Bachelor Thesis Log 19

November 9th, 2010

Welcome back to the latest bachelor thesis report, a few hours early today for a change. This will probably be the fourth last one, since I have to hand in the thesis on December 6 at the latest. That actually calls for a new schedule. But before that, here's a general status update.

Completed Bits

In the last few days I have presented a detailed teachlet procedure on four to five pages (including some illustrations) as indicated two weeks ago. This should especially help teachlet newbies to get an idea of the concept when reading it for the first time. So that point of the list is now checked off.

I also took some time to explain the typical role of the class diagram in teachlets better, in the section about the average teachlet. That's where it makes the most sense in my eyes, since working out class diagrams on the whiteboard is not a teachlet requirement and has nothing to do with the definition, but is more of a tradition from the teachlet workshop.

I also came up with an example of a teachlet that does not require an initial system. I described what a teachlet for the observer pattern could look like, such that a system is developed from scratch in BlueJ. The example is about cameras that can automatically track a soccer ball – I found the idea quite amusing. To put the code into the appendix I was able to use my experience with listings and PDF file attachments in LaTeX again.

Concerning style and small corrections: Axel had noted a few wordings that I had not managed so well, I have now corrected them. In the glossary I had forgotten the “code inspection,” this term is now also explained. I also solved an old problem, which I think I mentioned here a long time ago: in my big diagram, the surfaces so far only differ by colors, which makes the whole thing completely incomprehensible in a black/white printout. That's why I introduced hatching in addition to the colors, which I hope will make it possible to distinguish the categories even in a grayscale world. People with color vision problems may also be appreciative. By the way, this was a lot (a lot) more work with Inkscape than I had expected. Are there any Inkscape hackers here who know a way to do this that doesn't take two hours?

The document has grown to a total of 56 pages at a hefty 820 KB.

Four Weeks Remain, What's Next?

I still need to find out whether my cover sheet contains all the important information and meets the formal requirements. I don't think there is a definitive template – the examination regulations state the info that must be on it, and there are guidelines for course papers from SWT that also say something about this. I'll have to consolidate that in the near future.

As far as possible content points are concerned, I now only have one on my list:

• Describe teachlet-related methods: at the end of the paper, some key points are evaluated as indispensable for teachlets. Perhaps ideas for teachlet-like approaches can be found from this, which I could then briefly illuminate. This should be possible, even if I won't write several pages about it – it's actually not part of my topic, strictly. But we'll see.

That could lead to a few more pages of text. I hope I can write it in time and still get feedback on it.

A New Schedule

Fortunately, I don't have many special assignments in any courses right before the deadline. That should give me enough opportunity to keep some buffer time in the last days of the thesis. Unfortunately, I might have to manage without external proofreading, but according to Axel's statement, my text is largely issue-free in terms of language. Right, I pay more attention to small details when writing my thesis than I do for a blog post like this.

Let's try a schedule for the next month:

• From now until approx. November 20: remaining writing tasks
• After that until November 30: feedback from Axel and possibly from other interested people
• From December 1 to December 5: buffer time, printing and binding
• December 6: submission of the thesis

That sounds quite feasible to me. I have to pull myself together now, but the bachelor's degree is not supposed to be a cakewalk.

Future

I'll be back next week at this time with the status of the rest of the work I've managed to get done by then. I'm curious myself how well this will work out. If you want to have a final pre-release for feedback purposes at the end of the month, feel free to get in touch now. Otherwise see you next Tuesday!