July 2009

Recently, one of my colleagues at Company started a story on how his closest friend’s daughter committed suicide. Obviously, very sad, and a tragedy. Then he continued that he blamed her mother because she was always a bit lost in this cloud of feminism, you see. Uhm, no… I don’t?
According to his reasoning, women can either have a career or a family, but not both because, you see, that’s how evolution made them. And men can, because you see, they are biologically different. And then he had the gall to call upon evolutionary psychology to support his theory.

*eyeroll* I told him flat-out he was talking bullshit and wandered out of the coffeecorner, because there was no way I was gonna get pulled into a long drawn-out discussion where he would barely listen to my arguments because he’s made up his mind.


One of my master students is almost graduating. It took a while, with lots of tears, sweat and blood (on both sides, I believe). I spent yesterday and today a grand total of 5 hours, correcting the last version of their thesis.
Now, is it fair to not want to give this student a high grade because of everything they didn’t learn in the whole process? Or should I reward them for the bit they did manage to pick up in the whole process?
We tend to grade students on three different things: the process they went through, their final presentation and their final thesis. When it comes to the process I’m highly minded to give them barely a pass, the final presentation is still to come, and so is the final thesis. However, considering all the help we’ve given over the past 1,5 year, I would have expected a better penultimate version. Not one where I had to spend another 5 hours correcting. Although, granted, 3,5 hours were correcting on paper and 1,5 hours were making sure they came in a constructive form in Word.

Still, can I detract that effort from their final grade? Or should I let that depend on how well the final version of their thesis is? I find this a difficult position to be in, really. Although,  at least now I’ve been through it, so maybe I’ll be able to handle this better in the future.

Jeff Atwood’s post yesterday stroke a very sensitive chord. He claims that nobody hates software more than software developers, because they actually know what’s going on. This is probably not true for every industry, but certainly something that rings true. So who is going to protect us from incompetent software developers?

I’ve never met somebody who hasn’t had issues with software (even Mac users, yes :P).
As Scott Berkun so aptly states:

If you look deeper, you’ll find that when people say “this sucks” they mean one or more of the following:
* This doesn’t do what I need
* I can’t figure out how to do what I need
* This is unnecessarily frustrating and complex
* This breaks all the time
* It’s so ugly I want to vomit just so I have something prettier to look at
* It doesn’t map to my understanding of the universe
* I’m thinking about the tool, instead of my work

Unfortunately, people who create something are also very sensitive about it (well, yes, I did recently get my teaching evaluations, why do you ask?). This is the part where it’s sometimes more useful to let anybody but the creator do the user-testing. When testers can only express their frustration through “This sucks” and “I hate the way it works” it can be very helpful to have a translator at hand. So, contact your friendly user-experience expert and ask them to help you out here, because apparently the feedback is so bad that all you can do is become defensive.
While this isn’t really the reason I started in Human-Computer Interaction, I do have to say it’s one of the more challenging and satisfying aspects of the ‘job’.

Sometimes I wonder whether I’m daft, trying to use a program from ’97 to assess whether or not my data are multidimensional. Especially when said program refuses to give understandable error messages! I’ve never known much about Fortran, but now I know more than I ever aspired to, and it’s still not explaining why I get a “memory allocation error”. Because which command couldn’t it load? After all, my output file seems complete?
Now I wish I was back at the run-time error=M6202, or run-time error=M1603 stage. At least I’ve got those solved!
Maybe this is one of the reasons I so much enjoy being a pilot-tester: at least there’s the option to give direct feedback and see the program/application change for the better.

It took a while, but I finally got feedback on my first draft from my other supervisor. It looks like the data fit a 3D structure much better than a 2D structure, and while it all makes a lot of sense, in essence it still might be hard to write this down in a comprehensible manner.

Tomorrow we have another meeting with all of us (i.e. both supervisors and promoter), and then I’d like this article to be on the agenda. Well, it’s up to me to make the agenda, so what do I want to discuss?
– my dataset, which generates errors and I don’t know what do to about them (prolly will make an appointment with one supervisor to figure it out)
– my online questionnaire, which is finally finished and now needs to be sent to several mailing lists. Do they (granted, mostly promoter) approve of the set-up, and what kind of announcement/begging test are they in favour of?
– the article I’m still writing on, mostly discussing the 3D figure, and the implications for my research (which are quite positive)
– the article that should have been written ages ago, but is being held up by the analysis of the giant dataset.
– the rest of my planning, which involves one more experiment somewhere in september and writing everything up (should I start filling in forms, or is it too early?)