Comments on 'Scientific Software Developers'

[bkatiemills]

Add your comments here for this post on the role of Scientific Software Developers.

[radaniba]

Thank you @BillMills for this post.

Although I agree with most of what you said, I think you need to open up to the end (may be as a follow up article) to how to express full recognition to what a Scientific Software Developer is bringing to his working environment, I am referring to value recognition through incentives and rewards as well as, which is more important here (may be the taboo no one wants to talk about) financial recognition through salaries out of the grid , grids always used in academia as a take-this-or-nothing.
I am talking from a perspective of a bioinformatician where I had enough time to see how this role is reduced to a "IT support" rather than valuable asset to help scientist discover insights out of the mountain of data their generating. They are simply super full stack developers that are daily dealing with big data management, writing software, developing models, investigating genomic information to answer big questions ... that are reduced to simple workers due to the lack of : recognition.

This applies I guess to other scientific programmers fields.

Without pointing to this very important point, am afraid any other effort towards defining a new label for what exists already under another umbrella, will be just a superficial change with no deep in depth flip of mentalities that are motivation killers.

[bkatiemills]

I definitely agree that scientific software developers need to be treated with more respect, like first-class members of the lab rather than IT-support, like we've all experienced. This is seated deep in the attitudes of the scientific establishment that imagine software development is somehow peripheral to 'real science' - the defense of which gets flimsier every year.

The trouble with money, is where is it going to come from? Our funding is already so strapped, that while I agree with the sentiment, the argument is usually a non-starter.

One way to approach the problem of finding a funding bucket and putting scientific software development on the same footing as research is to recognize the centrality of programming to science by putting scientific software developers on the same career tracks as traditional researchers; staff scientists and faculty all have their specialties, and software development should (and even could) be recognized as just another one of those specialties.

Being a serious contender for a faculty position as a scientific software developer would be satisfactory to me, even though the pay scale differs dramatically from a private industry developer. If we insist on competitive salaries, a whole new funding model needs to be explored; on this topic, I've been toying with the notion of a core + contract scheme: a scientific software developer is attached to a department or lab on a base salary (core), and they get a salary top-up out of the grants of the groups they work with on a contract basis. This at least spreads those salary demands across a few funding sources.

[abbycabs]

Thanks for the shout-out :)

[bkatiemills]

@acabunoc np np :) - btw, I'm taking a crack at roadmapping ala the WoG over here - feedback (on the project management or, y'know, the actual project) very welcome!