[fpc-pascal] Re: fpc-pascal Digest, Vol 9, Issue 39

[Angelo Bertolli]

> Let's now briefly touch the statement that 'Pascal is a 
> well-established language'...
>
> The Pascal I used first did not know a thing about objects,
> dynamic arrays; operator overloading, nor did it know how to
> treat strings, among a number of other things..
>
Sure, but we have to accept that no matter how well-established it was 
before Borland, it was the Borland compilers which usurped the throne 
for the sake of a "universal" standard.

> And, on the topic of 'adherance to certain rules and philosophies'..
>
> Luckily, people then were not so adamanat about --so called---
> Wirthian nature of Pascal... including Wirth himself; just look
> at the things Wirth altered in Modula/Oberon series.
>
> So... Not only the 'strict adherance to certain rules' thing
> is not cast in stone, but also --it seems-- can change from
> one version of the same compiler to the other...
>
> Finally, about 'Pascal is the philosophy, not the syntax'..
>
> As far as I gather, the only philosophy behind Pascal was
> to remove ambiguity in written code. And, that's it.
>
> If there is no ambiguity, why worry about whether it
> conforms to some mystical/celestial strictness..
>
Ok, I'll be more specific for those of you who don't see the value in 
keeping a type declaration as one unit.  I think it greatly reduces 
readability to not enforce this type rule.  This type rule allows 
programmers to expect certain things when reading someone elses or their 
own programs.  Don't get me wrong--I won't be too broken hearted if this 
is changed in an appropriate way, but I'd rather not see code like this:

type

var

procedure

type

And especially if var or procedure uses something defined in the second 
type--that breaks top-down design which I think IS an essential 
philosophy of Pascal.