Heb je er ook zo'n pesthekel aan als je de was gedaan hebt en je sokken niet meer te sorteren zijn? Dan is er nu een fantastische uitvinding! Wat kan je er allemaal mee doen? Je kan ze decoratief op de grond leggen, zoals op de volgende foto:

Maar je kan deze handige stukjes onbreekbaar plastic nog veel nuttiger inzetten! Echt waar! Net als iedereen trek je 's avonds voor het slapen gaan je sokken uit, tenzij het natuurlijk heel koud is, maar we hebben een fijne nazomer, dus ze kunnen gewoon uit. Als je vervolgens je sokken wast en ze daarna van de waslijn haalt, dan zijn ze meestal niet meer als stelletjes terug te vinden. Natuurkundigen hebben een moeilijk woord hiervoor, entropie. Waar kan je nou deze plastic vriendjes voor inzetten? Heel simpel! Je klemt de bek om een paar sokken heen, et voilà, ze kunnen niet meer uit elkaar:

Dit plastic is helemaal fantastisch. En wasbaar. En het eet je voorraad pindakaas niet op! En zorgt er dus voor dat het sokkenmonster niet meer z'n ding kan doen! Bestel dus snel op http://www.cyclopssockclip.com/. Tenzij je de hele dag op slippers of sandalen loopt.

If you want to suppress the generation of legend lines for certain elements in your graphs while using pylab, use the following trick:

pylab.plot(y, z, '-k', alpha=0.4, zorder=1, label='_nolegend_')

Elements with label '_nolegend_' will not show up in the legend!

(Note that only one underscore is required around nolegend ; serendipity seems to mangle this stuff even after inputting html )

First of all, a happy new year, gentle reader.
Here's a tiny snippet of code that shows how to skip a line in a formatted data file in Fortran:

    CHARACTER :: dummy

    OPEN(UNIT=fileunit, FILE=filename)
    ! bogus read call to skip the first input line
    READ(unit=fileunit, FMT='(A1)') dummy

As the tap in my kitchen water heater is leaking I set out to find some new parts for that thing. First try: last weekend, I went to a high-profile plumbing store (that shall not be named), and they were not able to help me as I hadn't brought the old parts, and they had no knowledge of my particular water heater. A more general parts store nearby also did not have my parts.

So, this morning, I disassembled my tap (after stocking some water), and set out to buy the wanted parts. The parts store did not have anything in stock, however, after showing my old tap to the plumber he went through some drawers and found a complete tap - costing €27.50.  I was needing two, so that would have set me back €55. That's a ridiculous amount of money for solving a problem that can be solved by replacing two pieces of rubber.

For keeping code quality high, at work we make extensive use of unit and functional testing. Today a small problem arose, as it seemed somewhat difficult to find links on icons. Luckily we have alt-tags on those icons, so in the end it was as simple as doing

 >>> b.find_link(text='Commit[IMG]')

Link(base_url='http://localhost:8085/cobes/@@index', ...)

Great isn't it?

I like nagios and software raid on my Linux boxes to make sure I don't lose any data, but I was unhappy with the script I used to monitor it. Attached is a new one that is shorter and works better, as it leaves all the work to the mdadm tool instead of trying to parse /proc/mdstat.

 
Continue reading "nagios and swraid"
            

http://jonathan.tron.name/articles/2006/11/05/multiple-identities-in-one-account-with-apple-mail-app

(WorksForMe™)

For several years, I have been using google adsense to recover the costs of running lkml.org, and to earn a tiny bit of beer money. However, per click revenue has dropped:

Earnings per click of lkml.org in Euros (gray) and Dollars (purple)

I've now found a nice alternative: adbard

As they are offering 40% of $2 per click, I'm hoping this number will rise.