clipka wrote:

> Not really the most modern of languages (unless you count VB/VBA, but as you say
> yourself these are actually totally different beasts), and indeed one of the
> most frowned-at languages (VB/VBA included in this case); I guess that's mostly
> undeserved, however: The frowns should go to the people who thought it a good
> idea to introduce "GOTO" as the second statement after "PRINT", thereby
> inviting virtually the whole Home Computer Generation to become hobbyist
> spaghetti code producers.

Of which I'm one. But to my mind any good computer language should allow 
one to produce
"spaghetti code" (I had forgotten that term) if one wanted to. The 
"GOTO" statement
deserves much of the frowning it gets, but not all.
> 
> Normally, the first two programs would be:
> 

I remember right the first "real program" I ever wrote was a program to 
calculate ozone
concentration on a Commadore Pet. It was the first computer I ever dealt 
with
and after a *very* short time playing with it I was able to write a 
program to replace
the one we had been using on a TI55(?) calculator (the one with the card 
reader --remember
that?) The Pet had a version of Microsoft BASIC in ROM. For years 
afterwards --
until Windows 95--Microsoft was my favorite software company! I had 
never even been
in the room with a computer before and could hardly use an adding 
machine. but I was
able to write a useful program with almost no experience. I don't see 
any good reason
why the basics of *any* high level computer language should not be 
almost that easy to
learn!
> It's really a matter of programming style whether a BASIC programmer should be
> despised or not. BASIC allows for a broad spectrum there. But then again, ever
> since I've seen the C source code of "NetHack", I can tell that at least C
> programs - even when not deliberately obfuscated - can be just as messy as any
> BASIC code. 

Much more so, I am assured. "Document the H* out of your C code!" was 
one of my
teacher's advice.
> 
> 
>> QuickBasic was it's highest development.
> 
> Just out of curiosity: Did they ever introduce such things as structs (records
> in Pascal)? That was the thing that bothered me most about BASIC ever since I
> first wanted to store a list of data tuples. Using a separate array for each
> "dimension" of the tuple just didn't feel right.

Not in QB as I recall. You can do it sort of in VB4, but ir fights you. 
That would
have been one of the next logical steps, along with C-like pointer 
capability and, of
course Windows graphic capability. If I remember right, the QuickBASIC 
and QuickC
IDE's were models of what a good simple IDE should be. Microsoft was 
good in those
days!

>> I was able to display an image and VBnet seemed to promise good image
>> handling capabilities if one could only figure them out.
> 
> If graphics (or any other non-textual user interaction, for that matter) is what
> you want to do, it comes as no surprise to me that you don't like modern
> computer languages. The major problem, however, is not an inherent feature of
> the languages themselves, but of graphical user interfaces (at least as far as
> Windows is concerned; 

Yes access to Windows graphic functions seem unnecessarily complicated and
poorly documented. I can't see any good reason for this.

I never got to have a look at X Window or the like),
> which follow an *event-driven* paradigm - i.e. *they* want to tell your program
> when to do what, which is almost perfectly incompatible with the classic
> application-driven paradigm. And although I guess this is exactly what you're
> wrongly attributing to OOP, you're not so far off the mark: As a matter of
> fact, an object in OOP can be regarded as an event-driven piece of software
> (each method call constituding an event), so OOP is a very good paradigm to
> model an event-driven application: All you need to do is consider your whole
> application an object, too - which I admit is indeed *not* a natural way of
> thinking.
> 
> (However, while the event-driven paradigm *is* an antithesis to the classic
> application-driven paradigm you undoubtedly favor, OOP can be used perfectly
> well for both.)
> 
> Unfortunately, modern mainstream programming environments indeed don't seem to
> come with any framework to support the classic application-driven paradigm,
> except for purely textual interfaces ("console application" in Microsoft terms)
> or a static sequence of dialogs ("wizards").

I don't understand a lot of what you say, but it fascinates me, I may 
have to learn OOP
and more about modern programming just for the fun of it.


> Speaking of Borland, maybe they still come with the good old "graph" library and
> the BGI graphics driver format, in which case I'd expect them to include a BGI
> to interface to Windows, too.

Maybe it does, if so I haven't be able to find it--or recognize it. The 
old "graphic.h" was for DOS and
won't work on XP, anyway, and I think the routines are incompatible with 
modern graphic
cards.
> 

> 
> C is really not a pretty language, by the way. 

No, but it's fascinating, and in my little experience addictive.

David :)

Post a reply to this message