.NOT My Views

[Jonathan West]

I found this:
"No, the transition to Visual Basic .NET 2002 was a one-time-only thing
that
allowed us to make the big leap from COM to .NET. If we've done things
right
with .NET, and I think we have, developers won't have to experience that
level of change again." - Paul Vick, Author of the Visual Basic .NET
Language Specification

I've had discussions with Paul Vick, and the impression I get is that he
regards it as justifiable to make language changes whenever there is a
platform change. Therefore, if he is still involved when .NET is superseded
(it will be, sometime) you had better prepare for further syntax changes.

You see, he would justify that on the basis that things are not as right
with .NET as they had thought, and so the changes are necessary.


--
Regards
Jonathan West - Word MVP
www.intelligentdocuments.co.uk
Please reply to the newsgroup
Keep your VBA code safe, sign the ClassicVB petition www.classicvb.org

 

[Herfried K. Wagner [MVP]]

you really need to do research before talking... all programs can be
decompiled...

They can at least be disassembled, but not necessarily decompiled.

its not something new with .NET VB6 apps

Mhm... I have never seen a working decompiler for VB6 native code
applications which restored the VB6 project files...

 

[Larry Serflaten]

If Hitler had won the war, than we would not have had *this* kind of
problems to discuss.

Just my thought,

....And if the sun goes into a super nova, this VB/VB.Net argument would
be the least of our troubles!

<g>
LFS

 

[Herfried K. Wagner [MVP]]

Is it not possible in vb.net to write desktop applications like vb6? What
I
think througth the petition, people want is that microsoft remove the
weeknesses of vb6, like object-orientation, threading, etc and maintain
the
appliction development that is Win32 focused.

That's definitely not the goal. The petition is /not/ about "programming
language X is better than programming language Y". It doesn't claim that
VB6 is better than anything. The aim of the petition is simply to ask for a
viable way for preserving customers' assets.

.Net offers a lot of things, lets say System.Convert. How would you
make that available in vb7? Will not the devs using vb7 want to use the
facilities of System.Convert, System.Text, or
System.Threading? I mean everything will be sort of existing in 2 forms.

I doubt that it has top priority for VB6 developers to be able to use
'System.Convert' et al.

Naturally in the new object.oriented vb7+ there would be string class
which
would offer methods when write "." instead of using mid or left etc. So
than
there would be some need for a class library like the .net's bcl.

VB.NET has a similar class library.

I think the vb.net is a pretty natural evolution of vb 6 and
earlier versions. We must accept it. Its beautiful.

Although I like VB.NET, I don't think that it is a new version of VB6.
Language stability was obviously not important for the designers of this new
programming language. The article below illustrates what language stability
means and why it should have top priority:

Language Stability
<URL:http://vb.mvps.org/tips/stability.asp>

 

[Phill. W]

You cannot use a gearbox from a Ford T as well on a Ferrari.

Oh; I'm sure someone /somewhere/ has tried it ;-)

... documentation, how to convert your vb6 program to Net.

Interesting how you say "convert", and not "upgrade".

Often the problem in this kind of situations is however that people
did not make their old programs in a good documented way and
don't even understand it themselves anymore.

All too true (even assuming the person who wrote the program is still
/with/ the company), but the fact remains that VB6 programs still run
and still support the businesses that chooses to run them. Rewriting
them just to "fit in" with a change of language is simply not an option
for all too many.

In my opinion is it not a challenge to make a program with
1000000 lines of code, the challenge is to make it in parts

Now /there/, I agree with you.
(Mind you, 1,000,000 lines of code must have an /awful/ lot of
repetition in it, mustn't it?)

Regards,
Phill W.

 

[Jonathan West]

Correct, but in this case the platform in question is just one and that is
.net and maybe they wanted visual basic to appear as #1 language (and i
remember them saying this #1-citizen-thing) and have a .net feel, which is
to have namespaces and classes, and everything being managed and so they
ended up with vb.net.

In that case, they put more effort into making VB.NET a #1 citizen than they
put into making VB programmers feel citizens of any kind. Personally, I
don't think that was a terribly sensible choice of priority.

Moreover, there would have been no difficulty in having the .NET classes as
*additions* to the language, alternative ways of doing things which people
could be encouraged to use for extending their programs.


--
Regards
Jonathan West - Word MVP
www.intelligentdocuments.co.uk
Please reply to the newsgroup
Keep your VBA code safe, sign the ClassicVB petition www.classicvb.org

 

[Cor Ligthert [MVP]]

Larry,

I was talking about a part of the history of mankind. You are talking of
future of the world.

We can change small things on the future of the world (with our current
knowledge not those big things as you describe). However we are able to
change the future of mankind. (Especially those simple things as
program-code).

I have not the idea that we both will see that mankind will be able to
change history.

Therefore let us talk about future and not about past, that was the
intention of my message.

Cor

 

[Jonathan West]

But if the move is from COM/Win32 to .NET. Do we still think it should
just
work? Can it be that easy?

The main platform change is the behaviour for object destruction. That is a
*behavior* rather than a *syntax* change. It might require some rewriting of
programs in order to account for the fact that you can no longer be certain
of the order in which object destruction occurs.

But to take a simple example, the syntax for property assignment in classes
was changed, on the assumption (true in the original VB.NET) that
mixed-visibility properties were not possible (e.g. Public Get and Friend
Let). Now that it *is* possible, you now have to assign the visibility to
the property and then modify it for the changed assignment. That was
unnecessary - the old syntax would have worked perfectly well and allowed
for the re-introduction of mixed visibility without a further syntax change.

Of course, you can argue that the new syntax is an improvement - it keeps
Let and Get together within the code. But even so, it mean that you can't
copy & paste from an old VB6 project or code sample and have it just work in
VB.NET. That's a great pity and a brake on productivity when extending
existing projects.


--
Regards
Jonathan West - Word MVP
www.intelligentdocuments.co.uk
Please reply to the newsgroup
Keep your VBA code safe, sign the ClassicVB petition www.classicvb.org

 

[alpine]

MS made it just that easy for their C++ code. They implemented IJW
("It Just Works") and, guess what, it just works!

If they had wanted to invest the resources in doing the same for
existing VB code, it could have been just that easy. However, they
don't own anything significant written in VB so, they didn't see the
need to invest the resources.

HTH,
Bryan
_______________________________
Bryan Stafford
New Vision Software
newvision_don'(e-mail address removed)

 

[mayayana]

What we /wanted/ was a compiler that took existing, VB6 syntax

oh ok, now its clear. I thought everyone from vb6 wanted to stick to win32.

Some of us do. What's wrong with Win32?
Underneath your .Net classes, somewhere, is
an equivalent to Win32 level.

Is it not possible in vb.net to write desktop applications like vb6?

Not very. It strikes me that this point just doesn't
seem to come up in the minds of most .Net users.
If it were not for Microsoft's strategy and tactics then
it would not even occur to anyone to discuss VB vs.
..Net. They're not for the same thing. This whole argument
is happening out of context.


For one good explanation of why .Net doesn't fit well
on the Desktop see Mark Russinovich's article:
http://www.sysinternals.com/blog/2005/04/coming-net-world-im-scared.html
]

My own situation: I taught myself VB, starting
out leaning heavily on OCXs and gradually
learned better, leaner coding. I write "shareware".
I don't make a lot of money at it but I enjoy coding
and I like making something that's useful to others.
My main program is an HTML and script editor
that's quick, light, free of dependencies (other
than the VB6 runtime) and has an installer of
about 800 KB. To my mind it's progress to cut
out the OCX middleman where that's feasible.
VB6 gives me the ability to do that, while still providing
easy methods where and when I want them.

What do I get if I move to .Net? First I lose
about half of my potential customer base right
off the bat, because dial-up users are not going to
download 25 MB runtimes and 12MB specialty
programs. Then I lose another 25% because many
people won't deal with the added complexity of
needing to download and install a runtime. I also lose
any ability to write for Windows 95. So now I've
got maybe 20% of my possible base left.

Beyond that, I'm
getting back into the bloat and inefficiency of objects
that are twice- or thrice-removed from the API. Why
would I want to do that? I like using the API. I like
using Windows. I don't want to be led gradually to some
sort of future X-Box/Vista crap hybrid that won't let me
access the system directly.

(And all of these reasons not to move to .Net don't even
address the common complaints, such as the issue
of easily decompiled code and poor speed, the
problem of continual changes that require new, giant
runtimes, and the question of how .Net will weather
the changes of Avalon, WinFX, etc. The thing is already
in its third rendition and now MS has plans for more
big changes with Vista.)

To my mind it borders on bizarre that so many people
using .Net keep hanging around here like religious
converts (or maybe MS plants) doing a .Net sales
pitch. I wish that people would try to just think for
themselves a bit, do what's best for their own case,
and not worry so much about what everyone else
is up to.

To anyone who says that the Desktop is old
hat and that I should be thinking about web-based
software: Maybe you're right, maybe you're not. Either
way, I have no intention of following such a path using
Microsoft software.

 

[DanS]

Value is created by architecture. If the Indians can do that as well
then good for them. Its their "competitive adavantage".

the competitive edge is strictly money based. period.

 

[Saga]

Haven't you heard? Much of the Managed Code hype has been
"decoupled" from Vista (as has the Windows File System) just so
that Our Friends in Redmond can get something to work and actually
sell some units of "Vista".

You forgot to mention "in a reasonable amount if time" <g>

Saga

 

[Abubakar]

Hi Herfried,

If you have not already done so and if you want to make Microsoft aware that
Classic VB should have a future, I encourage you to sign the petition at
<URL:http://classicvb.org/petition/>.

Is it not possible in vb.net to write desktop applications like vb6? What I
think througth the petition, people want is that microsoft remove the
weeknesses of vb6, like object-orientation, threading, etc and maintain the
appliction development that is Win32 focused. But how would you facilitate
multiple language versions? I'm trying to understand what would microsoft
have to include in vb7, for instance. .Net offers a lot of things, lets say
System.Convert. How would you make that available in vb7? Will not the devs
using vb7 want to use the facilities of System.Convert, System.Text, or
System.Threading? I mean everything will be sort of existing in 2 forms.
Naturally in the new object.oriented vb7+ there would be string class which
would offer methods when write "." instead of using mid or left etc. So than
there would be some need for a class library like the .net's bcl. Than the
community will start asking for "how can vb7 and vb.net code interop with
each other?" leading to what I imagine will be known as a "visual basic
mess". When every one knows that everything in Vista os will be available in
managed form why does visual basic community want to go back to an
environment where you have to pass function address for call backs and deal
with raw apis? I think the vb.net is a pretty natural evolution of vb 6 and
earlier versions. We must accept it. Its beautiful.

(these r just my thoughts on vb.not)

Ab.
http://joehacker.blogspot.com

 

[Abubakar]

Hi Phill,
first of all thanks for the reply.

What we /wanted/ was a compiler that took existing, VB6 syntax
and generated CLR-compliant code that would run atop the

oh ok, now its clear. I thought everyone from vb6 wanted to stick to win32.

"Rewrite/retest and redeploy everything because the language has
changed (again)"

what do u mean again? When did this last heppen other than the .net change?

Haven't you heard? Much of the Managed Code hype has been
"decoupled" from Vista (as has the Windows File System) just so

what i have heard about winfs, is that its not going to be available in the
initial release of vista that is comming next year. It doesnt mean thats not
going to be there forever. Work on winfs is going on and it'll be there with
all that it promisses. Even its beta 1 is here. And the managed code, its
all there. Much of the new things are available in managed code (avalon and
indigo).

IMHO, Rubbish!

I dont know what to say. It just looks good to me, but you have all the
write to say that.

VB*7* (and, yes, it did get written and /very/ nearly released) was
the natural evolution of VB6 - almost identical syntax, simple (or,

Really? Please give me any link on the web where I can read about it. Its
sounds interesting.

at least, extant) upgrade path. "VB.Net" grew out of a desire not
to waste all the effort Our Friends in Redmond made in their
attempted (and failed) theft of Java.

Now that is something I can never agree with. Sorry Because than Java
would become a failed theft of earlier written, less known, vitual machine
systems also.

Whatever you're taking must be Good Stuff ... ;-)

hey I dont understand what u mean by this. My english is not that good

Cheers.

-Ab.
http://joehacker.blogspot.com

 

[Karl E. Peterson]

It *hasn't* - that's my point - but now that it's happened /once/,
who's to say it /won't/ happen again ... ?

Oh yes it has! They redefined the fundamental String datatype with VB4/32, thus
breaking *every line* of binary file i/o code ever before written with MSBasic.

See http://vb.mvps.org/vfred/unimess.asp for (deep) background.

The following September, many in that thread were brought to Bellevue to discuss VB5
at its pre-alpha stage. In those discussions, this topic turned rather heated. They
said "We get it!", and promised that such redefinitions would never happen again.
Six short years later...

 

[Abubakar]

It *hasn't* - that's my point - but now that it's happened /once/,

who's to say it /won't/ happen again ... ?

I found this:
"No, the transition to Visual Basic .NET 2002 was a one-time-only thing that
allowed us to make the big leap from COM to .NET. If we've done things right
with .NET, and I think we have, developers won't have to experience that
level of change again." - Paul Vick, Author of the Visual Basic .NET
Language Specification

cheers,

Ab.
http://joehacker.blogspot.com

 

[Karl E. Peterson]

VB6 Code To Call API

Are you suggesting that "best practices" in VB.NET regressed to using addition for
combining bitflags?

 

[Roger Rabbit]

(And all of these reasons not to move to .Net don't even
address the common complaints, such as the issue
of easily decompiled code and poor speed, the
problem of continual changes that require new, giant
runtimes, and the question of how .Net will weather
the changes of Avalon, WinFX, etc.

This makes little sense complaining about an OS thats not even been
commercially released as reason not to switch over to a development platform
that been in use for around 5 years. Especially when you claim to be still
writing for Win95 customers. Your hardly pitching your software to an early
adopter market. This issue of decomplied code is not very compelling either
in that anyone whose ever taken a trip to China/Bali/or one of a million
websites in this world knows that exisitng software, whatever language its
been written in, is clearly fairly easy to deconstruct. Poor Speed. Thats
highly dependant on the application at hand but a quick Ngen on deploy so
the app doesn't JIT compile every time its boots would solve the vast
majority of problems.

To my mind it borders on bizarre that so many people
using .Net keep hanging around here like religious
converts (or maybe MS plants) doing a .Net sales
pitch. I wish that people would try to just think for
themselves a bit, do what's best for their own case,
and not worry so much about what everyone else
is up to.

Maybe they do think for themselves and they simply disagree with you.
Hanging around "here" is actually a dotnet newsgroup. Its odd that you
should come to one and complain that there are so many pro dotNetters on
board. As you've already said each to his own and do whats best for you. In
your case you clearly believe that dotNet is of no use to you. So what? Just
keep writing your VB6 apps. No one here is going to stop you.

To anyone who says that the Desktop is old
hat and that I should be thinking about web-based
software: Maybe you're right, maybe you're not. Either
way, I have no intention of following such a path using
Microsoft software.

I dont think anyone with any credibility is claiming that at all. Theres no
doubt over the next 5 to 10 years we are going to see a fundamental change
in the way software is licensed and delivered but the Desktop has quite a
long future ahead of it yet. ABove you complain that dotNet reduces your
overall potential platform market share (large download/Win95/etc) and now
here you claim that the "supposed" future of web based will see you reject
Microsoft software?

If that's the case then i really dont understand your initial complaint
given that IE is running at around 90 % + of overall marketshare. Your
position is to reject the supposed "omnipotent" platform of the future as a
response to a complaint about possible reduced market share today? Aren't
you just contridicting yourself?

RR

 

[Herfried K. Wagner [MVP]]

Karl,

Are you suggesting that "best practices" in VB.NET regressed to using
addition for
combining bitflags?

Using '+' to combine bitflags is not a "best practice" in VB.NET .

 

[Roger Rabbit]

Why? If it works as is, why should it have to be changed?

The whole point is he doesn't. Its been constantly suggested that unless
there is a very specific need there is no need to convert "legacy" code to
dotNet.

Also, why are you making assumptions that his million line application is
not already well-structured?

I think its a fair assumption. If he has a million lines of code that can
only be converted a million lines at a time then clearly its a monolith
thats not been very well constructed. Assuming the is a need to convert at
all, any reasonable design should allow a change in one part of the
application to have minimal effect on the remaining parts. Its fair for Cor
to assume spagetti code whether it is not.

RR