Isn't it weird that CERN and high energy physicists develop things like C++ interpreters and the world wide web?
I'm really glad these things exist (well less so with cling, as I'd pick a different language if speed wasn't my top concern but I understand how useful this is to play around with code you later plan to use at a larger scale).
Anyway my point is that I'm not sure this was how it should work - this isn't really physics? Why does it the up being paid for with physics funding?
Because of the fairly unique requirements of their setup (I have heard them describe the LHC, with its storage ring of about 30km, as doing “10,000 experiments per second”. They throw away most of their measurements within a ms, but still keep about a petabyte each day (https://home.cern/science/computing/storage)), they build their own hardware, network, storage arrays, etc.
I think that builds a culture where it’s normal to build your own tooling.
It also may help that they have huge budgets, making it relatively cheap to develop this kind of stuff.
I'm really glad these things exist (well less so with cling, as I'd pick a different language if speed wasn't my top concern but I understand how useful this is to play around with code you later plan to use at a larger scale).
Anyway my point is that I'm not sure this was how it should work - this isn't really physics? Why does it the up being paid for with physics funding?