An “ass” out of “u” and “me”

I’m off-white, so I’m used to people assuming rather strange things about me. The big one I am forever explaining is that no, I really am Canadian. Yes, that’s where I’m really from. That’s where my ancestors are from too. I’m biracial, but even the “immigrant” side of my family has been in Canada for well over 100 years, and the other includes united empire loyalists. Really.

Sometimes people assume I don’t speak English when they see me teaching at the university. Sometimes they assume I speak French after proclaiming my Canadian-ness. (I do, but most Canadians don’t.)

When people see my name, especially associated with my geeky pursuits or with my academic or professional work, they sometimes assume I’m male. Hilarity ensues.

But the weirdest assumption I ever encountered was that because I was a girl, I would somehow know something about computer usability. People would get me to test things. People would ask me questions. People would assume I could do design. People would assume I wanted to work at the interface level.

You think I know what?

When I was a fledgling programmer, I stayed as far away from interface as I could get. My first real programming job had me working on an SQL query optimizer and tracing bugs through threading libraries on old unix systems. I worked with email at the text-only, server level in my spare time. I lived off the command line. I didn’t know anything about interfaces, and was irked that people kept assuming.

But… I watched how people used things. I saw how sometimes, it wasn’t their fault that they couldn’t unsubscribe from the mailing list when you had to do it by clicking a button labeled “Edit Options” (yes, as a Mailman developer, I made sure this was fixed.)

Hm… maybe I should learn some usability after all

I wanted to be able to help prevent this sort of nonsense. So I fit some human factors and user interface courses into my schedule. I read books. I angsted over whether this was “girl stuff” that I shouldn’t be doing lest I never be seen as a core programmer again, as Mary mentioned in her post. (There’s other girl stuff, but that’s not part of this story.)

But then I got into computer security, and in the course of my study, started to realise that a lot of “security flaws” come through people misusing products. I read Why Johnny Can’t Encrypt: A Usability Evaluation of PGP 5.0. (Yes, the SSL cert fail you get when you click that link is pretty funny in context.) I realised that it didn’t matter if usability was “girl stuff” — it was core, it was important, and people who ignored it were just being foolish. I finally embraced my “girl stuff” as part of security.

The happy ending

I won’t claim to be a usability expert, but I’ve got a lot more training and a lot more curiousity. It’s made me a much better software developer, and it’s made me a significantly better security researcher. (Most people in my research lab would argue that without understanding what users do, you can’t really understand computer security!) My open source project of choice, Mailman, is slowly benefiting from my increased expertise. And all starting with a really strange little assumption that usability was girl stuff. I still don’t understand why it’s girl stuff, but I understand that it doesn’t matter.

Turns out, when you assume, you don’t have to make an “ass” out of “u” and “me” after all.

11 thoughts on “An “ass” out of “u” and “me”

  1. Chris

    I wonder if I could find a usability class or 2 at a local university that I could take. Too bad Boston isn’t a big college town.

    1. Terri

      heh. Nope, no academia at all in Boston. ;)

      On a related note, for those of you who might be interested in taking courses but aren’t sure about doing so, here’s a tip: If you’re not sure if a course will be worth your time, try just showing up for the first couple of lectures. Most universities don’t care if there are a few extra people in class, so you really can “try before you buy” when it comes to university courses. I did this all the time in graduate school, and no one seemed to notice or mind.

      Many universities also offer and “audit” option, where you pay less money and attend the class, but don’t get marked or any credit for it. If you don’t have time for a full course load but still want to learn stuff, this can be another option for you.

  2. Skud

    I’m going through a similar process wrt Javascript. I should probably blog about it :)

    (In short: I’ve resisted JS for a decade because I don’t want to get stuck doing front-end web stuff when I much prefer back-end programming. Now my workplace has a major product which is a hosted server-side Javascript app platform, I’m having to get over my internalised sexism to some degree and just sit down and learn it.)

    1. Terri

      If it makes you feel better, JavaScript doesn’t have to be just client-side web code. I use JavaScript to write Firefox security extensions. It’s not quite back end programming, but it’s not “Oh, you make pretty web pages?” either.

      Ditto for using it to write grease monkey scripts. Again, not back end programming, but suddenly it becomes like the perl glue I use on the command line to save myself hassle. Totally cool.

      JavaScript has a fair bit of stigma that’s not even a sexism “you do the interface” issue. It was just such a poorly supported, poorly used language when it first came out that it’s taken a long time before I was willing to hold my head high and admit that I use it regularly.

  3. Mackenzie

    “Why Software Sucks” is supposed to explain how to do UI to programmers.

    And UI can’t be Girl Stuff. I have no eye for design (or maybe I’m not a girl?). I can point out “this UI sucks, because it takes me 12 clicks with the mouse to do what would only take 2 commands in the shell,” but suggesting a better way is problematic. I’m the sort that will give up on a GUI application if it’s too annoying and use the CLI instead. Screw OpenOffice Impress; LaTeX is easier. The users who can’t just switch to LaTeX though…they’re problematic. They require that UIs actually be fixed.

    As for me & code…if it’s higher level than the compiler and lower level than button.setText(“Click Me!”) it’s magic. No idea how it works. Something in there draws some pixels and figures out when the mouse is touching it and then…ok then I get a callback signal and can write code to handle it. The part with the pixels and the figuring out the mouse, though? Uhh… *backs away slowly* *runs*

    1. Terri

      The usability book I’m now recommending is Don’t make me think by Steve Krug. It’s about the web, but a lot of that stuff will follow to other types of interfaces.

      It’s short, it’s funny, and it’s positive about interfaces, rather than cataloging disasters. But perhaps most importantly, he makes usability sound like something anyone can do. Or at least do better. None of this “you need a huge test group and well-defined experiments” but more “buy your buddy a beer and ask them to try something for you” — you can learn a lot with fairly simple experiments.

      (And if anyone wants a more extensive reading list, feel free to ask — I’ve been re-reading a lot of usability and design stuff for my PhD background section, so I could easily name a few more of interest!)

      1. Mackenzie

        Ah, yeah, that’s the sort of testing Celeste had our LoCo team doing at one point. Grab someone that’s wandered into the library or bring a friend. Hand them a digital camera, sit them in front of an Ubuntu machine, and tell them to get the pictures off the camera and onto the computer…do a bit of tagging…blah blah. If they get stuck at the import screen or go “hey! I only told it to import these 5 pictures but it pulled all of ‘em in and now I have to delete the stupid things…grrr….” well, there’s an answer for you. Oh, and I’m pretty sure I’ve done both of those when trying to use F-Spot.

      2. wahiaronkwas

        That’s the best way to test anything – operating system, program, website. Grab User Anybody, and User Anyone and sit them down. I think that’s what the Drupal Usability group did as well in D.C.

  4. spz

    one thing I always wondered:
    why is usability only always about the user who never saw the product before and will only use it every blue moon, and not (at least also) about not driving the people nuts who use it four hours a day every day?

    In my book, mutt -is- a lot more useable than all the clicky MUAs. So there! :)
    You were an expert on useability for yourself already and just didn’t notice. ;-P

    1. Terri

      You’ve clearly been really misinformed about what usability is. For a product, it is typically heavily about core users. For some web pages, it is sometimes the casual users who are more important. Determining who your users are and which ones matter most to you is actually an essential part of usability work.

  5. Asad

    Why do we need users anyway? Shouldn’t the abstract Platonic world of ideal forms be good enough for us?

Comments are closed.