- From: Larry Masinter <masinter@adobe.com>
- Date: Wed, 21 Jan 2009 17:31:49 -0800
- To: Marcos Caceres <marcosscaceres@gmail.com>, "public-pkg-uri-scheme@w3.org" <public-pkg-uri-scheme@w3.org>, "public-webapps@w3.org" <public-webapps@w3.org>
- CC: Tim Kindberg <timothy@hpl.hp.com>
I'm not sure how "tag:" adds any benefits over "cid:" http://tools.ietf.org/html/rfc2392 as specified in MHTML http://tools.ietf.org/html/rfc2557 especially given the wide deployment and use of cid: within MHTML for this purpose. Yes, using Zip is a different overall serialization than MIME multipart, but aren't the problem spaces similar enough that differences from what is already widespread practice? I have nothing against "tag", except that "tag" goes out of its way to deny any intrinsic semantics for what it identifies, leaving that for some contextual discovery, while "cid" was described as explicitly identifying content which has been labeled as having the content within the package, and thus matches the usage profile better. Larry -- http://larry.masinter -----Original Message----- From: public-pkg-uri-scheme-request@w3.org [mailto:public-pkg-uri-scheme-request@w3.org] On Behalf Of Marcos Caceres Sent: Tuesday, January 20, 2009 12:15 AM To: public-pkg-uri-scheme@w3.org; public-webapps@w3.org Cc: Tim Kindberg Subject: FW: tag: uri scheme Hi All, WebApps-WG recently asked Tim Kindberg, editor of the TAG URI scheme, if could make some comments about the widgets' requirements for a URI scheme. Below, Tim agrees that tag: is probably suitable for widgets, but raises some concerns about i18n and having a resolution algorithm to go from a tag: to a resource inside a widget. Kind regards, Marcos ------ Forwarded Message From: Tim Kindberg <timothy@hpl.hp.com> Date: Wed, 17 Dec 2008 09:15:48 +0000 To: Arthur Barstow <Art.Barstow@nokia.com> Cc: Argh <marcosscaceres@gmail.com>, "www-archive@w3.org" <www-archive@w3.org> Subject: Re: Fwd: tag: uri scheme Hi Art, I've had a quick look through the resources you referenced. It does seem that tag: is a good candidate for your needs (although I have a couple of questions below). You need an id scheme that isn't quite matched by http etc., and it's better to avoid an entirely new scheme. My points are: (1) Don't you say you want internationalised ids? We (I) gave up on IRI-compatible tags some time ago. It's do-able but I got mired in the mud and ran out of time. (2) Somewhere I saw a reference to using a fixed date, 2008, for 'security reasons". Well that's OK as long as any domain name (or email address) that comes before it was indeed assigned to the minter of the tag on 2008-01-01. (3) From my glance at your documents, it seems that you need a resolution algorithm, to go from a tag: to a resource inside a widget. That's OK by the tag spec: there is no default resolution mechanism but any tag usage protocol can define its own. AFAIK, Atom 'id' elements (the biggest user of tag URIs) are chosen to be resolvable internally to feed generators. I hope that that is of some help, even though I've only had a little time to look at your requirements. I'm actually on leave until January now but may be able to respond occasionally in the meantime. Cheers, Tim. Arthur Barstow wrote: > Hi Tim, > > I am a Chair of the W3C's Web Applications WG [WebApps] and in the > context of our Widgets spec [Widgets], particularly Requirements #6 > [Req-6] and #37 [Reg-37], we are interested in the tag: scheme. > > Marcos Caceres, the Editor of our Widgets spec, asked me to contact > to you re this scheme. For a bit of context, please see his email > below as well as a summary of the Widget scheme issue in [Widget- > scheme]. > > If you have any info to share, we would greatly appreciate it. > > Here is one the latest discussion threads we've had on this scheme: > > <http://lists.w3.org/Archives/Public/public-webapps/2008OctDec/ > 0299.html> > > -Regards, Art Barstow > > [WebApps] <http://www.w3.org/2008/webapps/wiki/Main_Page> > [Widgets] <http://dev.w3.org/2006/waf/widgets-reqs/> > [Req-6] <http://dev.w3.org/2006/waf/widgets-reqs/#r6.-addressing> > [Req-37] <http://dev.w3.org/2006/waf/widgets-reqs/#r37.-> > [Widget-scheme] <http://tinyurl.com/69yegh> > > Begin forwarded message: > >> From: "ext Marcos Caceres" <marcosscaceres@gmail.com> >> Date: December 6, 2008 11:56:24 AM EST >> To: "Arthur Barstow" <Art.Barstow@nokia.com> >> Subject: tag: uri scheme >> >> Hi Art, >> I'm wondering if you could do me a huge favor. In light of the tag: >> uri scheme potentially meeting our needs in regards to a widget uri, >> I was wondering if you could take the time to contact Tim Kindberg >> (timothy@hpl.hp.com), who created the tag: scheme, and possibly ask >> him to weight in on the discussion on the packaging list, web apps >> list, and the TAG list. It would be great if you could point him to >> our requirements R6 and R37 and maybe some of the relevant emails. It >> would also be a big help if you could also ask him if he managed to >> resolve i18n issues with tag:; And if no, what would need to be done. >> >> Kind regards, >> Marcos >> >> -- >> Marcos Caceres >> http://datadriven.com.au > -- Tim Kindberg hewlett-packard laboratories filton road stoke gifford bristol bs34 8qz uk purl.org/net/TimKindberg timothy@hpl.hp.com voice +44 (0)117 312 9920 fax +44 (0)117 312 8003 ------ End of Forwarded Message
Received on Thursday, 22 January 2009 01:32:49 UTC