Is Microsoft's .NET a better place for the development of Web Services? IBM thinks not. In response to a recent Microsoft paper on about creating Web Services with .Net versus IBM WebSphere 4, IBM has released a competitive review debunking Microsoft's claims and showing the Websphere and J2EE advantage.

Read WebSphere competitive review

If I am not mistaken this topic was also addressed in the discussion "Microsoft claims.NET version of Java Petstore is superior" on The Serverside

This actually came up in one of the .NET vs. J2EE threads, and Microsoft has a response to the IBM response at http://www.gotdotnet.com/team/compare/ibmrespond.aspx. Now I'm just waiting for the response to the response to the response. ;)

I'll be honest and say I haven't downloaded the samples and walked through them yet. I do find it somewhat hard to believe you could write a web service that actually does anything with 0 lines of code, but if IBM can pull that off that's pretty cool...

Mark

That is correct. Microsoft has an official response to the IBM competitive review, which was originally posted in response to our paper comparing building Web Services in Visual Studio .NET vs. IBM Web Services Toolkit. We have since updated our paper to reflect the release of IBM's new tool for Web Services, the IBM Websphere Studio Application Developer released 11/21/2001. Just as important, IBM's competitive review is repleat with tremendous factual errors and misstatements, including the code count pointed out above, as well as numerous other issues. For example, the license cost calculations IBM makes for the Microsoft Web Services platform are simply false. These are covered in the Microsoft response at http://www.gotdotnet.com/team/compare/ibmrespond.aspx.

Also check out the full updated paper at http://www.gotdotnet.com/team/compare/webservicecompare.aspx.

We invite customers to read the paper, download the code, compare the deployment costs, and try out both tools for building Web Services to judge for themselves.


Greg Leake
Group Product Manager
Microsoft Corporation

Hi Greg,

As a product manager for Microsoft I am interested;

Intersted in your thoughts.
How does Microsoft expect the Java Community to embrace .NET technology when Microsoft have not embraced Java?
And how does Sun expect the .NET community to embrace Java technology when Sun have not embraced .NET?

The two technologies appear to be in conflict; not designed to work together but as seperate spaces doing the same thing. This is a pity, but somebody must be hoping for a winner but is this all based on the $ and not evolution? fear.

IMHO if we are to make the global advances required in the 21st century to get to the next stage in our evolution we cannot have this divide, it is counterproductive. On the bright side Web Service technology is about bring things together. However the current standing of .NET and Java is divisionary not visionary. It starting to breed more developer coun'ter-cultures and will slows progress as people are forced to make choices, choices that are presented in such a way that time is needed to decide. At this stage in the information evolution I am disappointed to see a .NET vs Java battle, I think it is wasting my time and I search for an environment that does not have these issues.
 
I would say to Sun and Microsoft;
Can you image what you guys could do together?

Regards,
Stephen

No they obcviosly can't :(

But it's refreshing to know that someone can!

 

"Using WebSphere, the lines of handcrafted code can be reduced from 106 using .NET to a single line of handcrafted code, a 99% reduction." -IBM

Oh boy, here we go. It's only a matter of time until we have a version of the Pet Store implemented in a single line of PERL.

Actually, how about APL? Do you remember those infamous "APL one-liners" and the wierd symbols on your key board?

    Seriously, I am gratified rather than appauled by the competition between IBM and Microsoft. It is unfortunate that the example (kind of like the playing field) is this game is so trivial. Could someone please propose a better one or suite of them? Small SLOC count is not the primary scoring mechanism in such a competition. Mistatements of the facts are like cheating in this game.

   It would be good if a collection of criteria and sub-criteria could be developed, with objective metrics against these. Everyone could weight the criteria to suit them and their organization.

-Tom