Re: A Rough Guide to Notation3

>>>Paul Prescod said:
> Dave Beckett wrote:
> > Blame XML for XML's efficiency, not applications of it.  
> 
> XML chose human readability over efficiency. Doesn't it seem kind of odd
> to build something unreadable on top of something inefficient so that
> you combine the worst of both worlds?

I don't think you are going to get anywhere trying to roll back the
use of XML for markup, semi-structured data on the web, whatever the
efficiency issues.  See the xml-dev flamewars.

It seems you have some issues with XML itself, unrelated to whatever
format it is used for.

> 
> >...
> > Apologies to N3 for that ;)
> 
> I find N3 reasonably easy to read.

All of it?  There's lots of punctuation that tends to dominate for
more complex stuff.  Hence the joking "perl" syntax comment I made
previously - if you want scribbleability of power in the language,
you lose some readability.

> 
> > ...
> > So how about protocols for machines such as email, HTTP which are not
> > machine-optimized binary?  
> 
> Those protocols are *very* human readable and are read by humans
> (programmers, administrators and sometimes system administrators) every
> day.

Yes, so like RDF syntaxes these protocols are for machines *and* people.
In which case, where does your efficiency claim lie?   Formats
that people can also read win over your request for binary formats
for machine efficiency.

> > ... That was a major reason the Internet and
> > Web won over earlier systems.  Such as those built with ASN.1 for
> > "efficiency" although having no "readability".  Promoting that is
> > going backwards.
> 
> I agree. That's why I think it is strange to say that RDF/XML is "not
> designed for human consumption." It would be a terrible mistake to
> create a web standard in 2002 and not design for human consumption.
> Making it ALSO inefficient would compound the mistake.

It wasn't me who said that, and I don't agree with it.  RDF was
created in 1998/1999, just after XML and we are revising now, not
replacing.  This format is no more inefficient than XML used in any
other specification.  It is an XML format, which is easy to read
about in any bookstore and it is trivial to find good XML tools that
can help with your efficiency claims.

Dave

Received on Friday, 23 August 2002 16:24:33 UTC