Or sprint and shatter?

For a while now I’ve been working on a new version of Freegle Direct, which is the way that most people use to Freegle (for valid reasons of history and democracy, there are others – Yahoo Groups, Norfolk Freegle, trash nothing…).  It went live today.

In a commercial environment, you’d take this nice and slowly – do your development, run your regressible unit tests and code coverage, do friendly user testing, test on all the standard hardware, operating systems, and browsers.  Then do a phased rollout to a few groups.  Leave that for a week or two, use Decibel Insight to do usability testing, tweak and repeat.  All nice and calm, with time for lattes and cake and cats and those fells which were beckoning me today.

Ha.  Let’s pause for a minute to dream about that.

Back to reality.  Freegle is a volunteer project, late evenings and weekends and half-snatched fragments of insomniac time.  We just don’t have the time or geeky resources to be able to do it in the same way as a commercial environment.

Nor, frankly, do we need to.  If people can’t reuse items, then who dies?  It’s not nice to annoy your users, it’s really not.  But, y’know, move fast and break stuff.  “Fit for purpose” doesn’t always mean perfect.

So with Facebook’s API changes breathing down our necks, we just had to take a deep breath, stick aforementioned neck out, and go for it.

So it’s been a high-wire-act kind of day.  Cunningly defer some function.  Be ruthless about what we need and what we don’t (no bad thing anyway).  Get other kind volunteers to test it, but accept that the testing will have gaps.

Then do the switchover, monitor everything, spot the howling bugs and get them fixed, wait for the first raft of problems to come in from our volunteers and users, then fix them too.  Development in real time.

It’s settling down now.  I might actually get some sleep tonight.