On the Slashdots today there was some stuff about how spreadsheets are evil and MUST NOT BE USED FOR RESEARCH EVER.

This is really sad. Instead of ranting about what’s wrong with office software, why aren’t the smart kids (ie us) working on ways to make it better? Surely one could develop some tooling to make code-review in spreadsheets better rather than complaining? Spreadsheets do not allow testing? Why don’t we make sure they do?

Anyway, the fact that this Piketty posted the data and the code, and others can re-run everything is the whole point of what we’re trying to do in eResearch. Matlab or R code could dodgy too. We want to maximise research integrity and impact. To maximise integrity, show your working out, listen to feedback, fix problems. Repeat.

The really sad bit? This is not constructive:

Piketty used spreadsheets instead of writing sane software

Sane software? Really? You think the software has mental health issues?

Me, I think the hard-core tech community has a real attitude problem. Maybe the text-file-comandline-distributed-version-control-way is the One True Way. If so do you think ranting about ‘sane software’ is the best way to win people over? Another way might be to, you know, try to think up a way to use spreadsheets in a sane way Or you could Show us the Code (ie post your own sane version).

Another example: just this week I’ve been playing with docpad, a promising looking website toolkit. Gotta love its about page:

WYSIWYG editors are sucky and stupid - why re-invent the wheel? We’re already trained with and love our desktop counterparts (Sublime Text, Vim, Byword, etc.)

Just between you and me, docpad itself turns out to be ‘sucky and stupid’ if you give it more than a few hundred pages to chew on, but I wouldn’t post that on the internet, I’d turn up at their github site and suggest a solution to the embarrassing out-of-memory issues (hint, maybe you do actually want to fire up a database or index of some kind kids).

[Update 2014-05-28 to fix a couple of typos]

Creative Commons License]( Elitist geeks considered hurtful (sic) by Peter Sefton is licensed under a Creative Commons Attribution 4.0 International License.


comments powered by Disqus