- From: Kingsley Idehen <kidehen@openlinksw.com>
- Date: Tue, 26 Feb 2013 10:57:01 -0500
- To: public-ldp-wg@w3.org
- Message-ID: <512CDB4D.7020608@openlinksw.com>
On 2/26/13 10:14 AM, Wilde, Erik wrote: > hello all. > > On 2013-02-26 10:37 , "Henry Story" <henry.story@bblfish.net> wrote: >> It seems accepted by the group as a requirement on the LDP specification, >> though it would be good to make it explicit, that LDP must allow >> implementations to follow what I would like to call the intuitive >> requirement, nameley that: > they should be able to follow whatever naming scheme the implementers > happen to prefer. we shouldn't make any statements about naming > conventions in the spec. we can, however, include recommendations and/or > best practices in accompanying documents. > >> URIs MUST be opaque >> ------------------- >> It is usually argued that this is bad because URIs must be opaque. >> But clearly that cannot be the case or else the above turtle would be >> illegal and the URI specification would be mistaken since it goes into >> great detail on this subject, as for example in >> http://tools.ietf.org/html/rfc3986#section-5.2.4 > this specific part of the spec is very detailed because it says how to > resolve relative URIs against base URIs. this indeed is important, and > you're correct that at this level, URIs do have structure: they have path > components and these have semantics in terms of resolving relative URIs. > but that's it, and it's different from prescribing URI patterns for > specific resources in specific media types. > > but like i said above, the point you're making is a useful one and would > be a good addition to best practices, so that implementations that want to > use hierarchical URIs (they don't have to if they don't feel like it) have > a good starting point. > > cheers, > > dret. > > > > Believe or not, s/URI/URL/g makes this all simpler to understand. A URL has specific characteristics e.g., the fact that URLs denote Web Documents. A file system is comprised of files and directories (a kind of file) [1][2][3]. URIs denote anything. Thus, using URI in this manner ultimately injects the breadth of *anything* into the mix. Net effect, it makes what should be simple and intuitive hard to understand. Thus far, this has been all about modelling a RESTful mechanism for creating files and directories all of which you denote using URLs. Of course, a URL is a kind of URI, but at this juncture, we are dealing with the URI subClass, so best to ground the narrative appropriately via use of URL. URLs should be hackable [4] while URIs should be opaque [5]. Links: 1. http://kingsley.idehen.net/DAV/home/kidehen/Public/ -- a directory associated with an assortment of files and sub-directories . 2. http://uda.openlinksw.com/data/turtle/ - directory associated with a collection of Turtle docs . 3. http://virtuoso.openlinksw.com/data/turtle/ -- ditto . 4. http://bit.ly/TZ7uNZ -- Roy's world view which is scoped to Web Documents bearing Web accessible content (the Data Resource) 5. http://bit.ly/SnuECw -- TimBL's world view which is scoped to the URI that denotes anything (including Web Documents). -- Regards, Kingsley Idehen Founder & CEO OpenLink Software Company Web: http://www.openlinksw.com Personal Weblog: http://www.openlinksw.com/blog/~kidehen Twitter/Identi.ca handle: @kidehen Google+ Profile: https://plus.google.com/112399767740508618350/about LinkedIn Profile: http://www.linkedin.com/in/kidehen
Attachments
- application/pkcs7-signature attachment: S/MIME Cryptographic Signature
Received on Tuesday, 26 February 2013 15:57:28 UTC