Why no MS products built with .NET?

[Rob R. Ainscough]

I was reading yet another book on .NET - VB 2005 Professional (wrox) and
read the statement;

"Microsoft has staked their future on .NET and publicly stated that
henceforth almost all their research and development will be done on this
platform. It is expected that, eventually, almost all Microsoft products
will be ported to the .NET platform."

So I thought about this statement for a minute -- then I realized that I
have ALL Microsoft's products and not a single one uses .NET framework. I
think you can figure out the obvious question -- why not? Going on 6 years
of .NET now, where are these Microsoft .NET migrated products? MS Office
certainly isn't -- it doesn't require me to install .NET framework when I
installed it (on a non-.NET machine), nor many other mainstream Microsoft
products.

Is this yet another case of "do as I say, not as I do"? Or was porting to
..NET platform a lot more difficult than they thought? Or is it the more
ominious, .NET anything just isn't really ready for prime time (certainly
not retail anyway).

Rob.

 

[Arnaud Debaene]

I was reading yet another book on .NET - VB 2005 Professional (wrox)
and read the statement;


So I thought about this statement for a minute -- then I realized
that I have ALL Microsoft's products

Really all? I bet not...

and not a single one uses .NET
framework.

You do not have got the good ones
Many server side tools working with and above SharePoint, for example, are
written at least in part in ASP.NET.
Visual Studio itself is written in part in .NET, as is Team Services.

I think you can figure out the obvious question -- why
not? Going on 6 years of .NET now, where are these Microsoft .NET
migrated products? MS Office certainly isn't -- it doesn't require
me to install .NET framework when I installed it (on a non-.NET
machine), nor many other mainstream Microsoft products.

Geee... Office is a multi-million lines source base that was written, tested
and qualified before .NET was invented. What would be the point to spend
many years-man (probablty even centuries-man) to rewrite it in .NET? There
will be strictly no benefit for the end-user. You must think about new
products of course, not historical ones.

Arnaud
MVP - VC

 

[Lebesgue]

I have ALL Microsoft's products

Do you really have all the 300 software products from Microsoft, or do you
think that ALL microsoft products include Office, Windows and VS?

 

[Rob R. Ainscough]

Yes I do have them all. But true, I haven't tested EVERY SINGLE one yet,
but certainly none of their main stream producst are written or make use of
managed or unmanaged .NET framework. If you know any of their main stream
products that are, please let me know. As mentioned it appears a few low
profile tools here and there, but certainly nothing I'd consider "prime
time".

I was just quoting the book, which was relaying what Microsoft sold to the
community. I realize it is perhaps a sales pitch for the book so as to make
one feel the investment in .NET is worth it -- but regardless, you're
diverting from the issue:

Where are the "prime time" .NET products? Microsoft expect/hope everyone
migrates so why aren't Microsoft? And I would NOT assume that Microsoft are
the only company with millions of lines of code for a retail product --
believe it or not, other companies exist that have products with millions of
lines of code also -- and if it takes 6 years to rewrite them from scratch
or migrate them to .NET then I'd suggest that is NOT good news.

But the question remains, why no prime time products based on .NET from the
creators of .NET?

I mean we're 6 years into .NET now, so I'd expect something -- even MS Money
2006 is not .NET based (a relatively simple product that retails for $30).

 

[Gabriel Magaña]

Where are the "prime time" .NET products?

Well, for what it's worth, SQL server 2005 does have the .NET built in...
Granted, it's for developer support, but I think at least MS thinks the CLR
is stable enough to be included in such an important product...

I am not too sure though, that I would want to have MS Excel written in .NET
just yet... There really is nothign faster than C/C++, and at this point
the existing code base is much more stable than anything MS could come up
with in a 1.0 version based on .NET...

I know what you mean, though, you want MS to eat its own dog food and start
to write actual released products on .NET... It would be cool to see one in
action...

 

[John Bailo]

Is this yet another case of "do as I say, not as I do"? Or was porting to
.NET platform a lot more difficult than they thought? Or is it the more
ominious, .NET anything just isn't really ready for prime time (certainly
not retail anyway).

The real issue is: Why bother porting Office to .NET

The reality is that people have long stopped using traditional
standalone client applications and now now moved to web applications,
and beyond those, to smart client applications.

 

[Gabriel Magaña]

asp.net then?

 

[John Bailo]

Microsoft makes money by selling:

Desktop OS
Server OS
Office Apps
SQL Server

If it can provide a killer language and easy to use development system
such as .NET for others to write apps that makes them buy and use
Windows, then they've achieved their goal.

In fact, I wish Microsoft would stop running rampant over 3rd party
developers and being so rapacious as to develop each and every class of
application themselves and leave a few scraps for their loyal developers.

 

[Rob R. Ainscough]

John,

I think you live a different reality -- most of my end users have little or
no interest in the "Browser" -- in fact most do NOT like it at all. If by
smart client applications, you mean ClickOnce or AJAX then I agree -- but
these are essentially stand alone applications with built in AutoUpdates
just accessed via a link.

Also, some end users do NOT want to be updated. But as long as we become
less and less dependant on the Browser the better -- maybe, just maybe
people will not fear the internet as much as they currently do and just
maybe we can start to reach those other 80% of the population that haven't
touched the internet.

What I see on the internet is toilet wear -- extremely un-user friendly,
over crowded with graphical crap (that is really of no aid to the user other
than slowing down their experience and getting in the way), full of trackers
and popups and flash ads and numerous other tricks forced on the
unsuspecting user.

But I can assure you, if we continue to think the "Browser" is our future,
we will be faced with limited growth. If we want to reach those 80% other
potential users, the current browser needs to die a quick death so we can
provide a far more secure and friendly user experience. Expecting a user to
know any of the following:

Security settings
Allowing popups
Cookie blocking or acceptance
SSL versions
TLS
Site certificates

....and the host of other browser settings that affect the user's experience

Is beyond stupidity -- it is ensuring that we (software industry) NEVER
reaches out to those 80%

Rob

 

[John Bailo]

John,

I think you live a different reality -- most of my end users have little or
no interest in the "Browser" -- in fact most do NOT like it at all. If by
smart client applications, you mean ClickOnce or AJAX then I agree -- but
these are essentially stand alone applications with built in AutoUpdates
just accessed via a link.

Boy, are you behind the times. If you want to use a term like "Smart
Client" read up on it first.

I will not read the rest of your post, because the 1st paragraph
bespeaks such great ignorance. Start here, read, then open mouth:

http://msdn.microsoft.com/smartclient/

 

[amaca]

According to the link you give, Microsoft Word *is* a smart client, so
'smart client' seems to mean 'desktop application that can also connect to
the net'.

Andrew

 

[Kevin Spencer]

Where are the "prime time" .NET products? Microsoft expect/hope everyone

migrates so why aren't Microsoft? And I would NOT assume that Microsoft
are the only company with millions of lines of code for a retail
product -- believe it or not, other companies exist that have products
with millions of lines of code also -- and if it takes 6 years to rewrite
them from scratch or migrate them to .NET then I'd suggest that is NOT
good news.

Microsoft has a huge problem in migrating their software to the .Net
platform, and that is backwards-compatibility. If they were simply creating
new products, it wouldn't be a problem. But having new .Net products that
are backwards-compatible with non-.Net previous versions is a nightmare. In
fact, Microsoft is working on this, and it seems that the solution is
evolution. Rather than making a sudden change in all of their software, they
are gradually migrating it.

Microsoft doesn't "expect" anyone to migrate to .Net. It's a matter of
personal choice. If you feel that the .Net platform is going to be
profitable to you, develop with it. The reasons for using it have nothing to
do with the products that Microsoft sells. It has to do with the features of
..Net, and what those features provide to the developer. .Net is a
high-productivity development platform. This means that it costs less to
build .Net applications, because it takes less man-hours. It uses managed
memory, which means a much smaller chance of having memory leaks. It is
fully object-oriented, which means that one can take advantage of all of the
properties of object-oriented programming with it. And Microsoft uses it
quite a bit. While the core components of their commercial software may not
be .Net yet, you should notice that the perifery of most of it is already
making extensive use of the .Net platform.

Still, bottom line is, who cares what some article says? Life is a game.
Play to win. Use whatever tools you have at your disposal to do so. If that
includes .Net, so be it. If not, so be it. In the end, it is not Microsoft,
your parents, the government, or anyone else that will determine your fate;
it is you. Accept responsibility for yourself, and take charge of your life.
Everything else is a distraction.

--
HTH,

Kevin Spencer
Microsoft MVP
..Net Developer
A brute awe as you,
a Metallic hag entity, eat us.

 

[Rob R. Ainscough]

John,

Did you actually read the link -- there is some learning to be had, but
you're the one that needs to be learned.

What do you think ClickOnce is??

Get back to me when you understand what I said.

Rob.

 

[Rob R. Ainscough]

It is ClickOnce -- you put a link on a web site, user clicks the link, the
ClickOnce deployment install necessary components (i.e. .NET framework) and
install the application on the client's PC (of course one has to configure
the access/security requirements in the ClickOnce deployment). From that
point on the application can communicate with web services or directly with
SQL servers or whatever one wishes.

I think John is a little clueless or hasn't actually implement ClickOnce to
understand.

MS Word isn't technically a "smart client" because it can't be installed
from a web link and doesn't use that web reference as a source to
automatically update or determine install components. But MS Word achieves
pretty similiar functionality as a "smart client" -- it runs stand alone on
the client PC, it can automatically determine if it needs updating, it
provides a much better user experience.

The end result is that a web site is used to provide the initial link and
that is all the web site does (thank goodness).

 

[Rob R. Ainscough]

Kevin,

By your own arguements, Microsoft SHOULD be using .NET to develop their new
products if the turn around is truely faster, but they don't -- MFC 8.0 VC++
seems to still be their prefered choice for old and new products. Like I
said, MS Money 2006 would be a classic candidate for .NET, but it's still
MFC 8.0.

My guess is that Microsoft operate something like this:

Develop application in the long standing traditional sense VC++
Any new or revised functionality that was built as a requirement for the new
application is potentially moved into a new MFC
This new functionality is then evaluated to determine if the new
functionality can be wrapped into next .NET framework version

This would explain why we don't see any .NET "prime time" applications from
Microsoft.

Rob.

 

[Guest]

Rob,
Microsoft Defender, msn desktop search, and every webpage in the microsoft
domain ending with aspx uses dot net.

DWS

 

[Chris Dunaway]

You might find this interesting. Go down to the part about "Express"
Applications.

This is not intended to answer the question you originally posted I
just thought that in the discussion of smart clients that you would
find this interesting.

 

[Jason Hunt]

Microsoft CRM is built on .Net

 

[Jon Skeet [C# MVP]]

So I thought about this statement for a minute -- then I realized that I
have ALL Microsoft's products and not a single one uses .NET
framework.

Have you used the SQL Server 2005 equivalent of SQL Enterprise Manager?
That's written in .NET, IIRC.

Parts (not sure how much - quite a lot though, I'd expect) of VS 2005
use .NET.

Microsoft CRM uses .NET.

I'm sure there are more - but three counter-examples to a blanket "not
one" are plenty to start with.

 

[John Bailo]

Let me try to educate him and the group once more.

Smart client means having the ability to run complex code, and to have a
complex UI, beyond the browser.

For example, I am working on a Warehouse locator application.

The back end database is Db2 (iSeries).

The middleware is c# web services. These encapsulate the standard
insert, update and retrieval features.

They only return type is XmlDocument.

The client is a c# windows form application.

All database updates and retrievals comes from the web methods. That
lets me run my windows form /anywhere/ that has port 80 open. Yet,
it's a complex windows form application using a treeview, multiple forms
an serialized xml for local storage. The dropdown boxes part
XmlDocument types which gives me high flexibility especially because I
bind my dropdowns to complex objects -- not just simple text values.

My next phase implementation is a peerRemoting feature that lets
individual clients update each other with refresh notifications.

If I wanted to, I could build in Office API features to my smart client.