From the department of "Dead But Not Forgotten" news...

[Karl E. Peterson]

http://www.codeproject.com/Articles/572061/Classic-Visual-Basic-s-end-marked-a-key-change-in

--
.NET: It's About Trust!
http://vfred.mvps.org

[Bob Riemersma]

"Karl E. Peterson" <ka...@exmvps.org> wrote in message
news:kjvg17$lfn$1...@dont-email.me...
> http://www.codeproject.com/Articles/572061/Classic-Visual-Basic-s-end-marked-a-key-change-in

Interesting (if long) read. Thanks.

I don't agree with some of it though. VB6 can be nearly as OOP-ie as
anything else within the limits of the model it offers (inheritance isn't
everything and constructors are foreign to COM, AFAIK).

But he may still have a point. VB really does let a shadetree plinker run
pretty far while hiding from writing and using Classes and UserControls.
Maybe that's what he was getting at.

[Karl E. Peterson]

Bob Riemersma laid this down on his screen :
> "Karl E. Peterson" <ka...@exmvps.org> wrote...

>> http://www.codeproject.com/Articles/572061/Classic-Visual-Basic-s-end-marked-a-key-change-in
>
> Interesting (if long) read. Thanks.

Painful grammar, too. That's gonna detract from the non-neanderthal
reader base, for sure.

> I don't agree with some of it though. VB6 can be nearly as OOP-ie as
> anything else within the limits of the model it offers (inheritance isn't
> everything and constructors are foreign to COM, AFAIK).
>
> But he may still have a point. VB really does let a shadetree plinker run
> pretty far while hiding from writing and using Classes and UserControls.
> Maybe that's what he was getting at.

Yeah, safe to say, the users "pushed the envelope" at both ends! :-)

[Jim Mack]

On 4/8/2013 6:26 PM, Karl E. Peterson wrote:
> http://www.codeproject.com/Articles/572061/Classic-Visual-Basic-s-end-marked-a-key-change-in

Right, mostly a rehash of what many have said, from a guy who could
seriously benefit from the services of a copy editor...

--
Jim

[Mayayana]

| Right, mostly a rehash of what many have said,

Many *have* said things to the effect that VB just had
too much power for a BASIC version, and that it's an
unsalvageable language, etc. On the other side, people like
Joel Spolsky essentially said, "It's the convenience and
the RAD, stupid."

But there seem to be more obvious issues that, for some
reason, never get noticed in these VB memorial services.
(Another tendency with such "analysis" is to define VB as
dead and gone.):
MS created .Net to compete with Java and assumed that
by killing VB they'd instantly have a massive .Net customer
base. (The same trick they're trying now with Metro.)
The .Net move dovetailed with a web services bubble
The original "tool for web services" press release is still
online:

http://www.microsoft.com/Presspass/press/2000/jul00/pdcdeliverspr.mspx

They weren't pretending, at that time, that .Net was
Windows programming. Presumably that was because
they expected that web services would be an easy sell as
the hot new thing. But the web services fad fizzled...
Internet keyboards disappeared from store shelves...
"thin clients" didn't sell. Most people didn't even have
a high-speed connection.

Web services/SaaS was an idea that stuck, however,
even if the logic really didn't. And even if .Net wasn't really
all that well suited for the actual web services that
eventually developed.

As time went on and the Longhorn
era happened, the managed style of .Net dovetailed with
the trend toward increased security, and both dovetailed
with Microsoft's vision of eventually restricting the OS, so
that Windows would cease to be a software platform as
such and become a services retailer, taking a cut of
software sales and usage. (And sure enough, Mr. Ballmer
now calls MS a "devices and services company".)

For some reason these articles always seem to be obsessed
with procedural vs OO and BASIC vs C, as though there were
some pre-destined line of maturation in programming that's
playing out.
Meanwhile, a group of profound changes gets ignored. OS
software platforms have been turning into services operations,
while pushing programmers into a once-removed role,
writing sandboxed games and services to run on the closed
systems. Actual software running on the platform is becoming
a corporate specialty. (If online subscription services are to
succeed over low-cost installed software then the latter must
be rendered difficult or impossible to use.)

We're moving from privately owned cars, that can go anywhere,
to taxis that provide various services and entertainments, while
tracking the customers and showing them ads. The role of the
programmer is to sell concessions in the taxi, while paying the
corporate cab company an extortionary fee. Simple, sensible
programming tools with relatively easy access to the platform
API simply don't fit in that world. Even .Net doesn't really fit into
that world. It's just being strung along as a usable but not
critical tool for writing to WinRT.

I'm curious to see where it all goes. Gartner recently made
headlines by talking about the downfall of the PC. But it's hard
to see what will really happen in the long run. The analysts
are always biased toward trends that might turn out to be big
money makers. To hear them tell it, the entire world has moved
from business software to Twitter, Facebook and Angry Birds.

[CoderX]

"Mayayana" <maya...@invalid.nospam> wrote in message
news:kk16sf$mir$1...@dont-email.me...

> from business software to Twitter, Facebook and Angry Birds.

Whoa, whoa! What's wrong with Angry Birds? <g>

[Larry Serflaten]

Mayayana wrote:

> I'm curious to see where it all goes.

They are all frantically racing toward the smallest mobile form factor which will (obviously) be the GPS, SMS, Vid-Phone enabled wrist watch, that uses wrist motion to keep the battery charged....

<g>
LFS

[DaveO]

Larry



Watches, really? Pah! We're no longer in the age of Dick Tracy - Smart
tattoos drawing power from your mitochondria with subcutaneous nano-tech
processing is the way to go.

After that, direct interface with your visual cortex to do the Google
glasses thing without the need for glasses.

You plainly don't read the right sort of science fiction.



DaveO


"Larry Serflaten" <serf...@gmail.com> wrote in message
news:fae50201-9f34-4485...@googlegroups.com...

[ralph]

On Tue, 9 Apr 2013 10:02:30 -0400, "Mayayana"
<maya...@invalid.nospam> wrote:

>
> For some reason these articles always seem to be obsessed
>with procedural vs OO and BASIC vs C, as though there were
>some pre-destined line of maturation in programming that's
>playing out.

In the case of "procedural vs OO" it is because of the general lack of
understanding of object-oriented programming, or in many cases a poor
understanding of the methodology they are already using. (Everyone
uses a methodology, whether they know it or not. <g>) The real
discussion should be data-oriented (data-centric) vs object-oriented.
Every program contains elements of "procedural" code. It's called
algorithms. <g>

The other issue is because of the over-emphasis on "language". Fully
understandable - we are programmers. VB (meaning the VB product)
definitely uses the BASIC language as its front-end script, but it is
NOT BASIC. It is a full-blown OOP RAD development platform that uses
an intermediate language called pcode.

There is a little addin that shipped with Visual Studio (also
available as a separate download) called Visual Modeler. It is a
graphical object modeling tool. One of its uses is to reverse-engineer
a VB project into a Class diagram. Used to have all my students run it
to demonstrate just how OO VB actually is - in spite of all their
efforts to be "procedural". <g>

-ralph

[DanS]

On Wed, 10 Apr 2013 13:03:44 -0500, ralph wrote:

> On Tue, 9 Apr 2013 10:02:30 -0400, "Mayayana" <maya...@invalid.nospam>
> wrote:
>
>
>> For some reason these articles always seem to be obsessed
>>with procedural vs OO and BASIC vs C, as though there were some
>>pre-destined line of maturation in programming that's playing out.
>
> In the case of "procedural vs OO" it is because of the general lack of
> understanding of object-oriented programming, or in many cases a poor
> understanding of the methodology they are already using. (Everyone uses
> a methodology, whether they know it or not. <g>) The real discussion
> should be data-oriented (data-centric) vs object-oriented. Every program
> contains elements of "procedural" code. It's called algorithms. <g>
>
> The other issue is because of the over-emphasis on "language". Fully
> understandable - we are programmers. VB (meaning the VB product)
> definitely uses the BASIC language as its front-end script, but it is
> NOT BASIC. It is a full-blown OOP RAD development platform that uses an
> intermediate language called pcode.

To be more accurate I think, VB (meaning the VB product), could be
whatever you wanted it to be.

It *could be* a very "basic" thing if you only used the intrinsic
controls and functions, or it can be nearly all-encompassing and do
virtually anything you want using the API and writing your own classes.

VBc 's "problem" was/is perception, not VB itself.

(Of course, it also let total idiots write programs too.)