How to build the mesh - #10: Standards
This final area in building the open mesh is the hardest.
No one wants to give up control of their technology - so (by definition) the open mesh must be made up of a combination of open, free protocols and technologies with proprietary APIs and technologies.
That doesn’t mean we should stop trying to develop compelling open platforms and experiences for end-users - it just means to be pragmatic you have to work with Facebook, MySpace and Google’s APIs - regardless of whether their technologies are truly open or not.
So when sitting back and thinking about the importance of open standards in the open mesh, one has to draw the conclusion that in fact it’s not about OPEN standards - but standards and redundancy. Its OK to have a hybrid combination of proprietary standards and open standards - just as long as we have choice.
Afterall - that’s what the Architect told Neo right “humans are about choice” (I’m paraphrasing here.)
So by having redundant infrastructure and choice over standards - we can achieve a harmonious open mesh.
IMHO
Now onto the summary post - which pulls it all together.
Much of this architecture has been built but simply has yet to be included in the major “web 2.0″ sites. LittleShoot (http://www.littleshoot.org) breaks down and includes the underlying protocols in some detail (you need a signaling protocol like SIP or XMPP, an offer answer protocol, ideally a NAT traversal protocol, a publishing protocol, a syndication protocol, etc etc).
You should also check out the work at the p2p SIP working group. Those guys are taking it too far into the realm of research IMHO, but they’ve got a sweet platform.
I couldn’t agree more the web needs to evolve into a family of much more robust protocols along these lines.
Much of this architecture has been built but simply has yet to be included in the major “web 2.0″ sites. LittleShoot (http://www.littleshoot.org) breaks down and includes the underlying protocols in some detail (you need a signaling protocol like SIP or XMPP, an offer answer protocol, ideally a NAT traversal protocol, a publishing protocol, a syndication protocol, etc etc).
You should also check out the work at the p2p SIP working group. Those guys are taking it too far into the realm of research IMHO, but they’ve got a sweet platform.
I couldn’t agree more the web needs to evolve into a family of much more robust protocols along these lines.
Much of this architecture has been built but simply has yet to be included in the major “web 2.0″ sites. LittleShoot (http://www.littleshoot.org) breaks down and includes the underlying protocols in some detail (you need a signaling protocol like SIP or XMPP, an offer answer protocol, ideally a NAT traversal protocol, a publishing protocol, a syndication protocol, etc etc).
You should also check out the work at the p2p SIP working group. Those guys are taking it too far into the realm of research IMHO, but they’ve got a sweet platform.
I couldn’t agree more the web needs to evolve into a family of much more robust protocols along these lines.