Old problem with performance

[Andrei Alexandrescu]

dsimcha wrote:
> == Quote from Christopher Wright (dhasenan at gmail.com)'s article
>> Kagamin wrote:
>>> Daniel Keep Wrote:
>>>
>>>>> You probably don't want D, you want ATS:
>>>>> http://www.ats-lang.org/
>>>>>
>>>>> Bye,
>>>>> bearophile
>>>> http://www.ats-lang.org/EXAMPLE/MISC/listquicksort.dats
>>>>
>>>> Dear god.  I think... I think I'm going to go cry in the corner...
>>> Is it common for functional languages to love 1-letter identifiers, I wonder?
>> Probably. Mathematicians use "ab" to mean "a multiplied by b", so they
>> tend toward single-character identifiers with subscripts. And functional
>> languages seem to cater more toward mathematicians.
>> I tend to find most examples of the remarkable efficiency of functional
>> languages entirely unreadable, partly because of this, and partly
>> because they seem fond of using many operators, some in ways that I find
>> bizarre.
> 
> <Rant>
> True.  Conversely, whenever I read something mathy, it always drives me crazy how
> hard it is to read something expressed in math notation that is actually quite
> intuitive to express in code in a sane language like D (or Python, etc).  If
> mathematicians applied their unreadable "style" to mainstream code, they'd be shot
> because people would believe they were obfuscating it on purpose.  Single letter
> variable names drive me crazy because when I read equations I constantly lose
> track of what the variables represent.  Worse yet, when mathematicians run out of
> good variable names, they use Greek letters that look almost the same as English
> letters, just to make it easier to confuse variable names.  When I do math myself,
> my programming side kicks in and I often find myself trying to name variables
> descriptively like I would if coding, rather than using one letter like a
> mathematician.
> 
> Similarly, I find linear algebra impossible to grok largely because some genius
> mathematician decided to overload a bunch of operators to mean completely
> different things when dealing with matrices than when dealing with scalars.  Heck,
> the multiplication operator isn't even commutative.
> 
> Frankly, yes, programming/comp sci may arguably be a branch of mathematics, but if
> it is, programmers have invented better notation than mathematicians.  The
> mathematicians should stay the heck out of all programming notation discussions.
> </Rant>

You're in good company. http://www.cs.utexas.edu/~EWD/ewd13xx/EWD1300.PDF


Andrei