The issue is software. It's a matter of protocol only, not hardware related; there's simply a dearth of proper software on all platforms except the tablets. Since the desktop client doesn't support video at all, it should be unsurprising that it does not support it with a specific remote endpoint. I'm not sure what you consider the "whole intent of the internet", but I don't see how this is against it.

While the native software was, AFAIK, only able to support video tablet-to-tablet (due to complete absence of jingle-video support in other software, including Google's own) till last year, with rtcomm comes SIP video interoperability with 3rd party software (including eyebeam and x-lite), so it's definitely fulfilled now.

So using Google's protocol, and having no interoperability because Google can't be troubled to implement their own protocol, is "profound proprietary hooks"?!

I say it is completely true because in actual use cases, it does function. I don't include Gizmo in this, because I don't use Gizmo; Internet Call works fine for me, and I think the use cases in which it doesn't work are not reasonably inferred from the marketing statements.

Well, picky's good, but I'd rather be correct; I'm actually gonna go for both here, though. Webcam is a word, and as such it has a meaning independent of its etymology. Generally, it refers not to videoconferencing at all, but to live uploading of images (or, lately, streaming) for viewing over the worldwide web (which, BTW, is a concept of interlinked hypertext, and doesn't properly apply to a Flash-based star-topology service). Still, proper webcam functionality also does not exist out-of-the-box on the tablets, which is actually a stronger argument that "webcam" is misapplied here. (Basic webcam functionality is available by third-party software, of course, e.g. motion.)

Clearly, though, the problem is an abuse of terminology, rather than intent to deliver a genuine webcam defeated by bugs or incompleteness.