View source for DistOS 2014W Lecture 17

== What is a Literature Review? ==
We shouldn't summarize everything. Instead, we should be doing a critical analysis/comparison of the all the work in a chosen area. We should cover all of the significant points in said chosen area. 

Tips for a literature review with Anil:
* Try an organize the papers thematically rather than presenting them all linearly (i.e., "This person did this, this person did this, etc.")
* Categorize papers into several themes/categories that relate to your topic and present the papers as part of a theme. This makes it easier to compare/contrast similar papers and to see how all the different chunks of knowledge/work relates.

== Point of lit review ==

With a research proposal, you would be asking, "What can be done to plug the holes here?" The point of a literature review, in this sense, is to try and offer new interpretations, theoretical approaches, or other ideas. 

A more traditional literature review comprises of providing a critical overview of the current state of research efforts.

A stand alone literature review of articles best fits what we're doing in this class. This typically involves:
	* Overview and analysis of the current state of the art in your chosen topic.
	* Evaluate and compare all the research in this chosen area.
	* It might be difficult to really show weaknesses and gaps in the area as you tend to need to be an expert in the area to find these gaps. But, you can criticize the overall body of work and say what's missing and give general areas of future work.

== Structure of a Research Paper ==
* Abstract
* Introduction
* Literature Review (about 1/4 to 1/3 of your thesis, typically, so if you can do the literature review for your thesis in a class, that's a great bonus)
* Methods
* Results
* Discussion
* Conclusion

== Finding Sources ==

A good way to go is to first look at tertiary sources (e.g., Wikipedia articles) to get a general idea of the area and what it's about. Following this, you can go into secondary sources to determine the history and general themes of the area. When you have a better understanding of the area overall, jump into your primary sources and get into the "nitty gritty" of the most current research.

A good way to find articles and papers to cite is "footnote chasing". In one of the papers you want to include, look at their related work and footnotes. These tend to be good citations that you should look up and include in your work--it helps you find out about the area as a whole, giving you a good base to start on. 

== Tips Writing the Literature Review ==

'''Step 1:'''
Have one document that has a list of all your sources. With each source in the document, have a short paragraph/blurb of what the paper is about, how it works, what they did well/didn't do well, etc. This will show that you understood the papers and will save you having to read them all again later.

'''Step 2:'''
Once again, thematic organization. Start to classify your papers into a few categories relating to your topic. This makes it easier to do things such as discoursing on the area overall in the introduction, abstract, etc. It also allows you to better organize the paper later and to compare/contrast them.

'''Step 3:'''
Once you have the categories, this is where the paragraphs you've written in Step 1 come in handy. You can write an introduction for each of your categories then just dump in the paragraphs from your document. Following this, you can polish your sections and better integrate them.

'''General Notes:'''

* Have a good introduction, outlining exactly what you're going to cover, why it's important, etc.
* Make sure to keep things high-level--you shouldn't have to be an expert in the sub-topic you've chosen to be able to read and understand your literature review.
* Make sure to be selective with what your'e including. Include all of the really current things and what is historically relevant for your topic (i.e., either very current work or very important, groundbreaking work for the topic). Intermediary work doesn't always need to be included. Don't include citations just to get your citation count up, this actually detracts from your literature review.