05
Mar
2019

Time limits in research

Faliure, as illustrated by theAwkwardYeti.com

“So that experiment didn’t go as planned? Congratulations, you are now officially a scientist”

This tweet (which I can’t find again to attribute), as well as many other things, claims that the secret to most scientific work is not genius, but persistence. “Scientists are people who keep going when most of us have given up”. There’s a lot of truth to that. But I can’t help feeling that it’s sabotaged by the way that science is funded these days.

If you get a surprising result, that’s great for science! Well, sometimes. Usually it means you made a mistake. If it’s real, it’s great for science… but it’s awkward if you have to get the work finished and the publication out for a deadline, and you have more things lined up to do after that.

When an experiment or a simulation doesn’t work first time… well, that’s normal, but a huge proportion of the research workforce is on short fixed-term contracts, and when you’re only in the job for a year, you can’t afford to explore too many dead ends. You can’t afford to take the failure and persistence approach that is idolised above.

I’ve just finished a one-year postdoc, and this has made me very anxious from time to time[1]. I’ve started to wonder how much of a correlation there is between people with successful academic careers (at least in STEM), and people whose experiments / models / observations / whatever happened to go right first time during their first postdoc. Obviously there is a strong element of researcher merit involved in whether things work, but there’s also a hell of a lot of luck. And if you need it to work first time in order to have the time to get everything finished and written up[2] before you’re in the next job and busy on something else…

[1] Along with all the other sources of anxiety that result from 1-year postdocs. That’s a much broader topic, but this is just one of the reasons that I’m really looking forward to my next job being a little longer!
[2] Or, just “in a sufficiently complete state that you can write it up in evenings and weekends while working on a different day job”, if you’re privileged enough to be able to do that.

Posted by simon in Reflective
01
Feb
2019

The end of one job… and waiting for the next

Cruise line poster advertising SS Leviathan on United States Lines, "Europe-America"
Not my intended mode of travel. Though I wish I had enough time that it could be.

My time at Marine Scotland Science came to an end yesterday… and then I went in today anyway for a meeting, because it was the only time that all the necessary people were free. But that’s at an end, and so is my much longer time with Heriot-Watt. Obviously, I hope to keep in touch with and collaborate with both.

My next role will be a postdoc with Dr. Zhaoqing Yang at the Pacific Northwest National Laboratory in Seattle. I’ll be building on my existing skills with FVCOM modelling, both for energy and other applications. That might hopefully be starting in about a month’s time, but it all depends on when my US visa comes through. In the meantime, I have plenty to occupy me with selling my car, sorting all my belongings into “ship, store, or throw away[1]”, and trying to see as many of my European friends – British and otherwise – as possible before I go. Oh, and that paper I promised to write 😉

Moving to America, and working over there, is both exciting and daunting. It will be an adventure!


[1] Give away, where possible.

Posted by simon in Professional updates
06
Jan
2019

New ideas in fusion

This is an interesting, if light-on-details, article about the new wave of private research in fusion power. Some are looking at new ways of building tokamaks, some at stellarators, and some at exotic new ideas. Some of them claim that they will achieve first power in 2030.

I’m not competent to judge these claims – although I will observe that the use of VC funding in the renewables industry has led to rather optimistic forecasts – so I remain hopeful, yet cautious. I’m really glad that there is a sprouting of new ideas around fusion.

However: No matter how well it goes, fusion is not going to help with our immediate mess. Even if somebody has a working reactor in 2030, and even if it has the potential to become economically viable, it’ll be decades more to optimise it, reduce the cost, and build a significant number worldwide – not just in rich nations. We need to decarbonise now, not after 2050, and for the moment that means renewables and, probably, fission.

Fusion is quite possibly the future, and it should be funded; but at the moment, it cannot take any attention away from shorter-term solutions.

Posted by simon in The wider world
03
Nov
2018

Teaching

Whiteboard markers and a whiteboard.This term, in addition to my modelling work at Marine Scotland, I’m the instructor for two masters modules at ICIT (Heriot-Watt’s Orkney campus). This is my first experience of teaching, beyond the occasional seminar here and there, and I’m really enjoying it. I have a small group of interested students, who want to be there (I realise that this is a privilege of teaching postgrad), who ask intelligent questions… and that makes it really rewarding.

It’s also very hard work. I was brought in at fairly short notice after a lecturer left, to fill in the gap before a new one could be recruited. I’m only going to be delivering this content once, yet I’ve chosen to put together my own material for it based on what the previous instructor did, rather than using his directly. That’s because the content follows a different logical order in my head to his, and… well, as anybody who has tried giving a presentation using somebody else’s slides will attest, it’s not a great experience for anybody concerned. So I’m talking to students for 2-2.5 hours most mornings, and spending the afternoons preparing future material – trying to stay 2-3 days ahead, but occasionally catching up with myself. It’s not a pace that I could sustain in the long term, but it works for a few weeks.

This experience has reassured me that, should I be successful in landing a long-term academic role in the future (and I realise that that is a very long way from guaranteed), then I would be able to embrace the teaching side as enthusiastically as the research.

Of course, this is only half of the job. My lectures finish next week, but towards the end of the year the marking will begin….

Posted by simon in Professional updates, Reflective
07
Oct
2018

The problem with acknowledgements

Academia doesn’t have a way of acknowledging contributions short of authorship that matters.

Yes, we have Acknowledgements sections, and I try to be very comprehensive in who I include, but while it feels good to be featured there, it doesn’t actually matter from a career perspective; nobody is going to sit in an interview or review panel and say “They didn’t author any papers this year, but they were acknowledged on seven really good ones, so they clearly did some good work”. Indeed, because acknowledgements aren’t indexed in the same way as authorships, nobody is likely to even know.

If somebody, say, allows use of a dataset that’s already been written up elsewhere[1], or is a technician involved in an analysis, their work has been important in enabling the study to be conducted, and thus should undoubtedly be acknowledged, and more so than a polite thank-you at the end of the paper that nobody will remember… and so they end up becoming authors, despite not having made the intellectual contribution that should mean authorship of the paper. And this is one of the causes of author inflation, and also of this sort of thing.

It makes me wonder if we need a third way. Let the authors just be the people who wrote the paper, or (perhaps) otherwise made intellectual contributions to the study. Leave the “acknowledgements” section for funders, companies, public data providers, whimsical mentions of friends, and so forth. Set up a new list of “Contributors” or some such, indexed as authors are, for people who need to be able to point to what they’ve done on PURE or Google Scholar.

 

[1] The move towards citable datasets that have their own DOIs should help here.

Posted by simon in Reflective, The wider world

Search

Archives