I've been spending a lot of time lately looking at video online. Part of it has been technical, and part of it has been thinking about what is needed and where it is headed. I was involved with IPTV / video over IP back at Nortel, and a lot of things actually haven't changed.
As if video on the Internet wasn't hot enough already, the whole Google buys YouTube has upped the fever pitch to...well, something hotter than fever :P
So, for starters, go and read (and link to!) Kevin Marks' five point plan to save use from Flash video. He included "crappy" in the title, and he's right, but...well, it's what works today.
And that's what will stop us from getting any further. Getting stuck in the "works today" mindset is a bad thing. When you sit down to build something new, you're really never building it for today...you're building it for today AND for the next little while. Somebody has to take a chance on figuring out the future.
In any case, I very much agree with all of Kevin's points. The phrase "detection script" makes me shiver a bit, and Flash is still the only platform solution if you want one integrated application that has interactivity built into it as well, but for straight up video, we can do better, and should be trying to.
A closer alliance between Google and Apple wouldn't surprise me...