There are many great ways to write your literature review (http://www.literaturereviewhq.com/3-great-methods-to-structure-your-literature-review/) and I think your idea sounds great.

However in my experience, the person who has the final say is you supervisor. If everyone thinks it's great but your supervisor doesn't, then it's not great :(

The trick is to get feedback as early as possible. Outline drafts of the other sections you want to write and schedule a time to go through them with your supervisor. Do this regularly and and you will end up with something great :)

That sounds like good advice from KB.

In terms of productivity I find the Pomodoro technique really helps http://bit.ly/j4q8uH. You basically do your work in 25 min bursts with 5 minute breaks but it's amazing what you can achieve when you focus for a short period of time.

Don't worry about your supervisor, I gather most of them are like that. The truth is that getting feedback is a good thing, some people have the opposite problem when they can't get in touch with their supervisor which is a SERIOUS problem. Don't take their comments personally but do as much as you can quickly and then when you get stuck send it back to them. Explain that you want to know if you are on the right track. The wait for more feedback and do it again. Believe me, after a few tries at this you will feel alot better. You will know what you have to do and you develop a thick skin to your supervisors comments!

Chin up, we'd all rather work in a cafe a lot of the time, you're not on your own :-)

I like medeley but I don't like to rely on reference management software as it has a tendency of messing things up. I download all of my papers into one file and name them like this "YEAR_AUTHOR_JOURNAL_TAG.1_TAG.2_TAG.3_TAG.20" the tags represent key words that I can then search for later without reference management software. Typical tags are which chapter I am writing for, the date I found the article and the name of the person who recommended it to me. For more info check out my literature review blog http://www.literaturereviewhq.com/

The amount of time you need to go back in your literature review really depends on the field. Some fields are only 3 years old, and other span decades. I think it’s best to go back to the very beginning, or at least acknowledge it. However, you can check this in other review, see how far they go back use that as a template.

As for using other review. I always use them to get started and find my first bunch of relevant references but I don’t cite them directly, it’s best to go back to the original research and draw your own conclusions. They usual way to cite other reviews is to forward people to background information that you don’t have the time or space to provide.

As for other tips a good one is to download all of your references to make it easier for them to read, save them all in one folder and give them a name like this “YEAR_AUTHOR_JOURNAL_TAG.1_TAG.2_TAG.3” Use the tags to put searchable keywords in your filename so you can search for the paper later without any problem. This way you don’t rely on reference management software either.

For more info check out my blog http://www.literaturereviewhq.com/