The Next Windows
OS/2 Finally Dead
[ Note: this version has been replaced by a newer version ]
Netscape's Andreson loudly (and very, very unwisely) boasted that Netscape would make Windows irrelevant. Microsoft believed him, and crushed Netscape out of existence to make the world safe for Windows (and lost a major anti-trust battle over how they did the crushing).
Now Microsoft finds they must make Windows irrelevant themselves. If they don't, others much stronger than Netscape will be more than happy to do it for them. IBM, Sun, Oracle, Hewlett Packard, Linux, AOL - each is a name to be reckoned with - together they resemble a steamroller - and they have friends, lots of friends (OK, maybe not AOL, but the rest of them do).
Wait! Didn't Microsoft already defeat IBM? Yes they did - riding a wave of new technology and new business methods. Even mighty IBM couldn't stop the winds of change (especially while tied down by a major anti-trust action). Now it's Microsoft defending legacy systems against new technology and new business methods (and tied down by anti-trust action).
Microsoft's long successful effort to freeze the evolution of computing is clearly crumbling - and denial just isn't going to cut it - so we have .NET. .NET embraces the concept of network computing, and extends it with the concepts of world domination and preserving the Windows and Office monopolies.
So, we can't know what .NET is, but we can talk about what Microsoft says they expect it to be.
Major components of .NET are XML, SOAP, C# and ASP servers.
Much of the structure of .NET imitates Java. You remember Java don't you? It's that "write once, run anywhere" programming language Microsoft licensed, "polluted" and destroyed. Any computer columnists will be more than happy to tell you how Java died at Microsoft's hands (even as the news section of his magazine announces one major Java based development project after another).
As a consequence of their ill-fated attempt to kill Java, Microsoft is exiled from Java development. Since Java largely defines computing in the new era, Microsoft found they had to duplicate its functionality, so they built a new language. C# (pronounced "C sharp"). It has remarkably Java like features. It even includes a cross platform CLR (Common Language Runtime) that interprets intermediate code, just like Java. Of course, C# itself runs only on Windows, unlike Java.
Please be aware that C# is NOT "Cool". Cool was an imitation Java project Microsoft launched in a fit of spite after Sun sued them for violation of the Java license. Do Not be fooled by the many C# variable names and internal structures that include the word "Cool". C# has nothing to do with Cool - Microsoft tells us so, so it must be true. They would know.
The major difference between .NET and Java, according to Microsoft, is that the .NET runtime is designed to interface with languages other than C#, including older languages like Cobol. If this actually works, it could save some companies a lot of money, but we're going to have to see it to believe it.
XML (eXtensible Mark-up Language) is the glue that is supposed to tie .NET together and make diverse systems (many not running Microsoft software) work together. Microsoft did not create XML, even though they sometimes claim to have done so. It is a W3C standard based on SGML, an ISO standard.
I won't discuss XML at length here because we have an XML article covering it.
Another component of .NET is SOAP (Simple Object Access Protocol), which allows one computer to ask another, possibly very dissimilar computer, to run a process and return the results to the requesting computer. The example usually given is a client computer asking an ASP server (through SOAP) to stream a stock ticker to it for continuous display on the desktop. The stock ticker program runs on the ASP server, but the result is displayed on the client's screen.
Microsoft did originate SOAP and has submitted it to the W3C committee for approval as a standard. IBM, Hewlett Packard and others also back SOAP, but the W3C has decided to study it (and alternatives) before issuing a standard.
All of these components (and more) are used to outfit a vast network
of ASP (Application Service Provider) servers, which dish up .NET processes
in response to requests from the client computers of .NET subscribers. Of
course the core ASP servers will be Microsoft's, but other companies will
fill in the gaps and provide auxiliary services (until Microsoft has time to
incorporate these into their own services and put the ASPs out of business
- haven't we seen this scenario before - software development or some such?).
Microsoft designed Windows 2000 to finalize world domination, but that just isn't going to happen. Incompatibility with "weaker systems" was designed in. Unfortunately for Microsoft, those "weaker" systems proved tough as nails, and those incompatibilities came right back to bit them - an example.
With .NET, Microsoft is touting interoperability and standards compliance - but isn't this just "embrace, extend, extinguish" all over again? Well, I'm sure Microsoft would like it that way, but I think they are becoming reconciled to it just not working this time. Everybody knows that game now.
With Balmer in command, we don't have to worry about Microsoft becoming nice, or ethics becoming a consideration, but reality is reality. In the business sphere Microsoft stands between the hard place of entrenched enterprise systems and the fast moving rock of Linux. They have to get along with very large companies that are years ahead of them in development.
In the consumer market, Microsoft rules - so expect to see them play
.NET hardball there - a very, very nasty game of .NET hardball.
Microsoft doesn't yet have a product, but don't expect that to slow down the selling any more than it has in the past. They've cranked up the vapor marketing engine full blast. Once again, Help Desk can assist you in understanding what's going on.
As in the past, Microsoft is playing software developers as their trump card. If they can build enough momentum behind .NET, they can starve competitors for developer interest. This move has failed only once (Java), but Java has so much momentum it's going to be a hard nut to crack - it's displaced Microsoft Visual Basic as number one in help wanted ads.
To build momentum, Microsoft started in early 2000 ordering their faithful pool of Windows developers to start migrating to .NET. They have been turning up the heat ever since. This has upset and confused many developers because it turns their comfortable Windows world upside down.
Microsoft is distributing pre-beta .NET development tools with Visual Studio, their suite of programming tools. They will be working some .NET features into Whistler, Windows 2000's successor, as soon as possible to try to build critical mass.
Elsewhere Microsoft is applying their vast wealth to tie hosting services to .NET. An example is the deal just completed with ASP USi (USinternetworking). For $50 million Microsoft dollars, USi has agreed to give preference to Microsoft applications and .NET services.
This deal could impact USi's future, because their clients tend to be large corporations using Unix and Linux back-end servers, and who require highly customized "best of breed" services. A "one stop shop" featuring "one size fits all" Microsoft solutions may not go over well. Oh well, the money is now, and .NET isn't for a while yet.
Microsoft has also invested about $150 million in cash strapped competitor Corel (publisher of WordPerfect Office) to encourage them to embrace .NET for their products. As a result of this investment, Corel is selling off their Linux development programs.
Given the current shortage of venture capital for .com startups, and the weakened condition of Microsoft's competitors, expect Microsoft to be buying a lot more .NET loyalty for a very modest cash outlay.
Now, in their biggest .NET move yet, Microsoft has
purchased Great Plains Software outright.
Great Plains is a leading publisher of Windows based accounting and business
management software. By this $1.1 billion purchase, Microsoft has put itself
in direct competition with its so called "Windows development partners" and
put them all on notice they will play a decidedly minor role in .NET, if any
(but do hurry and port all your stuff from Windows to .NET anyway).
We still think Microsoft will be unable to complete deployment of .NET; it's just too big, too complex, and too poorly defined. We do expect them to implement major parts of it, but companies that commit to .NET services will find themselves under performing compared to competitors who take different paths.
We still expect high speed Internet access will not be universally available nor uniformly fast enough to support .NET. This will force Microsoft to support two completely different architectures, and combinations of the two. This will be expensive and confusing.
We still warn that the combination of .NET and UCITA could prove deadly. Missing a .NET payment or two could put you out of business by depriving you of access to your critical business data and the software you need to interpret it. Under UCITA, you have no rights.
The one big change since our original .NET editorial is Microsoft's
acquisition of Great Plains. This puts Microsoft on a collision course with
it's "partners" and will result in very limited choices in the Windows
business management software market. We suggest you carefully review our
editorial on the Great Plains buy,
particularly the section on alternatives.
Andrew GrygusNightmares for Microsoft - challenges from all sides.
- Automation Access
Velocity Networks: Network Consulting Service - Internet Service Provider - Web Page Design and Hosting
All trademarks and trade names are recognized as property of their owners