Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

John McCarthy Has Died

en.wikipedia.org

Post reply on HN
Print view View original

201–210 of 231 posts

Re: John McCarthy Has Died

#201

Post by phillmv »

papa_bear wrote: post #172
phillmv wrote: post #156

Earlier quoted context omitted.

It's how any species gets to iterate. The old have to die off to make room for the young.

Right, but if there is actually any selection for aging and death in organisms (maybe there is, I'm not sure), it's not useful anymore. There's a lot of evidence that there's no longer genetic selection acting on the human population, and it's therefore no longer true that our children should be genetically superior.

>There's a lot of evidence that there's no longer genetic selection acting on the human population

Sure there is. The fitness function is just different from what you or I consider 'optimal' or 'superior'. We're still becoming more resistent to disease, even if the future may have more in common with trailer park people than urban elites.

Re: John McCarthy Has Died

#202

Post by InclinedPlane »

TruthElixirX wrote: post #105
0x12 wrote: post #92

What a pity. I sense a black bar coming. The last few weeks have been terrible, one icon after another has fallen. What is different from other fields is that computing is still a young enough field that almost all of its luminaries except for the earliest ones are still alive. I fear the avalanche has only barely begun, and at some point the frequency of these will become high enough that either we will stop to noti…

Eventually, I think, there will cease to be icons within the industry. What we have now are the people who invented the internet/computers and the basic components. The people who are doing the inventing now are just as smart, but they aren't creating fundamental things, they are improving on an existing system. We don't say, "engineer team y invented the iCore series of processors." we sort of just say "Intel is com…

I disagree, here's why:

First, the software engineering field is still very immature. The tools and terminology in use are still very primitive. Studies seeking to identify best practices via evidence are still few and far between. Worse yet, known best practices with substantial evidence behind them are widely ignored in the industry on average. I believe there is still a lot of room for new techniques, and new models of software engineering, which will serve to guide developers for generations.

Second, we are at a critical juncture in programming languages. There are a lot of new languages gaining popularity, and a lot of older languages seeing substantial change. The fundamental paradigms in use by the average programmer are changing (becoming increasingly more functional, generally). Meanwhile, there are even more advanced concepts on the horizon (monads, for example) which are still not accessible to most programmers. There's a huge opportunity there for the next generation of languages to increase the accessibility of those advanced techniques. Additionally, we're in the midst of a revolution in terms of compiler technology. We are seeing things like modular compilers, very advanced run-times capable of incredible speed improvements, cutting edge garbage collectors, and the potential for massive parallelization on the desktop. Again, there's a huge potential there for some very fundamental innovations.

Third, we are no where near the end of the line in computing hardware. Over the next few decades we could see some very crazy things. 3D micro-chips putting trillions of transistors into every cpu. Even more massive parallelization than what we've seen today. RSFQ processors that could put petaflops computing power on a single chip. MRAM married to processor cores leading to breakthroughs in performance. Non-Von Neumann architectures enabled by memory and processing power being mixed together. FPGAs with billions of gates with the capability to operate and reconfigure at multi-gigahertz speed. Asynchronous CPUs.

And these are just things at a very fundamental level. There's still a huge potential for lots of innovations in the software and hardware space closer to the end-user. There's plenty of room for more icons in the future.

Re: John McCarthy Has Died

#204

Post by scott_s »

bad_user wrote: post #180
wisty wrote: post #154

Earlier quoted context omitted.

I hate to sound negative, but things won't get better. OK, this month has been a shocker, but next year might be just as bad. The real fundamentals of computing happened a long time ago. Hero of Alexandria, some Chinese and Islamic scholars, Albertus Magnus, Roger Bacon, Charles Babbage, Ada Lovelace, Joseph Marie Jacquard, Turing, non Neumann. But progress was linear, not exponential - the tools they made were so im…

It's easier to write git if you have a CVS repo to store your work in Not to diminish your point or anything, but regarding Git, I'll just quote Linus Tolvards: For the first 10 years of kernel maintenance, we literally used tarballs and patches, which is a much superior source control management system than CVS is Basically Git started as a bunch of scripts written by Linux for dealing with "tarballs and patches" an…

I think it's worth noting that git did not replace tarballs and patches. Git replaced Bitkeeper. Bitkeeper, as far as I know, replaced tarballs and patches. And as far as I know, git did not start as a bunch of scripts for dealing with tarballs and patches. The design was inspired by his knowledge of and experience with Bitkeeper.

Re: John McCarthy Has Died

#205

Post by Udo »

phillmv wrote: post #156
dlss wrote: post #125

Earlier quoted context omitted.

> Death is necessary for human ecology Ugh, not now. References or stop talking. I have my fingers crossed he was signed up for cryonics. This is tragic.

It's how any species gets to iterate. The old have to die off to make room for the young.

We're not Drosophila, we're people with minds capable of exploring and understanding a world that goes beyond biological evolution. We don't have to use death to iterate, we do it using general intelligence (it outperforms evolution by a factor of a bazillion). Also, not all organisms use preprogrammed death as a mechanism to iterate, ask an amoeba sometime.

All this talk about the inevitability of death and how that's supposdly a good thing is basically just pseudo-religious reasoning in defense of a status quo that is rapidly crumbling away.

Re: John McCarthy Has Died

#206

Post by scott_s »

cpher wrote: post #170
elliottcarlson wrote: post #167

Earlier quoted context omitted.

While there may be a bit too much bickering going on about it - having a factual source is important instead of just believing a single tweet or Wikipedia edit.

A bit too much? Half of the posts in this thread are debating this very topic. It's like everyone missed the forest for the trees.

It's not about "who knew what first." It's about figuring out if it's actually true.

Re: John McCarthy Has Died

#207

Post by bad_user »

scott_s wrote: post #204
bad_user wrote: post #180

Earlier quoted context omitted.

It's easier to write git if you have a CVS repo to store your work in Not to diminish your point or anything, but regarding Git, I'll just quote Linus Tolvards: For the first 10 years of kernel maintenance, we literally used tarballs and patches, which is a much superior source control management system than CVS is Basically Git started as a bunch of scripts written by Linux for dealing with "tarballs and patches" an…

I think it's worth noting that git did not replace tarballs and patches. Git replaced Bitkeeper. Bitkeeper, as far as I know, replaced tarballs and patches. And as far as I know, git did not start as a bunch of scripts for dealing with tarballs and patches. The design was inspired by his knowledge of and experience with Bitkeeper.

The workflow was indeed inspired by Bitkeeper, nothing is truly revolutionary, but Bitkeeper itself was not the status quo in a world dominated by CVS, SVN and Perforce.

On "tarballs and patches", I was referring to this famous quote from http://marc.info/?l=linux-kernel&m=111288700902396 ...

     So I'm writing some scripts to try to track 
     things a whole lot faster. Initial indications
     are that I should be able to do it almost as 
     quickly as I can just apply the patch, but 
     quite frankly, I'm at most half done, and if I 
     hit a snag maybe that's not true at all. 
     Anyway, the reason I can do it quickly is that
     my scripts will _not_ be an SCM, they'll be a 
     very specific "log Linus' state" kind of thing. 
     That will make the linear patch merge a lot more
     time-efficient, and thus possible.

Re: John McCarthy Has Died

#208

Post by agavin »

For nearly two decades I was a diehard LISP advocate. I even forced all my programers to code three Crash Bandicoot and four Jak & Daxter games in custom LISP dialects that I wrote the compilers for (an article on one here http://all-things-andy-gavin.com/2011/03/12/making-crash-ban...). But by the mid 2000s I started doing the kind of programming I used to do in LISP in Ruby. It's not that Ruby is a better language, but mostly it was the momentum factor and the availability of modern libraries for interfacing with the vast array of services out there. Using the crappy unreliable or outdated LISP libraries -- if they worked at all -- was tedious. Plus the LISP implementations were so outmoded. It was very hard to get other programers (except a couple enthusiasts) to work that way. Ruby struct a decent compromise. And it's type system and object model are better than CL anyway. The syntax is more inconsistent, and the macro model nowhere near as good. But it turns out. Libraries and implementation matter a lot. Still, you can feel lots and lots of LISP influence in all the new runtime typed languages (Ruby, Python, etc). And 30 years later, listeners still rule!

Anyway I bundled up some of my thoughts on this on my blog: http://all-things-andy-gavin.com/2011/10/25/lispings-ala-joh...

Re: John McCarthy Has Died

#209

Post by crpatino »

0x12 wrote: post #168
icandoitbetter wrote: post #164

Earlier quoted context omitted.

Not sure why you're being downvoted. Somehow people have this picture of mathematics as this universal body of knowledge, but that couldn't be further from the truth. They are as earthly as anything else we've been doing as a human race. For an interesting perspective on mathematics from a cognitive-scientific point of view, check out George Lakoff's book "Where Mathematics Comes From."

> Gates, Zuckerberg, Larry and Sergey, Torvalds, RMS, Knuth It's not too hard to imagine a world without facebook, but I'd find it very hard indeed to imagine a world without Microsoft, Google, Linux, Open Source or the Art of Computer Programming. It's almost like this is one of those trick questions: "which name does not belong in this list". Zuckerberg pops out at me immediately. All the rest is timeless, Facebook…

I have to disagree on this one. If there is one who does not belong in the list, it would be Knuth. Zuckerberg, as well as Gates, Larry and Sergey, has created a technology company that changed the lives of millions. They are all well known by the "civilians". Torvalds less so, but still, the Linux project has had a relevance that trascends beyond mere technology.

Mr. Knuth's contributions, on the other hand, are much more fundamental. But, because of this, his name remains unknown for the general population. For us, he is a titan amongst men; for the world, he'd be an old, retired professor who cannot even figure out email.

Post reply on HN