Visual Studio .NET, the Tipping Point

Whether you’re ready or not, Web services is set to make a lot of
noise in the developer community, and Microsoft Corp. intends to lead the band with the general release of Visual Studio .NET — the first major piece of its .NET Web services strategy — at VSLive! on Wednesday.

Visual Studio .NET is intended as a single, unified development
environment for the creation of XML Web services. It automatically
creates the necessary XML and SOAP interface needed to turn an
application into an XML Web service. Visual Studio .NET features
Visual Basic .NET, which now includes new object-oriented programming
features; Visual C++; and C#, a hybrid of C and C++ intended
to compete with Sun’s Java language. C# is an object-oriented
language and boasts type-safety, garbage collection, simplified type
declarations, versioning and scalability support, and other features
that make developing solutions faster and easier, especially
for COM+ and Web services.

Its core is the .NET Framework, which consists of the Framework
classes, ASP.NET and the common language runtime.

The .NET Framework Class Library, in turn, lies at the core of the
.NET Framework. It supplies the syntax, code examples and related
information for each class contained in the .NET Framework
namespaces. In other words, it is a comprehensive collection of
objects that form the starting point for the creation of any .NET
application. Database access, XML manipulation, standardized user
interface features and other functionality is intended to handle all
of the “plumbing” and technical details, allowing developers to
concentrate on actually developing business applications.

Meanwhile, the common language runtime allows developers to create
applications using any modern programming language. Companies are
already working on modules that allow developers to write .NET
applications using COBOL, Ada, Haskal, SmallTalk, Java, Perl, Python
and others.

Developer Perspective
Excitement in the Web services space is building as the release of
Visual Studio .Net approaches because the environment will allow
the estimated 6 million to 8 million developers now programming in
Visual Basic — arguably the largest single developer
community — to leverage their skills to create Web services applications.

“It’s the largest community of developers on a single product,” said
Mike Clark, senior analyst at Lucin and principal designer for
Web services brokerage Salcentral. “If they can get that going, the
whole idea of Web services and technology will just take off.”

Clark, who has been using .NET for the past 8 months, said Visual
Studio .NET should not be a difficult transition for VB

“At this moment in time, everything that we’ve known and loved from
Visual Basic has been reproduced,” he said, adding that those
things that have changed have resulted in clear benefits. In fact,
Clark’s major concern is that Visual Studio .NET will make it so
easy to deploy Web services that quality control will become an issue.

He added, “We’ve cut 1,000 lines of code in the past month and we’ve
shoved it into ASP.NET and we got it working in 24 hours.”

Is Microsoft on Easy Street? Certainly not. For the details, see Page 2.

That’s not to say Microsoft doesn’t face challenges in its .NET strategy.

“Visual Studio .NET is a pretty slick tool, and it’s a tool that I
think is really going to open [Web services] to the masses,” said
Gary Hein, analyst with The Burton Group.

However, Hein noted that Microsoft must address security for .NET to
succeed. The company has recognized this, as the much
publicized Bill Gates memo which called on Microsoft employees to
make security their number one concern attests.

“I think Microsoft has really realized that security is the lynchpin
of the success of this strategy,” Hein said. “Security is the
number one concern and this is an area where they’ve been pretty weak.”

Security is so essential when it comes to the .NET strategy (or any
Web service for that matter) because so much sensitive
information would exist online under the Web services model. In a
July 2001 report called “Deciphering Microsoft .NET,” Hein said,
“Businesses, consumers, and governments must trust Microsoft with
private information; yet, given Microsoft’s history, this may
prove difficult. Security, privacy and stability are critical, but
these are not perceived Microsoft strengths. In essence,
Microsoft must earn the public’s trust in these areas for .NET to succeed.”

But that will take work. In fact, as of January, the “Security
Considerations” section of the SOAP specification (SOAP, or Simple
Object Access Protocol, is one of the core technologies in Web
services, allowing for exchanges of messages and data between
applications by encoding and wrapping the data in XML) was empty.
That’s not just a problem for Microsoft, it’s a problem for IBM,
Sun, and other proponents of Web services. There are methods around
it; SOAP over HTTP could utilize HTTP basic authentication or
SSL, but neither method enforces access controls or provides for
non-repudiation of the SOAP message content, strong authentication,
or full data integrity.

And beyond the security issues, Microsoft will face competition from
multiple directions, primarily from companies like IBM , BEA Systems , Hewlett-Packard and Oracle , which have rallied
behind Sun Microsystems’ Java 2 Enterprise
Edition (J2EE) platform. The J2EE platform consists of a set of
standards for developing enterprise applications using Sun’s Java.
Sun’s recently released Web Services Pack adds key technologies
for building Web services using the platform.

In this area, Microsoft has a bit more of an edge, because of its
strength in all three of the pillars upholding the vision of Web
services. Those three pillars are:

  • The framework (and the developers to go with it)
  • The platform
  • Services.

Not only does Microsoft have the framework for building services in
Visual Studio .NET, but in four to six months it will release
.NET Enterprise Server, a platform for those services. It also
already has bits of the services pillar — like aspects of .NET My
Services and Passport — on the table with more coming down the pike.
Individually, Microsoft may not have the best offerings in any
of these areas, but together they make a compelling end-to-end
package that is hard to beat, Hein said.

“In terms of the main three pillars of .NET, in any one of those
areas I think Microsoft is second or third place, but it’s the
whole complete strategy,” Hein said. “There are already leaders in
each of those respective areas, but Microsoft’s sweet spot is
that it combined all those areas into a unified strategy.”

Web services, whether built on .NET, J2EE, or some other framework,
have a considerable value proposition for businesses. In theory,
Web services offer developers the ability to create modular
applications or components that can be delivered over the Web and can
reused in multiple applications or even resold to multiple users. Web
services also promise applications independent of hardware
platforms, operating systems, development environments or
applications services, potentially making them an ideal bridge between
dissimilar systems.

The application server is key to the model.

“A tax calculator (delivered as a Web service) must execute
application logic and may require access to multiple databases or
repositories,” Hein wrote in a January 2002 report, “Web Services
Basics: Separating Hype and Reality.” “Application execution and
connection to external systems is the realm of the application server
platform. Exposing these applications to the world at large
through standards formats and interfaces is the realm of the Web
services framework. So, while it is possible to place Web services
interfaces on any software, most Web services software
implementations are extensions of underlying application server
Therefore, for most organizations, any Web services framework
discussion will usually involve application server strategies as

.NET Enterprise Server is essentially Windows 2000 Server, but it
also embeds an application server — much like Sun’s iPlanet

The strategies of both Microsoft and Sun when it comes to their
application server platforms will likely focus on each of their
relative strengths: Microsoft dominates desktop and office
applications while J2EE has the greatest traction in the enterprise
Internet-facing software solutions.

But, as it has done in the past, Microsoft may well be able to parley
its strength on the desktop into a powerful advantage. The
company is already offering a number of B2C-type tools through .NET,
whereas competitors are focusing on the infrastructure for
connecting businesses through Web services. Microsoft is offering B2B
capabilities as well, giving it a more expansive solution.

“I do think it is a potential edge,” Hein said. “They have the
developer tools and platforms and they also have the B2C. They have
the solution end-to-end. Not only will it address developers and the
enterprise, but it also will bring consumers into the mix.”

To catch up with Microsoft then, Hein said competitors would have to
form a partnership or alliance that brings all three aspects of
Web services to the table.

“It’s going to take a union or partnership,” he said. “We have yet to
see anything like that really emerge.”

So, who are the competitors? See Page 3.

Potential Dangers
Such a union may spring from the Liberty Alliance, an organization
that presents a real danger for Microsoft’s plans to dominate the
Web services market. The Liberty Alliance is a large organization,
which currently sports 38 Charter Members like Sun Microsystems,
AOL Time Warner , Sony , RealNetworks
, Hewlett-Packard, etc. The alliance formed
in response to Microsoft’s push to turn its Passport authentication
service into a federated service which other companies could

As an alternative, the Liberty Alliance has proposed a universal,
open standard for single sign-on with decentralized authentication
and open authorization from multiple providers. That poses a danger
to Microsoft’s Passport, which the company sees as the core of
its billing system for Web services.

“Liberty Alliance presents a danger,” Hein said. “Microsoft has lain
out this business model that includes developers and consumers
but what wasn’t clear at that time was the payment mechanism. As a
result of the Liberty Alliance, a lot of the players that
Microsoft could have or should have partnered with can almost be seen
as being in the enemy’s camp.”

While Microsoft could conceivably alter that by joining the alliance,
Hein said he expects the company to bide its time.

“They’re way ahead of Liberty Alliance,” he said. “They have
products, they have customers, and they have market share. I don’t
anticipate Microsoft to make a move anytime soon.”

He added, “Although the Liberty Alliance has assembled a pretty
interesting mix of businesses, Microsoft has the service in place
today with 150 million identities…You still need adoption and
Microsoft has the ability to bundle this service with their
desktops. As we see in other areas, until Microsoft supports the
standard, you wind up with two standards.”

And while Microsoft dukes it out with the Liberty Alliance over
authentication, .NET will duke it out with other Web services
frameworks built on J2EE. While the vision of Web services promises a
“write once, deploy anywhere” capability, that’s not likely to
be a full reality — at least on the framework layer — anytime soon.

“It’s possible,” said Fred Hoch, director, eBusiness Division,
Software & Information Industry Association (SIIA). “That’s like
asking if America and Russia are going to war; it’s possible. .NET
and SunONE are major players. I think most of the services will
be developed in a Java or .NET framework. The promise of Web services
is that it shouldn’t matter what platform you use because
you’re basing them all on open standards. That’s the promise. On
paper, it’s unclear whether anyone will adhere to that. I mean, how
many different kinds of UNIX are there?”

Certain third-parties are offering tools that could overcome that
sort of “walled garden.” One is Halcyon Software, which has
developed iNET, a Java-based implementation of the .NET framework
which would allow developers to create applications with Visual
Studio .NET and then deploy it through J2EE architecture. To do this,
iNET utilizes an IL2JAVA converter, which generates Java class
files or Java source code from the .NET Intermediate Language (IL).

However, such solutions are not likely to get support from the major
players, like Sun and Microsoft.

Still, Hoch has no doubt that Web services are more than just hype.

“[Web services] are easy to do in a simple format,” he said. “It’s
easy to create a simple Web service that connects two systems,
but you also have to think of the larger ecosystem you are creating.
It’ll take time to make it really efficient so they become
pervasive, but looking at it another way, I’ve never seen such a
convergence around one idea. I mean, how often do you get IBM,
Microsoft, HP, Sun and Oracle to agree on one thing?”

In any case, 2002 is shaping up to be a decisive year for Web services.

“I do think this year is going to be very important because you’re
going to see the developers have access to a lot of these Web
services tools and frameworks,” Hein said. “As this year progresses,
it’s really about the developers.” Next year, Hein said, expect Web services applications to really get rolling.

— Prepared by Thor Olavsrud with contributions from Steve Kapsinow and Robert Liu.

News Around the Web