I want to write a scientific master thesis. I dont want it to look like a technical report.

I had to configure a system: It contains roughly 2 chips. Then this system has to send data to another system with same architecture. I had a lot of work to programm one of those chips. Looking at other scientific theses, I don't see that they even mention how they programmed their chips.

I want to show the difficulties of programming it. Maybe only some important code.What do you think about it? I dont understand the scientific way of writing a thesis. So maybe you can explain it easy.

I tried to look at other scientifc work:

E.g. a white paper shows everything in one page, or here on three:

For a normal human being, it would not be possible to recreate this, without the basics and also test environment. Its like someone crazy on the street tells me what they have done. For understanding I should recreate their experiment. This would take a lot of time.

E.g a thesis:

It looks the same but bigger. It is also not possible to recreate it without knowing more basics. But I already saw that the author points to some literature.

But it is like someone on the street tells me, what they have done and tells me I can look into books to find out, how to do it in real.


Is it right to use language like:" (...)we have already discussed in section (...)" Shouldnt it be like" (...)it was already discussed (...)"?

UPDATE: I am reading a book about scientific writing: The work shouldn't show the chronology. It is no experience report. It is a result oriented summary. So keep in mind that you choose a logical structure, that has no repetition of results. Don't structure it in that way that you need to present results before they are done in the thesis. The process of creating the thesis means often that you work iteratively, with trial-and-error. The steps are irrelevant only the planned methods how to do it and the results are relevant. Only present results in text that are relevant.

The whole work process/effort doesn't play a role. Talking to people, reassembling a working peace or reproducing components are not relevant for the result. Same is for long looking for an error in the program code. A big effort is usually treated with little to no scope in the thesis. The advisor will take it into account....

  • 3
    Please clarify your specific problem or provide additional details to highlight exactly what you need. As it's currently written, it's hard to tell exactly what you're asking. Jun 26 at 9:35
  • thank you for your answer! I tried to change it. Hmm I just wonder why in scientif thesis they only describe roughly the methods. they also show some results and discuss it . but they dont show in detail how they have done it...
    – t1fpga1
    Jun 26 at 10:04
  • 1
    Understood- I'm thinking it may also help to know for what purpose you're writing the thesis (for degree requirements, presumably?) and in what subject. This may help find someone who has an idea of what to suggest. Jun 26 at 10:06
  • 1
    ok I added master thesis .
    – t1fpga1
    Jun 26 at 10:21
  • 1
    What does your advisor say about those points? Jun 26 at 11:07

2 Answers 2


I'm interpreting the question here, for the sake of potentially being useful to other situations, to be: "How does one write an academic report on a project whose major output is a piece of software?"

  1. You mention that this is a thesis. You should have a thesis adviser. It's a tired trope, but you should ask your adviser first for their view.

  2. Your target audience is not "someone off the street", it is specifically the person examining your thesis, and more abstractly someone who is reasonably acquainted with your field of study but does not know about the specifics of your project. A common and useful piece of advice is to imagine that you are writing a document that would convey as much useful information to yourself before starting the project as clearly as possible, and within the space requirements.

  3. For software based projects, it's typical to have a written report (that is, your thesis) which explains how the software works at a high level, why it was made, any novel contributions compared to the state of the art, discussion and interpretation of specific results etc. One then accompanies it with an implementation. There are different levels of robustness you'll find for these across disciplines. In an ideal world, an implementation is comprised of open-source software an interested reader could obtain and use themselves, including documentation that would allow them to use the software. Separately one provides data inputs and software which uses the implementation to reproduce any results or re-perform any tests mentioned in the written report.

    In practice, unfortunately, many researchers provide just the software they used but not in a form usable by others, along with any output data products presented in the report. Some provide no code at all and just the output data products, which seems to have been a standard for many years that thankfully seems to be changing.

Part of the training you're receiving in a Masters degree is learning to communicate about technical subjects in a way other people can understand. This is not easy. Code is written in some language that has been developed as a compromise between what a human writer and a computer reader can process. English is usually more efficient for human-to-human communication.

  • thank you really for your advice! 1.) advisor is key. but he wants also that I understand it. 2.) ahh so it will be like my advisor. and useful information ok this is like introduction, looking for similar work, and the state of the art and my work environment 3.) yeah the problem was i had to program the hardware(chip) with software ( some sepcial code) and it was not that easy to do it becasue i had to check a lot of things. e.g.: I had to add some code and in general make the code run and make sure that it connects well to another chip.
    – t1fpga1
    Jun 26 at 12:30

A technical report states how something was done.

A scientific report states a hypothesis and how it was tested. Your explanation so far does not state a hypothesis.

  • CS is a bit different from classical science. Not all theses have that structure, actually, though many do.
    – Buffy
    Jun 26 at 20:23
  • That's true. The question is tagged computer science, but the question asks about "scientific work." Definitely not the same thing. Jun 26 at 20:46
  • 1
    Actually, I think the OP is a bit confused on this issue. Not unexpected in a novice.
    – Buffy
    Jun 26 at 20:55
  • Hey Anonymous Physicist thank you. This what I also wanted to know. A hypothesis and how it was tested. What should I do if the task was to check if I can improve a speed connection by using a specific protocol ? Finally I implemented the connection and tested it. What is scientifc about it?
    – t1fpga1
    Jun 27 at 6:10

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .