BV> BTW, do you know what M$ used to use to write their C compiler
BV> ? Their Pascal compiler (sic)
BL> What did they use to write their Pascal compiler?

Their Assembler. Or at least, they used to. You could always tell
when a new version of something was coming out by the product before
it :)

BV> In the meantime, you have become a better programmer. What have
BV> you lost ? Nothing. What have you gained ? A wealth of
BV> experience and knowledge that you wouldn't have gained if you
BV> had been playing with VB.
BL>  I lose the time learning Delphi and Pascal that I could have spent
BL> doing something else. I'm a lazy bastard. I learned enough of C to
BL> understand what it was about, and then I switched onto VB because it
BL> is the best way to learn programming tricks... and then I switched to
BL> VB/DOS because all the pretty interfaces were only getting in the way.

But Bob, you are forgetting one very important point. VB for DOS is DEAD!

BL> I know quite a few tricks now.

ROFL...You've been playing with programming for 3 months Bob.
You THINK you know quite a few tricks :)

BL> I can see the limitations of VB quite clearly, and also its strengths.

Every product has limitations and there will never be one which
doesn't. The goal is to find a product which you feel comfortable
with. In your case it sounds as though VB is it....for now.

BL> Delphi may be ideal as you say - a VB on hormones - but you have a
BL> different attitude to user interfaces than me. To me, it is the
BL> *most* important thing.

Err, scooze me.
I think you may be confusing me with Paul (I don't do U/I's Edwards)
Remember me, I'm the bloke who make sure that every bloody pixel
lines up properly and that you don't have to press one more key than
is absolutely necessary.

BL> I don't want to spend my time being annoyed when I could be psyching
BL> myself up to do something impossible. Nothing breaks the spell quicker
BL> than a dopey program that makes we wait 10-seconds while it stuffs
BL> around... and leaves me in the wrong window.

Sure, I understand that. However, no matter how much you rant and
rave about it, you are going to have to learn to live with it when
using a real compiler. VB isn't a real compiler and as soon as you
learn to accept this then you might be able to move on.

BL>  I did it the other day. I invented a whole new system (probably
BL> re-invented it), and now I can write it in C if I wish... but I
BL> could never have created it in that awful Borland compiler.

Perhaps. I *DO* understand where you are coming from Bob.
An interpretive language can be great for development and sometimes
it can also be ideal for production code as well. It depends on a
few things. In general though, you won't ever get to exploit the
full potential of the machine with an interpreter like VB. However,
the real problem with VB for DOS is that it will NEVER be updated or
improved. What you have now is what you will have in the year 2010.

In the meantime, compilers and linkers will become faster and the
development environment will become more programmer friendly.
From the quick look I have had at it, it certainly seems that Delphi
is a lot closer to what you are seeking than anything else. Yet, you
seem to be dismissing it because it is based on Pascal and not BASIC.

BV> You buy Delphi and they end up teaching you Pascal and you
BV> don't know it.
BL> ROFL! They teach you Pascal without a function reference manual!
BL> What a clever trick.

Serious (sic) From my first look, I think they may just do it.

BV> In the meantime, they get the pull-through from the largest
BV> base of developers they have. Their Turbo Pascal market. If
BV> this is what they have planned, then they are very smart.
BL> I agree, but it leaves me having to buy Pascal, too... to do DOS.
BL> Your idea of writing DOS in VB, and Windows in Delphi is silly.

Why ? They are two totally different platforms in the way that
Digital Electronics is different to Analogue Electronics. Each
requires a different development system and strategies, Windows and
DOS are the same in this regard. What works for one may not be the
best for the other.

BL> I'm not saying that this is a bad idea - just an expensive one: a
BL> combined $600 compiler. Pascal is *nearly* an alternative to C, and
BL> it may be possible to learn just one language: Pascal and Delphi with
BL> dBase, and that's it! I can see why Frank wanted Delphi.

Sure, for Frank it was an easy choice. For me it is worth the cost
to have a look at it. If it is a product that I can use to make
money, I will keep it. If not, I'll sell it. No matter what the
final outcome, I will learn something from it and earn money from it.

BL> I *still* don't have a clue if the compiler is a patch on VB.
BV> ROFL...You'd better believe it Bob. The compiler is FUCKING
BV> FAST. ..[chomp]..
BL>  This is what I wanted to hear. I don't have the patience to plan out
BL> a program, write it, and then spend an hour compiling and degbugging
BL> it. This is the classic approach and VB broke that nexus. It makes an
BL> impatient, creative bastard like me work so well. I don't make logical
BL> steps - I jump to the end and see if it works. VB lets me do that.

I think Delphi will also. This is what I have been trying to tell
you but you aren't listening. Your just shouting " IT'S NOT VB ".

BV> I'd even say that it is faster in making the final EXE than VB is.
BL> See? Now you've got me worried! VB is *slow* writing its EXE. It
BL> takes 2 seconds to compile and 20 seconds to link. I can tolerate the
BL> 2 -seconds, but not the 20.

ROFL...Fuck me dead Bob. You only have the 20-second delay when you
are trying to produce a production executable. So fucking what if it
takes 30 minutes. You should have done all your testing using the
internal interpreter and you KNOW that they EXE will perform EXACTLY
the same. What fucking difference does 20 seconds make.

BL> Delphi would *have* to be faster compiling than VB. It would have
BL> to be *10-times* faster if it makes an exe every time you run it
BL> in the development environment. VB doesn't make an EXE.
BL> That's its strength.

Groan.

BL> You are watching this happen on a DX24/100. Try it on a 386/20...
BL> or try it on a larger program.

I'd be happy to. I'll make a disk set from the CD and bring it
around one day.

BV> The development environment is laid out a lot better than VB as
BV> well. You don't have these fucking windows everywhere which end
BV> up hiding behind other windows.
BL> Yair. That really shits me. I'm all the time moving things around
BL> trying to find the *one* spot where I can get at them.

See, this is another area where VB is full of holes. I've told you
about a product that seems to address these problems, but yet again,
you are screaming " IT'S NOT VB ".

BL> Dearie me... VB is aimed at the dickhead market. *I* am
BL> dead-centre on the market Borland should be aiming at, so why
BL> do I feel so disappointed?
BV> Because you are a real dickhead. You haven't even looked at it
BV> and you have dismissed it out of hand.
BL> No, I haven't! I haven't dismissed it at all. I've decided not to
BL> make a decision until I've seen it... and I *still* haven't seen it!
BL> I'll make a decision when I have more information.

This is one of the inherent problems with computer software. You can
NEVER have enough information Bob. Even after you've been using the
thing for 3 months, you will still find problems, things that don't
work, or things that you feel should be better. It's a fact of life.

BL> You are the dickhead, Brenton. You bought it sight-unseen when there
BL> was no hurry; you don't know Pascal so you are going to have to learn
BL> a new set of functions (without a book), and you are going to have to
BL> buy Pascal so you can program in DOS, or mix it with C that doesn't
BL> work because they are similar enough to stuff you up.

Nope. I bought it after listening to glowing reports from others in
the industry that I respect. Some who knew Pascal and others who
don't. Learning Pascal isn't going to be a big deal. It's just a
different set of inbuilt functions and some semantics of the compiler.

I won't have to buy Pascal for DOS because I am quite happy with the
tools I have for DOS. I am unhappy with the tools I have for Windows
and it seems that Delphi may be of assistance in that regard.
For the $250 it cost me, I will get a return on my money even if it
is no good. If it turns out to be a total dog, I will know and I
will have an idea as to why it is no good. Then when a customer or
potential customer asks me to write something for them using Delphi,
I can tell them that it isn't the way to go and I can turn them
around to use a product that I'm more comfortable with.

The **_REAL_** problem with any tool is that you don't even begin to
see it's weaknesses until you try to build a real product with it.
I've tried doing this with VB and I reckon I could have written the
bloody program faster in C. I spent so much time fighting with VB
that I lost all of the supposed productivity gains. Delphi may turn
out to be the same, I won't know until I try to use it. But if I
don't try, I will never know.

BL>  Frank's decision is sensible, and in his place I may would have done
BL> the same. He uses Pascal, likes Pascal, and is so good at it that he
BL> can shit on C. Now he can do it in Windows, too.

True, but if Delphi does turn out to be a dog, Frank will be in the
shit. He will be worse off than someone like myself, because he will
then have to wait until it's fixed or someone does something better.
Of course, the above should not be directly applied to Frank, but is
meant in a more general manner.

Regards, Brenton
@EOT:

---