Nice post. One comment though: "There are so many browser bugs, and oddities in JavaScript, that its always a long slog to debug it"

What about jQuery? It's supposed to remove the browser oddities by programming using one API.

In all the of the debates I see people are making a case for HTML 5 or Flash/Flex they all failed to consider one thing: integration

Neither HTML 5 nor Flash/Flex will ever be able to integrate at the level Silverlight will. The real case for Silverlight is its integration model (WCF on the server and Silverlight on the client). Microsoft will push Silverlight into its core products including SharePoint, ASP.NET, Windows Mobile 7, and the biggest bets of all its cloud OS, Windows Azure. I think Silverlight is here to stay regardless of what happens to HTML 5 or Flash/Flex.

You are correct that Flash has a thriving community and a large ecology of add-on functionality, whereas HTML5 right now has pretty sparse support.

However... that all grew after Flash gained a large browser market share... Right now, there is nobody "pushing" HTML5 other than geeks. But, once HTML 5 has a killer-app that everybody wants -- like Google Wave -- then that could trigger the same kind of thriving community.

I'm just sayin... if I were making a killer web app that wouldn't be finished for a year, I'd probably pick HTML5 over Flash.

On the surface your statement sounds great; "The problem boils down to this: there are millions of people dedicated to making the web better; but only one small part of Adobe is dedicated to making Flash better".

But then why has Flash been vastly superior in contrast to HTML, and even AJAX, for so long? You're in a sense saying that a group of guys have been far outpacing the mob? Delivering audio/video/3D/complex animation/file-reference-uploads/and so much more, and the "millions of people dedicated to HTML are just now nearing agreement on audio/video element attributes? Will that element have a computeSpectrum method? Web 2.0 was bubbling with the concepts of "cloud computing", "online OS", and RIAs, and the Flash Player has been the closest reality to a programming platform to deliver apps that rival photoshop. A Flash App was touting the new bitmap smart-scaling technology before Photoshop itself. Flash Player 10 is entirely more comparable to Cocoa than HTML/DOM scripting. And best of all the Flash Player elegantly side-steps the tremendously ridiculous render/display differences between the multitude of browsers. The best thing about HTML5 is that it may continue to convince Adobe to further open-source the Flash platform. If that were ever to occur I'd be surprised to see an HTML 6. Flex is adapting the niceties of element tags, but As a developer I'll always choose rendering consistency, and sprites over browser-display-differences and divs.

"Web standards take much longer than they need to"

Amen! You're opinion that using HTML5, when it has the same market-penetration as Flash, makes more sense doesn't make sense to me. Your 'what if Adobe tanks' statement is a valid concern, but one that is shared in similar ways by open-source tools/languages. Proprietary software runs the same risk of being deprecated, or mishandled by new management, as does open-source software.

"But each one (Webkit, Ajax, XHTML) was an attempt to add one small thing that made a big difference to the web. Flash is a BIG THING that makes a medium difference."

That's opinion. How did you measure the impact?

Anyhow, HTML is great, I'm glad to see it's advances. It is the browsers display inconsistencies I have real issues with. I'd be happy to see a one browser world, or one where any browser deviating from display-standards is fined, sued, and flogged by an international jury of disgruntled web design/developers = ] I also look forward to the day when the Flash Player is open-sourced, and becomes an integral piece of the browser's rendering engine, not just a plugin. It's a small world afterall, and we are the people.

Silverlight has a great integration story with MS core products, but it has much more than that. Silverlight does not require the use of MS products beyond the client, indeed it can be served from any server and communicate with other servers and the tooling is very good. Microsoft's presentation layer technologies have always been 1st rate and compilable. Flex is also a good platform and the two will live on and compete with HTML 5. As an architect, prematurely advising clients to take the HTML 5 path seems unsupportable at this time.

amen, brother. amen...

I was thinking the same thing while reading this.

(btw, either the CAPTCHA's on this site are too hard, or I need to enroll in some type of CAPTCHA class... I hate always trying to prove I'm human)

jQuery is the only open source javascript framework shipped with the Microsoft stack. That should make it easier to swallow.

Which one doesn't seem stable? jQuery or Yahoo?

"The problem boils down to this: there are millions of people dedicated to making the web better; but only one small part of Adobe is dedicated to making Flash better."

And yet, HTML is JUST getting around to being able to play a simple video, something Flash has had for nearly a decade now. How about Fullscreen? Closed-caption support? Streaming video? 3-D? Bitmap manipulation? Sound spectrum? etc...

Yes, the iPhone and iPad don't support Flash, but that is by design. Jobs is being pretty public about his dislike of Flash, and while he can sugar coat it and say it's because it's unstable, I would venture that it's more likely because Flash is not something he can control.

One need only look at the iTunes DRM and the App store to see that Apple doesn't have much interest in nonproprietary development, and I believe one of the main reasons they are backing html5 over Flash is because they can at least exert a small amount of control over the development of html5. For instance Canvas was originally developed by Apple and at it's inception they patented it and had planned to license it.

Now i'm not going to say that the days of the plugin aren't numbered, because they are, eventually open standards have a way of winning. However I don't see the end of plugin architecture for a while. That leaves a lot of time for Adobe to improve on Flash and if Adobe ever really did feel backed into a corner they could always make Flash open source.

While the concern of "what if Adobe goes under" is valid, lets not forget that just a few years ago Flash was Macromedias baby and Adobe seems to have done pretty well with it sense its acquisition. Generally when a company buys another they don't abandon major revenue generating platforms.

I am excited for the new html5 spec, but like everyone else here my biggest concerns are support and rendering. At this point in time my clients prefer not to pay for hours of extra work debugging on multiple platforms (though jquery et al are god sends they are not perfect). Flash/AS is still a viable platform, and will continue to be for a while, though a little competition never hurts the consumer.