March 15, 2004
Why the Whidbey Slip IS a Big Deal
While I respect Scott Hanselman, I have to disagree with just about everything he writes in his post entitled, "Yukon and Whidbey Slip, and your life goes on. Film at 11."
Let's analyze his arguments.
- The slip is no big deal because the platform has not been fully exploited by many developers.
Even if you buy his premise, this argument doesn't address the impact of the slip on development shops that are exploiting advanced features of the platform and trying to build modern websites on the .NET platform.
- The slip is no big deal because people are creating "fantastic mature software" with the current platform and tools.
This depends on your definition of mature. If by mature you mean old-as-dirt markup, then I agree. If by mature you mean moved out of the "Gee this is neat" phase and into the modern, standards-compliant world, then you're way off base. Try pointing the W3C Validator at www.asp.net and amaze at the fact that you're likely to find at least one error for every bone in the human body once YOU explicitly declare the DOCTYPE.
The fact is that the slip
is a big deal. Not because those of us trying to revamp old, outdated public-facing websites want to use generics or ObjectSpaces or any of that crap. I want the platform I've invested in to allow me to write modern web-based software that doesn't produce bloated, invalid, and antiquated markup which has a real cost both in terms of user experience and bandwidth consumption. Whidbey was to deliver on that promise -- and still will -- we just have to wait an unacceptably long time for it. Furthermore, you can hardly fault the
developers customers when MS marketing has been hyping all the problems Whidbey is going to solve with their currently deficient product for months. The fact is that Microsoft needs to do a better job of
listening to its customers. And for those who claim that using a different editor is the solution, you're dead wrong. What good will it do to produce nice self-closing break tags when the first ASP.NET server control you drop on a web form breaks your validity? The IDE is merely the tip of the iceberg. Think about it another way. Would you tolerate VC++ emitting non-standard C++? Uh...maybe I should try another anology...
At least on the XML front MS has the likes of and friends working hard on implementing standards-compliant software. Now if only the infection would spread more quickly.
Posted by Christian at March 15, 2004 09:21 AM |
Thanks for your post, Scott. As for the EAP/beta scenario, I'm not too comfortable with it for the particular sites I'm in charge of. I definitely fall into the Early Adopter category, but given our extensive use of Xml, I'm nervous about deploying technology that is still in such a state of flux right down to the class names. On a smaller site, I'd definitely consider it.
Good stuff...I do appreciate your points on the "State of The Markup." Forgetting about the Yukon this and ObjectSpaces that, the Markup thing IS a shame.
True, I don't address shops that exploit advanced features, and that's valid. We've insultated ourselves in some ways by hiding generics with code generation, and ASP.NET 1.0 with our own Personalization and MasterPages implementation. I agree that the slip hurts "advanced" shops more than the average J6P.
But, you have a valid point about the XHTML thing...it truly does suck that ASP.NET 1.1 doesn't even come CLOSE to XHTML. We are fairly "quirks mode" at my shop, as we attempt to support IE 5.0 and up, so we're "transitional" HTML 4.0. The XHTML 1.0 transition has been in the works for almost 2 years, and it looks like we'll still be in limbo for a year.
I guess I've resigned myself to "markup that works" and maybe focused more on the elegance of the C# than the elegance of the resultant HTML.
Have you thought about the Whidbey EAP (Early Adopter Program) and going live with Betas of 2.0?