@Andrew
definitely a nice article and the benchmark results are exactly what I was trying to explain across many of my points. no issues on self-promotion on my blog as long as the articles are relevant to the topic.

The focus should be more on building fast, interactive media that works everywhere, and less on platform. End users don't care what it's built in, just that it works.

This is a (transparency: my) blog entry RE: web-based charting in Flash vs. HTML5. We found that even based on side-by-side performance, there is no single winner between the two. It depends on platform, browser, and even chart complexity.
http://www.zingchart.com/blog/2010/04/07/zingchart-plots-to-end-the-war-between-html5-and-flash-in-web-based-charting/

A call for developers to code stuff that "just plain works."

Open-ness and closed-ness are usually confused and indistinguishable by the naked eye, your big argument about HTML5 being an open format can quickly become an argument on your iPhone having a proprietary version of Safari to run it, or Opera being closed-source.

As for the rest, I already stated that misusing Flash is the biggest issue, maybe you are following the discussion with some preconceptions about why I'm stating certain things and that leads you to wrong conclusions. Don't jump the gun, we all want better web.

Flash is not a standard, nor is it an attempt at one. It is a proprietary plug-in run on top of the web stack that happens to play nice with HTTP and XML.

Flash is not open. Flash is not platform-agnostic. A 3rd party cannot come along and write their own flash player based on any sort of public specification. There is no way for an entity like Google or an individual developer to apply their expertise to a competing closed and compiled stack in an effort to improve performance or cross-platform compatibility.

That said, you're spot on that Flash *is* good for animation and rich media and it enables richer interaction when used well. Unfortunately, Flash as a platform was designed from the ground up to be a walled garden.

Is it getting better? Yes.

Was it designed to support and run standards-based markup, presentation, and behavior in an open and accessible manner? I'd be hard-pressed to find a qualified professional who could answer with anything but a No.

It's a simple fact that the lion's share of the content on the web that is imprisoned inside flash's walls today could succeed at the same goals with much better accessibility and semantic relevance in HTML/css/js.

As a standards-based web developer, I have developed a successful SEO-friendly semantic flash platform as well as implemented flash video platforms. The advanced effects and animations can be put to tastefully good use, but I have also seen the conflation of web video and flash result nightmarish app architecture, simply because the back-end developers made the assumption that video would always be served within flash.

I see the HTML5/Flash debate as much larger than the narrow scope of delivering standards-based video content. It's a movement toward a more semantic and open interactivity layer… and one that Adobe should embrace instead of try to overshadow.

Yes, flash is good for some things… but it should be used as a last resort. Anything along the lines of navigation, text, or image content that is walled up inside flash without a standards-based fallback is just simply *doing it wrong*. If you think otherwise, take a long hard look at why you do.

Second thing, Flash had at least 2 major completely different dev teams behind it, one was from Macromedia, the other is from Adobe. It's also suspected by many (me included) that there were at least 3 teams in the "Adobe era", first being the transition team, second being the CS4 team and currently a more talented team that is clearly doing things right.

Also, what you call a "crappy job" is currently part of every Firefox browser out there, as much of the TraceMonkey code is based on the Tamarin code open-sourced by Adobe in 2006 (Tamarin is the AS3 VM on Flash Player).