RE: New Specification Published!

Rob-

For what it's worth I also favor yours and Herbert's suggestion for
namespace:

  http://www.w3.org/ns/openannotation#

>From the outset we have wanted to keep the size of the vocabulary small,
i.e., limiting ourselves to specialized, annotation-specific semantics -- no
inventing or reinventing semantics beyond the bare minimum necessary to
generically model and share annotations. 

I also do not think we need to preserve the potential for an extension
namespace. Over time, there may be value in adding ancillary classes and
properties for narrow use cases or for use in specific domains in
conjunction with the Open Annotation, but if not core to Open Annotation
(generic modeling & sharing of annotations) I've concluded from our
experiment with 'extension' over the summer that these should be developed
and maintained separately under distinct rubrics.

This logic also suggests 'openannotation' rather than 'annotation'. We know
there are features and aspects of annotation important in one specific
domain or another that are left out of what we've done. The 'open' makes
clear that we are limiting our model and vocabulary to what is common to
annotation in multiple domains and what is needed to share and interoperate
over annotations spanning multiple domains and use cases. Using simply
'annotation' would be a bit arrogant, actually.

Preference of 'openannotation' instead of 'oa' is more a matter of personal
preference and acronym overload. (And so, not a strongly felt position.)

Tim Cole
University of Illinois at UC


-----Original Message-----
From: Ivan Herman [mailto:ivan@w3.org] 
Sent: Thursday, February 07, 2013 10:59 AM
To: Robert Sanderson
Cc: James Smith; public-openannotation@w3.org; Antoine Isaac
Subject: Re: New Specification Published!


On Feb 7, 2013, at 17:15 , Robert Sanderson <azaroth42@gmail.com> wrote:

> To try to summarize, the following properties seem desirable for the
namespace:
> 
> * Stable
> * Memorable
> * ... (via) Brand name recognition
> * Short
> * Ends in #
> * Only one / (eg not /ns/openannotation/core#) if I understand Raphael
correctly
> 
> And if I may throw one more consideration in to the mix, it would
> cover the memorable and short desiderata if the namespace followed
> exactly the pattern of a TR/ path for the specification.
> Thus if the spec were to have its final home at www.w3.org/TR/OA/ then
> it would make sense (to me) for the namespace to be www.w3.org/ns/OA#
> 
> And then replace OA above with whatever is deemed most appropriate for
> both paths.
> 
> The suggestions:
> * OA
> * oa
> * OA-core
> * oa-core
> * annotation
> * openanno
> * openannotation
> 
> Herbert and my preference would be:
>    http://www.w3.org/TR/openannotation/
>    http://www.w3.org/ns/openannotation#

As I said in one of my previous mails, it should not be both. It should
either be '#' based, ie

http://www.w3.org/ns/openannotation#

meaning you guys give me (or Phil) one file with the full vocabulary and we
install it, or it is

http://www.w3.org/ns/openannotation/

meaning you guys give me (or Phil) one file per term/concept and we install
them.

But it cannot be both. And it is obviously way more simple to manage the '#'
format.

(I do not have a strong opinion on the oa vs. OA vs. openannotation)

Ivan


> 
> 
> Dewitt Clinton once (good naturedly) chided me that OpenSearch
> succeeded where SRU/CQL did not was not due to technology, but due
> simply to the name.  I think that we have a reasonably well known
> "brand" with Open Annotation, and that it would be shame to not
> capitalize on it for the sake of saving a few bytes.
> 
> Rob
> 


----
Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
FOAF: http://www.ivan-herman.net/foaf.rdf

Received on Thursday, 7 February 2013 17:22:42 UTC