5 ms·
My impression: Europe does pretty well in non computer-science areas.
by tpmx 6y ago
My impression: Europe does pretty well in non computer-science areas.
- andrewflnr 6y agoThis is funny to hear, because a lot of the CS research I care about, namely compilers, comes from Europe, particularly IIRC ETH Zurich.
- tpmx 6y agoThere will always be exceptions to generalizations. One or a few exceptions dont't make the generalizations less valid.
- andrewflnr 6y agoFair enough. I wasn't being sarcastic when I said it was funny. :)
- domnomnom 6y agoException:INRIA.
- tpmx 6y agoI searched for INRIA success stories. I found the MLdonkey peer-to-peer client written in OCaml. (I do remember using it many years ago. It was odd.) Where are the successful, world-changing company spinoffs?
- andrewflnr 6y agoOk, that's the wrong measure. But didn't they sponsor basically the whole OCaml compiler, which has gone on to be hugely influential?
- tpmx 6y ago> Ok, that's the wrong measure. Is it though? I guess that's the difference between Europe and US in CS. In the US they find a way to make money on the innovations.
- andrewflnr 6y agoYes, it is wrong. Financial success is only one proxy for impact, and not an especially good one in my view. Lots of things that really maybe the world better will make no money, and in fact reduce a lot of companies' incomes by making purchases unnecessary. Open source software is a prime example.
- themacguffinman 6y agoThe biggest and most useful open source projects are often either funded or created by financially successful corporations. The rest of open source is generally rife with useless junk that benefits almost no one (except maybe the author) precisely because there is no pressure to validate ideas and outcomes with money. This isn't an argument or discouragement against open source, but any financially successful product is a billion times more likely to be impactful and useful than sampling by almost any other factor like open source.
- periheli0n 6y agoWhereas in Europe university driven innovations are often placed in the public domain, especially in France. What is better for society? Depends on the viewpoint I guess.
- tpmx 6y agoNow pause to think about how e.g. France has made the EU spend many, many billions to create a search engine that can replace Google? (Unlikely.)
- periheli0n 6y ago
- ur-whale 6y ago> Exception:INRIA. Name one successful (as in: >1G$ cap) company started by INRIA researchers.
- periheli0n 6y agoI thought the question was about research strength, not money-making strength. Although capitalisation is not really about making money rather than promising to make money later. Pretty much orthogonal to the aim of research.
- tpmx 6y agoIn CS it's kind of combined.
- periheli0n 6y agoHm. One can do awesome CS research without having to make tons of money. One can also make money with other people’s/universities’/countries’ CS research while doing zero CS research in-house. Research and money making are pretty much orthogonal. With research I refer to fundamental CS research, not how to squeeze 0.00053% more out of imagenet or making photos of young people look old, or solving problems of logistics in driver scheduling.
- tpmx 6y agoOkay, let me rephrase this: The current strategy is great for individual researchers - when they become successfull they can get a great $400k/year job in the US. It sucks for the europeans who paid for them to get there though.
- periheli0n 6y agoFunny enough, the way Germany’s Max-Planck Society is successful in hiring top US scientists not through outrageous salaries, but by offering them guaranteed, generous research budget in perpetuity. Not all researchers care about getting rich personally.