On 13 December 2011 21:50 Julian Reschke <julian.reschke@xxxxxx> wrote: > On 2011-12-13 20:07, "Mykyta Yevstifeyev (М. Євстіфеєв)" wrote: >> >> ... >>> >>> Maybe it's worth pointing out that this does not apply as verbatim >>> instruction, but as HTML example. >> >> >> I have that - "in source code". Maybe I should add "in HTML source code"? > > > The latter sounds good. > >>> It would be good to confirm with the W3C that this actually is a >>> requirement and not only a suggestion (cc'ing the author of PUBRULES). >> >> >> And I have this as well, as I've introduced this extract with "must look >> like". > > > I'd like to see this confirmed by the W3C. > >>> Such provisions existed in previous versions of Publication Rules as >>> well, so such source text is often found in different W3C documents >>> that predated publication of RFC 5988 significantly. However, >>> 'disclosure' relation type has not been mentioned in RFC 5988 when >>> creating the registry for relation types; nor was it registered >>> separately. >>> >>> I think the paragraph above is misleading. It was not the point of RFC >>> 5988 to define all current link relations (it *did* add existing HTML4 >>> relations and Atom relations to the new registry but that's a separate >>> thing). >> >> >> This may be read as if RFC 5988 is to determine all the rel types that >> were used at the time of its writing, but I don't think the paragraph >> directly implies this. I mean that it hasn't been registered either >> centralized (in RFC 5988) or separately here. > > > Well, it doesn't say anything interesting, and might confuse people to think > this was an oversight in 5988. Just drop it. OK, I will. > >>> 2. 'disclosure' Link Relation Type >>> >>> Whenever the 'disclosure' relation is defined, the target IRI MUST >>> either >>> >>> (1) designate a list of patent disclosures, or >>> >>> (2) refer to a particular patent disclosure made with respect to the >>> material being referenced by context IRI. >>> >>> I think in both cases the patent disclosure(s) apply to the context, no? >> >> >> Yes. I may change to: >> >>> Whenever the 'disclosure' relation is defined, the target IRI MUST >>> either >>> >>> (1) designate a list of patent disclosures, or >>> >>> (2) refer to a particular patent disclosure >>> >>> made with respect to the material being referenced by context IRI. >> >> >> to improve readability. > > > OK. > >>> This section provides several examples of possible use of >>> 'disclosure' relation type. >>> >>> If the page <http://example.org/ipr/meta-spec/> contains a list of >>> patent disclosures made with respect to the specification found at >>> <http://example.org/specs/meta-spec/spec.html>, the latter would have >>> the following fragment of HTML source code: >>> >>> <html> >>> ... >>> Please visit >>> <a rel="disclosure" href="http://example.org/ipr/meta-spec/";> the >>> IPR page</a> for the list of patent disclosures made with respect >>> to this specification. >>> ... >>> </html> >>> >>> Or, in the case of Link header field, the HTTP response would contain >>> the following header field: >>> >>> Link: <http://example.org/ipr/meta-spec/>; rel="disclosure"; >>> title="Patent Disclosures List" >>> >>> (Please note that the actual header field will not contain the line >>> break after 'rel' parameter.) >>> >>> It could if the second line was indented; maybe adjust the example? >> >> >> I have: >> >>> Link: <http://example.org/ipr/meta-spec/>; rel="disclosure"; >>> title="Patent Disclosures List" >> >> >> both lines are indented with 5 Courier white spaces, and so I think my >> explanation in parentheses is correct. > > > Well, HTTP header fields start at column 0 in a message. > > Of course you can indent your examples, but my point was that if you indent > the second line *more*, it would be valid. Like this: > > Link: <http://example.org/ipr/meta-spec/>; rel="disclosure"; > title="Patent Disclosures List" This is OK as well. Mykyta Yevstifeyev > >> ... > > > Best regards, Julian _______________________________________________ Ietf mailing list Ietf@xxxxxxxx https://www.ietf.org/mailman/listinfo/ietf