Live data from Hacker News

Peter Higgs: I wouldn't be productive enough for today's academic system

theguardian.com

171–180 of 204 posts

Re: Peter Higgs: I wouldn't be productive enough for today's academic system

#171
post #11

Earlier quoted context omitted.

He is asking why only 10 papers in 40+ years. What else was taking up his time? I have the same question.

Not a physicist, but writing one good paper every four years sounds like a very good thing. Would you rather read 40 papers, each of which is a lowest-publishable-unit, or 10 which noticeably move science ahead? Because of his fame, Higgs had the luxury few do--- he could contemplate quietly, and people wouldn't dare say he was doing nothing. If you do that as a mere tenure-track professor, I'd figure the university…

>Would you rather read 40 papers //

That would depend on whether the atom of scientific knowledge in one of those many published papers was what was needed to progress another's/one's own work or not.

I'm not suggesting quantity and regularity are better - I don't think you can force scientific progress to adopt a given release schedule - but neither am I happy to say that publicly funded results should be sat on pending a larger breakthrough.

Re: Peter Higgs: I wouldn't be productive enough for today's academic system

#172

I think this sort of problem starts at the top, with our time limited democracies. Democratic governments find it hard to implement long term thinking because its unlikely the current government will ever get credit for it. This results in policies which are short term enough to promote re-election. This bleeds down to the rest of society. Where it does happen, it tends to be compromised because one side has to make…

Can we please stop promoting the "democracy sucks" meme? It's not as if the Chinese are actually planning in that long a term; they're using markets now like everyone else.

>It's not as if the Chinese are actually planning in that long a term; they're using markets now like everyone else. //

So, how long term are they planning, in years?

In the UK we get [a few] plans that encompass maybe the next 10-15 years at most. So presumably they've got that at least.

Re: Peter Higgs: I wouldn't be productive enough for today's academic system

#173
post #129

Earlier quoted context omitted.

He was a consultant. FYI: It's extremely rare for anyone from academia to make a commercially successful language. Their incentives are different. They want publishable novelty, and they don't care about building a useful suite of libraries and a vibrant community. Jones (Haskell), Syme (F#), and Odersky (Scala) are the only designers w/ academic backgrounds I can think of for marginally popular languages.

Academia might have been the wrong word. Do you know of any other language designer (except Larry Wall) who doesn't have a PhD? Btw, what's the source of that consultant bit?

C#, Python, PHP, Javascript, Perl, and Ruby. Obj-C, because the guy had a doctorate in biology. They are all very smart people.

As the other comments said, Tcl and Lua are designed by phd guys. As were C++, Java and Erlang. I guess we've got a pretty good track record. I should use my phd in programming languages to do something useful. ;-) maybe later.

Re: Peter Higgs: I wouldn't be productive enough for today's academic system

#174
post #129

Earlier quoted context omitted.

He was a consultant. FYI: It's extremely rare for anyone from academia to make a commercially successful language. Their incentives are different. They want publishable novelty, and they don't care about building a useful suite of libraries and a vibrant community. Jones (Haskell), Syme (F#), and Odersky (Scala) are the only designers w/ academic backgrounds I can think of for marginally popular languages.

Academia might have been the wrong word. Do you know of any other language designer (except Larry Wall) who doesn't have a PhD? Btw, what's the source of that consultant bit?

Robin Milner of ML is the most prominent example I know. One of the two Turing award winners without a Ph. D (the other being Floyd.) He introduced polymorphic type inference to a full-blown programming language - this is the most jaw-dropping language feature I have seen to date:

http://en.wikipedia.org/wiki/Robin_Milner

http://en.wikipedia.org/wiki/Type_inference

Re: Peter Higgs: I wouldn't be productive enough for today's academic system

#175
post #125
post #69

Earlier quoted context omitted.

The main thing I don't like about BBT is that it's generally not funny. Look how the show holds without its laugh track: https://www.youtube.com/watch?v=jKS3MGriZcs

It's not fair to just remove the laugh track, you need to re edit to change the pacing to avoid the silence.

If you watch actually funny shows that have a laugh track, such as Seinfeld, and removed them - they're still funny. Idk I just find BBT humor very bland, like the punch line to a joke may be just someone screaming out a nerd reference just for the sake of saying "Hey were nerds!"

Re: Peter Higgs: I wouldn't be productive enough for today's academic system

#176
post #76

The issue is Volume versus Quality. The conservatives (not political but professional) administering our research infrastructure find it much easier to greenlight an incremental (or inconsequential) project versus anything truly revolutionary. We need more Higgs and less incremental academic researchers - industry can do that better.

Genius is one percent inspiration, ninety nine percent perspiration. -- Thomas Edison

If smart people are not expected to get incremental results we may lose both incremental progress and ingenious leaps.

Re: Peter Higgs: I wouldn't be productive enough for today's academic system

#177
post #158

Earlier quoted context omitted.

I was about to write a longer post, but this is so incredulously complicated, and I am so biased, that I changed my mind halfway through. I'm honestly too biased about some of its points to provide anything marginally better than flamebait if I start going into details, but I can offer some hints. 1. The funding for research is increasingly scarce, and the ones who decide how big they are and to whom they should go a…

You seem to, maybe fairly, denigrate startups and industry in general as well as academia. What should an intelligent person do? What did you settle on? What advice would you give to young people?

> You seem to, maybe fairly, denigrate startups and industry in general as well as academia.

Yeah, this is exactly why I changed my mind halfway through that long post :-).

I honestly love the startup environment. I'd much rather work in start-up than a large company. What I do denigrate is starting up for the sake of starting up. Finding an (often otherwise legitimate) need, pounding at it for six months while it's still hot, and coming up with a technologically half-assed product to sell for a reasonable price.

This is, in my opinion, destructive both intellectually and technologically. It teaches bad habits and gives little time to learn both adequate technologies and the fundamentals of their trade to programmers. I'd be a very rich man if I had a penny every time I told a colleague who was enthusiastic about a new technology he'd discovered on kickstarter or here on HN that or had this. You know something is wrong when so many new things are so similar to old things -- so similar to the point that they repeat the mistakes.

As for academia, I still keep in touch with some of the people I worked with there. It's the most intellectually stimulating place I've ever worked in, and contrary to popular belief, one of the most refreshing feelings a professional can have is walk into a roomfull of people and realize they're the dumbest in there. I miss being the dumbest man in the room.

> What should an intelligent person do?

I honestly have no idea.

> What did you settle on?

I didn't settle yet, but if I were to look back, I'd say what I'm doing now is better than what I had back then.

Now I'm working for a large-ish company. Their business is entirely software, but they want to start doing hardware and they brought me in to help. They pay me well and I can come in the office at 10 AM, which is good (I have some sleep issues). The work itself is shit; there's a long ladder of managers who are increasingly clueless about what embedded development means, but each of them has to deliver results (no matter how irrelevant) because they made promises. Consequently, most of what I do is pointless, but not entirely uninteresting. In the last month or so, I dabbled in a USB driver in the Linux kernel, hacked on an HTTP proxy, helped a colleague build a PCB... it's useless, but not entirely devoid of fun. I intend to leave as soon as someplace where they actually need me to build stuff, not massage egos, shows up, but until then, I can bear it.

That being said, it also leaves me enough free time to do some hacking of my own, enough time for my hobbies, enough time for family and friends. I have enough time to brew beer with my girlfriend, read whatever books I want, learn Go and post crap on Hackernews. I'm far more unhappy with what I do than I was in university, but overall, I'm a happy person.

Re: Peter Higgs: I wouldn't be productive enough for today's academic system

#178
post #158

Earlier quoted context omitted.

I was about to write a longer post, but this is so incredulously complicated, and I am so biased, that I changed my mind halfway through. I'm honestly too biased about some of its points to provide anything marginally better than flamebait if I start going into details, but I can offer some hints. 1. The funding for research is increasingly scarce, and the ones who decide how big they are and to whom they should go a…

For #3, I am starting to think we should split off a portion of computer science and call it software engineering. People who are interested in the theory of algorithms, etc can do computer science. Those that want to learn how to build fault-tolerant, scalable systems can do that in software engineering. It is similar to how chemistry and chemical engineering are split. Of course you still need to get a basic unders…

> Of course you still need to get a basic understanding of how computers work with a software engineering degree, but knowing how to mathematical prove some algorithm is O(log n) is pretty pointless for most (not all) jobs in industry.

I agree, but IMHO most jobs in the industry that don't require you to prove that an algorithm is O(log n) shouldn't need a degree at all.

In the area of the world where I live, if you want to be an electrician, you can do that after you finish high school. You have to take a course and get a certificate for it as a legal requirement (which is true of any profession where you can get people killed) but it's pretty straightforward, and the course only involves pretty basic stuff.

And don't look down on electricians. My degree is in EE and when I need some work done on the installation in my house, I call an electrician. I could do what they do, but sloppily and with far more dangerous results.

Re: Peter Higgs: I wouldn't be productive enough for today's academic system

#179

Earlier quoted context omitted.

Can we please stop promoting the "democracy sucks" meme? It's not as if the Chinese are actually planning in that long a term; they're using markets now like everyone else.

> It's not as if the Chinese are actually planning in that long a term; they're using markets now like everyone else. // So, how long term are they planning, in years? In the UK we get [a few] plans that encompass maybe the next 10-15 years at most. So presumably they've got that at least.

The problem is not with doing planning, the problem is with implementing tradeoffs after such planning - in democracy, you can't take actions that hurt you in this year but will benefit 10 years afterward, if it is not completely obvious to uninformed laymen. If you do, they get reversed after the next elections anyway.

Re: Peter Higgs: I wouldn't be productive enough for today's academic system

#180
post #102
post #100

Earlier quoted context omitted.

"Brian Greene has remarked that today’s grant-driven academia would not have allowed Einstein the luxury of a decade in which to develop his General Theory of Relativity." Academia has been a force for stagnation well before the 1900s. e.g. Einstein wasn't working in Academia when he released his 1905 General Theory of Relativity paper. He spent 1901-1908 working in the patent office while releasing papers outside of…

> 1905 General Theory of Relativity paper Special relativity. On general relativity he worked from 1907 to 1915.

Yeah. Sorry. Copy pasted badly. The link was right though...

Speaking of which:

https://en.wikipedia.org/wiki/History_of_general_relativity#...

Seems he had most of the concept nailed down by 1907 but was struggling with the math. So, he worked the module step-by-step only adding gravity in 1915 after learning about Riemannian manifolds.

Post reply on HN