MathGroup Archive 2004

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Re: Mathematica language issues

  • To: mathgroup at smc.vnet.net
  • Subject: [mg53195] Re: [mg53151] Re: Mathematica language issues
  • From: Andrzej Kozlowski <akoz at mimuw.edu.pl>
  • Date: Tue, 28 Dec 2004 23:12:12 -0500 (EST)
  • References: <BAY1-F20F726EB03FF724DBC191EC69A0@phx.gbl>
  • Sender: owner-wri-mathgroup at wolfram.com

I suppose I don't disagree but at the same time I fail to see your 
point. I mean, either Mathematica is good enough for this purpose, 
which is great, or it is not good enough, which, as far as I am 
concerned, is  just as good. I simply don't care since it does not 
affect me in anyway. If it is something important to you and you think 
Mathematica should be better in this respect then by all means let WRI 
know about it, but why do you expect others, like myself to support 
you?
Actually, come to think of it, it probably would be in my interest to 
oppose any work in this direction, since inevitably it would divert 
resources form the sort of things I want to be done. Nothing is free 
and in most cases when something gets done another thing doesn't or 
gets delayed.

Andrzej Kozlowski



On 29 Dec 2004, at 00:02, Joseph Oswald wrote:

> Andrzej--
>
>  History has shown that the same users/programmers who have "above 
> average intelligence" wish to use software environments far beyond the 
> original conception of the developer. In particular, the use of 
> abstraction and well-defined interfaces to allow the solution of 
> complex problems is one of the great intellectual developments of 
> human history. Software tools that are too brittle to support it are 
> poor tools.
>
>  In particular, every time that a tool-developer must pause to 
> investigate the limitations of a particular Mathematica primitive 
> wastes precious time; not just the time to discover and understand the 
> limitation, but the time required to avoid tickling the bugs while 
> developing code, and the time of end-users who must suffer tools which 
> have less functionality due to these limitations.
>
>  Tools are meant to reduce effort and amplify the results of effort. 
> In fact, there are precious *few* programming languages that support 
> this kind of tool development, not a huge number as you state.
>
>  I believe the listings on this page: 
> http://www.wolfram.com/products/field_specific.html are almost all 
> examples of tools written to translate specifications of specialized, 
> complex problems into lower-level Mathematica primitives. That Wolfram 
> links to them tells me that they *are* indeed part of Wolfram's target 
> market. That Wolfram does not address these bugs, on the other hand, 
> does not surprise me: historically, these kind of strange behaviors 
> are either ignored, because they are the result of deep design 
> decisions (such as the use of pattern matching as the fundamental 
> evaluation mechanism) or treated by the development of 
> yet-another-special-case operator. (Consider Block[] and Module[], 
> neither of which truly deals with the problem of free variables in 
> expressions.)
>
>  In fact, Mathematica's stated goal is "to create once and for all a 
> single system that could handle all the various aspects of technical 
> computing in a coherent and unified way."
>
>  Manually typing expressions one-by-one and checking each result, 
> using "above average intelligence" before entering the next is not 
> sufficient to handle the problems of technical computing, broadly 
> defined.
>
> --Joe
>
>> From: Andrzej Kozlowski <akoz at mimuw.edu.pl>
To: mathgroup at smc.vnet.net
>> To: josephoswaldgg at hotmail.com
>> CC: mathgroup at smc.vnet.net
>> Subject: [mg53195] Re: [mg53151] Re: Mathematica language issues
>> Date: Sat, 25 Dec 2004 19:52:23 +0900
>
>>
>> The simple answer to that is that I don't think  Mathemtica was ever 
>> written for this sort of purpose and WRI as a company does not aspire 
>> to a share of this market (if it exists). Otherwise obviously they 
>> would do something about this but the fact is that they don't, which 
>> is perfectly all right with me though may not be with you.
>> Mathematica was written for human beings and in fact for human being 
>> of above average intelligence.  At least for the time being this 
>> disqualifies it as a tool for the sort of purpose you have brought up 
>> here. If you are interested in that kind of thing, Mathematica is 
>> obviously not the right language for you. Why on earth does it have 
>> to be one? There are countless other purposes it is not suitable for 
>> and a huge number of other programming languages to choose form.
>>
>> Andrzej Kozlowski
>>
>
> _________________________________________________________________
> Don?t just search. Find. Check out the new MSN Search! 
> http://search.msn.click-url.com/go/onm00200636ave/direct/01/
>


  • Prev by Date: Re: shuffling 10^8 numbers
  • Next by Date: Re: Titles disappear on SuSE 9.1
  • Previous by thread: Re: Re: Re: Re: Mathematica language issues
  • Next by thread: creating exact numbers, strange behavior