As someone who joined the IE dev team relatively recently, I must say I had much of the same ambivalence you express in your first sentence. To add to that, I was a Mac user and moved from the Bay Area -- not exactly the bastion of Microsoft love -- and still have friends who are skeptical of my move. For me, the decision boiled down to two things:
(1) Web browsers have intrigued me for about a decade, and this gave me a chance to help push the web platform ahead. I love how through a process of public discussion and consensus some of the largest technology companies come together to build these (more or less) mutually compatible and complex platforms (a.k.a web browsers). Notice how I skipped past much of internet history there? You might say that, well, there are other browsers out there that I could have contributed to, and I will then ask which one do you think needed the most help? ;)
(2) Where would my small contribution be able to have the most impact. As you rightly point out, IE remains the most used browser (on the desktop at least) and every line of code I write may some day be exercised by hundreds of millions. I cannot emphasize enough just how fulfilling this is to me. When I was graduating and interviewing, Facebook liked to say that their user to dev ratio was a million to one, and our team easily tops that.
How's it like to be an engineer on this product? Lots of fun and very challenging. Much (as my limited view sees it) boils down to adding new features and/or rewriting older subsystems for performance reasons. Simple enough, but when you factor in about 15 years of legacy code, much of whose functionality has to be maintained (or emulated even in the new subsystems), life becomes very interesting very fast. I _suspect_ Firefox and Chrome are running into some of these challenges with the Electrolysis project (FF) and adding in graphics hardware acceleration (Chrome), which is perhaps why they're taking longer than anticipated.
Do other web developers treat me guilty by association? I don't know :)
From my perspective, the IE team's priorities are weird. That's all. Yes, the race for performance is super-important as we're pushing our browsers to do more and more but if IE can't DO the same things other browsers do I don't really care how fast it can not do them. ;)
EDIT: grammar.
Microsoft can attack on that front very aggressively. So they're taking the one thing that most people know is different or special about Chrome, and saying, 'IE is faster. Faster canvas, faster svg.' Seems like a great strategy to me.
MS have stated time and time again that they pay more in licensing fees to MPEG LA than what they make from it. Therefore it'd only be detrimental to other members of the cartel if they go after Microsoft for implementing WebM.
The reason for not implementing WebM are as simple as MPEG LA and Microsoft not wanting the format to succeed.