blog.zacbrown.org - just run away, now.

I was browsing my feeds in Google reader and came across this blog entitled “Signs You’re a Crappy Programmer (and don’t know it)”. For the most part I thought it was fairly accurate with the exception of not following the rule of trying to make your functions less than 20 (30/40/10/whatever) lines of code.

Now I’m not hardline on it, I think one should try to make all functions as short as they possibly can, but I do understand that some functions just can’t be made that way. I think preaching hard line either way is a bad plan but one as a general rule should try to keep functions short and sweet. Writing big long functions makes it hard for people besides yourself to follow the code quickly. I’ve dealt with my share of crappy functions that were a hundred lines long and totally unclear. Thats what you get when you work with people that aren’t programmers first though.

It would seem this week that everyone is out to get me… Guess thats life. Who wants to take my marine science exam? Any takers? Shouldn’t be hard… guess thats a good reason for me to take it.

===============

On a different note, Algorithms & Data Structs was interesting, lots of probabilities. Hurrah!

Wanna know what really chaps my ass? People like Louis Savain and his COSA project. In science and math I tend to take a fairly hard line when it comes to most things. That is: If you can’t show it, you don’t know it. Now thats not to say that theoretical physicists or mathematicians are wrong because they can’t show something, but at some level you have to have some sort of substantial product or outcome to what you purport.

Now the problem with Mr. Savain is he goes about calling all programming languages crap, blaming software models on the reason why “stuff isn’t as good as it could be” and saying that hardware models are the solution. Now this is all good and well, everyone’s entitled to their opinion; so long as they do not redefine words or invent bunk terms. My favorite word that he’s redefined is algorithm. The definition of algorithm from webster.com as follows: a step-by-step procedure for solving a problem or accomplishing some end especially by a computer. Now I don’t know about everyone else but I’m pretty sure that regardless of paradigm (functional, concurrent, imperative, procedural, object oriented etc.) at some level you must have a plan on how your program is going to accomplish its goal. I know in Prolog, thats a bit more blurry, but at some level you must describe a mechanism for which your program will solve the problem.

All in all, if you spend enough time reading his posts as well as his website, you start to see that he’s a complete crack pot. He has no substantial evidence for what he writes, no clearly defined product that he’s published or created. All in all, just a lot of fluff.

Can a brother get an amen?