Assorted links

by on February 26, 2012 at 7:52 pm in Uncategorized | Permalink

1. The importance of IQ and trust for investing.

2. The new George Dyson book on the origins of the computer, I just bought mine here.

3. Alan Turing’s school reports.

4. The problem with collateral, and what could possibly go wrong?, and a Google count on Target2 searches.

5. Chris Blattman’s links.

Jack February 26, 2012 at 9:06 pm

Two papers by the same authors on IQ and investing are available here and here.

Andrew' February 26, 2012 at 9:11 pm

4. Volunteer tax collection to go with volunteer tax payment?

Name Nomad February 26, 2012 at 9:21 pm

I love how Turing’s teachers kept harping on his weaknesses — ones which played no part in his adult life whatsoever. Rather than encouraging him to develop his obvious strengths, they just kept wanting him to waste time shoring up his French and “tidiness.”

Adrian Ratnapala February 27, 2012 at 2:32 am

Actually I was quite impressed with the reports, they seem much more frank useful than the ones I got when I was a kid. Whether and how much to worry about those weaknesses is up to the kid and his parents. He got good marks in the stuff he was good at.

Andrew' February 27, 2012 at 5:53 am

“how much to worry about those weaknesses is up to the kid and his parents”

We give grades here.

dearieme February 27, 2012 at 9:41 am

‘Rather than encouraging him to develop his obvious strengths, they just kept wanting him to waste time shoring up his French and “tidiness.”’
He needed to strengthen his weak subjects if he was to pass the exams needed to get into university.

It was as he turned 15 that they began to realise that they had someone special on their hands, and they were obviously delighted too by how he matured thereafter. As AR commented, the old style of school report was actually useful.

CH February 27, 2012 at 1:11 pm

“a man’s gotta know his limitations”
- dirty harry

“a great seducer knows to maximize his strengths, minimize his weaknesses”
- CH

successfulbuild February 26, 2012 at 9:30 pm

#2. This is apparently a topic that brings much confusion and incorrect statements. I saw somebody at John Quiggin’s blog state that von Neumann was the chief engineer of the ENIAC.
http://johnquiggin.com/2010/12/10/state-of-innovation/. Of course it was Eckert, although apparently he may be right in some sense. Another funny statement I’ve seen on an economics blog is that the Internet was invented for physicists to share information (presumably the commenter meant the WW).

Many US computer science textbooks start out with a timeline of generations and seem to imply that the ENIAC was first. If you look at the computer tree from the US army you see the same thing, yet if you remove one of the adjectives from “first electronic, general purpose, large scale, digital computer” other computers (such as the Colossus, which they mention, the Z3, and the ABC) become candidates. (The First Computer: History and Architectures).

Theoretical computer science has developed notions like \mu-recursive functions, primitive recursive functions and the Chomsky hierarchy but as McCarthy once said “Computer science must study the various ways elements of data spaces are represented in the memory of the computer and how procedures are represented by computer programs. From this point of view, most of the work on automata theory is beside the point.” I’ve always been interested in the first computer engineers and how they dealt with memory issues and so on. (The Z3 could only store 64 values and the ENIAC had 20 accumulators), so that definitely looks like an interesting book, from my perspective.

Hugh February 27, 2012 at 1:55 am

It’s easy with hindsight to scoff at Turing’s teachers. But tidiness isn’t worthless in mathematics and his teachers seem to encourage as much as criticize. Also, if you’re a French teacher, you report on his performance in French (and even in English he is encouraged as a promising mathematician).

Turing did become a great mathematician. I don’t see much evidence his teachers are culpable for limiting his potential.

Adrian Ratnapala February 27, 2012 at 2:36 am

When I do derivations I cover pages and pages with un-ordered, scrawled formulae, some of them deliberately left incorrect. Occasionally I tidy up later. Maths is like that, you have to keep throwing ideas at the wall until one knocks out a brick.

I only found tidiness useful in mathematical work when I started marking the work of students. Teachers don’t want to spend half an hour decyphering scrawl on the off chance that it contains the right answer.

Dan Weber February 27, 2012 at 11:54 am

My son makes mistakes when he doesn’t keep his signs and columns well-ordered. Maybe Turing was capable of keeping track of that stuff in his head and his teachers were uselessly concerned with neatness. Maybe he wasn’t capable of keeping track of that stuff and neatness would have improved his math.

Slocum February 27, 2012 at 7:31 am

@3. Those school reports would now (in the U.S. at any rate), translate to high school grades that would keep a future Turing out of any elite university.

dearieme February 27, 2012 at 9:43 am

Any system that rates an 18 year old by how well he did at (say) 14 is bonkers – as is a system that rates 22 year olds by how they did at 18.

Andrew' February 27, 2012 at 1:56 pm

It’s sequential. I’m not sure exactly how bad it is at the end results, but now we have STEM elementary schools, we are definitely going down a bad road.

Newt February 27, 2012 at 6:22 pm

In the USA, they would make it hard for Turing to even make it to a decent state school, but he could start in community college and transfer to a state school or go to a second tier state school. HSPY, MIT, CMU, and Caltech would be out of the question and Cal, Madison, Urbana, or Boulder would probably be impossible even as a transfer.

On the other hand Cambridge, Oxford, the Ecole Polytechnique, the University of Tokyo, the University of Kyoto, the Ecole Normale Superieure, and Seoul National University would eagerly take a student with Turing’s marks once he passed their exams (as he easily would). That list includes most of the top 20 research schools outside the USA.

The extreme priority on getting along with the secondary system, sports, knowing the right people, and fitting in has become a uniquely American passion. Likewise the lack of any prestigious academy in the USA that considers an admission test as a principal criterion of admission is a distinctive failing. The decline of native born Americans in academe, even in the USA, continues apace and is no mystery.

Nigel February 27, 2012 at 9:17 am

You’re just trying to start another IP debate with the Dyson link, aren’t you. :-)

“…JN Another significant moral of the tale is the importance of open publication. The documentation for the IAS machine was all published, which meant that the machine could be cloned elsewhere (and indeed was by commercial companies such as IBM, as well as other research institutes), whereas the guys who built the ENIAC lodged patents, started a company and in due course became enmeshed in litigation. In our time, the computing industry is increasingly enmeshed in the same kinds of patent wars, so maybe there’s a lesson here for us. Is there a correlation between openness and innovation?

GD Yes, indeed. And what is amazing – and would horrify Abraham Flexner [the founding spirit of the IAS] – is that academic institutions are now leading the way in proprietary restriction on the results of scientific research! Of course there are arguments that this will fund more science, but those arguments do not make sense to me…”

successfulbuild February 27, 2012 at 2:53 pm

I don’t have a link right now but as I understand it Eckert and Mauchly lost those cases and computer engineers were able to develop without fear of violating the dreaded “ENIAC patent.” However, they had just as much a claim to their work as Microsoft does, if not more so.

In Scott Aaronson timeline of computer science he lists Eckert and Mauchly twice for their developments, one for the release of the first commercial electronic computer. From my readings I always felt their work was not appreciated enough.

Travers Naran February 27, 2012 at 11:20 am

For those of us who got a Computing Science degree, Von Neumann’s contributions are well known. We were taught, over and over again, that the architecture of the computer in front of you is a VON NEUMANN ARCHITECTURE. He literally wrote the book on how a computer should do its business . Although as the link points out, he was one of many contributors.

Turing contributed too , but due to security reasons and Turing’s later legal problems, he was elided from Britain’s official histories of the machines.

successfulbuild February 27, 2012 at 2:42 pm

I agree but still the first computers are considered to be the ENIAC etc., and Eckert perhaps should have gotten credit for the von Neumann architecture. Perhaps that is misleading though if this is ture:

“After I’d finished the book, I went back to look at the accepted “popular” histories of digital computing, and it seems that the IAS machine has effectively been airbrushed out of the picture. In most accounts, the story starts with the ENIAC machine in Pennsylvania and the Colossus machine built at Bletchley Park. But these were not stored-program machines and so were not really ancestors of the computers we use today, whereas the IAS machine was.”

philip[dot] greenspun [dot] com / ancient-history / history-of-computer-science

IAS is not there, and the EDSAC is referred to as the first von Neumann machine.

Travers Naran: “search for Manchester_Small-Scale_Experimental_Machine…”

Yes… More non-market involvement in the creation of the computer.

Travers Naran February 27, 2012 at 11:21 am

Links got lot: Search Wikipedia for Von_Neumann_architecture and Manchester_Small-Scale_Experimental_Machine.

Ken Rhodes February 27, 2012 at 11:33 am

In #2, Dyson wrote: “The inner sanctum of the IAS is the climate-controlled Rosenwald rare book room in their main library, which holds priceless classical manuscripts and later texts. A full set of the bound volumes of the Electronic Computer Project Interim Progress Reports are now shelved there, next to first editions of Newton and Euclid, where they belong.”

Fred Brooks wrote the wonderful little book, “The Mythical Man-Month,” which I regard as the best anecdotal textbook on project management I ever read. Those first editions of Newton and Euclid are priceless because they’re first editions, but their contents are available to all of us. I wonder if the contents of those project progress reports could also be made available to all of us. In paperbook they sould surely belong on my bookshelf alongside my Mythical Man-Month.

egl February 27, 2012 at 12:34 pm

The discussion in the Guardian does not seem to mention Atanasoff, and I would be wary of any history of computers in the 1930′s and 40′s that does not mention his work.

Kenneth W. Regan February 27, 2012 at 1:41 pm

My recollection, which I’ve been unable to trace on the Net or otherwise, is a source that pictures von Neumann attending lectures by Alonzo Church and maybe Kurt Gödel on formal logical systems of recursion that define computation, and coming away thinking they are nothing like his own idea of computation. He then attends a lecture by the visiting young Alan Turing, and reacts, “now that’s more like how I picture a computing device!” The source implied that Turing was the inspiration for his architecture and machine-step vision.

Andrew' February 27, 2012 at 1:55 pm

1. To me, trusting the stock market is largely about distrusting most everything else.

Comments on this entry are closed.

Previous post:

Next post: