Re: Experiments With Abbreviated EARL As An Authoring Tool

Chaals wrote:-
> One question is whether you can convert RDF EARL into this
> form. (Most EARL is likely to be produced by tools, so the exact
> details of the syntax are somewhat less important than they are
> for people authoring by hand). Beyond a more compact syntax,
> what would that enable?

It would be possible to convert *some* EARL instances into the abbreviated
form, but I was thinking of it mainly in an Authoring=>Publication
scenario. I agree that since most EARL is likely to be produced by tools,
it means that there is less (i.e. virtually no) use for an abbreviated
syntax, but perhaps some of the spin-off applications (thinking about the
EARL model, and porting the approach over to other RDF languages that *are*
authored by hand) will prove more productive.

The other main benefit for abbreviated EARL is that you can easily validate
it: validation of EARL has been an issue for quite some time, and linear
XML subsets of EARL--whilst not being as expressive as M&S XML RDF--are at
least validatable.

Nick wrote, in reply to me:-
> > So there you have it. If anyone finds this useful and wants me
> > to come up with some sort of schema for it (check out the
> > transformation for the basic semantics of the abbreviated form
> > for now, or ask me), I may cobble one together.
>
> Not sure about that.  I wouldn't use it directly, because it doesn't
> express valet's internal workings (at least, not now).

Right: I wasn't proposing the abbreviated format for storing EARL: it's not
as expressive as XML RDF and N3, and it's not as efficient for reports with
an {o{s, p}} optimal model (as you pointed out to me). In other words, I
don't expect people to use it for interchange; rather, I think it would be
useful for people producing a lot of EARL by hand, or even just a lot of
EARL that they don't want to have to store using such a verbose format.

Nadia wrote:-
> As for the reification thing, I agree with Nick in that it seems to have
> lost its structure.  How is the machine reading that supposed to know
> what the relevant bits of your statement are?  Is the date failing or the
> Language?  Is the id or the label being failed?

Right; since it's XML, it has no particular model, which is why we use RDF
for EARL in the first place. However, all of the answers to the questions
asked above (and therefore the "specification" to some extent for the
abbreivated syntax) are in the XSLT sheet that I attached to the original
email. As I said in the original email, if people expect to actually use
this for anything, I don't mind writing up documentation for it, and some
sort of schema to validate it against.

--
Kindest Regards,
Sean B. Palmer
@prefix : <http://purl.org/net/swn#> .
:Sean :homepage <http://purl.org/net/sbp/> .

Received on Tuesday, 25 June 2002 13:58:07 UTC