11,881 reputation
52742
bio website
location
age
visits member for 3 years, 9 months
seen 2 hours ago
Very senior embedded real-time systems programmer (Texas law is very picky about the use of the term "engineer"), unsuccessful candidate for U.S. Congress, and unreconstructed Cold Warrior.

Jul
7
answered Full signed part of two complement long
Jul
3
answered Is it a good idea to return a default value if a field in a query cannot be found?
Jun
30
comment Is it safe to rely on static analysis to “reproduce” concurrency bugs reliably?
Something to add: 6) NEVER acquire and hold more than one lock at any given instant. 7) If you absolutely, positively MUST acquire and hold more than one lock at any given instant, make damned certain that they are ALWAYS acquired in the SAME order. If process A and process B must both acquire locks Red and Green, they can deadlock if one of the acquires Red first and the other acquires Green first.
Jun
30
comment Is it safe to rely on static analysis to “reproduce” concurrency bugs reliably?
@delnan: EXTREMELY cogent, perceptive comment. +100 if I could.
Jun
26
comment Which are the alternatives to using a stack to represent function call semantics?
@LorenzoDematté: Precisely. That was how the CDC 6600 Return Jump (RJ) subroutine call worked. "RJ xxx" stored a jump to PC+1 in xxx and then jumped to xxx+1. When the subroutine finished, it jumped to xxx, which jumped back to the calling routine. This works fine for nonrecursive languages like FORTRAN, but it doesn't work for recursive languages like PASCAL. PASCAL had to implement a call stack in software.
Jun
17
comment How to define different names for the same type and have the compiler check them?
The question is whether you want to catch the error the moment you utter it, or when you try to store it and discover the types don't match up. Over fifty years of computer science seems to suggest that you want to catch the error as early as possible. A complicated expression with two errors, in the presence of implicit type calculus, could wind up generating total garbage with a correct result signature, which is the worst-case scenario.
Jun
17
comment How to define different names for the same type and have the compiler check them?
@kevincline, the explicit instantiation requirement is arguably a Good Thing. As I recall, it was Tony Hoare who proposed a set of default typing rules, analogous to the FORTRAN I-N integer rule, to the ALGOL committee. They chastised him most severely, explaining that requiring explicit declaration of variables, and their types, reduced programming errors. (The author of that tale added that this happened BEFORE anyone knew the possibly-apocryphal story of the lost interplanetary probe, caused by a typo combined with default typing in FORTRAN.)
Jun
17
answered How to define different names for the same type and have the compiler check them?
Jun
14
comment Looking to simulate the rolling of a ball around a roulette wheel, while the resulting number is already known
37 for a Monte Carlo wheel, 38 for a US wheel. (All wheels have 36 numbers, divided evenly red/black. Monte Carlo wheels have one zero, colored green. US wheels have zero and double-zero green, to increase the house take. The Wikipedia article traces the history, and it is actually quite interesting.)
Jun
12
comment Why are so many languages passed by value?
@supercat: You're on your own. At best. Some places, you'll get a free spa treatment, complete with tar and feathers.
Jun
12
comment Why are so many languages passed by value?
@supercat: Deliberate aliasing. Undefined behavior. You'
Jun
12
comment Why are so many languages passed by value?
@supercat: No. In/out parameters are by definition readable and writable by the called routine.
Jun
11
comment What methods exist for assessing an organisation's development capability?
SEI CMM, albeit dated, and its more recent flavor, CMMI, is in fact exactly what you are looking for.
Jun
11
comment Is my work on a developer test being taken advantage of?
Here's a suggestion, for people in the US. Whenever a firm asks you to submit code samples, or do example work for them, every chunk you submit should carry a copyright notice: "COPYRIGHT (year) (your name). ALL RIGHTS RESERVED." If it generates code that will be run in a user's browser, make sure that the generator also generates a comment line containing that copyright notice. Honest outfits will have no problem with this. Dishonest ones will scream.
Jun
8
answered How can you predict the time it will take for two processes in two different machines in a cluster to communicate?
Jun
4
comment are programmers more forgiving of buggy software?
+1000 if I could.
Jun
4
comment Are there advantages for using recursion over iteration - other than sometimes readability and elegance?
@KChaloux: Tail call optimization is not an attribute of a language, but of the language translator (compiler). In this, The Year of Our Lord Two Thousand Fourteen, a production-quality compiler that does NOT do tail call optimization must be regarded as brain-damaged. (We used a PowerPC compiler at Nortel Networks in about 2000, as I recall. I had occasion to test tail call optimization, ON REAL CODE FOR A REAL SYSTEM, and proved that the compiler did do it correctly. More recently, 8051-derivative and PIC compilers have surprised people by doing TCO correctly.)
Jun
4
comment Are there advantages for using recursion over iteration - other than sometimes readability and elegance?
@delnan: No, a function call does not NECESSARILY "need to do more than a loop backedge". Read "Debunking the 'Expensive Procedure Call' Myth, or, Procedure Call Implementations Considered Harmful, or, Lambda: The Ultimate GOTO", by Guy Lewis Steele, Jr. MIT AI Lab. AI Lab Memo AIM-443. October 1977. BRIEFLY, Steele shows that a procedure call is just { push RETURN_ADDRESS; jump routine }, and, in some cases, the "push RETURN_ADDRESS" step can be optimized out of existence.
Jun
3
revised Are there advantages for using recursion over iteration - other than sometimes readability and elegance?
typo
Jun
3
answered Does anyone still use logic analyzers for debugging real time systems?