Stian Håklev’s Design Proposal

Stian Håklev offers an intriguing proposal for a graduate student’s knowledge workstation in a screencast mashup of Papers, Skim, Devon Think, and Tinderbox.

The mistake here, in my opinion, lies in covering up the seams between programs, in the aspiration that everything should happen inside one grand package. Lots of people wish for this, because the friction between two systems is so obviously irritating. But smoothly integrating everything in one huge package means we all use one huge (and, inevitably, widely-disliked) package: Microsoft Word, say, or Photoshop. (If you think there will be a vibrant ecosystem of competing integrated scholarly workbenches, I have a bridge to sell you.)

I think we’re far better off accepting occasional nuisances in order to be able to mix and match smaller, more interchangeable tools. Want to use Acorn instead of Photoshop? Fine! Swap Papers for Scrivener? OK! Want to try FreeMind instead of Tinderbox today? Go right ahead. Besides, if you can put up with occasional context switches and perhaps write some scripting glue, you can have almost all of this today without waiting for someone to build it for you.

Håklev’s screencast is a great example, however, of clearly showing what one might want to do, and why. This is indeed the trickiest part of the craft of software.

Guess Mueller’s image editor Acorn was recently trounced – well, mildly tut-tutted – by a formerly-significant computer magazine because it’s not Photoshop. This never happens to big companies, presumably because they might be big advertisers and because software reviewing on the whole is so inept that nobody notices petty corruption. Software is the great art form of our age, and we treat it less seriously than children’s music or food writing.