[SAM] draft-kolberg-sam-baseline-protocol-00

[Gustavo Carneiro <[e-mail]>]

Hi,

I am trying to understand
draft-kolberg-sam-baseline-protocol-00<http://tools.ietf.org/html/draft-kolberg-sam-baseline-protocol-00>

<http://tools.ietf.org/html/draft-kolberg-sam-baseline-protocol-00>One thing
that confuses me is in Sec. 5.  Group Management API.  First it says,

While native multicast is bound to IP addresses, ALM uses arbitrary strings
> as multicast names, which will be mapped to the overlay identifier space.

The aim of this API is to implement group-oriented data
> communication independent of the underlying distribution technologies.


But then the API described is based on Address types, which are described
as:

Address is any address structure suitable with respect to the technology
> available at the host. This may be an IPv4 or an IPv6 address, or any
> overlay identifier.


So it seems that the Group Management API does not work with high-level
abstract identifiers, it already works with transport-specific identifiers.
 This is contrary to what was show in the  presentation
http://www.ietf.org/proceedings/10mar/slides/SAMRG-1.pdf

In page 11 of the slides:
             Sec 5. Group Management API
• API between Application and Group stack
• init(out Handle s)
    – This call creates a multicast socket that is bound to some virtual
multicast
       interface and provides a corresponding handle to the application
programmer,
       which will be used for subsequent communication.
• join(in Handle s, in URL g)
    – This operation initiates a group subscription for the name g,
including the
       corresponding tree access.

Here we see a different API, where URLs are used instead of identifiers.

Another not so clear item is how the API in Sec. 5 articulates with Sec. 6.
 Sec.6 starts like this:

In this document we define messages for hybrid overlay multicast
> tree creation...


I get the feeling this used to be a separate draft that was converted to a
section of another draft.  I suppose that the Protocol described is
triggered by API calls of the Sec. 5 API, but I get the feeling I am
guessing what should be explicitly stated in the document.  In that
sense, draft-waehlisch-sam-common-api-02 (Fig. 2) is clearer.

And what is exactly the relationship
between draft-waehlisch-sam-common-api-02
and draft-kolberg-sam-baseline-protocol-00.  They were published almost at
the same time (for the meeting?); are they competing proposals?

Thanks,
-- 
Gustavo J. A. M. Carneiro
INESC Porto, UTM, WiN, http://win.inescporto.pt/gjc
"The universe is always one step beyond logic." -- Frank Herbert