicking through the Apple Store, ready to order my dream configuration. So far I have managed to restrain myself, but I fear it’s only a matter of time before this cute little thingy comes to replace my clunky old desktop PC.

vs.

Well, since I’m just into dissing corporations today, I’ll hand out another
slap to my bank, Bank Austria Creditanstalt [DE].
I just stumbled across a statement that I have over 2000 bonus points for their
reward program. Since its intent AFAIK is to get people to use online banking
instead of pestering branch office personnel, which is the thing I do anyway, I
asummed that that number would be a lot, and indeed, it is. Have a look at their
list of rewards
[DE] – holy crap, nothing in there consumes more than 60 points.

Which would be really cool, but the problem is that there is nothing you get
for your points without paying extra cash. Hell, I’d settle for free pens
or children’s toys. About the best thing you can get is a 20 percent discount on
some chocolate, and even that is scarily expensive chocolate, and they throw in
a mug with each set to spoil my idea of ordering the 150 sets and justing having
enough chocolate for the rest of my life. The following list is a quick estimate
of just how useless this is:

So, yes my, 2000something points could save me some money, but that’s only
nominal, since it assumes that the alternative is buying all of the
pointless crap rewards undiscounted. I think I’ll get by without them a bit
longer.

Also, and more importantly, BA-CA don’t offer any kind of standard compliant
online interaction with their accounts. I could buy an expensive business
account, but that still won’t produce anything that I can import into href="http://www.gnucash.org/">Gnucash [EN]. The system that they use, href="http://www.mbsip.info/">MBS [EN], is some kind of proprietary
development of Austrian banks, which of course no one else in the world
supports, thus no free software for it (it’s probably also shrouded in layers of
lawyerese and intellectual property issues).

They don’t do href="http://en.wikipedia.org/wiki/HBCI">HBCI [EN], either. So, I had to sit
down and automate what I need from the web interface using Perl und href="http://curl.haxx.se/libcurl/">libcurl [EN]. It’s not pretty, but it
does fetch you statement data in CSV or IFX XML format, and combined with the
thing I wrote for href="http://home.themel.com/weblog/archives/permalinks/2004-10-31T21_28_21.html">this
post [EN], it produces useful QIF files that Gnucash understands without too
much bother.

However, contrary to href="http://home.themel.com/weblog/archives/permalinks/2005-01-08T21_42_10.html">the
eBay case [EN], these issues don’t bother me enough to switch to a different
bank. The pointless-points-program thing is really not an issue, since I didn’t
expect anything until I noticed that number on the statement. The other thing
would interest me, but I haven’t yet found a bank in Austria that actually
offers some kind of cool, automatable customer interface that also works with
free software (if you know one, please comment!).

Update 2006-08-03: Look for the bach-client project in there, since a couple of updates have come along.

Since there’s a lot of junk in my apartment, I thought I’d sell some of the
crap on eBay. I have had an ebay account for years, but it seems you need an
extra registration to sell stuff. All I get is a lapidary one-line page that
says

The specified module could not be found.

Using a slightly different way through the link jungle, I end up with

The filename, directory name, or volume label syntax is incorrect.

Sometimes, when ebay is in a particularly nice mood, I get to a page that
lets me choose whether I want my identity verified via credit card or via postal
mail. However, either of these options redirects me to href="http://pages.ebay.com/messages/DE_page_not_responding.html">this [DE]
slightly more colourful error page.

When this happened for the first time, I thought it was a temporary glitch.
When it happened again, a month later, I sent a detailed error description to
eBay support. The response I got was this:

Yeah right. So they want me to scan a number of personal documents and send
them via postal mail. Plus they’re asking me for information that only a
resident of Germany can provide, conveniently ignoring the fact that I’m in
Austria and using their .at domain.

While this might be a feasible approach if you’re a small company and your
web site is a bit of a mystery to you that ocassionally generates an email
telling you what to do, I’m not taking this bullshit from a company that has href="http://finance.yahoo.com/q?s=ebay">70 billion USD market cap [EN], and
pretty much all they do is running web sites. If they want my business, they’d
better fix their product. I’m sticking with href="http://www.onetwosold.at/">onetwosold [DE], in the mean time.

Just before I moved to Vienna, I bought a 20GB third generation iPod as a
consolation gift to myself for leaving Villach. It’s been my trusty companion
for many hours of public transit and various other situations ever since, and
I’m quite happy with it. I use gtkpod
[EN] to manage its contents, and though the program has some quirks, that is
not a major problem since I don’t update my music collection that often.

Yesterday, I stumbled across href="http://hackaday.com/entry/1234000147025394/">an iPod hack [EN] to
allow sound recording without buying the href="http://www.amazon.de/exec/obidos/ASIN/B0001WW398">expensive [DE] href="http://www.amazon.de/exec/obidos/ASIN/B0001USHA4">add-ons [DE] that
third party vendors offer (found via href="http://www.emergentchaos.com/archives/000771.html">this Emergent Chaos
post [EN]), and it required installing Linux on my iPod. Now, I never really
wanted to record audio on the iPod, but at the same time I have a hard time
resisting the urge to install Linux on anything I own. So, I went through the
crystal clear href="http://www.ipodlinux.org/index.php/Installation_from_Linux">installation
instructions [EN], and got a dual-boot capable iPod, which defaults
to the Apple OS, but can be rebooted into Linux at will. The recording hack
mentioned on the web page worked after a ‘mkdir /Recordings’ on the Linux
partition.

Now, the weirdness began. While I played with the iPod Linux loader, I
rebooted the iPod numerous times, all the time leaving it in the dock. It seemed
to me that the driver for the firewire interface on the iPod would sometimes
hang when the iPod was in the dock at Linux bootup time. While investigating
this, I pulled the iPod out of the dock while the Linux system was starting up,
when suddenly the PC the iPod was attached to as well as the iPod
itself rebooted.

This in itself was strange enough and would have been highly embarassing if
witnessed by any MS supremacists, given all the bragging about the supreme
stability of Linux that is spread all over the web. The really interesting thing
is the situation I have after both systems came up again: the PC refuses to
detect the iPod as an SBP-2 device (and consequently, as a SCSI disk). Before
the mysterious reboot, I used to get something like this in my kernel log on
connecting in the iPod:

Dec  5 18:24:26 sophokles kernel: ieee1394: The root node is not cycle master capable; selecting a new root node and resetting...
Dec  5 18:24:26 sophokles kernel: ieee1394: Node changed: 0-00:1023 -> 0-01:1023
Dec  5 18:24:35 sophokles kernel: ieee1394: Node changed: 0-01:1023 -> 0-00:1023
Dec  5 18:24:37 sophokles kernel: ieee1394: Error parsing configrom for node 0-00:1023
Dec  5 18:24:37 sophokles kernel: ieee1394: Node changed: 0-00:1023 -> 0-01:1023
Dec  5 18:24:46 sophokles kernel: ieee1394: Node added: ID:BUS[0-00:1023]  GUID[000a2700026ae376]
Dec  5 18:24:47 sophokles kernel: sbp2: $Rev: 1219 $ Ben Collins <bcollins@debian.org>
Dec  5 18:24:47 sophokles kernel: scsi0 : SCSI emulation for IEEE-1394 SBP-2 Devices
Dec  5 18:24:48 sophokles kernel: ieee1394: sbp2: Logged into SBP-2 device
Dec  5 18:24:48 sophokles kernel: ieee1394: Node 0-00:1023: Max speed [S400] - Max payload [2048]
Dec  5 18:24:48 sophokles kernel:   Vendor: Apple     Model: iPod              Rev: 1.51
Dec  5 18:24:48 sophokles kernel:   Type:   Direct-Access                      ANSI SCSI revision: 02
Dec  5 18:24:48 sophokles kernel: SCSI device sda: 39063024 512-byte hdwr sectors (20000 MB)
Dec  5 18:24:48 sophokles kernel: sda: test WP failed, assume Write Enabled
Dec  5 18:24:48 sophokles kernel: sda: asking for cache data failed
Dec  5 18:24:48 sophokles kernel: sda: assuming drive cache: write through
Dec  5 18:24:48 sophokles kernel:  sda: sda1 sda2
Dec  5 18:24:48 sophokles kernel: Attached scsi removable disk sda at scsi0, channel 0, id 0, lun 0

Now, all I get is this:

Dec 30 22:35:56 sophokles kernel: ieee1394: Node added: ID:BUS[0-01:1023] GUID[000a2700026ae376]
Dec 30 22:35:56 sophokles kernel: ieee1394: The root node is not cycle master capable; selecting a new root node and resetting...
Dec 30 22:35:57 sophokles kernel: ieee1394: Node changed: 0-01:1023 -> 0-00:1023
Dec 30 22:35:57 sophokles kernel: ieee1394: Node changed: 0-00:1023 -> 0-01:1023

My first guess was that I had probably managed to break the iPod by playing
with Linux. However, this was quickly refuted when I connected the iPod to my
ThinkPad, and instantly got a SCSI disk device for it. At that point, I assumed
that I had forgotten to load some kernel module or other, but I couldn’t find
any notable differences between the two systems, and I didn’t really have to do
anything special to access to the iPod from the desk machine before (I rebooted
the iPod once, back in June, in response to a “Error logging into SBP-2 device -
login failed” message, but I don’t get that message any more).

Since time was running out at that point (I had to get up at 05:30am, and
wanted to catch close to six hours of sleep), I got desperate and tried few
reboots of both the PC and the iPod, and some variations in module insertion
order and device plugging order, but to no avail – all the cases are pretty much
the same in that it works flawlessly with the laptop and not at all with the
PC. I thought that perhaps I had broken the firewire card in the PC somehow, but
I think I can rule that out, too, since a) gscanbus shows the same information for
the PC/iPod combination on the PC and the laptop and b) eth1394 networking
between the PC and the laptop works.

So, it seems I’m missing something, but I don’t quite see what it could
be. Both machines are running vanilla 2.6.9. The SBP-2 driver is loaded on both:

sophokles:~# ls /sys/bus/ieee1394/drivers/
eth1394  raw1394  sbp2

sophokles:/home/themel# ls /sys/bus/ieee1394/drivers/
eth1394  raw1394  sbp2

Both machines seem to have SCSI disk support, since their /proc/devices list
numerous entries for sd.

sophokles:/home/themel# cat /proc/devices 
Character devices:
  1 mem
  2 pty
  3 ttyp
  4 /dev/vc/0
  4 tty
  5 /dev/tty
  5 /dev/console
  5 /dev/ptmx
  7 vcs
 10 misc
 13 input
 14 sound
 21 sg
 29 fb
116 alsa
128 ptm
136 pts
171 ieee1394
180 usb
195 nvidia
 
Block devices:
  1 ramdisk
  2 fd
  3 ide0
  8 sd
 11 sr
 22 ide1
 65 sd
 66 sd
 67 sd
 68 sd
 69 sd
 70 sd
 71 sd
128 sd
129 sd
130 sd
131 sd
132 sd
133 sd
134 sd
135 sd
254 device-mapper

sokrates:~# cat /proc/devices 
Character devices:
  1 mem
  2 pty
  3 ttyp
  4 /dev/vc/0
  4 tty
  4 ttyS
  5 /dev/tty
  5 /dev/console
  5 /dev/ptmx
  6 lp
  7 vcs
 10 misc
 13 input
 14 sound
 21 sg
 81 video4linux
 89 i2c
116 alsa
128 ptm
136 pts
171 ieee1394
180 usb
216 rfcomm
226 drm
 
Block devices:
  1 ramdisk
  3 ide0
  7 loop
  8 sd
 22 ide1
 65 sd
 66 sd
 67 sd
 68 sd
 69 sd
 70 sd
 71 sd
128 sd
129 sd
130 sd
131 sd
132 sd
133 sd
134 sd
135 sd
254 device-mapper

I don’t get this. The next thing I’m going to try is a KNOPPIX CD on the
desktop to find out whether it works with that. But that’s for next year, since
it’s currently assigned to a time-critical task of crucial importance:

ophokles:/home/themel# grep mplayer  /var/spool/cron/atjobs/*
/usr/bin/mplayer /mnt/local/mp3/mp3/Eins\ Zwo\ -\ Gefaehrliches\ Halbwissen/Eins\ Zwo\ -\ 09\ -\ Frohes\ Neues.mp3

Update:The problem is gone, as mysteriously as it appeared. I rebooted
into Knoppix, which recognized the iPod just fine. Then, after booting back into
my homemade 2.6.9, that also recognized it. Damn you, cosmic rays.