Course Surveys 2009

Gernot's Comments on CATEI Survey

This was impressively positive, across the board better than last year's (which already was very good). However, with a return ratio of 40%, not too much can be read into this.

Free-form comments

Nice to hear the positive comments on the tutors. They did indeed work very well.

I appreciate all the improvement suggestions, and think we can address most of them easily next time round. Specifically:

• Give demo mark during demo: This is supposed to happen, there is no reason why the tutors should hold this back. We'll issue clearer instructions next time. Also, I agree that marks should be entered in SMS ASAP.
• Three-hour lecture right after the demos: This is a perennial problem. There is very little we can do about the timing of the lecture (the alternative this year was 11–2 on Wed, which I think would be worse). At least the Fri-arvo lecture block means we can all feel good about going to the pub right after.
  
  However, we can do something about the milestone demo time, and I'm somewhat surprised that this hasn't come up earlier. I intend to have the milestone demo on Mon in the future. This actually helps scheduling m0, and means students can pull their all-nighters far away from the lecture ;-)
• Scheduling lecture material is always a bit tricky, particularly with me travelling so much. But I agree that caching should come earlier. There is no real need for having the Week-3 material so early. Will fix next time.
• The issue about final docs vs code submission hasn't come up in this form before, and we'll need to think about it. The idea was to leave time after debugging to do the docu, as else it's likely to be de-prioritised to debugging. But maybe it's educational to require both to be done at the same time? We'll consider this.
• Provided docu: This one will be hard to fix. Trust me, the level of docs we provide is generally of good standard compare to what's around elsewhere. I suspect this relates more to particular bits, such as the network stack (which is a horrible hack...)

Gernot's Comments on LiC Survey

Many thanks to all students for taking the time to answer the course survey. A return rate of >80% is very good and makes the results highly representative.

The results mostly speak for themselves. The course seems to be in good shape, overall satsifaction was better than the previous year and possibly the highest ever (but given the small size of the class, the statisitical significance of this result is low, so should not be overrated).

Specific observations:

• The biggest gripe was the quality of the documentation. Which is interesting, as previous years that was also the biggest gripe, but then we were using a research platform, while for the last two years we used the commercially-supported and OKL4 platform. I guess part of the problem is that we're programming directly to the kernel API, which isn't the recommended way of using OKL4, and thus not so well supported by docs. (But making more use of the library API would have hidden some of the stuff I wanted you folks to learn).
  We'll keep improving docs, but we also provide a lot of support in face-to-face consults and on-line forum...
  We're considering moving to seL4, but that's unlikely to improve the documentation situation. But in real-life, docs are usually much worse.
• Workload is heavy — sure. It's meant to be that way. You won't learn building real systems without that. But we're always very up-front about this, so students know what to expect.
• Timing of demos vs lecture, etc: As discussed above, we'll try to change this.
• More about internals of main-stream OSes: Fair point. This is some of the stuff that's been squeezed out by the 12-week session. We'll see what we can do about it.

The rest are one-off comments. I'll look at them again before the beginning of next year's offering.

All up, that was a lot of useful feedback. Thanks to all those who submitted!

Gernot