Hello,

There seem to be a few issues with the current RDF output from OpenLink
Data Spaces.

1. In the RDF/XML SIOC output, like [1], the value for the sioc:content
   property is incorrect. sioc:content is for unformatted (plain-text)
   content[2], but ODS exports the HTML and *doubly* encodes it (ending
   up with things like "<"). The same applies to wiki pages like
   [3]. The SIOC spec recommends that AtomOWL or the RSS 1.0 Content
   module be used for formatted content.
2. When comments to blog entries are exported, their authors are given
   their OpenIDs as URIs. For example, see [4] and the foaf:maker
   assertion therein. This is probably incorrect, as OpenIDs are
   generally URIs for web pages, and thus cannot identify people.
3. The XHTML+RDFa generated for the pages is not well-formed[5], making
   it difficult to use with off-the-shelf RDF libraries (e.g. RDFLib
   refuses to deal with it).

[1]
http://www.openlinksw.com/dataspace/[email protected]/weblog/[email protected]%27s%20BLOG%20%5B127%5D/1613/sioc.rdf
[2] http://rdfs.org/sioc/spec/#term_content
[3]
http://virtuoso.openlinksw.com/dataspace/dav/wiki/Main/VOSDownload/sioc.rdf
[4]
http://www.openlinksw.com/dataspace/[email protected]/weblog/[email protected]%27s%20BLOG%20%5B127%5D/1530/4646/sioc.rdf
[5] http://validator.w3.org/check?uri=http%3A%2F%2Fwww.openlinksw.com%
2Fweblog%2Fkidehen%40openlinksw.com%2F127%2Findex.vspx&charset=%28detect
+automatically%29&doctype=XHTML+%2B+RDFa&group=0

-- 
Vasiliy Faronov


Reply via email to