Will it be graded on a curve?
Is there a make-up date?
Can we use a cheat sheet?
Are all the questions from the book or in class?
Sorry, I've taught too often :-(
1,804 karma · joined December 14, 2011
Is there a make-up date?
Can we use a cheat sheet?
Are all the questions from the book or in class?
Sorry, I've taught too often :-(
Just calling everything a "variable" seems imprecise.
I've been programming for 50 years. There are no "revolutionary models". Nor is there a silver bullet.
The only trend I see that has merit is proving programs correct. CMU (Harper, Pfenning, Crary, etc.) teaches the next generation programs how to write reliable software.
Write specifications so you know what is required. Show formally that the program meets specifications.
Most companies have bought into the old "Frat Boy" meme, albeit in a different form these days. The idea is to only pick "the best and the brightest" judged by the criteria chosen by the "best and the brightest" (Dunning-Kruger effect : -) https://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect)
Back in the day, MS decided you needed to be able to figure out a puzzle, e.g. how many ping pong balls fit in a truck, how many piano tuners are there in Detroit, etc. Eventually books arrived with all the "good" puzzles and people stopped using that as a gate keeper.
These days you'll find that your post-phone interview is conducted by 20-somethings who just graduated. The questions are all related to their data structures course. How fast can you reverse a linked list? What's the O(n) of this search? I'd be curious to see if they can perform Knuth's line break algorithm or Forgy's OPS5 algorithm.
Of course your resume is never read.
And, of course, those data structure experts have no clue how to prove programs correct, create an expert system, write a JTAG device driver, etc. The "gate keeping interview" is the new IQ test, conducted by people who think they are the "best and the brightest".
It used to be that the manager who needed a person would interview the person, examine their resume and references, and see if the person had the required skills for the job. Now it seems you don't meet a project manager until you are hired.
The primary blame is on Google/Microsoft/etc. Their HR (sorry, People department) is badly broken. I can't find the reference but I believe Google's head of "Persons" said the same thing.
Could someone suggest how this could possibly work?
Do you mean assembler language programmers?
Or do you mean programmers who understand cache?
Or do you mean programmers who understand the semantics of the ISA?
Or do you mean programmers who know compiler optimization for RISC-V?
Or do you mean FPGA-Verilog programmers who know how to build a soft-CPU?
Or do you mean programmers who understand the latest RISC-V Crypto extension ISA?
Or do you mean programmers who know how to build instruction extension hardware?
Or do you mean a Bipolar chip designer for high-speed circuits?
Or do you mean an ASIC designer?
Or do you mean...
Or do you even know what you mean? Because there are many aspects of the RISC-V ecosystem. I've played in all of the above areas and I still have no idea what a "RISC-V High Performance Programmer" could mean.
I only use fundamental-mode.
Emacs, like lisp, is "clay for the mind". You can adapt it to any task easily.
But requiring a security review is a bit much. I did security reviews for a living. The consulting company was paid $250/hour for my work for the client. When open source pays me $250/hour I'll do a security review. They are hard, they are subtle, and they require you to be "up to speed" on the latest, greatest, CVE announcements. You did read, understand, and are able to find the Qualcomm hole, right?. You do know how to find an attack against your software that is embedded in the hard drive controller, right? Keeping up with the flood of attacks is REALLY a full time job. I don't have the time to keep up anymore.
As for "legal notice in every file"... no. The license is included with the source. If you choose to ignore it or copy single files (allowed by my license but...) or any other circumvention then that's your problem, not mine. I gave the code away.
As for legal issues... I've studied copyright and trademark issues. I've had discussions with an IP lawyer. I know "all the words". But being my own IP lawyer is a fools idea. IP law is at least as complex as my program and subject to a lot of side-conditions as well as site-issues (aka courtrooms). Did you remember to include a severability clause? (In law, severability refers to a provision in a contract or piece of legislation which states that if some of the terms are held to be illegal or otherwise unenforceable, the remainder should still apply.) No? Well, then ANY part of your legal claim throws out ALL of your legal claim. Unless you have a law degree AND specialize in IP law, you don't know what you're talking about. Oh, and lawyers charge big bucks which you won't earn in open source. So, no, I'm not gonna be my own IP lawyer.
"Software does not live up to security review and legal standards"... issue closed. won't fix.
This is basically a high-temp version of a cracking tower used in the oil industry. The difference is that the temperature is high enough to "crack" anything.
Do they pay for an ASCAP license?
I'd still be confused.
I read research papers on tablets all the time and not one of them seems to be able to show the whole page as large as a piece of letter paper.
Tape is wildly expensive for some reason. This seems odd to me. I wrote a tape-based archive system in the 1970s. Most technologies have gotten cheaper. Tape has not.