Monday 23 September 2013

Unavailable by request - Do you have to ask?

I've decided to no longer review papers where the software is "available upon request from the authors". I'm sure the world will keep turning, but I just don't want to spend time reviewing this type of software.

You see, to my mind "available from the authors on request" means "not available at some indeterminate point in the future". If they actually wanted to make it available they would have put it up on the web somewhere. Ergo, since they haven't, it means...what exactly?

I don't get it. Ok, so there's probably an innocent explanation. Maybe this is how it was done in the good ole days ("Hey, the 80s are calling. They want a copy of that software you wrote"). All I know is that I don't want to spend an hour or whatever reviewing a manuscript that describes software which may vanish into thin air.

I'd be interested to know if you have a different view on this.

Image credit: Available on request by Iain Farrell (CC-BY-ND)

Sunday 22 September 2013

Marketing and Scientists

Having left academia just over a year ago and joined NextMove Software, I am reminded of a comment that a postdoc made to me about why she would not like the idea of working for a software vendor: "Marketing". I once had the same idea, but after several years postdocing I realise that a big part of a scientist's job, and perhaps especially a postdoc's, is all about marketing. We can call it self-promotion or networking, if that's more palatable, but it's essentially selling oneself and one's ideas.

Let's start with the obvious stuff. Papers: they promote your ideas, and also yourself. To get the paper in, you first need to market it to the editor with the submission letter. To encourage people to read it, you need a punchy abstract and title. For further encouragement, you need some marketing: a poster, a talk, a write-up on the webs.

And yes, blogs do help. Sure, marketing is not the only reason to have a blog, but it is a positive side-effect (it's a mystery to me why more cheminformaticians don't write them). Similarly, putting talks and posters up on the web ensures a wider reach. As I've said before, more people will read it online than will ever hear you give it in person.

And then there's the pure marketing: grant applications and job applications. If like me, you are somewhat reluctant to write a page of text on how awesome you are, you'll need to overcome this handicap pretty quickly in order to fill in the various bits of grant and job applications. As for the rest of the sections, it's a fine balance between promoting your ideas as solving all of the world's problems and being scientifically realistic.

"Marketing" may not be a scientist's favourite word, but if your scientific study is buried in the literature and forgotten, it neither advances the field nor your career. So if you feel you have something worth talking about, get out there and get marketing!

Image credit: Social Media Marketing Madness Cartoon by Hubspot (CC BY-NC-SA 2.0)

Wednesday 11 September 2013

Poll answer: Time for Python 3?

Results are in for the Python poll: 33 are using Python 2 (of which 15 are thinking about moving to Python 3) while 8 are already on Python 3.

So still very much a Python 2 world, at least among respondents.

It'd be interesting to know what exactly is keeping people at Python 2. Inertia? Dependencies on legacy code? Deep-seated aversion to 'print' as a function?

Tuesday 3 September 2013

ASCII depiction for Accelrys Draw

At work I've been creating some plugins for Accelrys Draw, and the thought occured to me, why not enhance the colourful razor-sharp depictions generated by the good folks at Accelrys with ASCII art versions beamed direct from the 80s?

So I wrote a plugin that uses Open Babel (through its .NET dll) to generate an ASCII depiction of the displayed structure and paste it as a text object. The image below gives some idea of it in action, and since you ask, I found that an aspect ratio of 1.7 worked well for Courier New.
Unfortunately, I'm not 100% sure that I can make either the plugin or sourcecode available, and so I'm going to err on the side of caution and leave it at the screenshots.