Things I wish I knew before I wrote my master thesis

A list of things I wish I knew before I wrote my master thesis:

• Define a super clear hypothesis and null hypothesis. Sit down with your supervisor and literally write “hypothesis: blah blah. Null hypothesis: blah blah”. This might not seem applicable to your subject, do it anyway. It will help you with a red thread throughout your thesis, and your supervisor should be able to help you with this. If not, maybe you both should look over what you are doing...

• Safety copy everything preferably by date, and make notes what differs between the different copies. Keep a spreadsheet or dokument of the different versions of what ever it is you're doing.

• If you work with a model or code and turn specific commands on and off, write a comment on when the change is made. Ex: turning on or off the albedo. Then you don’t have to wonder how long it been of when you suddenly find it off last week of writing and wonder how many model runs were made without albedo.

• Save your articles in the format “Author Year - Title”, as you will need to be able to find them based on author name. This is because when another article refers to them it will be by author name and year. Title is convenient for when you look for a specific subject.
  Ex: Assman et al 2013 - Variability of Circumpolar Deep Water transport onto the Amundsen Sea continental shelf through a shelf break trough

• Bring a printed copy of your thesis to the meeting with your supervisor, so that you can make your own notes in the text as you go through it. Pen and additional paper is good as well. 

• Ask your supervisor to make notes with ink, it is difficult to pick out small pencil corrections.

• Print your paper and read through it before handing it in. A lot of formatting mistakes and weird sentencing is easier to discover when reading a printed copy.

• Mark of the supervisor comments you fixed with a contrasting colour pen. That way you don’t have to wonder what you fixed.

• If you use some data and find a big change in data over time, for example an regime shift in wind, double check it towards other databases.

• When making the presentation, you can put text over "unimportant" parts of pictures/plots. 

• The presentation text should be short and highlight important points of your pictures/graphs/plots.