Re: making our HTML+RDFa queryable

I the documents would be using XHTML instead of HTML, you could parse
the RDFa metadata using a simple XSLT stylesheet. Yet the browser
vendors went for HTML5 and made parsing hard.

Interestingly, those are the same companies that are able to harvest
documents on Web scale, write own parsers, and manage large graphs of
structured data.

We are giving away the dog food yet have a hard time eating it
ourselves. Which brings us back to the need for decentralization.

On Tue, Jan 24, 2017 at 12:09 AM, Ruben Verborgh
<Ruben.Verborgh@ugent.be> wrote:
> HI Kingsley,
>
>> Do  we need to stop at RDFa?
>
> Probably not, lots of things to do.
> But we'll need adequate parsers as well;
> I wasn't really successful with any23 for instance,
> it it didn't work well on HTML5.
>
>> Fundamentally, use of <script/> for
>> embedding structured data islands in HTML is a new pattern that should
>> be added to the Linked Data deployment mix.
>
> Although this can be useful indeed,
> I'm not a big fan myself of this pattern.
> The advantage of RDFa is its structural coupling to content,
> which can be exploited in lots if interesting ways.
> I personally would use conneg for large blurbs of RDF.
>
>> As an example, our Structured Data Sniffer [4] browser extension doesn't
>> have a preference with regards to HTML+RDFa, HTML+Microdata,
>> HTML+JSONLD, HTML+Turtle, or raw Turtle, JSON-LD, RDF-XML
>
> Cool, is there also a standalone parser?
> The pipeline would become more powerful with this one,
> as it currently only supports raw Turtle and RDFa.
>
>> BTW -- I was actually going to make a demo using your page but I hit a
>> TLS issue that can be recreated via:
>>
>> curl -kIL https://ruben.verborgh.org/
>
> Thanks for reporting, but I can't reproduce with curl 7.51.0.
> My TLS settings purposely avoid weaker algorithms
> to obtain an A+ grade on https://www.ssllabs.com/ssltest/.
>
> Best,
>
> Ruben
>

Received on Tuesday, 24 January 2017 14:43:32 UTC