Situational illiteracy

I’ve often wondered over the phenomenon that error messages apparently instantly remove the ability to read from the person viewing them.
“Aaah, my program crashed! Now I'll have to randomly click icons until the problem goes away!”
“This popup says ‘Connection lost, please check your Ethernet cable’. Would it be this cable that’s fallen to the floor?”

sigh OK, that’s enough smug superiority for now. There are of course reasons for this behaviour. Too much software generates too many useless error messages:

• A common method is by not catching the error at all and instead getting a stack dump from the underlying runtime system, which may be useful to a debugging programmer, but not to a working user.
• Or, if the programmer has bothered to trap the error, the error message is still mostly intended for debugging (“Stack overflow in function getResCompFlow().”) and does not suggest to the user either how to avoid it or how to proceed from a bad state.
• And then we have the opposites: the too brief message (“Syntax error”) vs dozens of long lines of information, with the possibly useful message hidden in the middle.
• Many error messages do a very bad job of analysing the problem situation and either suggest something very vague (“Error, please retry”), or something completely incorrect. (“Couldn’t connect, please check your Ethernet cable.” Actually I mistyped my password.)
• And so on…

So, the end result is that users are conditioned to believe that error messages will not contain any useful information–even when they actually do so. Now what still surprises me is that methodical checking of possible error sources apparently is not even the last resort for many people, who rather prefer wildly clicking at anything in sight in the hope that the error will go away. Is there anything in computer systems that conditions people to believe this will work?

Bad site, good sight

I booked tickets for an upcoming train excursion on the SJ web site. Through the perversity of capitalism or whatever, it turned out to be considerably cheaper to ride first class than second. This meant that breakfast was included on the way out. However, if you have “special needs”, e g being a vegetarian, you can’t just check a box for this, but have to call separately to order your special food. This I did today, and after having been shunted around a bit by different call centre persons (because of course the voice menus did not indicate which menu option would give you the menu options, as it were), I ended up with the right person, who, with some audible effort, managed to adjust the food orders in some, as it sounded to me, less than obvious user interface. Now, I thought that maybe I could also add dinner reservations for the journey home, which I had missed to do earlier. The CCP excused herself and disappeared for ten minutes or so, trying to work out how to add dinner reservations and eventually couldn’t come up with anything better than booking new seats for us, this time with dinner reservations. I didn’t think we’d eat so much that we’d need two seats each, so I thanked her for her efforts and left it at that.

Now, this conversation had been long enough and required sufficiently little of my attention that I’d managed to do other things in parallel: to begin with, look up on the SL site how to get to my ultrasound examination. They have the very useful checkbox “I don’t mind walking a bit if it gets me there faster.” and turning it on suggested I could get to my destination in a third of the time it would otherwise take me, just by taking a ten-minute walk through the woods behind my house to a bus stop by the big road. This was a nice enough little outing, though the little narrow and winding road through the woods turned out to be frequented by huge articulated lorries every few minutes. I’m not quite sure where they thought they were going.

In the time since the Google car had passed, somebody had carefully smashed all the glass walls of the bus shelter, but it was a sunny (though chilly) day, so it didn’t affect my immediate comfort too much.

Then I got to the care centre with a radiology department and got probed. I was fascinated to see that the bottles with conductive gel were kept in a little heater, so as not to be cold and nasty. Then I got to see the inside of my leg, and ye gods how thick the femoral artery is! Still, the ultrasound did not give sufficient resolution for a definite diagnosis, so next up: magnetic resonance tomography.

Jobshead

My new 15" MacBook Pro has arrived! Somehow every Mac generation manages to be better-looking than the previous one. The touchpad is soft as silk and the screen so sharp.

The one blemish so far has been the step in the installation process where I was to transfer all the information from my old Mac: thanks to the still-living connector conspiracy the FireWire cable I’d borrowed from the office only fit the old Mac. The little manual implied that you could transfer wirelessly, but it wasn’t entirely obvious that I was supposed to press the little button “Use Ethernet” to get to another screen where I could select “Use wireless”, especially when the on-screen instructions exhorted me to press “Previous” in the case that I didn’t have a FireWire cable. Then it took all night and a bit to transfer all the data, but now it’s done and I’m happy as a lark with my new machine.

---

I take that back: it’s not just silky, the touchpad is like an antigravity surface and when I moved my fingers apart it scaled the contents of my browser window! At this rate I’ll have to buy an iPhone soon.

Misguided striving for perfection

When people say that something works “works like a machine” the implication is that this work is not only ceaseless but flawless. In particular this seems to appply to machine intelligence, intelligent robots and computers are, not only in fiction, assumed to have all information—and correct information only—available and then flawlessly proceed to the correct conclusion. Certainly often evil conclusions, but still the only possible conclusion.

Well, of course real software doesn't work that way. I have not worked with AI per se, but any interactive systems should be as “intelligent” as possible, where this in practice means that one studies users and figures out what they want done most of the time and then try to make the interface anticipate what the user wants in every given situation. In many cases this turns out not to be what the user wanted and bad user interfaces tend to do their anticipation in such a way as to annoy the user. Good user interfaces on the other hand are unobtrusive and smoothly let the user continue with whatever was actually intended, silently withdrawing whatever suggestion might have been proposed.

Machine translation has always been an important task for AI, and it seems the applications I have tried go for the ideal of the all-knowing computer. Thus if you submit a text for translation, you get the output all at once, unalterable, regardless of how bizarre it ends up. Shouldn't it be possible, in this day and age, to have an interactive translation application which presents alternative interpretations of the input and lets the user guide the translation? Certainly, even when I as a person translate text I end up having to make notes in the output, stating that a particular interpretation is dependent on a previous term having meant this and not that.

A lesson in interface design

I got a promotional ticket for a free cruise to Helsinki on Silja Line and thought it would be an excellent use of a weekend. So I brought up the Silja web site, clicked on “Cruise” and started filling out the cruise booking form. This was divided into multiple pages, requiring input of lots of things and only on the last page did I get the chance to input a discount code. This made no difference to the price. Maybe the code was wrong. Instead I tried the serial number of the ticket. This was also accepted by the form, but didn't cause any change in the price either. Weird.

The next day I called the online booking helpdesk and explained my problem. The helpdesk person explained, in a tone suggesting he did this for nine out of ten calls, that I had to do an “online booking”
Well yes, that's what I did?
Push the button on the right marked “Online booking”.
On the right? Oh, that's a button?
Clicking the marked region brought up a form similar to the earlier one but with the difference that it had an extra field for entering a discount code at the bottom (visible by scrolling).

I leave it as an exercise for the readers to enumerate the ways in which this interface design can be improved.