In Lars and Kasper Google I/O talk, they explain that V8 has become an extremely complex code-base, now weighing more than 500k LOC, with advanced compiler tricks to achieve the leading performance it currently has. The V8 designers think they're nearing the max performance ceiling possible with JS where it's unlikely they'll be able to get a further 2x perf improvement: http://www.youtube.com/watch?v=huawCRlo9H4
Which is in contrast to Dart which was developed with the input of VM engineers, is a much smaller and simpler code-base that is able to support proper classes, real integers and SIMD support which is already capable of delivering 10x faster startup times (with snapshots) and 2x faster and more predictable runtime performance which is still in beta and has yet to receive anywhere near the engineering effort V8 has.
Google is big enough to attempt 2 competing strategies to improve web development and performance, i.e. a faster JavaScript being led by the Standards committee that's enhancing JavaScript at a comittees pace and by introducing a completely new language for the web built from the ground up to be faster, structured and more consistent: http://infrequently.org/2011/09/google-the-future-of-javascr...
The real story is that Google wanted to go it alone rather than improve JS, and that's why there's Dart.
Dart wouldn't exist or have anywhere near the technical excellence it has today if it wasn't a sponsored project. Its existence also doesn't prevent any other browser vendors from trying the same thing - it may even encourage it if it yields noticeable end-user improvements.
The fact that others could try to make their own competitors is an exact description of the situation where the web breaks down in favor of proprietary solutions and the strongest companies end up owning the web.
The web doesn't have to worry about the strongest companies making too good compelling open-source solutions that people fear will end up owning it, we're in 2013 and web development is still fragmented, inconsistent and progressing at a glacial pace. The biggest threat is coming from proprietary native apps in walled garden mobile platforms being developed behind close doors without a standards body in sight and no desire for any open or interoperable initiatives.
Without any attempts at any new initiatives trying to make web development better, in the next 5 years we'd still be building buggy web apps on a broken language running at similar to todays performance - but hey, at least we waited for the standard bodies to create all possible solutions. All other attempts that stray from the script be damned.
If standards bodies are as feeble as you suggest aren't going to be a serious custodian for Dart anyway.
Ossified standards bodies or unilateral action are not the only options. Google could have engaged in cooperation with other interested parties throughout the development, so that it was a truly collaborative effort.
Whilst I agree there are technical problems that Dart is solving, they have chosen an approach designed to maximize their control over the web.
I read a lot of criticism of Google's centralized incubation of Dart, that it has not been developed by some standards committee. How would this prevent the nightmare scenario of a Google overlord (it's a _language_, for Pete's sake)? What was the last successful committee designed language you've used?
tl;dr Hate Google for all kinds of reasons, but judge the language on its merits.
Your TLDR says it all - you tell us to look at Dart as if it were some disembodied platonic entity independent of the goals of the corporation that controls it.
But Dart isn't independent of Google, so can you explain why we should limit our thinking like this?
You see, it's possible to separate the product from its origin.
A language has even less to do with advancing corporate interests (other than Google's professed goal to "make the web faster"), since it's an agnostic platform rather than a product designed to further a specific goal.
No. JavaScript has some warts that are more easily fixed by going a different direction, hence Dart. Diverting effort away from improving JavaScript isn't serving their goals, having a nicer language to develop in is.
None of Google's goals are harmed by Dart. That's the point - it's there to serve their goals by diverting effort away from improving JavaScript - a project that benefits everyone towards supporting Dart - a project that benefits Google.