The Future of Unmanaged Code

[Tim Romano]

As a newcomer to the discussion, I apologize if this subject has been worked
to death. But it seems to me that .NET and its "managed code" approach is a
way of killing off competing application development tools/languages such as
Delphi and Java. What is the future of "unmanaged code" under Windows?
What is the projected lifespan of the Win32 API?

Tim Romano
www.aimsdata.com/tim

[Mike Mitchell]

On Thu, 22 Feb 2001 09:49:54 -0500, "Tim Romano"
<tim_romano@yahoo.com> wrote:

>As a newcomer to the discussion, I apologize if this subject has been worked
>to death. But it seems to me that .NET and its "managed code" approach is a
>way of killing off competing application development tools/languages such as
>Delphi and Java. What is the future of "unmanaged code" under Windows?
>What is the projected lifespan of the Win32 API?

If you're wondering whether there will come a time when VB6 apps will
no longer run on the then current version of Windows, I do not know.
But what is clear is that Borland, for example, *have* produced a
Delphi look-alike in Kylix which obviously relies on absolutely no
Windows platform. So if one vendor can produce a RAD package without
basing it on Windows, so can others.

MM

[Karl E. Peterson]

Hi Tim --

> As a newcomer to the discussion, I apologize if this subject has been worked
> to death. But it seems to me that .NET and its "managed code" approach is a
> way of killing off competing application development tools/languages such as
> Delphi and Java. What is the future of "unmanaged code" under Windows?
> What is the projected lifespan of the Win32 API?

Oh, it's far more sinister than that! A friend of mine who shall remain nameless
(but has a popular east-coast radio talk show about computing <g>) and I were talking
about this yesterday, and I thought his comment below was particularly inciteful:

>>You think MSFT will compile Word or Excel using .NET and rely on copyright
>>enforcement to protect them? The day they ship this port, will be the day
>>you can assume your intellectual property is somewhat safe.
>
>Nah, that ****'ll be coded in native C++. And they probably wish
>they never ditched the ASM.
>
>That's the REAL story here: How MS and Sun end up controlling
>high-performance computing because they're the only guys left standing who
>will still be coding in ASM and C.

Hear them black helicopters? <g>

Later... Karl
--
http://www.mvps.org/vb

[Jonathan Allen]

X86, C and C++ are unmanaged languages.

LISP, Basic, Pascal, VB, Java (and probably Delphi) are managed languages.

Keeping that in mind, please rephrase your question.

--
Jonathan Allen


"Tim Romano" <tim_romano@yahoo.com> wrote in message
news:3a95267d@news.devx.com...
> As a newcomer to the discussion, I apologize if this subject has been
worked
> to death. But it seems to me that .NET and its "managed code" approach is
a
> way of killing off competing application development tools/languages such
as
> Delphi and Java. What is the future of "unmanaged code" under Windows?
> What is the projected lifespan of the Win32 API?
>
> Tim Romano
> www.aimsdata.com/tim
>
>
>

[Tim Romano]

Mike,
"Whether there will come a time when VB6 apps no longer run on the then
current verison of Windows" is precisely what I'm wondering. I'm not a
computer scientist and have tended to avoid learning about the innards of
Windows, and so invite correction by those here who know these things-- but
it seems to me that if the Win32 API were to become frozen in time, and COM
as well, then that stability would usher in a golden age of desktop Win32
computing--hardly what Microsoft would want if it is trying to move the
desktop forward onto the .NET framework. So, my mind jumps to this
conclusion: how long could the Win32 API and the .NET framework peacefully
coexist, given such marketing pressures?

I take it as a given the Microsoft OWNS the desktop after having
successfully killed off Java there, if not on the server. Microsoft, as
owner of Win32, is probably capable (technically, if not legally) of doing
subtle things to the Win32 substratum that could make life quite difficult
for Win32 legacy applications.
Tim Romano


"Mike Mitchell" <kylix_is@hotmail.com> wrote in message
news:3a957ee7.3899021@news.devx.com...
> On Thu, 22 Feb 2001 09:49:54 -0500, "Tim Romano"
> <tim_romano@yahoo.com> wrote:
>
> >As a newcomer to the discussion, I apologize if this subject has been
worked
> >to death. But it seems to me that .NET and its "managed code" approach
is a
> >way of killing off competing application development tools/languages such
as
> >Delphi and Java. What is the future of "unmanaged code" under Windows?
> >What is the projected lifespan of the Win32 API?
>
> If you're wondering whether there will come a time when VB6 apps will
> no longer run on the then current version of Windows, I do not know.
> But what is clear is that Borland, for example, *have* produced a
> Delphi look-alike in Kylix which obviously relies on absolutely no
> Windows platform. So if one vendor can produce a RAD package without
> basing it on Windows, so can others.
>
> MM

[Tim Romano]

Jonathan,
Are you making a distinction between interpreted and compiled languages
here? If so, then that is not what I meant by "managed" versus "unmanaged".
With .NET, the operating system -- or the "framework" controlled by the
operating system vendor-- becomes the interpreter. What would happen to
third party legacy programs that depend on Win32 if that API were to become
destabilized? I expect it to become destabilized because, as I said in my
reply to Mike Mitchell, a stable Win32 "frozen in time" would usher in a
golden age of Win32 legacy programming, which is not what Microsoft wants,
is it?

Or did I misunderstand you with respect to Delphi? Is there going to be a
..NET version from INprise?
Tim Romano

"Jonathan Allen" <greywolfcs@bigfoot.com> wrote in message
news:3a95a1d9@news.devx.com...
> X86, C and C++ are unmanaged languages.
>
> LISP, Basic, Pascal, VB, Java (and probably Delphi) are managed languages.
>
> Keeping that in mind, please rephrase your question.
>
> --
> Jonathan Allen
>
>
> "Tim Romano" <tim_romano@yahoo.com> wrote in message
> news:3a95267d@news.devx.com...
> > As a newcomer to the discussion, I apologize if this subject has been
> worked
> > to death. But it seems to me that .NET and its "managed code" approach
is
> a
> > way of killing off competing application development tools/languages
such
> as
> > Delphi and Java. What is the future of "unmanaged code" under Windows?
> > What is the projected lifespan of the Win32 API?
> >
> > Tim Romano
> > www.aimsdata.com/tim
> >
> >
> >
>
>

[Tim Romano]

Hi Karl,
I hear those helicopters, but it seems to me they're hovering over Sun. I
wouldn't expect Sun to be very happy about .NET as its purpose seems to be
to get Microsoft ownership of the server space as they now own the desktop.
Tim Romano

> >
> >Nah, that ****'ll be coded in native C++. And they probably wish
> >they never ditched the ASM.
> >
> >That's the REAL story here: How MS and Sun end up controlling
> >high-performance computing because they're the only guys left standing
who
> >will still be coding in ASM and C.
>
> Hear them black helicopters? <g>
>
> Later... Karl
> --
> http://www.mvps.org/vb
>
>

[Jonathan Allen]

> Are you making a distinction between interpreted and compiled languages
> here?

No, the distinction is wither memory is managed by the runtime or the
programmer.

Wither the language is compiled or not rarely factors into programming.
Especially when you consider that many languages off both modes.

--
Jonathan Allen


"Tim Romano" <tim_romano@yahoo.com> wrote in message
news:3a95d93e@news.devx.com...
> Jonathan,
> Are you making a distinction between interpreted and compiled languages
> here? If so, then that is not what I meant by "managed" versus
"unmanaged".
> With .NET, the operating system -- or the "framework" controlled by the
> operating system vendor-- becomes the interpreter. What would happen to
> third party legacy programs that depend on Win32 if that API were to
become
> destabilized? I expect it to become destabilized because, as I said in my
> reply to Mike Mitchell, a stable Win32 "frozen in time" would usher in a
> golden age of Win32 legacy programming, which is not what Microsoft wants,
> is it?
>
> Or did I misunderstand you with respect to Delphi? Is there going to be a
> .NET version from INprise?
> Tim Romano
>
> "Jonathan Allen" <greywolfcs@bigfoot.com> wrote in message
> news:3a95a1d9@news.devx.com...
> > X86, C and C++ are unmanaged languages.
> >
> > LISP, Basic, Pascal, VB, Java (and probably Delphi) are managed
languages.
> >
> > Keeping that in mind, please rephrase your question.
> >
> > --
> > Jonathan Allen
> >
> >

[Jonathan Allen]

Eventually Microsoft is going to have to let Win32 go. The CLR may contain
the seeds of this transition, but it is still a long way off. Some day we
will have a safe and clean OO based operating system and API. (I wonder if
it will be in my lifetime.)

--
Jonathan Allen


"Tim Romano" <tim_romano@yahoo.com> wrote in message
news:3a95d77e$1@news.devx.com...
> Mike,
> "Whether there will come a time when VB6 apps no longer run on the then
> current verison of Windows" is precisely what I'm wondering. I'm not a
> computer scientist and have tended to avoid learning about the innards of
> Windows, and so invite correction by those here who know these things--
but
> it seems to me that if the Win32 API were to become frozen in time, and
COM
> as well, then that stability would usher in a golden age of desktop Win32
> computing--hardly what Microsoft would want if it is trying to move the
> desktop forward onto the .NET framework. So, my mind jumps to this
> conclusion: how long could the Win32 API and the .NET framework peacefully
> coexist, given such marketing pressures?
>
> I take it as a given the Microsoft OWNS the desktop after having
> successfully killed off Java there, if not on the server. Microsoft, as
> owner of Win32, is probably capable (technically, if not legally) of doing
> subtle things to the Win32 substratum that could make life quite difficult
> for Win32 legacy applications.
> Tim Romano
>

[Zane Thomas]

On Thu, 22 Feb 2001 21:16:36 -0800, "Jonathan Allen"
<greywolfcs@bigfoot.com> wrote:

>Some day we
>will have a safe and clean OO based operating system and API. (I wonder if
>it will be in my lifetime.)

Not if the DOJ has its way.


---
Ice Z - Straight Outta Redmond

[Tim Romano]

So, now that I know what you meant in your correction to my question, and
now that you know what I meant to ask with my question, what IS your take on
the future of programs that rely upon the Win32 API? At present, my
software licensing agreements warrant that my programs are "Windows 9.x
compatible". I'm concerned that Microsoft might introduce changes to the
Win 9.x platform, to better support .NET, that could destabilize VB6/ActiveX
and invalidate that contractual promise. I would be relieved to learn that
my concerns in this regard are unfounded. By others I have been told that
..NET and VB6/ActiveX will peacefully coexist on Windows 9.x -- is that your
understanding as well?

Tim Romano


"Jonathan Allen" <greywolfcs@bigfoot.com> wrote in message
news:3a95f72e@news.devx.com...
> > Are you making a distinction between interpreted and compiled languages
> > here?
>
> No, the distinction is wither memory is managed by the runtime or the
> programmer.
>
> Wither the language is compiled or not rarely factors into programming.
> Especially when you consider that many languages off both modes.
>
> --
> Jonathan Allen
>
>
> "Tim Romano" <tim_romano@yahoo.com> wrote in message
> news:3a95d93e@news.devx.com...
> > Jonathan,
> > Are you making a distinction between interpreted and compiled languages
> > here? If so, then that is not what I meant by "managed" versus
> "unmanaged".
> > With .NET, the operating system -- or the "framework" controlled by the
> > operating system vendor-- becomes the interpreter. What would happen to
> > third party legacy programs that depend on Win32 if that API were to
> become
> > destabilized? I expect it to become destabilized because, as I said in
my
> > reply to Mike Mitchell, a stable Win32 "frozen in time" would usher in a
> > golden age of Win32 legacy programming, which is not what Microsoft
wants,
> > is it?
> >
> > Or did I misunderstand you with respect to Delphi? Is there going to be
a
> > .NET version from INprise?
> > Tim Romano
> >
> > "Jonathan Allen" <greywolfcs@bigfoot.com> wrote in message
> > news:3a95a1d9@news.devx.com...
> > > X86, C and C++ are unmanaged languages.
> > >
> > > LISP, Basic, Pascal, VB, Java (and probably Delphi) are managed
> languages.
> > >
> > > Keeping that in mind, please rephrase your question.
> > >
> > > --
> > > Jonathan Allen
> > >
> > >
>
>
>

[William Cleveland]

Zane Thomas wrote:
>
> On Thu, 22 Feb 2001 21:16:36 -0800, "Jonathan Allen"
> <greywolfcs@bigfoot.com> wrote:
>
> >Some day we
> >will have a safe and clean OO based operating system and API. (I wonder if
> >it will be in my lifetime.)
>
> Not if the DOJ has its way.
>
It doesn't have to come from Microsoft.

Bill

[Sjoerd Verweij]

> my concerns in this regard are unfounded. By others I have been told that
> .NET and VB6/ActiveX will peacefully coexist on Windows 9.x -- is that
your
> understanding as well?

I'd say you'd be safe in that assumption for at least five years or so.
Microsoft is not suicidal.

[Zane Thomas]

On Fri, 23 Feb 2001 09:15:06 -0500, William Cleveland
<WCleveland@Mediaone.Net> wrote:

>It doesn't have to come from Microsoft.

They'll do it if the government stays out of things.


---
Ice Z - Straight Outta Redmond

[Jonathan Allen]

> So, now that I know what you meant in your correction to my question, and
> now that you know what I meant to ask with my question, what IS your take
on
> the future of programs that rely upon the Win32 API?

See my commits on the other branch of this thread.

--
Jonathan Allen