So, we seem to have drifted from the topic of test groups over to this topic of future proofing against broken versions and TAP producer / parser version negotiation. I have a simple solution for this.
If all we do is argue about TAP extensions and never actually produce one we will never have to worry about new versions! Brilliant! I'm a god damned genius. And it seems that we're steady on track on this one, batting 1.000 since May 2001. [1] Point is, its a little premature to worry about future proofing TAP against versioning mistakes when we're not producing new versions! I said last week that I'm worried about how much time has been spend talking about TAP extensions and so little actually writing code and I'm serious. There are four tasks to be done. Andy Armstrong has done one. Andy, Ovid and Eric are working on TAP::Parser. There's been one edit to the TAP spec, which I suspect was done by Andy. And nothing done to work on the TAP diagnostic syntax, easily the most pressing new TAP feature. I'm not even going to address the issue further, consider it shelved until we get a TAP extension implemented. [1] Technically Andy Armstrong went ahead and actually implemented the TAP version spec in TAP::Parser breaking our perfect record. Boo! ;)
