Among those demands are an adherence to "stateless interactions" - i.e. where there's no obligation from any user's system to remember the state of another - along with a dynamic reaction to network conditions and interoperability with existing standards. It must also support multiple codecs in case of future changes.
Unfortunately, Microsoft points out, WebRTC in its current form doesn't meet with all those demands. There is, for instance, no ubiquitous deployability, with the standard today showing "no signs of offering real world interoperability with existing VoIP phones, and mobile phones, from behind firewalls and across routers and instead focuses on video communication between web browsers under ideal conditions."
It also builds on the legacy of SIP, Microsoft argues, which doesn't make for a stateless system. Applications would be forced "to resort to trial-and-error and/or browser-specific code" Microsoft says.
The company's solution is CU-RTC-Web, freshly submitted to the W3 organization, adding a real-time, peer-to-peer transport layer, and building on the existing W3C getUserMedia API - something that Microsoft has already been playing with for integrating video authoring and voice commands to HTML5 apps.
Part of Microsoft's motivation may well be Outlook.com, a preview of which recently launched to replace Hotmail. Although not currently functional, part of the Outlook.com promise is in-browser Skype support with no local install required.