Why no serious MS Application in .NET yet ??

  • Thread starter Thread starter Herr Lucifer
  • Start date Start date
Perhaps you're right... but it just might behoove them to do a "gigantic"
rewrite.... MS is currently having their cake and eating it too and have
taken a serious beating in the reputation department. They can't keep
extolling the virtues of .NET while continuing development on the inherently
less-stable, less secure, Win32 / COM platform. It just doesn't make sense to
me.

Sean Hederman said:
Alvin Bruney said:
phew, good blog. pure poison. bitter stuff.
Click to expand...

Poison? Bitter? Are you talking about
http://blogs.msdn.com/danielfe/archive/2004/11/02/251254.aspx? Cause I
didn't see any nastiness there. Merely a point by point refutation of the
assertion that Microsoft isn't writing stuff in .NET.
now, i am totally confused. i'm just not buying these .net arguments at
all. even the "there's no business case for re-writing components in
.NET".
Well wouldn't rewriting some core products improve
visibility/marketability/consensus on .net?
Click to expand...

60% of MS revenue comes from Office, which at a total revenue of $38bn,
comes to $22.8bn. Have you ever heard the term "If it ain't broke don't fix
it". Especially when you're gambling with 22 billion dollars a year.

How much do you think just rewriting the Office Suite would cost? I figure
at an absolute minimum you're looking at about a thousand man years per
product in the suite, so around a half billion dollars.
hmmm, strange. real strange.
Click to expand...

Not really. I don't think any company on earth would take such a massive
gamble. I personally think it's strange to expect MS to throw away billions
of dollars of existing investment and revenue, and spend billions more in
order to arrive where they already are. And if you think the Office division
should take orders from the .NET area, think again. Not going to happen.

Office will gradually migrate pieces of itself to .NET (as is already
happenning). To expect it to happen in some huge gigantic billion dollar
extravaganze is unrealistic however.
--
Regards
Alvin Bruney
[Shameless Author Plug]
The Microsoft Office Web Components Black Book with .NET
available at www.lulu.com/owc
--------------------------------------------------


Junfeng Zhang said:
http://blogs.msdn.com/danielfe/archive/2004/11/02/251254.aspx

If you are talking about re-writing office in managed code, I guess that
will be a few decades away.

--
Junfeng Zhang
http://blogs.msdn.com/junfeng

This posting is provided "AS IS" with no warranties, and confers no
rights.

As the founder of .NET framework, Microsoft claims that it invention
will be
the next best platform for programming in a near future. Now it is 2005,
.NET is 5 years old, and can talk and walk for himself with some help of
his
mum.
However, we see the same native office applications are coming out
again,
and many other tools in SP2 of XP which could be in managed code....but
are
not. So, as the inventor of .NET , why doesn't Microsoft itself use
"DOTNET"
in its applications? Is there any concern over the baby's runnung
performance inside Microsoft itself, or they gonna teach the baby how to
run
like a C kinda guy in future, so that they'll be able to use it for
themselves?
Click to expand...
Click to expand...
Click to expand...
 
many of the other adjunct components completely. Unfortunately, MS has always
been hellbent on 110% backwards compatibility... which is why Windows suffers
from so many problems today.
Click to expand...

.... and enjoys a market share somewhat higher than Apple... ;-)

Joe.
 
CMM said:
Unfortunately, MS has always been hellbent on 110% backwards
compatibility... which is why Windows suffers from so many
problems today.
Click to expand...

Perhaps. But their concern about backwards compatibility and the same
concern that is shared by the dominant chip makers - Intel and AMD - enusre
that the platform and its slew of applications run on more desktops than
anything else on the planet.

Regards,
Will
 
I agree... and I'm not saying it's a totally bad thing. But, I think at some
point MS has to show that they're true innovators and not just evolutionary.
..NET is probably the most ambitious and courageous thing to come from Redmond
since Windows 95. Now we have to wait to see if that courage extends to the
rest of the Windows world.
 
CMM said:
Perhaps you're right... but it just might behoove them to do a "gigantic"
rewrite.... MS is currently having their cake and eating it too and have
taken a serious beating in the reputation department. They can't keep
extolling the virtues of .NET while continuing development on the
inherently
less-stable, less secure, Win32 / COM platform. It just doesn't make sense
to
me.
Click to expand...

I agree in principle. I think you'll find that completely new products will
tend, more and more, to be written in .NET. As for existing products, I
think you'll find that various of their components will slowly be migrated
to .NET, and they will certainly be enhanced for better .NET support. The
fact of the matter is that applications like Office cannot remain static.
Reskilling all the developers to .NET and identifying areas where .NET can
assist will take time.

In the Office suite, the one section really screaming out for .NET is the
VBA/ActiveX support. It's so insecure right now that it's insane. I wouldn't
be too surprised if in the very near future MS notify their clients that
non-.NET scripting will no longer be supported.

I've heard horror stories about what the Word and IE code looks like
(particularly in the rendering engines). There's two approaches they can
take as far as this is concerned: rewrite and hope to hell it doesn't mess
something up, or carefully carry on adding to the tottering framework for as
long as it'll take it.

In any case, if I was a betting man, I'd wager that Microsoft won't start a
..NET rewite of the Office products until at least after Longhorn ships.
Sean Hederman said:
Alvin Bruney said:
phew, good blog. pure poison. bitter stuff.
Click to expand...

Poison? Bitter? Are you talking about
http://blogs.msdn.com/danielfe/archive/2004/11/02/251254.aspx? Cause I
didn't see any nastiness there. Merely a point by point refutation of the
assertion that Microsoft isn't writing stuff in .NET.
now, i am totally confused. i'm just not buying these .net arguments at
all. even the "there's no business case for re-writing components in
.NET".
Well wouldn't rewriting some core products improve
visibility/marketability/consensus on .net?
Click to expand...

60% of MS revenue comes from Office, which at a total revenue of $38bn,
comes to $22.8bn. Have you ever heard the term "If it ain't broke don't
fix
it". Especially when you're gambling with 22 billion dollars a year.

How much do you think just rewriting the Office Suite would cost? I
figure
at an absolute minimum you're looking at about a thousand man years per
product in the suite, so around a half billion dollars.
hmmm, strange. real strange.
Click to expand...

Not really. I don't think any company on earth would take such a massive
gamble. I personally think it's strange to expect MS to throw away
billions
of dollars of existing investment and revenue, and spend billions more in
order to arrive where they already are. And if you think the Office
division
should take orders from the .NET area, think again. Not going to happen.

Office will gradually migrate pieces of itself to .NET (as is already
happenning). To expect it to happen in some huge gigantic billion dollar
extravaganze is unrealistic however.
--
Regards
Alvin Bruney
[Shameless Author Plug]
The Microsoft Office Web Components Black Book with .NET
available at www.lulu.com/owc
--------------------------------------------------


http://blogs.msdn.com/danielfe/archive/2004/11/02/251254.aspx

If you are talking about re-writing office in managed code, I guess
that
will be a few decades away.

--
Junfeng Zhang
http://blogs.msdn.com/junfeng

This posting is provided "AS IS" with no warranties, and confers no
rights.

As the founder of .NET framework, Microsoft claims that it invention
will be
the next best platform for programming in a near future. Now it is
2005,
.NET is 5 years old, and can talk and walk for himself with some help
of
his
mum.
However, we see the same native office applications are coming out
again,
and many other tools in SP2 of XP which could be in managed
code....but
are
not. So, as the inventor of .NET , why doesn't Microsoft itself use
"DOTNET"
in its applications? Is there any concern over the baby's runnung
performance inside Microsoft itself, or they gonna teach the baby how
to
run
like a C kinda guy in future, so that they'll be able to use it for
themselves?
Click to expand...
Click to expand...
Click to expand...
 
I agree... but would add that once the Framework matures to the point of
being able to do everything the direct API's can do (even if it's just in the
form of "managed" wrappers around the API's), MS would have no reason to not
completely rewrite their apps using the Framework... even the big apps and
even the heavy duty code like rendering engines. I mean, at some point MS had
to "rewrite" Word for DOS right? Why can't they do it again?

Sean Hederman said:
CMM said:
Perhaps you're right... but it just might behoove them to do a "gigantic"
rewrite.... MS is currently having their cake and eating it too and have
taken a serious beating in the reputation department. They can't keep
extolling the virtues of .NET while continuing development on the
inherently
less-stable, less secure, Win32 / COM platform. It just doesn't make sense
to
me.
Click to expand...

I agree in principle. I think you'll find that completely new products will
tend, more and more, to be written in .NET. As for existing products, I
think you'll find that various of their components will slowly be migrated
to .NET, and they will certainly be enhanced for better .NET support. The
fact of the matter is that applications like Office cannot remain static.
Reskilling all the developers to .NET and identifying areas where .NET can
assist will take time.

In the Office suite, the one section really screaming out for .NET is the
VBA/ActiveX support. It's so insecure right now that it's insane. I wouldn't
be too surprised if in the very near future MS notify their clients that
non-.NET scripting will no longer be supported.

I've heard horror stories about what the Word and IE code looks like
(particularly in the rendering engines). There's two approaches they can
take as far as this is concerned: rewrite and hope to hell it doesn't mess
something up, or carefully carry on adding to the tottering framework for as
long as it'll take it.

In any case, if I was a betting man, I'd wager that Microsoft won't start a
..NET rewite of the Office products until at least after Longhorn ships.
Sean Hederman said:
"Alvin Bruney [Microsoft MVP]" <www.lulu.com/owc> wrote in message
phew, good blog. pure poison. bitter stuff.

Poison? Bitter? Are you talking about
http://blogs.msdn.com/danielfe/archive/2004/11/02/251254.aspx? Cause I
didn't see any nastiness there. Merely a point by point refutation of the
assertion that Microsoft isn't writing stuff in .NET.

now, i am totally confused. i'm just not buying these .net arguments at
all. even the "there's no business case for re-writing components in
.NET".
Well wouldn't rewriting some core products improve
visibility/marketability/consensus on .net?

60% of MS revenue comes from Office, which at a total revenue of $38bn,
comes to $22.8bn. Have you ever heard the term "If it ain't broke don't
fix
it". Especially when you're gambling with 22 billion dollars a year.

How much do you think just rewriting the Office Suite would cost? I
figure
at an absolute minimum you're looking at about a thousand man years per
product in the suite, so around a half billion dollars.

hmmm, strange. real strange.

Not really. I don't think any company on earth would take such a massive
gamble. I personally think it's strange to expect MS to throw away
billions
of dollars of existing investment and revenue, and spend billions more in
order to arrive where they already are. And if you think the Office
division
should take orders from the .NET area, think again. Not going to happen.

Office will gradually migrate pieces of itself to .NET (as is already
happenning). To expect it to happen in some huge gigantic billion dollar
extravaganze is unrealistic however.

--
Regards
Alvin Bruney
[Shameless Author Plug]
The Microsoft Office Web Components Black Book with .NET
available at www.lulu.com/owc
--------------------------------------------------


http://blogs.msdn.com/danielfe/archive/2004/11/02/251254.aspx

If you are talking about re-writing office in managed code, I guess
that
will be a few decades away.

--
Junfeng Zhang
http://blogs.msdn.com/junfeng

This posting is provided "AS IS" with no warranties, and confers no
rights.

As the founder of .NET framework, Microsoft claims that it invention
will be
the next best platform for programming in a near future. Now it is
2005,
.NET is 5 years old, and can talk and walk for himself with some help
of
his
mum.
However, we see the same native office applications are coming out
again,
and many other tools in SP2 of XP which could be in managed
code....but
are
not. So, as the inventor of .NET , why doesn't Microsoft itself use
"DOTNET"
in its applications? Is there any concern over the baby's runnung
performance inside Microsoft itself, or they gonna teach the baby how
to
run
like a C kinda guy in future, so that they'll be able to use it for
themselves?
Click to expand...
Click to expand...
Click to expand...
 
I don't really need a managed Word today. What I need is a complete managed
APIs to get at all features of Word. Don't they have some of that now? As
for scripting, I am sure the first retro will be to play nice with MSH
(which is all managed). But now that I think about it, if all the .Net
wrappers are there, they don't need to do anything special to support MSH.

--
William Stacey, MVP
http://mvp.support.microsoft.com

CMM said:
I agree... but would add that once the Framework matures to the point of
being able to do everything the direct API's can do (even if it's just in the
form of "managed" wrappers around the API's), MS would have no reason to not
completely rewrite their apps using the Framework... even the big apps and
even the heavy duty code like rendering engines. I mean, at some point MS had
to "rewrite" Word for DOS right? Why can't they do it again?

Sean Hederman said:
CMM said:
Perhaps you're right... but it just might behoove them to do a "gigantic"
rewrite.... MS is currently having their cake and eating it too and have
taken a serious beating in the reputation department. They can't keep
extolling the virtues of .NET while continuing development on the
inherently
less-stable, less secure, Win32 / COM platform. It just doesn't make sense
to
me.
Click to expand...

I agree in principle. I think you'll find that completely new products will
tend, more and more, to be written in .NET. As for existing products, I
think you'll find that various of their components will slowly be migrated
to .NET, and they will certainly be enhanced for better .NET support. The
fact of the matter is that applications like Office cannot remain static.
Reskilling all the developers to .NET and identifying areas where .NET can
assist will take time.

In the Office suite, the one section really screaming out for .NET is the
VBA/ActiveX support. It's so insecure right now that it's insane. I wouldn't
be too surprised if in the very near future MS notify their clients that
non-.NET scripting will no longer be supported.

I've heard horror stories about what the Word and IE code looks like
(particularly in the rendering engines). There's two approaches they can
take as far as this is concerned: rewrite and hope to hell it doesn't mess
something up, or carefully carry on adding to the tottering framework for as
long as it'll take it.

In any case, if I was a betting man, I'd wager that Microsoft won't start a
..NET rewite of the Office products until at least after Longhorn ships.
:

"Alvin Bruney [Microsoft MVP]" <www.lulu.com/owc> wrote in message
phew, good blog. pure poison. bitter stuff.

Poison? Bitter? Are you talking about
http://blogs.msdn.com/danielfe/archive/2004/11/02/251254.aspx? Cause I
didn't see any nastiness there. Merely a point by point refutation of the
assertion that Microsoft isn't writing stuff in .NET.

now, i am totally confused. i'm just not buying these .net arguments at
all. even the "there's no business case for re-writing components in
.NET".
Well wouldn't rewriting some core products improve
visibility/marketability/consensus on .net?

60% of MS revenue comes from Office, which at a total revenue of $38bn,
comes to $22.8bn. Have you ever heard the term "If it ain't broke don't
fix
it". Especially when you're gambling with 22 billion dollars a year.

How much do you think just rewriting the Office Suite would cost? I
figure
at an absolute minimum you're looking at about a thousand man years per
product in the suite, so around a half billion dollars.

hmmm, strange. real strange.

Not really. I don't think any company on earth would take such a massive
gamble. I personally think it's strange to expect MS to throw away
billions
of dollars of existing investment and revenue, and spend billions more in
order to arrive where they already are. And if you think the Office
division
should take orders from the .NET area, think again. Not going to happen.

Office will gradually migrate pieces of itself to .NET (as is already
happenning). To expect it to happen in some huge gigantic billion dollar
extravaganze is unrealistic however.

--
Regards
Alvin Bruney
[Shameless Author Plug]
The Microsoft Office Web Components Black Book with .NET
available at www.lulu.com/owc
--------------------------------------------------


http://blogs.msdn.com/danielfe/archive/2004/11/02/251254.aspx

If you are talking about re-writing office in managed code, I guess
that
will be a few decades away.

--
Junfeng Zhang
http://blogs.msdn.com/junfeng

This posting is provided "AS IS" with no warranties, and confers no
rights.

As the founder of .NET framework, Microsoft claims that it invention
will be
the next best platform for programming in a near future. Now it is
2005,
.NET is 5 years old, and can talk and walk for himself with some help
of
his
mum.
However, we see the same native office applications are coming out
again,
and many other tools in SP2 of XP which could be in managed
code....but
are
not. So, as the inventor of .NET , why doesn't Microsoft itself use
"DOTNET"
in its applications? Is there any concern over the baby's runnung
performance inside Microsoft itself, or they gonna teach the baby how
to
run
like a C kinda guy in future, so that they'll be able to use it for
themselves?
Click to expand...
Click to expand...
Click to expand...
 
They'll do it when the pain of not doing it exceeds the pain of doing it.
That's the point at which anything gets done, by anyone ;-)

Seriously, I believe that every code base needs a thorough (and probably
complete) rewrite every 5 to 10 years, even in the absence of a major
platform change -- just to blow out the lint and gain the benefit of
hindsight and experience and bring things more into sync with all the other
technologies that the software has to interface with. Office is approaching
that point anyway. Currently, MSFT's public stance is that it's perfectly
good code and they aren't going to "throw it away" but I'm sure that
sometime in the next few years that various forces will overcome the
inertias at work here and suddenly all you'll hear about is the sleek new
Office v13 or 14 that is all managed code.

In the meantime I don't believe that the lack of a publicly acknowledged
crash Manhattan Project to port Office to .NET says anything at all one way
or the other about whether MSFT "believes" in .NET or is willing to eat its
own dogfood. With Office they are in fact eating their own dogfood today
anyway, even if it's yesterday's dogfood. At about the point that Office is
finally redone in .NET we will no doubt be having a similar debate about why
they haven't rewritten it in Super-Avalon-Plus or whatever is supposed to
save the world of 2010. So I am just going to take it as it comes and not
read too much into the tea leaves. It always takes quite a bit of time for
the marketplace to fully absorb any new technology, and we are always in a
transition state. It's not a perfect world.

--Bob
 
I just thought of this after reading your post:
Before they can (or should) rewrite office, maybe they should practice by
completely making the .NET IDE as an all "managed" application rather than
the COM application that it is today (as much as I love the .NET IDE).
 
CMM said:
I just thought of this after reading your post:
Before they can (or should) rewrite office, maybe they should practice by
completely making the .NET IDE as an all "managed" application rather than
the COM application that it is today (as much as I love the .NET IDE).
Click to expand...

That sounds like a lot of work for not much gain.

The future of .NET is not big apps, it's the SOA.

Perhaps even the IDE should be an SOA...a set of very small components,
like say, a notepad, that can call a /syntax/ web service.

Or a messaging system for sending debug information back and forth to a
little service app with a popup window.

In other words, build a brand new type of system for creating code that
reflects the SOA nature of .net
 
CMM said:
I just thought of this after reading your post:
Before they can (or should) rewrite office, maybe they should practice by
completely making the .NET IDE as an all "managed" application rather than
the COM application that it is today (as much as I love the .NET IDE).
Click to expand...

I would just settle for it being stable and bug-free. Then I wouldn't care
if it were written in COBOL.

--Bob
 
Um yeah right! Like I really want my IDE to be SLOWER. You're kidding right?

Case-in-point, I'm reading this post through MSDN's *online* newsgroups
reader (most likely not very workable in anything but IE)... which is
impressive, but is like 1% of what any desktop newsreader is (OE, Free Agent,
whatever).

The idea of SOA is great but no one really understands it and it will
eventually find its more modest place. XML, UML, DNA, CORBA, and Sun's
"The-Network-Is-The-Computer" all were big "all-encompassing" notions that
have always and will always fall short of its great big promises.
 
yeah im a diffrent Joe, but this raises some good points. I guesse we
willhave to see what takes place after the .NET framework is "baked into" a
M$ OS
 
Case-in-point, I'm reading this post through MSDN's *online* newsgroups
reader (most likely not very workable in anything but IE)... which is
impressive, but is like 1% of what any desktop newsreader is (OE, Free Agent,
whatever).
Click to expand...

.... and it can't even post your comments _after_ the quoted part, like
any capable newsreader ;-)

Joe.
 
No, it's a COM application. That probably reflects that they needed to be a
fair way along in developing it before the framework was functional enough
to support a complex WinForms app. And just like Office, it's a perfectly
functional app and no pressing reason to rewrite it just for religious
purity.

--Bob
 
to support a complex WinForms app. And just like Office, it's a perfectly
functional app and no pressing reason to rewrite it just for religious
purity.
Click to expand...

Actually, I would beg to differ (somewhat). As much as I love the VS.NET IDE
(and I do!), it is a little sluggish at times... and not just at startup.
COM... now matter how fast the machine... always seems to drag everything
down. COM object instantiation and marshaling is just way too expensive
compared to .NET's blazing's object architecture. However, this might be
offset by Windows Forms somewhat sluggish event response architecture. I
don't know.
 
If I had to guess, I'd say it's because Microsoft has way too much legacy
code in their major applications.

Microsoft Office probably has at least 30 million lines of code in it. I
wouldn't be surprised if it's 50 million. Imagine you are the manager in
charge of deciding what to do for the next version of Office. Whatever you
do, you know that you have to be done in three years, at most. Your
decision is, "do we spend three years re-implementing and re-testing all of
Office in managed code, to end up with a version that looks just like what
we have now, or spend three years working in our existing code adding
features that people might actually want to buy?"

If I were that manager, I know which one I'd choose. I'd be sad that I
wasn't working in managed code, but I'd understand the business reality that
I need to give people some reasons to upgrade. But I'd probably also set
three or four developers aside and say "hey, guys, see if you can't build
some sort of conversion tools to convert all that legacy code into C#"
because I'd know that I can't realistically work in unmanaged code forever.

And also, there is some code that just can't be written in managed code.
Windows, for example. After all, the CLR has to run _on_ something. So
until the essential features of managed code are implemented in hardware,
Windows is gonna continue to be written in unmanaged code.
 
Back
Top