Maybe there's a division here between "SIP the abstract protocol" and "SIP the API implementation reality". The problem Skype has is that everything is so dynamic and unstable in its environment; all the proxies can come and go, timeouts might take forever, nodes may partially fail. So perhaps to be able to set up a call and keep it up they've had to re-implement the same basic concepts, but also optimise it to take account of the rotten network environment. Cached data might be passed up and down chains of proxies and supernodes, multiple responses/options ("he could be here, here or here") included where standard SIP might only offer one, etc.

This is just a SWAG based on my Oracle experience and I have zero evidence.

What if the whole thing was deleberatly done for misleading the user?
How can you say to have an innovative technology and then using what all others used since long?

In principle Free World Dial up was there much before them and so VoIP.
I do not think they invented the hot water, they just used it in a most clever way and built a very good market product out of it.

How long it will last that has still to be seen as how much revenue it will produce.

Patrizia

Martin:

Pursuing your line of thought, data exchanges between proxies and supernodes could have been done outside of SIP (after all SIP is "immensely extendable"). Also, won't forking have handled the scenario you are alluding to in your comment? My puzzle is given that under normal conditions Skype provides directory service and assists in NAT traversal, why SIP is not sufficient. IF I want to replicate Skype, then I need to understand this.

Patrizia:

They are probably misleading the industry players and not users, for users are getting the service they were promised. Even though FWD and other VoIP players have preceeded Skype, we must recognize that Skype is the first one to use wideband codec. All the others were busy imitating POTS (not just PSTN) both in the technical architecture and in their business models. I agree with the uncertain economic viability; but then it is true for all VoIP service providers.