[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: s-exprs + prototypes




On Mittwoch, Juni 25, 2003, at 10:53  Uhr, Michael Vanier wrote:

>
>> Date: Wed, 25 Jun 2003 19:52:46 +0200
>> From: Pascal Costanza <costanza@web.de>
>>
>> [1] These are descriptions of extreme points of views. Of course, 
>> there
>> are more mild options in between. For example, one can think that OO 
>> is
>> suitable in 90% of all the cases, so it's safe to opt for it.
>>
>
> I was with you up to this point.  IMO it's exactly *because* of these 
> 10%
> of cases that relying on objects as the overarching paradigm is 
> dangerous.
> Sooner or later, many applications grow from the 90% realm to need 
> features
> in the other 10%.  How often have you used a language with a particular
> object system and then ended up butting your head against certain
> unchangeable rules that you could code around (or at least, could only 
> code
> around with great difficulty)?  That's why I think the discussion isn't
> bogus, not because of lambda-fundamentalism ;-)

I agree with you - I am actually in the Lisp camp. ;-)

I have just tried to describe what the fundamentally different 
positions are, and then add that some people are, of course, not that 
fundamental (for various reasons).

I am convinced that there won't ever be a unifying paradigm, mainly 
because of philosophical reasons. The point is: I am also trying to 
take into account that other people have a right to believe otherwise, 
and maybe they will find something some day.

However, the two beliefs (multi-paradigm / paradigm-less vs. 
single-paradigm) cannot be reconciled. You are either on one side or 
the other. And because of the terrible state of computer science we are 
largely left to our own experiences and beliefs.


Pascal