Re: Motivation as Property or Value

Hi, folks–

Someone mentioned to me offlist that it seemed like I was trying to 
disrupt consensus in this latest thread, so I wanted to be perfectly 
clear in my intent.

Personally, I can live with the change to the data model that Rob is 
proposing; while it differs in some areas to what I'd actually proposed 
as a compromise, I feel that it is simple and usable enough that it's 
good enough.

I think we should integrate these changes into the data model spec, and 
publish an updated draft right away. I suspect that we are pretty close 
to a general data model that could be adopted widely (or at the very 
least, one we could credibly shop around to major vendors).

I truly appreciate all the effort from everyone in bringing us to this 
point.


All of that said, as Tim has noted, there was an outstanding concern 
that an annotation implementer (OpenGov Foundation) brought forward, and 
from a short off-list discussion with them after last week's telcon, I 
found out that the compromise solution I proposed, and the final version 
that Rob solidified, was not acceptable to them, and that as a 
consequence, they were unlikely to adopt the data model.

This is a big deal to me! First, it means that we don't have consensus 
in the working group (and I don't mean the pie-in-the-sky "unanimity" 
consensus, but the more pragmatic "I-can-live-with-it" consensus that 
W3C practices); second, if OpenGov Foundation can at all be looked upon 
as representative of an implementer who is not using the RDF/Linked-Data 
stack (and I'm not claiming that they are representative, only that they 
may be), then their aversion to this aspect of the data model might be a 
canary in a coalmine.

However, in a separate off-list conversation, Benjamin suggested to me 
that he had a point of view that might convince OpenGov Foundation that 
this formulation data model is actually acceptable; when he explained it 
to me, I came away fairly convinced.

My goal here is to air these different points of view on an archived 
public forum, rather than in ephemeral off-list discussions that are 
only passed around as rumor. If we can provide a calm and reasoned 
rationale, on this list, for why this data model is not only good 
enough, but indeed why it might be better than Bill's model, then people 
who later ask us to defend our technical decisions (and if we're lucky, 
we will be questioned on them, because it will then mean that someone 
outside our small group is paying attention), we can provide not only an 
answer that satisfies the RDF/Linked-Data folks, but also the people who 
avoid RDF/Linked-Data.


Rob made a Call for Consensus. This is not simply a pro-forma request 
for +1s, but is the time to drive toward consensus (which we currently 
don't have) through conversation, until 1 September, when the CfC ends.

I honestly believe that we can get consensus here. If we don't… well, 
that will be unfortunate, but fortunately, we have a path forward though 
the W3C Process [1]:

[[
3.3.1 Managing Dissent

In some cases, even after careful consideration of all points of view, a 
group might find itself unable to reach consensus. The Chair MAY record 
a decision where there is dissent (i.e., there is at least one Formal 
Objection) so that the group may make progress (for example, to produce 
a deliverable in a timely manner). Dissenters cannot stop a group's work 
simply by saying that they cannot live with a decision. When the Chair 
believes that the Group has duly considered the legitimate concerns of 
dissenters as far as is possible and reasonable, the group SHOULD move on.

Groups SHOULD favor proposals that create the weakest objections. This 
is preferred over proposals that are supported by a large majority but 
that cause strong objections from a few people. As part of making a 
decision where there is dissent, the Chair is expected to be aware of 
which participants work for the same (or related) Member organizations 
and weigh their input accordingly.
]]


My understanding is that Chris (and through him, Bill) are the only ones 
who've said they can't live with this solution (and they've done so 
graciously, saying they won't block the group moving forward). Again, 
I'm not blocking it, and I'm even in favor (albeit with some 
reservations) of the new proposed model.

So, my suggestion is that let the conversation play out until the end of 
the CfC, and unless we get other serious objections, we should integrate 
the new proposal into the Data Model spec, and publish an updated draft.


[1] http://www.w3.org/2014/Process-20140801/#Consensus

Regards–
–Doug

On 8/28/15 10:03 PM, Robert Sanderson wrote:
> Hi Bill,
>
> Thanks for your input into the discussion!
>
> On Fri, Aug 28, 2015 at 5:33 PM, Bill Hunt <bill@opengovfoundation.org
> <mailto:bill@opengovfoundation.org>> wrote:
>
>     I understand that the consensus is that my suggestion wouldn't work
>     because you'd have to define every role as it would appear - which
>     is hard since we can't predict everything anyone would ever want to
>     do.  My argument was that 99% of roles could be defined in a very
>     narrow set of a half dozen or so; anyone who's doing something else
>     or wants to create a new role wouldn't be handled by a "standard"
>     consumer so they'd be off the map anyway.
>
>
> I think that, happily, we can easily put this to rest.
>
> There are already a dozen motivations defined in the model [1], and your
> use case did not fit into those 12, as there isn't a motivation that's
> explicitly the text to replace target text with.  So in order to fulfill
> your use case, you would be outside of the standard too.  The 12 were
> already trimmed down from longer lists of motivations by looking across
> many use cases and previous research in the OAC and OA-CG days.
>
> In the IIIF use case, we had to define a new motivation (painting), and
> that's likely insufficient and we'll need at least 3 further motivations
> as narrower concepts of it to clarify expected client behavior
> (transcribing, translating, editing [in the sense of creating a textual
> edition of a work]) [2].
>
> So, I hope that you can see that 99% of roles really won't fit into half
> a dozen, and hence property-as-role really doesn't work for
> interoperability without the significant overhead of RDF inferencing to
> determine how they relate to hasBody or hasTarget.
>
> Thanks,
>
> Rob
>
> [1] http://www.w3.org/TR/annotation-model/#motivations
> [2] http://iiif.io/api/presentation/2.0/#image-resources
>
>
>
>
>
> --
> Rob Sanderson
> Information Standards Advocate
> Digital Library Systems and Services
> Stanford, CA 94305

Received on Sunday, 30 August 2015 09:46:04 UTC