Iva Laginja

I’m a Croatian-Austrian astronomer and work on high-contrast imaging. I work as a CNES postdoctoral fellow at LESIA in Paris, France.

PhD starter pack

Introduction

Welcome to my page on recommendations for all fresh PhD students in astronomy and/or their advisors!

Embarking on a PhD means there’s a lot to learn ahead and I consider the ownership over and mastering of certain tools an absolute necessity to render that learning process as effective as possible. This post does not contain any groundbreaking insights or hidden gems of the research world. In fact, the tools I suggest here are used fairly standardly in the international research community. The main goal here is to give you an exposure to them early on with the main incentive being to avoid the famous: “If I had know this earlier, it would have made my life so much easier.”

As all advice, always take it with a grain of salt and most of all, don’t forget to do your own research on it.

To get you started, I invite you to have a look at this “illustrated guide to a PhD”:

The illustrated guide to a Ph.D.

Fast summary

PhD starter tools

During your PhD, you will learn a lot of new technical and organizational skills that will carry you through the rest of career, no matter what way it takes you. This will include learning to use some of the most important tools in astronomy research, and the ones listed below are the some of most important ones to get you started as a responsible researcher.

I recommend that during your first couple of weeks as a PhD student, you take your time and learn about all these tools, and set up accounts for them one by one

ORCID iD

When you get into doing proper bibliography, if not earlier, you will find out very quickly how it can be hard to track down certain papers if all you have is an author name. This can happen when people tell you to “look up that paper by XYZ that talks about this”. Not only will have one researcher most likely authored more than one paper in their life, but there is also some ambiguity on finding one specific person when they share a common first or last name with others, or have changed their name over the course of their life. If only every astronomer had a unique ID that would make it easy to find their, and only their publications with a single click….

This is where ORCID comes in. Read everything about ORCID and ORCID iDs here, and make sure to create your own ORCID iD asap. Many journals will ask you for your ORCID iD when you are a paper co-author, so it’s good to have one from the start.

git

I don’t care what language you code in, I don’t care what client or editor you use, but you absolutely need to use version control. The most widely used and user friendly tool for this currently is git. You need to learn it, nowadays it is as integral a part of being an astronomy researcher as it is to know LaTeX. It will let you (a) version your code, making sure you can go back to an older version if you need to, (b) set you up for collaborations, because nothing is more disgusting than to send bits of code back and forth per email, and (c) provide you with a free tool to back up the code you write during your PhD.

If you don’t know where to start, have a look at this tutorial:

https://swcarpentry.github.io/git-novice/

I also recommend that you create a GitHub account, as this is the open source platform many researchers use to share their code and collaborate.

LaTeX

You have probably already worked with LaTex in your undergrad. If not, it’s about time. Most astronomers write their papers in LaTeX, and so will you, most likely. Now, you could install it on your local machine, and also install an editor for it on your local machine, but why bother if you could just use an online editor that saves your documents for you, rids you of the need to install anything at all and lets you share links and documents for easy collaborations? Creating an overleaf account will make all of this significantly easier. Handily enough, you can create an account directly with your ORCID iD which you surely already created!

Also, it’s worth checking with your group or institution whether they hold a license, which allows you to share your overleaf/Latex project with a larger number of persons, which is a very useful option when writing papers or applications.

Bibliography management

It is very likely that you will start you PhD by reading a ton of papers. It is even more likely that you will need to remember those references when you are writing your first paper, and especially when you are writing your thesis. This means that it is super useful if you find a good way of organizing a bibliography early on. There are two main parts to this:

Saving a copy of the paper, with permanent access
Saving the BibTeX references of the papers

Even if we live in the marvellous world of the 21st century, there is no good tool yet that allows you to do both in the same place. But having hundreds of paper PDFs scattered about on your computer also doesn’t seem like a good solution, so we can only minimize the damage.

Find a reference management system that suits you (e.g. Zotero, Mendeley) and set it up. Collect your papers in there, as sometimes you will get papers sent directly from authors or colleagues per email, as many are behind a paywall (RIP all those SPIE papers…). This will allow you to read any paper at any given time without having to scramble to find where exactly you had saved it (or left in a random email).

As for number 2, the BibTex entries, there are many solutions out there; one you could use, is ADS libraries (see below).

When working on bibliography, you will not get around using ADS, the Astrophysics Data System. It is a “digital library portal for researchers in astronomy and physics”, allowing you to find abstracts and references of almost all research papers written in the domains of astronomy and physics. While ADS might contain links to the publisher website of a paper, it is not guaranteed that you will be able to access a paper.

If you need help on using and navigating ADS, you can consult the ADS Help pages. It might also pay off to set up an account, as this allows you to generate custom ADS libraries, which let you make collections of papers, and find their references quickly when you need them.

You can also connect your ORCID identifier directly in your ADS account and “claim” your papers in ADS (also see this step-by-step guide by ESO), in which case they will show up automatically in your ORCID account.

arxiv.org

arXiv is a free distribution service and archive for papers in the field of physics, mathematics, computer science, quantitative biology, quantitative finance, statistics, electrical engineering and systems science, and economics. While ADS is an indexing service, arXiv is a place that actually saves the paper files. You always have to keep in mind though that submissions to arXiv are not peer-reviewed, which means that just because something is on arXiv, it doesn’t mean that it is trustworthy. For the astronomy community though, arXiv is mostly a place to make author’s papers available without a pay wall. This is why you should absolutely, if the rules of the journal submitted to allow it, upload all of your papers to arXive once they are published. And to do so - you might have guessed it - you will need to make an arXiv account. Many astronomers say: “It’s only really published when it appears on arXiv.”

A special note for you astronomers working in instrumentation: SPIE explicitly allows the upload of conference proceedings pre-prints to arXiv, even if the spiedigitallibrary.com keeps them behind a paywall. Do your fellow researchers who do not have paid access to SPIE a favor and upload your SPIE papers to arXiv.

Personal website

We believe there is no point in reproducing information that is already very well explained, so please let us point you to what we consider an excellent summary about why every researcher needs to have a personal website, and why it is ok is you invest only half an hour into it: See here! (M. Kenworthy)

This was written by Dr. Matthew Kenworthy at the University of Leiden, and while it is completely unsolicited advice, I believe it is good advice. Additionally to all the reasons Matt Kenworthy puts forward on his page as to why you need a website, look at it this way: We are in the 21st century now, and towards the end of your PhD, you will be looking for a job. Having a correct online presence will make it easier for people to find you, find all the information they might be looking for in a potential new colleague and be able to reach out to you if they wish to. Don’t hide, have a website! If you are having troubles hosting a website on your institute’s home page, consider hosting it on GitHub (which also spares you having to migrate your website each time to start working at a new place.