2

I am doing a degree in Computer Science. I am not in my final year, however my lecturer has asked me to do a dissertation over the whole of this year as a sort of dry run. This is so when I go to do my actual dissertation next year - the whole process won't be foreign to me.

My project (which I have had signed and approved) is to write a piece of software which converts SVG images into front end application code (converting a button design into an actual coded button which can be used in c# applications).

My problem is that I'm struggling to find areas to research for my literature review. My project isn't innately about any topic of research, it's just creating a piece of software.

Methods and results of what I've tried:

  • From what I've read, this part of the essay should include literature for/against (or opposing thoughts) of the topic at hand. The only way I can break this down to fit my project is - Should/Shouldn't User Interfaces be created from deigns. Could this possibly lead to less accurate representations of what was needed? Will the controls created be as efficient as a control created by a person based off of the designs
  • What is known or already understood about this topic. I've not yet found anything suitable that another person has done with regards to converting SVG's to code.
  • 3
    I would suggest you simply ask your lecturer what they are expecting. – Nate Eldredge Nov 22 '15 at 16:34
  • @NateEldredge Yes I have emailed him however he is away for two weeks on paternity leave and I would like to start witting it up now. – Ralt Nov 22 '15 at 16:42
  • 1
    @Ralt, I'd be surprised if a CS professor didn't check his/her email occasionally over their leave. – Ric Nov 22 '15 at 17:46
6

Figuring out what to put in a literature review usually involves taking a giant step back and thinking about (1) what someone needs to know in order to understand your work, and (2) how your work fits into the overall body of research. For example, you might want to address the following:

  • What are vector graphics (like SVG) and how do they differ from bitmap graphics (e.g. PNG).
  • Why is there an interest in automatically generating code?
  • What other attempts have been made to automatically generate code (not just from UI designs, but also from other types of designs like UML diagrams)?
  • How successful were these attempts?
  • What problems have been encountered?
  • What is the state of the art in automatic code generation today?
  • Why isn't automatic code generation more widely used?
  • What benefits and drawbacks are there for automatically generating code from a graphic, as opposed to from a drag-and-drop approach as is typically used in Visual Studio.
3

Ultimately you need your professor's take on this. However, a literature review doesn't have to just be about research. It would probably be helpful to find similar programs and study how they handle this problem. In doing this, you can learn what works, what doesn't work, and areas where you can add improvements. If writing a program like this is common for dissertations at your school, you can approach more senior students or other professors who are familiar with the process and ask them how this is handled.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.