Why I like VB.NET

[Richard Curzon]

> might be better. The debugger in VS is much, much better, and

Your appreciation of the VB debugger is limited by your experience. Same
trap you warned us about, right? <g>

Try this in VS6/ C++ .

Stop at a function breakpoint Check some locals, oops... you see something
is already wrong in the function here.

You read it from the top.... OOPS, there's a questionable line early in the
code. Set a breakpoint there. Go to the immediate window, and invoke your
function from the Immediate pane, bang it reenters this function, and stops
at the new breakpoint. Yes, definitely a change to be made here. Type in
the changes, while both calls are still running on the stack.

Now you see a ripple effect down the function, past both breakpoints. Fix
it. The IDE is stopped at BOTH breakpoints still. Try the function again
in Immediate mode again.

Now there are three "programs" running this one function at the same time,
with 3 copies of locals, three instruction pointers. Let them finish now
from the top down. All functions complete. And the original call even uses
the new code you put in from the second level of debugging.

Sound a bit useful? Yeah but it's a hack, ya know.

We are told we are going to lose this, cuz it's a "hack". Fine, "handy",
okay, "productive", but you developers gotta realize... it's just a hack,
you were bound to lose it sooner or later.

What does this mean, a "hack"? Unknown to the baseline tools like
C/C++/Java?

Excuse us, w're just VB programmers here, but ... we don't see see the hack
here. Just a prime example of excellent, productive tool design.

And no excuse for dropping it in vb.net, or not having it in c-sharp for
that matter! The old mechanism may not be feasible, but it should be no
harder than the other proclaimed miracles of dot-net, JIT among them. At
least tell us it's coming in V2 <g>.

regards
Richard.

[Andy Chevin]

> "New languages [Java, C-sharp] always proclaim their simplicity and
denounce
> the needless complexities of older languages [C++]. Much of that
> "simplicity" is simply immaturity. A language's complexity comes from
> dealing with the rather nasty and varied complexities of the real world."

I remember reading that and seeing his point 100%.


> Here're are more AT&T/Bell Lab engineers who can perceive the emperor's
bare
> butt is hanging out!

That's quite a jump from what Koenig & Lippman are saying.
There are a lot of problems with VB4 onwards that would benefit from more
publicity from authors. Only Mckinney has really pinpointed VB's pitfalls,
yet still come out with one of the best books on VB.


Andy.

[Steve Dee]

Richard,
I doubt if my VB debugger experience is any more limited than anyone
(almost) in this newsgroup. I would never even dream of doing what you just
suggested though...In any language. What's the point? I don't see it as
useful at all...sorry about that.

[Steve Dee]

Richard,
who the heck was talking about curly braces? Oh and...

> Turns out, only the people who created the durn thing, and nobody else,
can
> see anything questionable about this thing!

Read the rest of it....Stroustrop (?sp?) also says that C++ will rule
forever. I don't happen to agree with him though.

Oh and...

>>** Andrew Koenig, "C Traps and Pitfalls", 1985. "The dangling else
clause".
>>Page 23, "Although this well known problem is not unique to C, it has
bitten
>>C programmers with many years experience".

Notice the date there? Trying compiling one of those problems now or for
the past several years....good luck......about as good luck as leaving off
an "End If"

[Joe \Nuke Me Xemu\ Foster]

"Steve Dee" <Steve_Dee@md.prestige.net> wrote in message <news:3a9b2f21$1@news.devx.com>...

> Richard,
> I doubt if my VB debugger experience is any more limited than anyone
> (almost) in this newsgroup. I would never even dream of doing what you just
> suggested though...In any language. What's the point? I don't see it as
> useful at all...sorry about that.

In 94 or so, I was involved in supporting an app done using Access 2.0.
Sometimes we'd have to Ctrl-Break, change some code, and resume, not
restart, execution. Restarting would have resulted in data loss and
even more Congresscritters breathing down our necks.

--
Joe Foster <mailto:jfoster@ricochet.net> Got Thetans? <http://www.xenu.net/>
WARNING: I cannot be held responsible for the above They're coming to
because my cats have apparently learned to type. take me away, ha ha!

[Joe \Nuke Me Xemu\ Foster]

"Richard Curzon" <ssi@interNOSPAMlog.com> wrote in message <news:3a994bb6$1@news.devx.com>...

> 2 nice quotes on this topic. (Guess the author, I'll tell you under the
> line below)
>
> "I don't see Java or C-sharp as being technically revolutionary or "from the
> ground up". Had Java fist been designed from technical principles, it
> wouldn't have the ugly and illogical C/C++ syntax."

Whaddaya expect from languages designed by C++ bigots? Both Java and C#
are really "C++ Lite", or C--.

> "New languages [Java, C-sharp] always proclaim their simplicity and denounce
> the needless complexities of older languages [C++]. Much of that
> "simplicity" is simply immaturity. A language's complexity comes from
> dealing with the rather nasty and varied complexities of the real world."

I thought many of C++'s "needless complexities" arose out of the early
attempts to maintain backwards compatibility with C. This is why you
have to recompile at least half your code whenever you change even a
private data member in a class, whether the public signature or even
the amount of memory used by instances has changed or not.

--
Joe Foster <mailto:jfoster@ricochet.net> Space Cooties! <http://www.xenu.net/>
WARNING: I cannot be held responsible for the above They're coming to
because my cats have apparently learned to type. take me away, ha ha!

[Richard Curzon]

>What's the point? I don't see it as useful at all...sorry about that.

In a word, the point is productivity. If you watched a good VB coder do it,
you'd probably see how it lets you examine details, understand more, and fix
quicker.

I'd bet on 10 times faster debugging vs the typical C programmer... partly
using techniques like that.
e.g.
- launching a function from the immediate window
- being able to launch extra frames on the fly, each with own values and
breakpoints
- judicious use of fixing code on the fly

But it frankly doesn't matter if you get it. It only matters, how many
users out there DO get it, how many will be upset by MS dropping it? VB is
the goose that lays the golden egg in the MS programming offerings... <g>.

Yes, VB needs CLR. Yes, there are some necessary changes in browser and
debugging.

But expert VB coders hope some dropped features (such as this one) WILL be
engineered in *whatever way they need to be*, into later versions of
dot-net.

So far, the VB golden goose looks a bit sick, in spite of the steroid
shot... proceed with care MS !

regards
Richard
(ex-C programmer)

--
-----
Live without dead time - Raoul Vaneigem
"Steve Dee" <Steve_Dee@md.prestige.net> wrote in message
news:3a9b2f21$1@news.devx.com...
> Richard,
> I doubt if my VB debugger experience is any more limited than anyone
> (almost) in this newsgroup. I would never even dream of doing what you
just
> suggested though...In any language. What's the point? I don't see it as
> useful at all...sorry about that.
>
>

[Rob Teixeira]

Yeah, that's been my point too all along.
On the whole, I like .NET. I don't particularly care for the direction VB
has taken in some key areas. Still, I do like the new power of VB. The more
I use it, the more I find it diffucult to go back to VB6.
Still, it does seem like the entire development effort is thrown at C# while
VB.NET looks more like an afterthought.
The little missing features begin to pile up and range from slightly annoying
to frustrating. Yes, we can probably live without a lot of these "little"
things or live until they get "patched in", but it still adds significantly
to the view that MS treats VB like a second-class citizen in the VS community.

I once again caution MS, as you do, to proceed carefully. MS has already
pissed-off a good part of the VB community by making extreme (albiet sometimes
necessary) changes to the language. All they would need to do to make VB
disappear in favor of C#-only dev shops is to continue to ignore the needs
and features that VB requires and deserves.

-Rob

"Richard Curzon" <ssi@interNOSPAMlog.com> wrote:
>But it frankly doesn't matter if you get it. It only matters, how many
>users out there DO get it, how many will be upset by MS dropping it? VB
is
>the goose that lays the golden egg in the MS programming offerings... <g>.
>
>Yes, VB needs CLR. Yes, there are some necessary changes in browser and
>debugging.
>
>But expert VB coders hope some dropped features (such as this one) WILL
be
>engineered in *whatever way they need to be*, into later versions of
>dot-net.
>
>So far, the VB golden goose looks a bit sick, in spite of the steroid
>shot... proceed with care MS !
>
>regards
>Richard
>(ex-C programmer)

[Mike Mitchell]

On 28 Feb 2001 12:10:57 -0800, "Rob Teixeira" <RobTeixeira@@msn.com>
wrote:

>..... but it still adds significantly
>to the view that MS treats VB like a second-class citizen in the VS community.

Only what I and a few others have been saying for the past two months.

MM

[Sjoerd Verweij]

> The debugger in VS is much, much better,

You can't be serious. The only thing wrong with the VB6 debugger was that in
any application using subclassing or large object trees, clicking "Stop"
would make the entire IDE go poof.

Oh, wait, no, that was because of that wonderful DF everybody wants back!

Sheesh...

[Sjoerd Verweij]

> > void (*p[10]) (void (*)());
> still a matter of style and the language in this case. I don't think this
> is an "Obfuscation" example.

That was the point.

> Yeah C is cryptic to those that do not know C.

My point is that it is easier to get cryptic in C (although Perl reigns
there in my book -- what an absolutely masturbatory geek language). And
especially with C++, the & :: ** ->... whatever! Who has time for that?

[dragon]

MSFT already invested in COREL For LINUX :P

"John Proffitt" <bogon@earthlink.net> wrote:
>
>"Sean Taffin" <sean@taffin.net> wrote:
>>
>>"Tom Cabanski" <tcabanski@oai.cc> wrote in message
>>news:3a91a738$1@news.devx.com...
>>> The bigger question, I think, is whether or not .NET will live up to
the
>>> promise of cross-platform portability. If someone, maybe even >>Microsoft,
>ports .NET to Linux, I think Java will be just another .NET >>language.

>If not, I think .NET might be a flop.
>>
>>Hmm. I remember Steve Balmer claiming that Linux is the threat #1 for
>>Windows. I still have doubts if they will port .NET to Linux and if they
>>or someone else ports .NET to Linux, if Linux user will welcome a MS
>>technology.
>>
>
>Microsoft needs to loosen up. The business of selling software (like most
>enterprises that involve human culture) is not a zero sum game. If they
>port software to Linux, or help with the WINE project, or with Samba, or
>enter a dialog with the KDE or GNOME folks about user interface issues,
it
>makes their bid for cross-platform interoperability look more credible.

>It gives the company a chance to make some of that unix money, too. And
>it makes the company look a little more tractable and sympathetic, something
>that the Microsoft corporate image could sure use.
>
>Who knows? If the Evil Empire started to look like the Nice Guys it might
>get the Justice Department off their case. It might make it harder to sell
>people on the benefits of an Open Source solution. It might make it easier
>for people to choose Microsoft over Apple or BE. There's more than one
way
>to compete and more than one way to measure success.
>

[Steve Dee]

I'm very serious. I've used both on a daily basis since VB 3.0. The VS
debugger is better. And it never goes "poof", and clicking stop on a
subclass has nothing to do with DF.

"Sjoerd Verweij" <nospam.sjoerd@sjoerd.org> wrote in message
news:3a9ed796$1@news.devx.com...
> > The debugger in VS is much, much better,
>
> You can't be serious. The only thing wrong with the VB6 debugger was that
in
> any application using subclassing or large object trees, clicking "Stop"
> would make the entire IDE go poof.
>
> Oh, wait, no, that was because of that wonderful DF everybody wants back!
>
> Sheesh...
>
>
>
>

[Steve Dee]

Sjoerd,

> there in my book -- what an absolutely masturbatory geek language). And
> especially with C++, the & :: ** ->... whatever! Who has time for that?

I just don't see this as "cryptic"...but as I have said, I "cut my teeth on
C". It's just the language. It's not hard to pick this stuff up. Any
decent programmer can do it. What's so much easier with "And" vs. "&"? "&"
has always meant "And". Same with the other stuff..

[John Proffitt]

"Sjoerd Verweij" <nospam.sjoerd@sjoerd.org> wrote:
>
>My point is that it is easier to get cryptic in C (although Perl reigns
>there in my book -- what an absolutely masturbatory geek language).

Yeah, I bought a book and was going to teach myself perl. I did fine until
I hit the chapter on regular expressions. I sez to myself, sez I, This is
not fit subject matter for a human being. Let the machine do this if it
will, but I'll be damned if I'm gonna touch regex with a ten foot Pole.

That was the end of my flirtation with perl. My brain rebelled. I can't
learn it if I can't see it. My eyes refuse to focus on that stuff.

John