[AFS3-std] Re: Standardization of the AFS3 protocol

[Jeffrey Altman]

Jeffrey Hutzelman wrote:
> On Wed, 22 Feb 2006, Jeffrey Altman wrote:
> 
>> If RX was to be documented, the intention would be to publish the
>> specification as an IETF Informational RFC.
> 
> As multiple documents, I think.  One describing the UDP-based transport
> later, another the session layer, and a third for the RPC layer (the last
> in terms of XDR, which is already an IETF protocol).  Plus one for each
> security class, and one for RXTCP, when that is mature enough.

I completely agree.  This is how we have published all of the Kerberos
and GSS-API protocols over the years.  It makes it much easier to
understand protocol and allow for updates to be published.  Not to
mention it makes it easier to get started since we can document the new
work as we develop it.

> Jeff, it is worth noting that the IETF's way of publishing protocol
> specifications is not the only way.  There is nothing that says we can't
> write all of our protocol specs in, say, MacWrite and put those documents
> up on grand.central.org (one of whose purposes has always been to publish
> AFS protocol documentation, if anyone ever wrote any).  

of course not but ...

> However, given the
> people who are and have been actively involved in AFS protocol work, it is
> much more likely that we'll want to do something sane, like develop
> documents in the form of I-D's and then publish them either as RFC's or
> independently.

this was the important point

> It is probably also worth documenting the .xg language, so that RPC-layer
> protocols can be documented partially in the form of definitions written
> in that language.

Absolutely.

Jeffrey Altman

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3288 bytes
Desc: S/MIME Cryptographic Signature
Url : http://michigan-openafs-lists.central.org/archives/afs3-standardization/attachments/20060222/c77b25fd/smime.bin