The UvA is introducing a new system for tracking expense claims (travel, conference fees, &c.). Actually, they’ve already introduced it, a fancy web app which only runs under IE on Windows machines, and they’ve already decommissioned the old-fashioned paper system. If you send your paperwork to someone who can do anything about it, these days they just have to send it back to you with a sharp note. (See note below regarding whose fault this is: not theirs, I know.)

The problem is, the new system has some usability issues. I just got an email explaining some of these problems, and the hoops we have to jump through to avoid them. For instance (emphasis in the original, terminology slightly altered):

Then you get a table. In this table, put 100 in the percentage column, delete the > ‘dingleberry’ and put the relevant dongleberry in the column ‘dongleberry’. Don’t try to > use the search function [intended to look up dongleberries –tikitu] but use the number > from the attached list with dongleberries. Then click ‘copy’.

I’m not sure what’s funnier: that we have to fill in the 100 percent by hand (there’s even a note in the usage handbook that this is the only value possible for this column) or that our poor long-suffering administrators had to send us a list of dongleberries because the search function that was intended to let us look them up instead sends an empty claim request to the Dean of the Faculty of Humanities and terminates your session.

But my absolute favourite was,

In general, do not use the help functions. They are not very useful and may cause the > system to crash.

(In case it’s unclear: I don’t know who to blame about this, but it’s not the long-suffering administrator who sent that email. In fact, I’m hoping that due to her efforts I might actually be able to submit a claim next week. I just can’t help wondering how much the university paid for this work, and what on earth they thought they were doing when they decided to make the entire staff alpha-test the software.)

Update 20/2: our faculty newsletter today had an entry about this system. It included the following sentence:

De ondernemingsraad FNWI signaleert dat veel medewerkers van de faculteit de moeilijkheden > tamelijk gelaten accepteren, in de redelijke verwachting dat kinderziektes bij een dergelijk > systeem normaal zijn en binnen afzienbare tijd overwonnen kunnen worden. (Emphasis mine.)

If I’m translating this right, the section I’ve highlighted says something like “the reasonable expectation that growing pains are normal for such a system.” And I’m left again gobsmacked at the level of either BS or plain incompetence that’s on display here. If our policy-makers don’t know the difference between “growing pains” and “widely deploying untested software” then they shouldn’t be setting policy about software use. And if they do know the difference, then they should know better than to do it.