Re: [core] htpp-coap proxy / review comments Peter on HTTP Mapping draft

["Dijk, Esko" <[e-mail]>]

Dear Peter,

thanks for your detailed review of the HTTP-mapping draft. This has been used, along with some general WG comments on last IETF meeting, to improve it.

Below are specific reply comments to your feedback in the Word document (using the original Word comment numbering [n...] where possible) with explanation how the authors applied the comment to the draft. The focus is here on Section 3-5. Any items for these sections not listed, were simply applied.
The new version is http://tools.ietf.org/html/draft-castellani-core-http-mapping-06

*** Section 3: Cross-Protocol Usage of URIs

Paragraph 3 & 5: Your suggestion was to re-write some of the sentences.  We reviewed the sentences and decided to combine the two paragraphs to make it clearer.

Paragraph 4: Your suggestion was to eliminate the paragraph (most probably because it was confusing). We agreed and have eliminated the paragraph.


*** Section 4: HTTP to CoAP URI Mapping
Various paragraphs: You had assorted editorial suggestions throughout the section.  We implemented most of your suggestions.  Also we re-worded the section to focus explicitly on the Reverse Proxy scenario more clearly.


*** Section 5: HTTP-to-CoAP
[n3],[n5]: your suggestion to remove items on timeout of HTTP request and HTTP connection pipelining;
Resolution: These items were moved to a separate section at the end, to make the section 5 intro easier to read.

[n9]: unclarity of text about multiple forward proxies; the issue was not resolved yet (pending more communication between the authors and/or WG). But the subject is indeed not clear enough yet.

[n10]: about detection of duplicate idempotent requests; here the text was updated to reflect that this is about duplicate requests made by the same proxy. In that case the proxy will know what other requests it is serving. (No communication with other proxies is needed.)

[n11]: unclear role of idempotency ; we removed the word here.

[n12]: observe on the proxy ; text slightly adapted. The cross-proxy will act as a CoAP-observe client here. It will not act as CoAP-observe server. Regarding the second question, we do not make recommendations currently about number of cross-proxy servers to be used. But the question is not entirely clear to us.

[n14]: proxy queueing and dropping behavior; we took over the suggestion to make this configurable.

[n15]: using coap-observe; correct there are no interoperability issues here in using –observe or not. Therefore it is also not part of our set of guidelines. This is just to inform proxy developers of a good technique to employ for caching. The section is shortened, though and only result heuristic is presented.

[n16]: about core-block mismatch of block sizes; text was added for case that CoAP endpoints do not support –block. Different block sizes support is dealt with by the core-block protocol negotiation. The transparent client-to-server blocking suggestion was not taken up; since the use of –block is only between proxy and coap server; client not involved.

thanks & best regards,

the authors of draft-castellani-core-http-mapping


-----Original Message-----
From: [e-mail] [mailto:[e-mail]] On Behalf Of peter van der Stok
Sent: Tuesday 11 September 2012 11:06
To: akbar rahman; [e-mail]
Subject: [core] htpp-coap proxy

Hi Akbar,

attached the comments on the cross-proxy draft.
I converted the txt document to word and edited the word document; that made my life easier.

I changed the abstract to make clearer that the draft gives guidelines for interoperability intead of alternative designs.
My comments on security are based on my sentiments and not on factual knowledge.
Text was added on the basis of my understanding and may consequently be off the mark.

Peter
--
Peter van der Stok
vanderstok consultancy
mailto: [e-mail]
www: www.vanderstok.org
tel NL: [phone]     F: [phone]

________________________________
The information contained in this message may be confidential and legally protected under applicable law. The message is intended solely for the addressee(s). If you are not the intended recipient, you are hereby notified that any use, forwarding, dissemination, or reproduction of this message is strictly prohibited and may be unlawful. If you are not the intended recipient, please contact the sender by return e-mail and destroy all copies of the original message.