Belatedly: Happy birthday, Debian! Stay around and keep being awesome! Happy birthday, Internet Explorer, remember when you were really big?

And currently: Happy birthday, emperor Franz Josef the First! Stay safely dead so that we can keep romanticizing your rule!

From the ever-excellent OKTrends blog, this graph:

I’m just glad my girlfriend has an Android phone as well.

So, we have continuing outrage across Europe, and apparently even some US states, about the fact that Google captured people’s WiFi data, including – gasp! – emails and passwords! I once again fail to be outraged, for two reasons: First, all the reporting that I’ve read so far, even from sources who should know at least as well as I do, has neglected the most likely explanation of how this came to be. The stated purpose of Google’s WiFi capture spree was to improve location sensing based on the available SSIDs, a nifty idea with minimal privacy implications (there is now a database that includes the position of my WLAN router, probably with SSID and MAC address, but what are realistic threats arising from that?). They recorded the data during their StreetView drives, and it’s most likely they didn’t want to write the software themselves, since Kismet already does everything they wanted to – it switches channels, looks for available networks and tags the networks with GPS data. In its default configuration writes all the traffic it receives to a pcap dump file. The output from a Kismet session usually consists of such a dump file and some XML files recording network and GPS data. It seems entirely plausible that the Google people would simply stick an extra antenna on the roof of the car and run unmodified Kismet during the recording session to then run a postprocessor on the easily parseable Kismet outputs to get their mapping information. This leaves them with pcap dumps of packet payloads which they probably don’t need for anything. Of course, this is not such a good idea when you’re already one of the foci of privacy paranoiacs, but I know that’s how I would have set out to map WiFi networks, without any evil intent at all. I don’t know why this simple explanation never seems to come up in reporting, since I strongly suspect it to be true – maybe because that would turn the entire thing into a bit of a non-story? I like that people are at least smart enough to question the handing over of the captured data to government agencies instead of taking Google’s offer to delete them, but overall, there would probably have been the least amount of damage if this had never become a public issue at all. These dumps would either have been discarded or sat around somewhere in Google’s huge data stores, and nobody would have cared.

The second reason, predictably, is that I fail to understand how anyone could still use unencrypted protocols over unencrypted networks to send and receive private data. Yeah, I’m more of a crypto adherent than most, but people who have their passwords and emails captured by a Kismet drive-by face much more serious risks than Google. I’m not quite comfortable arguing the “if I can read it, don’t complain when I do” position, but I think the common sense solution here is that the outrage gradient should definitely run upwards to “deliberately cracked my neighbor’s WiFi encryption” from “accidentally read what my neighbor beamed all over the place for everyone to read”, and I have yet to see any coherent explanation of the evil designs Google would have on the people who had their data captured.

The 0x100th post on this blog, and, coincidentally, my first as an officially minted physicist (thanks, alma mater!).

So, what are my options as a freshly minted science graduate? Well, there’s the perennial favourite, a dissertation. What guidance do the Interwebs have to offer on that? From my recent readings, I especially liked The Real Science Gap – once you get past the US-centrism, I think it offers a rather plausible summary of the current situation of the classical academic career path, which doesn’t differ that much in Europe. TL;DR summary: Prospects are sad, and the reason is probably an oversupply of scientists, contrary to what regular public lamentation would make us believe.

For a couple of pre-crash years, I thought I could always move from “the statistics of particle collisions” to “the statistics of financial transactions” and become a quant if nothing else worked out. Sadly, this also doesn’t seem quite as attractive and pristine after all the post-crash abuse laid on the profession (surprisingly, however, demand still seems to be there, and the base salaries are slightly above ten times what I’m currently being offered for a dissertation).

Another routine destination for people who find the prospects of an academic career too daunting is a gig in the upper layers of the consulting business. Here, I really liked the recent four-parter from The Tech, describing the experience of an MIT nuclear engineering graduate as a consultant in Dubai (I,II,III,IV). The money quotes about the consulting industry are mostly in part III:

What I could not get my head around was having to force-fit analysis to a conclusion. In one case, the question I was tasked with solving had a clear and unambiguous answer: By my estimate, the client’s plan of action had a net present discounted value of negative one billion dollars. Even after accounting for some degree of error in my reckoning, I could still be sure that theirs was a losing proposition. But the client did not want analysis that contradicted their own, and my manager told me plainly that it was not our place to question what the client wanted.

Reading the entire thing isn’t exactly a motivational aid for prospective strategy consultants. I’m totally open to the suggestion that the experience of one disgruntled ex-employee is atypical, but it’s a bit hard to be sure since the entire industry seems very concerned about confidentiality, and the recruiting people who show up at university job fairs are really quite admirable in how they politely manage to avoid revealing any specifics of what consultants actually do, even under repeated questioning from multiple sources. While this mystique is intriguing, it does imply a certain vulnerability to bad press of the above type.

All in all, plenty of things to be cynical about, but also – interesting times!

My X sessions usually last for some time, and they usually devolve into terminal hell:

themel@kallisti:~$ ps uwax | grep x-term | wc -l
49

It’s always fun to go looking for that one terminal that contains that background job whose output you want to read right now or where the history of that one task you were working on is stored, and ion3 helpfully names all those windows “x-terminal-emulator “, so that going through them is a bit of a pain. I spent quite some time digging through the system in the hope of finding the window associated with a given pty, but so far a solution has eluded me – the terminal emulator does not hold an FD for the device, and the hell of finding out anything useful about an X window through all the layers of abstraction is worthy of a blog post of its own.

After some frustration, I stumbled on this simple, yet effective hack: My .bashrc now contains

if [ "$PS1" ] ; then
  xtitle xterm $(tty)
fi

This puts the pty in the window title, where it is easily acessible using ion’s builtin navigation functionality. I’m still interested in a solution for the problem above, though.

Or so they say. Slashdotted or fooled? You never know.

So here I am, enjoying a cold beer on a Friday afternoon with my Humppaball buddy, and suddenly, this crazy car that is bristling with cameras drives by, and now my pixelated face is all over the Interwebs! This is AN OUTRAGE!1! (for a second, I even considered putting a <blink> in there)

(I don’t really see why our privacy agencies are getting so upset about StreetView while people fritter away the actual private details of their lives on Facebook and are about to have their entire communications history stored in arbitrarily secure lockers at telecom companies.)

1980 flashback edition!

Der Kurier konstatiert in verfehltem Zynismus: “Bald wird uns nichts mehr verblüffen
können. Kein nationales Heiligtum bleibt verschont. Die Schlagzeile Sängerknaben zahlen
Schmiergelder an Lippizaner wird uns nicht mehr aufregen.”
— gingrich, oliver: Der Fall Rabelbauer. Der Mann mit dem Koffer. Seminararbeit, 2002

So, at least, things aren’t going down the drain, they have been solidly there for my entire life (and long before that). Btw, though all my examples seem to cluster in one party, there are plenty of examples from the rest of the political spectrum, I just can’t be arsed to waste more time digging through historical accounts I find deeply depressing just to provide a balanced view here.

Commentary is here.

Yay, data mining the US dating scene – the OKCupid blog! 7% of white females agree with “Interracial marriage is a bad idea”, while 54% agree with “I would strongly prefer to date someone with my own skin color/racial background”. “You know, racism is bad, but colored people? They’re just not that good.” (and ha, it’s even worse with non-whites, so maybe all the anti-racism conditioning at least has some effect).