Sebastian, all: I went through the server discovery document in
preparation for the shepherd writeup and moving the work ahead.
I have some small suggestions on improving the document, and if
you can attend to these suggestions quickly and upload a new version,
I can then move the new version towards Publications Requested.
Major issues: 0
Minor issues: 1
Nits: 10
Minor:
- S5: The IANA section in the document asks IANA to register a U-NAPTR
application service tag. RFC4848 provides a template to register
a U-NAPTR service tag. The information in the draft matches what
is required in the template with the exception that interoperability
and security considerations, listed in RFC4848 template, are not
provided in the document.
However, this is easy to fix as follows: the security considerations
of the U-NAPTR service tag are described in the document itself, in
S6.2; so just mention that. For interoperability considerations, you
can just say that there aren't any known interoperability
considerations to list since this is an ALTO-specific service tag,
and ALTO itself is a new protocol.
Nits:
- S3.1.1: s/DNS suffix on its own,/DNS suffix,/
- S3.1.1: I would suggest s/myaltoprovider.org/example.org/
The example.org and example.com domains have been reserved
by IETF for documentation (c.f. rfc2606).
- S3.1.1: s/In case no ALTO NAPTR/In case no ALTO-specific NAPTR/
- S3.1.2: For the same reason listed a couple bullet points above,
s/myisp.com/example.com/
(Note that you have it right in S3.2, where all your examples are
rooted in the example.com domain.)
- S3.2, first paragraph: I would suggest starting a new paragraph with
the line "The first step ..." It improves readability.
- S3.2, last paragraph: The text says, "...wait a period of time ..."
before retrying. Is the expectation that a retry may eventually
succeed? What will change to make the retry succeed, besides someone
configuring the SOA with the required information.
- S4.2: s/it SHOULD ask the ALTO server(s) for guidance that have been
discovered for this specific interface and address family./it SHOULD
ask the ALTO server(s) that have been discovered for this specific
interface and address family for guidance./
- S6.1: The sentence "But given that users cannot rely on the
availability of an ALTO server, this results in no significant
additional security risk." can be perhaps improved. The issue is not
that the users cannot rely on the availability of an ALTO server,
but that the presence of an ALTO server enhances their application in
some manner.
IMHO, you can simply take that sentence out without loosing any
generality or context. You may want to:
s/performance and traffic distribution will correspond to/performance
and traffic distribution will subsequently correspond to/
- S6.1, third paragraph: s/it may provide/the discovery service may
provide/
- S6.2: I realize that this is an extremely pedantic nit, but grammar
calls. It would be good to end bullet items 1 and 2 with the ";"
punctuation symbol and bullet item three with ".". Sorry.
Thank you for your attention.
- vijay
--
Vijay K. Gurbani, Bell Laboratories, Alcatel-Lucent
1960 Lucent Lane, Rm. 9C-533, Naperville, Illinois 60563 (USA)
Email: vkg@{bell-labs.com,acm.org} / [email protected]
Web: http://ect.bell-labs.com/who/vkg/
_______________________________________________
alto mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/alto
