Re: objects? we don't need no stinking objects!




"Tony Gravagno" <g6q3x9lu53001@xxxxxxxxxxxxxxxxxxxxxx> wrote in message...

I have a problem with people who don't understand "the other side",
and claiming that their coding model is still superior, to a point of
saying everyone outside of the Pick market is blind or stupid for
wanting to use something primitive like OO.

I'm not sure many of those in CDP aren't familiar with OO programming. It
is probably true that OO offers a model perfectly acceptable for some things
while MV BASIC and good design offers a model perfectly acceptable for other
things.

To suggest one development model is superior to all others in all
circumstances is irrational. MV development has produced long term, cost
effective solutions in the business market for years. This is evidence of
its efficacy as a development model. It is, however, by no means perfect as
many can attest to.

The problem with OO development in the business market is it's difficult to
see the benefits through the massive costs. Many who'd be glad to use this
model simply find many tasks very difficult. There's a certain amount of
complexity that is unnecessary in many circumstances for those not willing
to make OO development a career.

No you don't "need" OO, but since this has been almost universally
adopted as the standard programming model, you'd think someone who has
spendt his life as a professional programmer would at least take the
time to understand what it is he doesn't like.

This isn't really true. There are huge amounts of code running that wasn't
built with OO toolsets. Look into the relational model and you'll get an
earful of the same claims that the relational model is used by everybody, it
is the only dbms worth considering, all the toolsets use it, etc, etc, etc.
But, we all here are using MV mostly and there's so much non-relational data
out there we mustn't forget that.

Pick isn't even a footnote in most texts on data processing, but the
rest of the world is speaking gobblygook. Doesn't it seem reasonable
that there must be something to this? That it's not gobblygook to the
people who use it but perhaps an elegant means of modeling our world
inside the machine? Perhaps when one is outnumbered by a factor of
millions it's time to consider exactly what it is that they support
which you don't support or even understand yet?

I doubt it's an "elegant" means of modeling our world any more than the
relational model is. It's probably more a particular method to emulate the
world within a particular intellectual framework. Don't forget, the entire
IT industry doesn't exclusively use the relational model for all dbms needs.
Nor do they use software exclusively developed by OO toolsets.

(As I wrote this I was reminded of my own comments about how the
Church of Codd has brainwashed the modern world into believing that
the relational model is the only true data model. I have to concede
this appears to be a vulnerability in my argument, but without going
into detail in this medium I still believe both statements are
accurate.)

Well, you're being open to a broader perspective here. :-)

BASIC itself is considered inadequate by the pundits in the industry,
and despite the success of VB through VB6 and VB.NET, BASIC is still
consdidered more of a language for Beginners and hobbyists, more than
a real language for higher development. And here we are trying to
defend a slightly extended version of the original procedural
Dartmouth BASIC? That's just insane.

Using an argument I read a few paragraphs ago, since VB is the most popular
language ever, that should tell us that it is the best and should be the
only language anyone should use. :-)

Again I find it amazing that you're saying people who use words that
you don't understand are obscuring issues. Our simplistic Pick jargon
would befuddle anyone who doesn't know anything about computers, but
that doesn't mean we're any more elitest than the OO people you refer
to who use so many acronyms and so much jargon.

This isn't really true, for the most part. The language of the MV model
comes from an already existing environment, that of office management. This
explains why it's so easy to understand and why non-technical people have
developed plenty of applications.

Despite many who denigrate development by those who know what the
application is supposed to do, I think it's almost a given that IT,
especially the development discipline, have invented words and acronyms that
go counter to their normal meaning and usage in the spoken and written
language. When I look at the various computer languages I can't help but
be struck that the syntax must have been designed by those who have no clue
how to type on a keyboard nor have a working knowledge of the English
language and its structure.

The procedural model has been dead for quite a long time.
It's not being dismissed, it simply doesn't exist for those who
have grown up with OO. It's a footnote in history because it's an
early step toward what actually works in the modern world.

This is grossly inaccurate. The financial world operates in a linear and
procedural way. Funds flow around the world in ways I'm sure most OO
aficionados would be astonished with. Many other disciplines, in the real
world, operate this way too.

Reading other's comments in this forum for the last several years tells me
that most in the MV world are employing many of the OO methods and "best
practices" methodologies in their MV development. In addition, many have
recognized the limitations of the MV model with respect to connectivity and
user interfaces and are taking steps to upgrade their solutions (combine MV
with other technologies).

It's interesting that this thread began with the praise for DesignBais
because those users, along with many more in this forum, are trying to apply
the same methodologies to their new development as you've been recommending
for years.

I guess we're all allowed to blow off steam from time to time. :-)

Bill


.



Relevant Pages

  • Re: Relational model versus object model
    ... > Every now and then, a question pops-up on this forum, that initiates ... > the discussion on how the relational model relates to the object ... > Handling redundancy by hand is not fun (and programming is supposed to ... > one of the language you work in, ...
    (comp.object)
  • Re: Databinding - Best Practice (object-oriented)
    ... >> decent database theory introductory book. ... >relational model throughout your application severely restricts the ... >inheritance, interfaces). ... IMO a good OO language must have relational features and a good ...
    (microsoft.public.dotnet.framework.windowsforms.databinding)
  • Re: Databinding - Best Practice (object-oriented)
    ... >> decent database theory introductory book. ... >relational model throughout your application severely restricts the ... >inheritance, interfaces). ... IMO a good OO language must have relational features and a good ...
    (microsoft.public.dotnet.framework.windowsforms)
  • Re: Object oriented database
    ... rather a database / distribution system that is tightly linked with ... The need for finding consistent cuts when persisting FSMs ... but OODBs also see uses when the relational model does not ... bring in more modern language support in some areas, ...
    (comp.databases.theory)
  • Re: Is linq the final straw for VB?
    ... from my point of view German is a much more strict language as Spanish ... Who said anything about 1 superior langauge? ... Just some rare examples who seem to ... about VB developers in contradiction to "other" groups we are pretty ...
    (microsoft.public.dotnet.languages.vb)

Loading