blog: Overlooked Essentials for Optimizing Code

retard re at tard.com.invalid
Wed Oct 20 09:51:20 PDT 2010


Wed, 20 Oct 2010 19:25:04 +0300, so wrote:

> Every language has at least one niche (that it why they keep coming
> right?), but the pain is that there are tons of them. You are not
> expecting someone know all of them to the fullest right? If he tells you
> now that he knows all of it, next time you would say "There is this
> other language got one awesome feature called donuts! You don't know it,
> you suck!".

No, I don't expect that and I do believe in the law of diminishing 
returns. But some languages are part of the basic skillset of every 
modern developer. E.g. one mainstream statically typed app/systems 
language (C/C++/D/C#/Java/Ada/Object Pascal/Scala), one scripting language 
(Bash/Python/Ruby/Perl/PHP/Javascript/Lua/...), one "pure" language (Lisp/
Scheme/ML/Haskell/Prolog/...). Am I wrong, you only need to know how to 
program in C/C++/Java/D and make/microemacs/.bat scripts?

> He has every right to talk about languages IMO. You know he wrote
> compilers to hardest languages out there to implement. If he is wrong,
> it us up to you to prove him wrong, though not many times i have seen on
> this board that someone actually did it.

I see your point, but

1) I didn't criticize his statements about languages in my post (Max 
Samukha mentioned it, I only pointed out the pattern)

2) I find it hard to believe he is qualified to criticize university 
degrees world wide. For instance Bruno probably comes from Europe (which 
is not a homogenic single country). Second, the quality of degrees and 
courses varies -- without active participation it's impossible to give 
accurate, objective statements about the educational system. In every 
place I've worked in the senior workers always mention how the young 
generations don't learn any useful stuff these days (polluting chainsaws 
instead of axes etc.). This is universal, it also happens outside 
software engineering. The rants of old men and women.

> 
> Sorry but it is ugly to see all these Walter bashings(not pointing you).
> And he says/does nothing about it, it is hard not to respect his
> integrity.

I'm not bashing him. It shows amazing dedication to the subject when one 
changes the field after graduating. This isn't even a rarity because many 
of the modern subfields of computer engineering/science simply didn't 
exist back then. It's just that I expect developers / computer scientists 
to improve their knowledge base constantly. There are no signs of the 
field not evolving anymore.

Some things can be extrapolated from the evolution so far:

- compilers will catch more errors and produce faster code
- in popular languages the abstraction become higher and higher
- more languages will appear
- in 10 or 20 or 30 years most systems are multicore

Now, how can you know whether language X (e.g. Haskell) will be suitable 
in the future environment? How do you know which languages are worth 
studying if you don't get the basics of programming language theory? How 
can we expect one to *design* relevant new languages without these skills?

> 
> On Wed, 20 Oct 2010 18:59:40 +0300, retard <re at tard.com.invalid> wrote:
> 
>> Wed, 20 Oct 2010 14:59:21 +0100, Bruno Medeiros wrote:
>>
>>> I don't mean to offend anyone, but if you [sic] CS degree (at least
>>> for the last decade or so), doesn't teach about points 1 and 2 above
>>> as part of core curricula, then it's a pretty crappy CS degree. The
>>> same is probably also true for other related degrees (*-engineering,
>>> maths), at least with regards to point 1.
>>
>> This reminds me of
>>
>>> That is funny. Now and then you and Andrei talk so confidently about
>>> Go, C#, Haskell and other D competitors, without having written more
>>> than a couple of lines in those languages.
>>
>> Walter also talks so confidently about CS degrees, without having
>> earned one. The experiences probably stem from his caltech times with
>> the smelly bearded hippie unix guys who wrote bubble sorts in some
>> deprecated assembler dialect.
>>
>> This is becoming a real problem. I gave an example of Scala fairly
>> recently. I've given examples of code in other languages earlier. So
>> has bearophile. I can't ever assume that you guys study these basics.
>> The discussion stays at this level. It takes enormous amount of effort
>> to teach simple concepts. How many knows now what a monad is? It was
>> discussed again recently.



More information about the Digitalmars-d mailing list