there was a huge chilling effect on both product and protocol design. In the 90s I had to fill out a form and submit it to RSA in order to get a copy of their library. Which I eventually got after waiting 6 months, but I had to agree not to redistribute it in any way.

Efforts to design foundational cryptographic protocols were completely hamstrung by the spectre of ITAR and the real possibility that designs would have to US only. Right around the time that the US gave up, the commercial community was taking off and they weren't at all interested in further standardization except was creating moats for their business - which is why we're still stuck in the 90s as far at the network layer goes.

AFAIK the Zimmerman case was quietly dropped instead of ruled

Tinfoil hat: it was dropped to prevent exporting code being 1A protected as case law

Seems not tinfoil and rather plausibly a pragmatic decision by the prosecution.

Would be a good day to have that enshrined in case law, maybe the US government would let me work on rocket GNC if code can’t be export controlled at all

when did local prosecution got a training in international politics and commerce interests?

Federal interests can easily tell the local prosecutor "hey, don't prosecute this, it risks setting bad precedent".