[foaf-protocols] is webid protocol a layer that can be designed as a component, for web services?
home_pw at msn.com
Sun Dec 26 16:06:38 CET 2010
Just as one MIGHT conceive of trust providers for the/a webid protocol
falling into 2 categories (browser-centric, or otherwise) can might ask: is
webid protocol mere a finished solution or can it be a layer that can be
designed as a component, for web services?:
Though SSL started out as a socket (allowing multiple sessions and multiple
connections from browser frames and browser instances to access a common
implementation of session management and security associations), it became
much more. Eventually, it even out grew the socket abstraction.
http://msdn.microsoft.com/en-us/library/ms678421(VS.85).aspx is a nice
example in which SSL was made into a simple provider for a subset of the
security semantics for COM (an infamous Microsoft componentware).
In conceiving of the webid protocol, perhaps its formulation, its definition
technique, it specification language and its notation design should be
selected with a view to it being a provider - for web services in the semweb
I think this gets to the heart of what we were discussing in formulating the
proposal: webid is more than linked data (and its about more than the
browser). It has to address the web as a whole (perhaps including acting as
a provider). Concerning integration, do things have to be limited to the URI
scheme as the integration mechanism, and the http scheme in particular?
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the foaf-protocols