Reference and Citation Management

Literature is the basis for research in any field. The past literature holds the key to what has been not done and has been never tried. With the new technological revolutions of the past two decades, we have noticed that number of papers being published has gone up in almost every field. This research publication explosion has led to a enormous stress on the researchers to keep track of the cutting edge research. So, there are software available to help us out in this messy jungle of journal articles.

Presently, many commercial and free citation management tools have emerged in the market. These software packages have their advantages and disadvantages. And of course, there is a bit of learning curve for each one of them. So, lets get on with our job of putting them together on one page! This is not the most thorough analysis and just based on my personal preference. Please pick the one you are most comfortable with and go on.

EndNote:
Proprietary Software from Clarivate Analytics

I have used EndNote from the very beginning in my college days until now. It is very reliable but doesn't have necessarily the best features. They release a new version every year. If your university has a subscription, it is worth having it. EndNote works best if you have not formed your library yet and are going to start fresh. I have had many problems with pdfs which I had already downloaded and making a working citation library out of them. It is well-established in the market so if you are downloading the citations from Web of Science and other related websites, you will get a standard citation format. It is easy to change citation styles and integrating it with MS Office. The pdf reader is a bit old-school and difficult to access as it is on the second tab of a reference.

Mendeley:
Free standard package with premium features at a cost
This software is amazing in terms of the importing pdfs and finding the relevant metadata. This works best if you already have downloaded different articles and want to bookmark them with their citations. They also have very good integration with MS Office and there is a good pdf reader with highlighting capabilities.

JabRef :
Open Source
Best for citation management using bib files and made specifically for the Latex bibliography management.Although it is a bit outdated and interface is not user friendly, but it helps in getting work done. You can very easily generate a bib file with auto-generated bibkeys.

Docear: 

This reference manager manages references along with a mindmap. It helps you find the connections between a number of references and understand what has been done and find the things which have not been done. This is for the super-organized researchers who need to know and track their references to do some Sherlock Holmes kind of reasoning as to why a particular topic needs to be explored.
You will find more details here :
Comparison between some reference managers

Zotero : Best for website bookmarking, storing online references, pdfs etc. Works best for people with extensive browser based research. The reference manager resides in your browser as an extension.
Allows to create citations in Word and OpenOffice and also tag your research. Also allows for collaboration

Burn the mathematics

I was reading through a economics book and then found a reference to Marshall's disregard for mathematics which did not relate to real life examples. I think this is very important lesson for physicists where the one needs to ensure that the implications of a simulation model have physical relevance. Sharing it for reminding myself and all the fellow researchers !

Alfred Marshall wrote about the correct use of mathematics in economics in a letter to A.L. Bowley dated 27 February 1906:

[I had] a growing feeling in the later years of my work at the subject that a good mathematical theorem dealing with economic hypotheses was very unlikely to be good economics: and I went more and more on the rules -

1. Use mathematics as a shorthand language, rather than an engine of inquiry.
2. Keep to them till you have done. 
3. Translate into English. 
4. Then illustrate by examples that are important in real life. 
5. Burn the mathematics. 
6. If you can't succeed in (4), burn (3). 
7. This last I did often.

Research Workflow

To understand how to be more productive in research, one must understand how the research flow works. This not only helps us understand which part we are lazing behind and correct our flow. If there is a disruption in any part of the research flow cycle there is a breakdown which leads to more and more procrastination. The flow cycle consists of mainly :

1. Identify Research Area :
2. Identify the research question 
3. Why is it interesting/ what motivates you and 
4. form a hypothesis/ experiment 
5. Analyze Data, Infer and conclude
6. Is hypothesis correct ? Yes ..then publish.
7. No ? Then correct hypothesis and go back to 4.

Identify Research Areas and questions

It happens quite often that in academia people are clueless after just submitting a paper. It is like an existential crisis, What should I do now ? Where will I go ? All these questions demand you to find a new vocation to engage your burning mind. Therefore it is always good to have ideas which strike you during the project and write them in a book/record.
Find a research question is to essentially find the gaps and loop holes left by researchers. You can find these open questions at the end of the latest review articles or papers. 

1. Set aside some thinking time : Find papers to read and try making a mind map as to what are their contributions to the fields. Go away from computers and have some thinking time. 
2. Keep track of ongoing research: Mind Maps is the way to go in keeping track of most esssential ideas. Summarize a paper in few key words and then write the conclusion in one line.
3. Record Ideas : Make an Idea Book to keep all your silly, crazy ideas in one place. Then review them as to when to convert them into papers. Assign them priority in terms of their do-ability and the impact they will have on the community.

From idea to paper 

Converting a simple thought into action is almost similar to converting energy into matter. It is creating tangible out of intangible. This conversion needs effort and persistence. Your ideas will get recognized only if they get validated and published. If you keep them in your idea book for too long, you get scooped which means somebody did what you intended to do. This is very frustrating if you were doing repeat experiments and hoped to make the research into a high impact paper but it finally ends up useless. So, the execution must be fast enough. If you are not able to drive the project, ask someone else to help you out. An external viewpoint always helps understanding what is stopping you from finishing the project. So go ask a colleague or a friend, what is going wrong.

Steps to follow :

1. Idea/Hypothesis 
2. Validation/Experiment/Simulation 
3. Correct hypothesis if not fitting the data
4. Repeat the experiments to confirm
5. Understand and write up

Importance of experimental log and periodic review

As a serial criminal in this regard, I can't just feel more regretful for not carefully documenting the experimental details and procedures followed during the course of your experiments. You may think that you will remember the reason you named a file abasas.dat on the day of your experiment, but the next time it will just appear as stupid data much like the others. Also, I have seen this a thousand times happening with me that the name of the experimental files begin with new,final. e.g. NewData.txt, Finaldata.txt, AmazingResult.txt. After a few days, you might forget what the context was. A good practice is to use a sample prefix for a set of experimental data, add a date and time if it is important, add the basic conditions at which the experiment was performed. This gives a clear usable file name which will be usable until eternity by your academic generations.

Periodical review is very important as well. However you named your file, you will still not be able to find it when you need critically while making a powerpoint. Make sure you analyse your data the following week and then make all the usable plots out of it. After that put them in a powerpoint presentation which makes your data more visible. You can add your inferences from the data.

Backup research data

Backup as much as possible. You always can delete the copies. If you lose the only copy that ever existed, you are pretty much busted. Signing up for cloud storage has its own advantages. Google drive and Dropbox have great starting packages and quite useful for storing your hard earned data. It always good to have access to your data from multiple computers.

Limited Working Hours/Time

One cannot stress the importance of time management enough in a research environment. We have to manage time to achieve your goals and also leave enough time for the life outside lab. Thinking that more hours in lab will lead to better results is a vanity. One must start the day with the aim to finish by evening and leave the lab for the next day. You could always read in your free time at nights but spending it in lab will not help out.

Will keep updating this with new ideas as and when I get better with my research life. These are few things to remind myself and be better at research. Please do share if you find some important missing here !!