Ivan, thanks for starting this discussion.

Dan Brickley & Danny Ayers from the SemWeb community helped us with
this at the time.

would it be possible for you to post this on the SemWeb list, I
believe I've seen more recent posts by them there & they pay more
attention to that? it would be nice for people there to be reminded
about metalink :) (been a few years) & they can probably address your
concerns there better.

please keep us posted tho, & we'll update any files necessary/add info
to the site.

On Tue, Sep 28, 2010 at 12:55 PM, Ivan Shmakov <[email protected]> wrote:
>>>>>> Anthony Bryan <anthonybr...@…> writes:
>
> […]
>
>  >> This way, some Metalink documents may be read as RDF/XML documents,
>  >> provided that the following two superficial changes are made:
>
>  >> • the top-level ‘metalink’ element be replaced with ‘rdf:RDF’;
>
>  >> • all the attributes are given namespace prefixes.
>
> […]
>
>  >> Unfortunately, most of the valuable elements of Metalink are
>  >> specified as both having attributes, and being direct descendants of
>  >> the ‘file’ element, which makes them incompatible with RDF/XML.  In
>  >> particular, these are: ‘metalink:pieces’, ‘metalink:hash’,
>  >> ‘metalink:metaurl’, ‘metalink:origin’, ‘metalink:publisher’,
>  >> ‘metalink:signature’, ‘metalink:url’.
>
>  >> While this particular issue could be solved by providing a separate
>  >> Metalink XML parser to the RDF tool in question (or by providing an
>  >> XSLT transformation), it isn't clear as to whether the elements
>  >> above should name (typed) RDF nodes or arcs (or both?)
>
>  > see also
>  > 
> http://groups.google.com/group/metalink-discussion/browse_thread/thread/566c588f85ed5b32/9d427439d57a3d4d
>
>  > (note: this covered metalink 3, not the current version/RFC)
>
>        Thanks for the pointer!
>
>        So, I see that that variant solves the issues above as follows.
>        (There, I imagine that the changes suggested are for the current
>        version of the Metalink XML specification.)
>
>        The top-level ‘metalink’ element is replaced with ‘Metalink’
>        (i. e., as per the common convention, a type; thus, the element
>        represents a typed node.)  Note that an ‘rdf:RDF’ element is
>        also present, while it's actually optional in this case (as per
>        2.6 of the RDF/XML specification.)
>
>        All the attributes are namespace-prefixed.
>
>        The ‘file’ element (under ‘metalink’) is replaced with a ‘file’
>        arc leading to a ‘File’ typed node.  The ‘pieces’ element is
>        treated in a similar way.
>
>        The ‘url’ element's destination is specified with
>        ‘rdf:resource’, so this element designates an RDF graph arc,
>        leading to a note, describing the resource the URI names.  That
>        way, the ‘location’ and ‘preference’ arcs are associated with
>        that resource (irrespective of the ‘File’ in question), which
>        is, to my mind, slightly inappropriate.
>
>        The ‘hash’ element is replaced with a stack of arcs, namely:
>        ‘md5_hash’, ‘sha1_hash’, ‘ed2k_hash’, ‘sha1’.  While the last
>        one leads to a blank node, containing an ‘rdf:value’ with the
>        ‘hexBinary’ representation of the hash, all the other ones lead
>        to the value directly.
>
>        Note that, thanks to this approach, the namespace for possible
>        hash algorithms is effectively the namespace of URI's.  It may
>        not be the most concise expression, but it for sure gives a room
>        for extensions.  That being said, I'd rather like a single URI
>        (Qname) per algorithm.
>
> --
> FSF associate member #7257
>



-- 
(( Anthony Bryan ... Metalink [ http://www.metalinker.org ]
  )) Easier, More Reliable, Self Healing Downloads

-- 
You received this message because you are subscribed to the Google Groups 
"Metalink Discussion" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/metalink-discussion?hl=en.

Reply via email to