owinebarger

Interesting, but there are significant differences between "software expert", "computer science students", and "software developer".

Your last paragraph is way off reality. People will use your talent to judge whether you're worth considering for a job. People will use your perceived character to decide whether you are worth doing a favor such as helping you in your career. If your attitude of "I'm talented enough to deserve your time and advice" showed, that alone would be sufficient reason not to help you.

A little humility while asking for help goes a long way with most people. That means if someone chooses not to help you out you recognize that their world isn't centered around you and your problems, and don't hold it against them. Instead, be grateful if they do give you the gift of their time and advice.

One of the best parts of this story is how crude a model can be and yet yield tremendous value. A dozen metrics on the underlying loans of the security and "most ARMs reset in two years" giving 3 years as the window for the market to fully recognize how bad the loans were.

Any con game relies on the greed of the mark.

Actually, they relied on rating agencies that said these were not very dangerous securities. They should have done their own homework, but there are regulations on valuing a pension fund's holdings. The government fell down by not ensuring the soundness of those regulations in practice.

I can only hope you're kidding. Those pension funds and state budgets are in shambles because they were underfunded and relied on the market to tell them the value of their holdings rather than the kind of fundamental analysis performed by Burry, because the market gave their investments ridiculous rates of return.

"Mark to market" is dangerous because it equates your option of selling an asset at the market price with the actual exercise of that option. It would be better to model the asset using the actual stopping times the firm uses to decide when to sell (on the portfolio as a whole).

I can't believe no one's mentioned:

  Computer science is no more about computers than astronomy is about telescopes.
  -- Edsger Dijkstra

Contrary to the beliefs or desires of some, colleges and universities are not vocational training institutions. In the actuarial world, there is a rigorous credentialing process involving a series of exams usually taking a minimum of 4.5 years to complete. However, even with all of the exams passed, without experience you'd be put into an entry-level position. If your experience is in one field of insurance and you want to get into another, that experience won't count for much in the new field and you'll probably be starting near entry-level again.

That's for good reason. Academia is for teaching you known facts and theories, and how to think for yourself. Experience in an industry teaches you how to do a particular type of work so that other people are willing to give you money in exchange for your labor. These are inherently different activities, though the former can weed out some inappropriate candidates for the latter.

Tried Doug Crockford's little Javascripter: (define loop (lambda (x) (loop x))) (loop 1)

Result in Firefox: "Too much recursion"

I'm still surprised I didn't flunk B521 when Dan Friedman put something like this in my Java implementation of a Scheme interpreter and it blew up after some finite number of iterations.

It might have worked if method calls were properly tail-recursive!

I posted this response in LTU, but it bears repeating.

Orthogonality (or representational independence), not size, is the relevant criteria. Harmonic analysis is the appropriate comparison. While every complete basis can represent any function in the relevant function space, it is better at representing some than others; notably the functions which are a finite linear combination of the basis functions. So trigonometric series represent one kind of function very efficiently, wavelets another. It's similar with programming languages.

Of course the orthogonal core may be mostly of interest to the language implementer,as orthogonality should, in the most faithful sense of the word, make the implementation of one construct independent of implementing another construct, particularly in optimization.

VBA when employed to do financial modeling.

Just how many decades does it take for a fundamental property of compiler implementation like "a proper tail call = goto + arguments" take to permeate conventional wisdom?

It's not like he's saying you need first class continuations - those can really add overhead to your computation. Properly implementing tail calls is a total win. If you need debugging help, add some keywords for it, but don't hair up your most convenient syntax for control transfer with unnecessary administrative overhead.

Behavioral interviewing techniques are supposed to be designed to induce stress well above and beyond nervousness when meeting someone new. The stress you feel from a deadline or a critical bug is also social stress. The deadline/bug are not able to care whether they are dealt with. The humans who expect you to deal with them are the source of the stress, whether they are your boss, management from other departments, or if you're running the show, your customers.

I'll make a non-mathematical observation about data-mining. Real world data sets are messy. Humans are just not the most reliable data entry machines. Data sets are not always gathered for the purpose you intend to use them. You have to validate/scrub/realign your data set before wasting time on an analysis that could be rendered meaningless by these issues.

It's the old "Garbage In, Garbage Out" principle, but it's easy to forget until the real world hammers it into you.

PG and you draw the wrong conclusion - to the extent hardware is like software it should only be protected as expression.

The sad fact about intellectual property law is that the courts have taken the "To promote the Progress of Science and useful Arts" part of the copyright clause and affixed it to the first amendment instead. The implicit message of including the clause in the constitution at all is that the right to copy exists and belongs to everybody. That right is only properly traded when the expression or invention fulfills that requirement.

That being said, I don't advocate acting as if patents are not where they are. You've got to take the world as you find it, and if the government makes patents available on loose terms you should use that to your business's advantage.

In this situation, I would recommend "Essentials of Programming Languages" before something like "Engineering a Compiler".

Shared processor, multiple memory sequential computation....

Everyone emails Excel spreadsheets around. The only problem is getting into email jail.

I think your problem is Sharepoint solves this problem (emailing spreadsheets) and is pretty standard these days.

It's the "literature review" that kills.

You're going to the wrong university/program. There are professors out there who exemplify the attitude you mention, though they also think getting it right is part of the fun.

I know you can find such professors at Indiana University, Bloomington from first hand knowledge. Northeastern's CS department seems another likely candidate.

Of course, I don't have a CS degree, I just took a fair number of courses while I was a math grad student.

Despite the purple prose, I generally agree with Aaron's sentiments about great programming ("the best", I can't say, but definitely "great"). He is, however, also very rude, at least in his online persona. Having "a forceful, uncompromising vision" is an orthogonal issue to how you choose to communicate.

It probably goes with being a recently credentialed actuary. It can feel like a job with the government at times, probably because a lot of the work is related to regulatory requirements.

Not to say there aren't entrepreneurial professionals, or entrepreneurial actuaries.

Hal Abelson's foreword is worth reading, if just for the footnote (about the AP CS exam that used to be based on Pascal):

   [1] Since Hal wrote this Foreword, they've switched the 
   AP exam to use Java, but the principle is the same.

Quibbling, perhaps, but "entrepreneur" is not a profession. Professions have codes of conduct, organizations that certify members as part of that profession, etc, etc. "Entrepreneur" appears to be on the opposite side of that spectrum.

Can anyone point me to a paper giving Clojure's formal semantics, if there is one?

You work for a large company with IT locked down, and this is the one "collaboration platform" available to you. Oh, but no Sharepoint Designer for you, my friend, if you want to make it do real tricks it will be with emacs (assuming you have jumped through the nine rings of fire to obtain permission for cygwin).

But I would leave out the "love it" part.

That idea has a name, and it is "Unix". qmail remains my favorite example of this philosophy put into practice.

I can't seem to find my copy of the 2009 Scheme Workshop proceedings, but Shriram Krishnamurthi's crew presented some work there that was pretty impressive. They are oriented toward teaching programming to 7-8th graders, before they learn algebra.

They use PLT scheme as the core of their system, but decided to implement it in JS to eliminate difficulties young kids would have installing software, and to simplify sharing of programs for bragging rights (for kids, of course - no adult programmers care about such things).

Not really. The flattening part is old hat in functional languages, though this particular formulation is restricted to the special case of "non-recursive" functions. I don't know about the particular optimizations they apply to the flattened code.

Excellent, excellent essay. There's even something for people seeking money (but you should read it even if you aren't):

  This was a typical Richard Feynman explanation. On the one 
  hand, it infuriated the experts who had worked on the 
  problem because it neglected to even mention all of the 
  clever problems that they had solved. On the other hand, 
  it delighted the listeners since they could walk away from 
  it with a real understanding of the phenomenon and how it 
  was connected to physical reality.

This extended flailing of Mark Helprin seems thorough, but it seems a lot longer than necessary for most of us.

Synopsis: Mark Helprin does not do his research and throws around a lot of unjustified and erroneous invective.