- From: Doug Schepers <schepers@w3.org>
- Date: Mon, 20 Jan 2014 15:42:15 -0500
- To: Bob Morris <morris.bob@gmail.com>
- CC: Robert Sanderson <azaroth42@gmail.com>, Paolo Ciccarese <paolo.ciccarese@gmail.com>, Leyla Jael García Castro <leylajael@gmail.com>, Ivan Herman <ivan@w3.org>, public-openannotation <public-openannotation@w3.org>
Hi, Bob– On 1/20/14 2:19 PM, Bob Morris wrote: > +1 > > As an aside quibble re Doug's path toward an example, and > acknowledging that it is synthetic: (1) I see no quotation of either > the cited wikipedia article or, within that, the given text attributed > to Vannevar Bush. (2) I doubt that the wikipedia \article/ should be > typed as a Person which is what it might come out to be, all other > things being equal. Well, I tried to link to Vannevar Bush's twitter account, but he doesn't seem to have one; maybe he only uses Facebook? :P More seriously, I admit that my approach and expression may have been flawed, but I see a solid use case in deconflating the source of a quote from the identity of the person who made the quote, and doing so in a distributed way. I other words, I was quoting an article written by Bush (the actual source doc is the Atlantic, of course), and separately pointing to a relatively canonical URL for Bush's identity (one of many such sources that may or may not be maintained by the author themselves). > It would be good to have a plain text explanation, devoid of > serialization examples, of who is annotating what resource, what part > the annotator is annotating, and whether they mean to be saying > something about the resource or something about the content of the > resource along with what they are trying to say. Then let the > serialization begin! Does the above satisfy that request? > The targeted audiences (web developers) not withstanding, the one > thing that is immutable is that something is not an OA Annotation if > it is not a graph. My goal was indeed to make something that could be used as structured raw material for a distilled graph. > Most (?) of the effort of the OA CG has gone into > RDF representations, so CG many CG participants could probably more > meaningfully comment when confronted with an RDF serialization, > leaving others to argue about the accuracy of HTML representations and > the semantic agreement between them. Thus my +1 is really +0.9, since > what I see emerging is more like working iteratively from both sides > toward a use case for HTML serialization. Yes, I am probably incapable of expressing in RDF what I want to represent... and I think that should be taken as significant. I think relatively few people would be able to. I'm aiming toward a syntax and expressive power that is simple enough for the average web developer to create. That's what we have computers for: to extract and extrapolate patterns that would be tedious to do by hand; I realize that computers are dumb, and we have to meet them halfway; but people are dumb, too, and they are more important than computers, so let's make the computers work a little bit harder than the people. Regards- -Doug
Received on Monday, 20 January 2014 20:42:34 UTC