It wasn't meant to run on their own hardware or software (and still hasn't been proven it actually does and really can't be.) They did what they had to do to stop clones and those clones will sure take their time from now on before releasing new firmwares or if they can't decrypt the whole thing hopefully not release any firmware unless they make it.
What reason isn't there to? Still been using 2.02b every day now playing games and I don't see why all these people are complaining of bricks, it doesn't.
Why should they make new stuff if a different team is going to steal it the day after and say it's theirs? Take them out first.
But why not brick with a 100% certainty on modified launchers then? That way the cloners would have bricked their test consoles (assuming they do at least one testrun before they release), scratched their head and never released their copied versions.
Also software can and always will have bugs. Doesn't even have to be that mystical "high energy photon" that some like to blame. It's just not possible to test this type of code on every possible system in every possible state (e.g.: Nintys own firmware might have a bug that results in jumping to the location where the brick instructions are (which would usually be empty and just result in a freeze) on European 4.2 systems that have Streetpass data for the game Mario Kart 7 containing Korean letters on every Monday the third). Therefore it' just plain stupid to ship code that contains any type of destructive behavior unless you have access to the complete code base and are able to test accordingly or have ways to guarantee code isolation (like running the dangerous/mission critical stuff on dedicated chips).
edit: if you've got time to waste try to get your hands for example on US RTCA DO-178B.
















