RE: Proposed Resolution to PROT I-D Issues List

["David B Harrington" <dbharrington@comcast.net>]

Hi,

Let me expand on my earlier comment.

The XSD will likely be distributed separated from the RFC, just as
happens with MIB modules, if certain conditions arise, which have
arisen with MIB modules.

Separate distribution is likely to occur if:
1) compilers and tools do not adequately distinguish the XSD from the
rest of the document programmatically, and require that the XSD be
separated from the surrounding text before processing,
2) multiple XSD specifications can exist within a document and
available tools are not able to deal with this,
3) XSDs are fairly easy to produce that are good enough for most use
cases, but RFCs are burdensome to produce because of boilerplates,
id-nits, MIB Doctor reviews, and other IETF-publication-process CLRs,
4) XSDs are fairly easy to produce that are good enough for most use
cases, and vendor-specific XSDs are produced without all the
unnecessary boilerplates, id-nits, MIB Doctor reviews, and other junk
that do not necessarily help vendors implement products, sell products
and get them to market before their competitors, and the extra junk
does not actually help their customers manage their networks, and
5) XSDs are unambiguous enough for operators to use, and operators'
tools to use programmatically, even if correct implementation might
have required referencing the surrounding text.

I think this last point is very important - I do not believe most
implementors work with only the MIB module; they also read the RFC.
Vendors distribute the stripped MIB modules to operators, because the
MIB modules alone are often good enough to meet the requirements of
operators and operators' tools, and the extra stuff is often available
in RFCs or product release notes or customer support.

But these points are all meta-discussion mostly unrelated to this WG's
deliverables.

I suggest that this WG develop specs that **support** the separate
distribution of the XSD from the RFC, and design the XSD to address
the potential issues caused by separate distribution. 

Let's make sure 
1) the XSD contains a clearly marked version, preferably
machine-parseable.
2) the XSD can be programmtically separated from the surrounding RFC
text reliably.
3) the XSD contains enough information to be meaningful to operators
and their tools.

David Harrington
dbharrington@comcast.net



--
to unsubscribe send a message to netconf-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/netconf/>