For many, the process for choosing a thesis topic looks like this:
- Browse open topics (e.g. on faculty pages)
- Choose the one that reads best (and is still available)
This process reminds me of ordering food in a restaurant:
- Browse food menu
- Choose the one that reads best (and is on offer)
If I am familiar with the restaurant, this process works great.
Things are different, however, if I try a new type of food for the first time. Say Afghan food. How hot is “hot” in this restaurant? What should I expect in terms of quantity—can I order starters, or will that ruin my chances for a desert? What is palau anyway?
I simply lack the experience to determine from the decription alone wether I will like the food.
If you choose your thesis topic, you are typically unfamililar with the context: the domain of computer science, the supervisor and the group he works in. It is thus hard to determine thesis writing experience from the topic description alone. How formal is “formal” in this context? What should you expect in terms of time requirements—can you continue your side job, or will it ruin your chances for free time on the weekends? And so on.
If you are unfamiliar with the context, the above process is like lottery: it can produce great results, but it heavily depends on good luck. For an evening eating out, this uncertainty can be a welcome twist. For the many months of your thesis project, however, you want more reliability.
To make a more reliable decision, you must use additional information.
For choosing food, ask friends who know the restaurant; take a look at some ingredients to get a better impression; if possible, try some dishes before ordering.
For choosing a thesis topic, try these tactics to obtain more information:
- Arrange (informal, e.g. coffe) meetings with the supervisors of the topics that caught your interest. Let them talk about the topic to get more information. Let them describe their style of thesis supervision. Only decide once you have seen/talked to several.
Read theses that this supervisor has supervised before. This tells you about typical thesis size, structure and focus (heavy on implementation? proofs?). Consider to also skim through theses from the same group but a different supervisor.
Talk to students who have been supervised by this supervisor. Get their names from their theses or ask your supervisor candidates for contact information.
Finally, you often get the best dishes from asking the waiter for advice. Often there are seasonal foods that are not on the menu. Similarly, there is a large pool of unpublished thesis topics that only exist in the heads of personal notes of supervisors. So if you like a supervisor, but are not convinced by the topic on his menu, ask for other topics, too.