Friday, April 27, 2007

How to write a thesis

I am currently reading a lot of PhD theses to see which I have to mention and were I can copy^h^h^h cite stuff. It also helped to figure out how a thesis shoud be written. The general structure seems to be:

1) Intruduction:

From the dawn of mankind people have always wondered about <obscure purely academic problem you write about>. This thesis will present the latest and best theory and/or data and will show mankind the way to a brighter future, where there is no hunger, no wars and everybody knows and cares about <obscure purely academic problem you write about>.

2) Theory:

I have read a lot more papers than you did, and I cite all of them. I even cite stuff I haven't read. I cite more stuff. I can use fancy words which nobody, even me, has any clue about what they mean.

3) Experimental Apparatus:

We have the best apparatus ever. You can't even imagine how great it is. The latest technology, the best ideas, blablabla.

4) Methods used:

I took some software, changed some parameters and let it run on a bunch of data. This took two years until I realised everything was wrong, because of a stupid bug and I did a fast workaround and got some results which might be correct.

5) Results

I got some numbers. Of course these are the best numbers, better than that of the other experiments. Some fancy plots and some vague statements where you have no clue about your data. Try to cover up the discrepancies by comparing plots with different scales and axes so that nobody notices that they show inconsistent data.

6) Conclusions and outlook

Paraphrase introduction.

Bibliography and Acknowledgements

Be sure to include advisor, referees and anybody who will have a say on your exam/defense.


Mariana said...


Fica é uma tese com poucas páginas e a minha faculdade gosta de tijolos... Mas acho que vou usar como base de trabalho! ;)

Kryptikmo said...

I was thinking about this, and have decided that PhD Acknowledgments are almost useless. Did your parents really help you? Or your wife? Or your supervisor? What you really want to write is along the lines of:

"I acknowledge the following contributions to my work:

Anadin - for making excellent headache tablets

The breweries of Germany, Britain and Belgium, and the vineyards of Italy, France and Spain for making it all more bearable.

My fellow PhD students who listened to me moan, and gave me somebody to laugh disbelievingly/cry hopelessly with.

The technicians.

The wee lady in the van/bakery that sells sandwiches. They were tasty."

But instead you end thanking people who secretly should end up in this section, detailing all the people who got in the fucking way:

"I want to blame the following people:

My neighbour - do you really need to play music at 2AM???

Friends from school - no, I don't want to know that you are married, happy and have a house. Yes, I still rent. No, I don't know anyone outside of work. Shut up.

My parents - yes I'll get a real job now. Yes, there's a ceremony.

My wife - yes I'll get a real job now. Yes, one with money.

My supervisor - yes I'll get a real job now. Thanks for all those "ideas".

The graduate school - when I have work to do, I don't want to be forced to go to a seminar given by the HoD's wee pal. I don't care about astronomy.

The wee woman in the van/bakery that sells sandwiches - I'm fat now, and it is your fault."

Your own may vary. I realise that's slightly harsh on the sandwich lady, but she'll never read it anyway.

Also: You missed one section. The endless appendices in which you include everything you ever did, regardless of relevance because it's taken fucking years and goddammit, its gonna be more than 150 pages if it kills me.