Hrmmm there's a few reasons why this quiz made me unhappy:

(a) you need a basic theory book, but Sipser's not necessarily better than Hopcroft/Ullman, and certainly less complete regarding automata and languages. anyone hoping to be less more than a theory tyro needs at least Computers and Intractability, Approximation Algorithms, Randomized Algorithms and a satchel's worth of math books

(b) I have to list APIUE because I felt we needed an operating systems book, but also a practicum, and also something on design; a thorough study of APIUE provides all 3, and also one of the most useful reference books around. but it feels a bit applied for this list.

(c) since you're trying to minimize your votes (so as not to list overly many), there's an emphasis on broad coverage (ie this felt more like "5-7 computer science books to bring to a desert island surrounded by a faraday cage and anechoic wall") -- hence the mandatory inclusion of Knuth and Leon-Garcia (Knuth covers algorithms broadly enough by himself to free up a CLR-ish slot, and you have to know something about networks or you won't be able to piss in modern buildings, let alone write code)

but what about...language design and lambda calculus? the former's too diverse. regarding the latter, only one book could possibly deserve the nod (Barendregt's The Lambda Calculus), but i doubt 5 people worldwide read the fucker in a given Year of our LORD.

what about architecture? hell, the single most important things any systems programmer will need to know for the next twenty years are:

- power wall and thus multi/many-core
- memory subsystems aka Keep your Willy in your Pants and your Data in your Cache

hell, the first question i ask almost every interviewee is "here is a data structure. optimize it for cache, which will be parameterized to you thusly". if they don't get that, they're shown the door.

what about GUIs and databases? well, frankly, GUIs nauseate me and databases are for accountants, the DMV and the sex offenders' office, certainly not elite coders (writing databases is something else entirely). so fuck 'em.

computer graphics ought just be called "graphics math" to cut down on confusion. books about triangles are beyond the scope of this enumeration, unless it's computational geometry for integer programming problems and loop optimization in compilers. numerical computing's the same kinda deal, just harder.

Nick wrote: "Hrmmm there's a few reasons why this quiz made me unhappy:

(a) you need a basic theory book, but Sipser's not necessarily better than Hopcroft/Ullman, and certainly less complete regarding automat..."

also: "Artificial Intelligence, a Modern Approach"? ummm, how about a 2 page handout on game trees, A* search and SVM's?

:D

databases are for accountants, the DMV and the sex offenders' office, certainly not elite coders GASP! relational databases are the backbone of nearly all dynamic websites! (it is true however that if you're using one, you definitely not doing any elite coding).


I feel like only 1 programmer in 100 can write efficient code these days. In the last 4 years, despite hardware getting faster, the same programs are running slower than they did and people are doing much less (computationally) than they used to with more powerful machines.

GASP! relational databases are the backbone

A story: My junior year, I took the intro to databases class at GT (CS4400), because they canceled my fucking graduate automata theory class in developments that can ultimately be traced back to the introduction of GUIs and people no longer a new assembly language every few years (in a Cormac McCarthy-like future, this will all change, and we'll also get to eat people). In one of those passive-aggressive bits of responsible behavior that so pockmarked my undergraduate career, I didn't bother attending until the first test. This wasn't so irregular as it was SOP -- what for textbooks, after all, if not autodidactically-freed time to explore massive SGI machines from the underwear-beclad comfort of one's weedful dorm?

Problem was, I'd forgotten to purchase the book, and knew no one in my class...it was an unpleasant night indeed, studying random things about databases, SQL, etc via undirected google searches. I entered with trepidation.

30 questions, multiple-choice (wtf? only CS class I can remember with multiple-choice) on the relational algebra. I read it through, not really knowing any terms, thinking I'll probably just drop the class...but as it turned out, assuming logical coherency in the questions and answers, there was only one likely, consistent assignment of semantics. I quickly reverse-engineered the selections, the projections, the joins and the closures, marked down my answers, and triumphantly departed for a nice Chianti and some fava beans.

Saying this list needs a databases book is like saying it needs a book on predicate calculus, which is like saying it needs a book on real analysis, which is like saying it needs The Elements of Programming and that rather than CS degrees we all ought get math degrees (one ought, of course, get both). Taken the other direction, it's like saying we need a book on scrotumtightening perl and a lab practicum with Windows NT.

What were we talking about? Sorry, I rant in my old age.

of nearly all dynamic websites!

I must admit to never having worked on a "dynamic website"; if I were to build such a thing, I'd almost certainly do it as a C apache module, and initiate the project by rewriting apache =D. Just for fun, from an old classic:

Where does the typical Real Programmer work? What kind of programs are worthy of the efforts of so talented an individual? You can be sure that no Real Programmer would be caught dead writing accounts-receivable programs in COBOL, or sorting mailing lists for People magazine. A Real Programmer wants tasks of earth-shaking importance (literally!).

* Real Programmers work for Los Alamos National Laboratory, writing atomic bomb simulations to run on Cray I supercomputers.

* Real Programmers work for the National Security Agency, decoding Russian transmissions.

* It was largely due to the efforts of thousands of Real Programmers working for NASA that our boys got to the moon and back before the Russkies.

* The computers in the Space Shuttle were programmed by Real Programmers.

* Real Programmers are at work for Boeing designing the operation systems for cruise missiles.

Some of the most awesome Real Programmers of all work at the Jet Propulsion Laboratory in California. Many of them know the entire operating system of the Pioneer and Voyager spacecraft by heart. With a combination of large ground-based Fortran programs and small spacecraft-based assembly language programs, they are able to do incredible feats of navigation and improvisation-- hitting ten-kilometer wide windows at Saturn after six years in space, repairing or bypassing damaged sensor platforms, radios, and batteries. Allegedly, one Real Programmer managed to tuck a pattern matching program into a few hundred bytes of unused memory in a Voyager spacecraft that searched for, located, and photographed a new moon of Jupiter.

The current plan for the Galileo spacecraft is to use a gravity assist trajectory past Mars on the way to Jupiter. This trajectory passes within 80 +/- 3 kilometers of the surface of Mars. Nobody is going to trust a Pascal program (or Pascal programmer) for navigation to these tolerances.

I feel like only 1 programmer in 100 can write efficient code these days. In the last 4 years, despite hardware getting faster, the same programs are running slower than they did and people are doing much less (computationally) than they used to with more powerful machines.

It's a damn shame. Then again, for that lucky 1%, you're now able to accomplish things you couldn't dream of 10 years ago -- combinatorial stuff like protein folding and tilings, thermonuclear simulations at the particle level in six degrees of freedom, moving around 10Gbps of Ethernet on COTS, supporting hundreds of unimaginably bloated ruby/php processes in just a few gigabytes of RAM LOL.... =D

Check out Network Algorithmics: The Design of Fast Network Devices and Hacker's Delight. Hackin' ain't dead punk fuckin' rock!

you're alright, MICHAEL