Dear Mike,
Sorry for the late response. I have been asking around in my research group and they have informed me that there are several options and one of them is very similar to the ones already mentioned.
•HTTP RangeThere are various protocols and conventions around <link> and HTTP headers, VOID descriptions for a domain, semantic site maps etc., but it really depends on the use case... (And whatever you do, if someone starts to say anything about HTTP Range 14 to you in response to this question... run)
• Schema.orgA page that you're in control of? Try something like schema.org.Microformats and RDFa are other keywords you'll want to search for.
• data.soton.ac.ukdata.soton.ac.uk uses the HTTP request to suggest a version of the information requested. For example, if you were to call Building 32's URI http://id.southampton.ac.uk/building/32 from a web browser you'd get a 303 (see other) redirect to the HTML page of information about building 32. If you were using an RDF library (eg Graphite) and you called the same address you'd get a 303 to the RDF data about building 32. Back in the old days people used HTML meta headers to indicate the existence of RDF data, eg <link rel="alternate" type="application/rdf+xml" title="RDF Description" href="./32.ttl" />
Some people at my university have been working hard promoting auto discoverable organisational profile documents to be a thing. See http://opd.data.ac.uk/ for information.
Best wishes,
Javier Pereda
> Date: Mon, 22 Jun 2015 10:18:51 +0100
> From: [log in to unmask]
> Subject: Re: Structured data: over here?
> To: [log in to unmask]
>
> Thanks Frankie, that's useful. Interesting that there isn't a kind of
> global thing which covers all "structured data" bases...
>
> tt
>
> Mike
>
>
>
> _____________________________
>
>
> *Mike Ellis *
>
> Thirty8 Digital: a small but perfectly formed digital
> agency:http://thirty8.co.uk <http://thirty8.co.uk/>
>
> * My book: http://heritageweb.co.uk <http://heritageweb.co.uk/> *
>
>
>
> Frankie Roberto wrote:
> > Mike said:
> >
> >> Is there a standard notation / way of flagging up on a web page that there is programmatic / structured data available for that page?
> >>
> >> AFAIK this doesn't exist as a thing - and it seems odd that it doesn't - not just our sector but for anyone. I suppose the nearest analogy I can think of is something like RSS which has an "alternate" rel tag which basically says "go here for a structured version".
> >
> > You can use the same syntax as for RSS / Atom autodiscovery: the<link> tag with rel=alternate. (see https://developers.whatwg.org/links.html#rel-alternate<https://developers.whatwg.org/links.html#rel-alternate>)
> >
> > ie:
> >
> > <link rel="alternate" type="application/json" href="http://www.vam.ac.uk/api/json/museumobject/O12345">
> >
> >
> > There aren't tons of tools which recognise that though, so I'd also recommend putting a link to the JSON / XML representation somewhere on each page (footer?) if you want it to be more discoverable (by humans).
> >
> >
> > Another alternative is to serve both the HTML and JSON representations from the same URL, using content-negotiation to determine which representation to serve (based upon the 'Accept' header in the HTTP request). See https://en.wikipedia.org/wiki/Content_negotiation<https://en.wikipedia.org/wiki/Content_negotiation>
> >
> >
> > Finally there's an emerging convention that the JSON / XML representations be served at a URL equal to the regular HTML page but with ".json" or ".xml" added to the end (as if it were a filename extension).
> >
> > Frankie
> > ****************************************************************
> > website: http://museumscomputergroup.org.uk/
> > Twitter: http://www.twitter.com/ukmcg
> > Facebook: http://www.facebook.com/museumscomputergroup
> > [un]subscribe: http://museumscomputergroup.org.uk/email-list/
> > ****************************************************************
>
>
> ****************************************************************
> website: http://museumscomputergroup.org.uk/
> Twitter: http://www.twitter.com/ukmcg
> Facebook: http://www.facebook.com/museumscomputergroup
> [un]subscribe: http://museumscomputergroup.org.uk/email-list/
> ****************************************************************
****************************************************************
website: http://museumscomputergroup.org.uk/
Twitter: http://www.twitter.com/ukmcg
Facebook: http://www.facebook.com/museumscomputergroup
[un]subscribe: http://museumscomputergroup.org.uk/email-list/
****************************************************************
|