Re: ACTION-920 completed

On 30 Sep 2009, at 02:10, Sandro Hawke wrote:

>
> Hmmm.  It looks like no one has reviewed Axel's edits.  Ooops.
>
> I think we're still okay to publish it, but the level of review is a  
> bit
> less than our usual standard.
>
> > http://www.w3.org/2005/rules/wiki/index.php?title=DTB&diff=11060&oldid=11006
>
> This appears to be just a few of the edits he did as part of  
> ACTION-920.
>
> > I went through the list predicates and functions and adapted them as
> > follows:
> >
> > * I added formal mappings where this was possible more or less
> > straightforwardly in my opinion;
> >    I agree with Sandro that it doesn't make sense to add a formal
> > mapping for the last 4 functions
> >    (union, distinct-values, intersection, except) but I checked  
> these
> > informal mappings again and am
> >    fine with them. I left the informal explanations as well, just to
> > remain (hopefully) least intrusive.
>
> Sigh.  Well, I can no longer read this section, and can no longer  
> vouch
> for its accuracy.  I hope someone who likes subscripts will check  
> Axel's
> work.
>

:-) the changes were meant to make the document more consistent in
itself, anyways, if this section is unreadable then the others are as  
well,
we had several long discussions in the past on whether formal mappings  
yes
or no, let's not go down this road again at this point.

> I think the bits that involve positions are wrong, though, since they
> leave out negative indexing (although they refer to it).
>

I think it is ok, since it says e.g.
"i corresponds to a position" ... where "corresponding possition" is
described in Section 4.11.1, including the negative indexes.


> > * I changed the examples to presentation syntax, since I wouldn't be
> > to happy an undefined
> >    syntax in the examples; even if PS isn't normative, it is
> > sufficiently specified.
> >
> > * I did some more editorial edits, like rewordigns in Section 4.11.1
> > and 4.11.2, which I'd appreciate to be checked.
>
> Those look fine.
>
> > * From my point of view, this is all fine now, except one open issue
> > noted with an editor's note in Section 4.11.4.4:
> >    It seems that the behavior of func:sublist for ?stop before ? 
> start
> > is not specifier.
>
> I think this is just like any other arguments-out-of-domain  
> situation; I
> don't see a need for special handling, except that the domain
> restriction should state that the adjusted-for-negative-indexing stop
> position must be greater than the adjusted-for-negative-indexing start
> position.

reformulated the mapping description, please check and if you are fine,
remove the Editor's note. I am a bit reluctant of packing it in the
domain restriction, since with the negative indexing I am unsure  
whether/how to
state the domain restriction as it would depend on the length of the  
list.

>
> > * BTW: Sorry about my forgetfulness, but did we decide on dropping  
> the
> > Editor's note on rdf:PlainLiteral (Section 4.10) already?
>
> I think so, but at this point I'd rather just leave it.

Can we really leave in Editor's notes?
>
> > Apart from the last two points, DTB should be ready to go, hope I
> > didn't forget anything more.
>
> I think, under the circumstances, it's okay to ship with these two
> editor's notes, even though we usually try to avoid it.
>
>     -- Sandro
>

-- 
Dr. Axel Polleres
Digital Enterprise Research Institute, National University of Ireland,  
Galway
email: axel.polleres@deri.org  url: http://www.polleres.net/

Received on Wednesday, 30 September 2009 05:46:38 UTC