Are there any recommended books or other sources about research in IT?

For example, I want to do some research on the subject 'Internet of things'. What would be the best way to approach this? What methods should I use?

  • $\begingroup$ A reference request like yours is too broad for Stack Exchange -- you ask for a survey of a whole research area! You need to narrow your focus considerably before a question of reasonable scope appears. Try talking to your advisor(s), search with Google Scholar and check out this guide to better (re)searches on Academia. If you want to learn how to do a literature search, see crypto.stackexchange.com/q/8316/351. $\endgroup$
    – D.W.
    Feb 12 '15 at 12:33

You do not do research on a topic out of the blue. You must have read enough, understand the basic concepts, the type of problems and the technology, and hopefully be proficient enough to form your own views of how things works, possibly how they should work, how they might be improved or completed. Work can be theoretical (devising new algorithms for faster or more reliable solutions). I may involve developing new concepts, evolving or unifying existing ones, finding new answers to old problems, finding new problems, and hopefully solving them. It may also concern new applications. It may be mostly abstract, or be mostly implementation (if it includes new techniques for achieving results), though it should ideally blend both abstraction and implementation. I probably forget many aspects.

In other words, it is essentially varied and undescribable, depending very much on you own relation with the topic, and the orientation of your necessarily deep knowledge of it.

It may occur that your adviser will serve you a good topic on a plate, with user manual and guidelines to achieve the research and get a PhD. But I would not count on it too much.

So I have 3 advices

  • read
  • experiment with existing systems
  • and read a lot more.

And while doing it, always keep an open and critical mind. Question any assertion you find, take notes, make sure you understand what is being said or implied. And do not expect everything to be true. It is not that people cheat (though a few actually do) but we all make mistakes or take unwarranted shortcuts, with the best of intentions.

If you are mature enough to do research in your chosen area, you no longer need to ask your question.

The way I started most of my research was unexpected. I wanted to learn on a topic. I started reading, and after a while thought I would do something differently and started developing my own idea. Sometime I was already very knowledgeable on the topic, but reading critically a paper, I thought that more could be done. New ideas come often from absorbed knowledge. When you have lots of knoledge on a topic, you can also try to unify it, condense it, and it may yield unexpected new results.

Regarding what you should read. I do not know. If you start working on the topic, you will quickly find out from citation and discussions on the Internet. But recall that there are also hidden gems that no one talks about, that may be the key to interesting work. Books are useful to get started on a topic, to get an overview and basic knowledge. But it is not likely that they will be enough to start doing research. For that, you generally need to read research papers, and that may be less easy (people are still struggling with new concepts), but it can be very rewarding.

So you may take any book with some reputation (and possibly pedagogy) to get started on basics. After the first, you should be able to choose the next on your own, in directions you are interested in. But start reading recent research papers as soon as you can, and look at their references. The rest should come on its own.

You can be a great teacher without being a great researcher. The converse is also unfortunately true. But both need to read a lot. Great teachers can write great textbooks.

  • $\begingroup$ This seems to be a decent meta answers we could give all book requesters. $\endgroup$
    – Raphael
    Feb 12 '15 at 13:07
  • $\begingroup$ @Raphael I know it is not done, but thanks. $\endgroup$
    – babou
    Feb 12 '15 at 13:11
  • 1
    $\begingroup$ I don't quite catch your tone from the text, but -- for the record -- I was thinking that we should maybe have a meta question explaining in detail to people asking "which book should I read?" and "what should be my research project?" and the like (we keep getting those) why/how they go about things the wrong way. Your answer strikes me (from skimming) as a good starting point. (As always, we can discuss this in Computer Science Chat if you want.) $\endgroup$
    – Raphael
    Feb 12 '15 at 13:13
  • $\begingroup$ I was more trying to answer the question: how do I start doing research? ... and my answer is by acquiring knowledge, getting it in your guts, and thinking about it. This is a bit different than what books should I read? The funny thing is that I am a bit mystified by people who ask, how do I do research? As I recall, I just started doing it, just by trying to understand and question, and answering my questions. You may have noticed I am not too good at chatting, but we can try. BTW, while answering you, I tried to complete my answer, so as to address the what should I read aspect. $\endgroup$
    – babou
    Feb 12 '15 at 13:40

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