What is the crowd's opinion of Wyvern?

From 3.5 billion comments

This is a recent public investment: http://www.cs.cmu.edu/~aldrich/wyvern/. However, considering it's certainly not state of the art (it appears those researchers don't know what they are doing(!)), I am not sure why the NSA would want to connect their name to such out of date research.

They probably have funded FPGA languages and tooling in the past. Chisel might also have been funded.

It's simple economics; if they can invest USD 1B to save USD 10B on hardware or staffing costs, they will. Since actually having extra people on NSA staff requires office space, etc. and research can be done by other parties in the open, it's a scalable method for innovation.

Consider the following scheme; let's say you want to do research on some good idea X*, instead of just funding X*, you also fund projects X1 , X2, .., XN. It might be that in X* there is only a small part which unlocks some new capability, which is your real secret. Other parties looking at what's happening, might be seeing that you funded some projects, but they don't know what to look for. In case you think this is too far fetched, there is plenty of research (e.g. mathematical research), which could be seeen as completely academic from even the point of view of a mathematician, but at the same time as practical from the point of view of some senior NSA official who actually knows the current capabilities of the NSA. (See the research on breaking SSL as an example.)

Additionally, just because there is a large body of research available, doesn't mean you have a working spy agency; that requires operational procedures and know how; knowledge and skills which isn't widely available.

Of course the above is all speculation, but all actions taken by the NSA as a whole still have to abide by the rules of efficiency as an upper bound.