Masters thesis - do I need chapter summaries?

AARRRGH

Hello

I'm writing up my masters thesis at the moment.

Do I need to add chapter summaries to each chapter?

Thanks

Tom Dunne

No, you don't.

You will have an abstract at the very start of the the document, that will be the only summary required.

Just make sure your chapters are well structured. For example, each one should have an introductory section, followed by the "meat" of the chapter and possibly a closing section.

AARRRGH

Thanks Tom. Can I ask you another question if you don't mind?

My thesis is on elliptic curve cryptography. Initially we had no angle on what I was trying to discover or examine -- it was simply a case of try to get the thing implemented and we'll figure out what the thesis should be on.

Well I discovered there are little to no readable non-beginner elliptic curve cryptography books, and as a result had to create all the algorithms from scratch myself. I also noticed my polynomial implementation turned out to be slower than my integer implementation. The reverse should have happened!

So in the end I managed to get it implemented using my own algorithms.

I still don't really know what the story of my thesis should be though. At the moment I'm just explaining every aspect of elliptic curve cryptography and mentioning things I noticed. But I reckon my thesis is too dry, and it doesn't really have any sort of narrative.

In my introduction chapter should I mention the lack of documentation, and how I had to kind of discover everything from scratch?

Is that enough of an angle to take?

Any advice greatly appreciated.

Thanks.

Tom Dunne

AARRRGH wrote: »

My thesis is on elliptic curve cryptography. Initially we had no angle on what I was trying to discover or examine -- it was simply a case of try to get the thing implemented and we'll figure out what the thesis should be on.

Ok, I was just about to say Dublindude was also doing that, but, well, duh. Quit confusing me.

AARRRGH wrote: »

So in the end I managed to get it implemented using my own algorithms.

Which in itself is an achievement. Can you prove conclusively that what you have implemented is elliptic curve cryptography? If so, mission accomplished.

AARRRGH wrote: »

I still don't really know what the story of my thesis should be though. At the moment I'm just explaining every aspect of elliptic curve cryptography and mentioning things I noticed. But I reckon my thesis is too dry, and it doesn't really have any sort of narrative.

Your narrative is what you set out to do, what you found out along the way and what you actually achieved, along with relevant proofs. Your supervisor should be reading the document along with you and be able to give you direction if the document is wandering and not focused enough.

AARRRGH wrote: »

In my introduction chapter should I mention the lack of documentation, and how I had to kind of discover everything from scratch?

Certainly, and I would look favourably on a student who teases out the main thrust of a subject for themselves. But, and it's a big but, you need some way of confirming what you achieved is not some tangential nonsense that bears no relevance to the topic at hand. How do you prove that what you have achieved is in fact elliptic curve cryptography? So you must have had some reference documentation.

AARRRGH wrote: »

Is that enough of an angle to take?

I can't conclusively answer that, it is up to your project supervisor (say hello to Tony for me ). But, from what you have said here, I would say yes. But again, I can't emphasise it enough, in a topic as specialised as this, you need to be convincing in what you are saying, providing conclusive proof of what you found and what you implemented.

On a side note, I would be very interested in reading it.

AARRRGH

Hi Tom

Thanks for your reply. The way I prove I implemented ECC correctly is by demonstrating encryption and decryption. I don't claim, however, to have implemented it perfectly, for example, I am sure my code is open to all sorts of attacks!

My conclusion is pretty much this is a difficult topic with a lack of documentation and the difficulty I had trying to decipher everything. Hopefully Tony thinks its ok

I'll let you know how I get on!

Cheers