Re: [alto] Review of draft-ietf-alto-xdom-disc and supporting WGLC.

[Sebastian Kiesel <[e-mail]>]

Dear Qiao,

thank you for the review.  We will post a new version of the draft
soon, which will reflect your comments.  For details see below.

Thanks again,
Sebastian


On Wed, Jul 19, 2017 at 08:06:54PM +0800, Qiao Xiang wrote:
> Dear Sebastian, Martin and WG members,
> 
> I reviewed the draft-ietf-alto-xdom-disc document. Overall the document is
> well written. It provides a smart solution to address Req. AR-33 in
> RFC6708. The idea to use reverse DNS lookup for cross-domain server
> discovery is elegant. I add some editorial comments for this document in
> the following. Other than these, it is my opinion that this document should
> be last called.
> 
> 
> ===Comments===
> Section 3
> Page 12, I might be extra picky here, but the title of this section is
> formatted as:
> "Using the ALTO Protocol with  ALTO Cross-Domain Server Discovery",
> while it should be formatted as:
> "Using the ALTO Protocol with ALTO Cross-Domain Server Discovery".
> with an extra space after "with" deleted.

Trailing space before a line break in the XML source...  Thanks for being
picky - we want to produce high-quality documents, both wrt. content
and formatting.


> Section 3.1
> Page 13, line 1: "in the subsection below." Based on the context, this
> subsection refers to Section 3.4. It should be stated explicitly.

We have changed it.

> Section 3.4
> Page 14, case 4:
> "...the ALTO client should split the list of source addresses, and perform
> separately for each source address the same steps as in case 1, as
> specified above.  As an alternative, the ALTO client could also split the
> list of destination addresses,..."
> 
> the expression should be consistent with case 3. I suggest it is changed to:
> 
> "...the ALTO client may split the list of source addresses, and perform
> separately for each source address the same steps as in case 1, as
> specified above.  As an alternative, the ALTO client may also split the
> list of destination addresses,..."

we have changed "should" + "could"  to  "should" + "may".

We think it is better to indicate a preference for one of these options,
therefore the "should". Note that we are using lowercase should and may,
not the RFC2119 ones, as this is not an issue of interoperability.


> Appendix A
> Page 24, para 1: "A solution for the problem described in the previous
> section". Which section is referred to should be explicitly stated.

This is a remains from moving that section from the main part to the
appendix. We have reworded it:

This appendix itemizes requirements that have been collected before the
design phase and that are reflected by the design of the ALTO
Cross-Domain Server Discovery Procedure.


> Appendix B
> Page 29, all the "m"s in the equation computing P(X=k) should be "M"

You are right. We have fixed that.

Thanks again,
Sebastian