On 1/24/14 7:16 AM, John F Sowa wrote:
> Krzysztof,
>
> No declarative or procedural language in the history of computer
> science and applications has as many notations as RDF and OWL.
> That multiplicity is a symptom of a bad design.
>
> In fact, even the two designers -- Guha and Bray -- admitted that
> RDF notation was bad. As Tim Bray said, "It's the syntax, stupid." (01)
RDF's abstract subject->predicate->object syntax wasn't the problem, it
was simply the poor decision by the W3C to use RDF/XML as the sole
notation for representing RDF statements. Even worse, RDF/XML and RDF
relation semantics became conflated; thus, anyone that ventured to
express subject->predicate->object statements in any other notation
ended up being rebuked (on a lucky day) as not actually representing RDF
semantics at all. (02)
Trying to reduce RDF Semantics to a specific Syntax Notation (or
concrete syntax in RDF spec parlance) is the sad legacy of RDF/XML that
wasted 13+ years !! (03)
>
> JFS
>>> That is why they [Google] use JSON instead of the XML-based notations.
> KJ
>> Just for clarification, RDF is not restricted to XML. You can also use
>> N3. There is also a JSON format for Linked Data called JSON-LD.
> JSON, by the way, is just LISP notation with brackets and curly braces.
> LISP was, in fact, Guha's preferred notation. And JSON was designed
> by Netscape -- where Guha and Bray were employed at the time. (04)
Yes, but JSON also has readability challenges. Nothing about it makes
RDF statements and relation semantics clear. It still leads to
programmers focusing on syntax notation without really understanding or
caring about relation semantics. (05)
Today, I tend to encourage people to understand RDF syntax and semantics
through Turtle Notation [1] and then (if they choose) write programs
that work with JSON bases notations (e.g., JSON-LD) of RDF statement
encoding and decoding. (06)
Personally, I do everything in Turtle because it enables me to be
extremely productive using a basic text editor. I write all my RDF by
hand using Turtle.
>
> But many voters in the W3C were in the grip of an untested ideology:
> edict XML for everything. I had been using GML at IBM since the 1970s,
> and I still use HTML for all my word processing. (I use OpenOffice or
> LibreOffice for converting HTML to .doc or .pdf format.)
>
> But a notation that's good for word processing is a disaster for
> logic, ontologies, data storage, data transmission, human factors,
> education, and official standards.
>
> John (07)
Amen!! (08)
Links: (09)
[1] http://www.w3.org/TR/2014/PR-turtle-20140109/ -- Turtle Notation
[2] http://bit.ly/1dUSAFG -- description of RDF (part of a glossary that
I wrote by hand using Turtle) . (010)
-- (011)
Regards, (012)
Kingsley Idehen
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter Profile: https://twitter.com/kidehen
Google+ Profile: https://plus.google.com/+KingsleyIdehen/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen (013)
smime.p7s
Description: S/MIME Cryptographic Signature
_________________________________________________________________
Msg Archives: http://ontolog.cim3.net/forum/ontology-summit/
Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontology-summit/
Unsubscribe: mailto:ontology-summit-leave@xxxxxxxxxxxxxxxx
Community Files: http://ontolog.cim3.net/file/work/OntologySummit2014/
Community Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2014
Community Portal: http://ontolog.cim3.net/wiki/ (01)
|