In his blog, the Canadian professor, programmer and reviewer is pessimistic regarding academic software while stating that “when I review a research paper or a thesis, I expect the software supporting it to be awful.” Among the explanations of such a state of awfulness he evoked professors spending little time programming and hence they do not enforce good practices: “Try to look up your favorite professor on GitHub,said that professor. Does he proudly display the code he produced?” before advising “you ought to spend more time on GitHub and less time in the classroom.”
Scholastica can be proud of its repository at GitHub. But as commented without answer in the blog: what’s can make researchers and professors spend time writing good software instead of studying the new big idea that might make them publish another paper?
“ Academic software is awful.” Is it?
2 Comments
I'm no computer programmer, I couldn't even "cook an egg" to quote Lemire. In fact, since I have no idea what a PHP script is, I guess that means I wouldn't even know an egg if I saw one (pretty sad). So I don't have an answer for you, but I find the question interesting because I'm now wondering what it is that all these people with computer science degrees 'do' if they're not writing programs for companies. And the people who are writing programs for facebook and google - do they have computer science degrees, some other degree, or no degree? It's a practical question, I know, and probably one that frustrating for computer science majors. I suspect it would be like asking a criminology major about forensic science (which actually falls within the biologists' domain).
The way I read it, the academics are busy innovating new programmatic ideas like a better web search algorithm or programs to recognize text or faces but their implementation is poor. I can see that it would be great if their implementation was better - faster/more efficient, bug free, and logically constructed/organized and documented. But I can also see why the focus is on innovation rather than quality control. They're not making money based on how neat and tidy the code is, but based on its novelty. Time spent tidying up the code is wasted time that could be spent on developing a new idea. As long as private companies like facebook or google exist then there will be someone there who earns a living transforming that raw idea into a clean, polished, and useful application for the masses. I'm not sure I see what's wrong with that - it's a division of labor that seems to work great.
Thank you very much dear Will Hauser for insightful answer.
Indeed, Lemire’s displeasure is addressed to computer science students. To understand his argumentations I had to make an analogy between a paper and a program. (e.g: one can “read” a php script for example and understand it just like he would understand a coherent simple English text). As such, and knowing he’s a reviewer, I think he’s insinuating a worst risky scenario of rejecting a program not respecting good practices the same way he would reject a paper for bad writing style. But I adhere to his indirect call to open source programming when he considered displaying one’s code in GitHub as a source of proudness.
I also share with you the acceptance of the reality of the division of labor while Lemine reminded me Ralph Waldo Emerson complaining this division in ‘The American Scholar’(1837). He may be right.