[w3photo] Alpha Spec

I've been reading back through all the emails. We are exploring 
interesting and important issues.

The following is the default specification of the *alpha* version of 
the repository, based on what I've gathered from the emails and my 
understanding of the decisions.

Feedback encouraged. If possible, identify feedback as: (a) correction 
for alpha version, (b) open issue, (c) pre-www2004 
correction/improvement, (d) post- WWW2004  improvement/extension, (e) 
brilliant observation and general commentary. Thanks.  --Greg

1. Document Info
2. Exec Summary
3. Alpha Specification
	a) Servers
	b) Seed photos, data
	c) Default Vocabularies and Namespace
	d) Default Policies
	e) Default User Interface

__1. Document Info__

Specification for Alpha version of W3Photo: A Semantic-Photo History of 
the W3 Conferences
January 25, 2004
Version 0.9
URL: http://esw.w3.org/topic/W3PhotoSpec

__2. Exec Summary__

The initial system will consist of two temporary "official" servers. 
W3photo.org (Greg Elin) will host the photographs. 
http://sw1.ilrt.org/discovery/2004/01/w3photo/ (Libby Miller) will host 
RDF annotations.  It is assumed with these two servers online, 
Southhampton can bring up an alpha AKTive Space for the project on a 
third server within their control.  The "front-door" website will be at 
w3photo.org

We are working only with JPEGs through www2004. We've identified a few 
links to existing photos online, some with associated RDFs. Everyone is 
welcome to contribute some photos to give us an initial set of 100+ 
photos. (It is expected these photos will eventually be contributed 
under the license, but we will also be prepared to delete them and/or 
re-contribute them.)

Default vocabularies (ontology definitions) have been drafted. Effort 
was made for compatibility with existing vocabularies.  When faced 
between defining something abstractly or specifically to the context, I 
think we've been decided to be specific to this exercise.  The drafted 
vocabs are:
	(Photos)
		- vocab for entire image
		- vocab for part/region of image (describing region + what's depicted)
	(Conference)
		- vocab for conferences
		- vocab for W3 conference materials, concepts

The proposed Creative Commons license is royalty-free, attribution, 
non-commercial, and share-alike. People contributing photos will need 
to provide some identifying information. The proposed privacy policy is 
to remove photos if a depicted party request removal.

AKTive Space is expected to be the default UI for browsing the 
repository by the time we get to www2004. A variation of Fotonotes will 
be the default UI for regions of images. A simple form will generate 
RDF for the image. (The server code will write data appropriately both 
into the JPEG via and to the annotation server.)  A description of UI 
issues can be found at http://esw.w3.org/topic/UserInterface


__2. Alpha Specification__
  ___a) Servers__

http://www.w3photo.org - front door website
http://www.w3photo.org/photostest/ - root directory of repository 
(online, but not finished)
http://sw1.ilrt.org/discovery/2004/01/w3photo/  - interface to add RDF 
to public database

The initial system will consist of two temporary "official" servers. An 
explanatory front door web site and an alpha set of images will be 
hosted at http://www.w3photo.org on a server in the US maintained by 
Greg Elin. RDF annotations for the images will be hosted at 
http://sw1.ilrt.org/discovery/2004/01/w3photo/  maintained by Libby 
Miller in the UK.

Other servers, and other applications like Annotea running on these 
servers, have been discussed. As most of the group has access to their 
own server resources, everyone is welcome to pursue alternatives on 
their respective servers with this single request: allow the agreed 
upon default servers, UIs, and vocabs to lead in the public face of 
this project as we ramp up towards the conference. We need an organized 
approach and public face to encourage involvement from others.

With these two servers online, Southhampton can bring up an alpha 
AKTive Space for the project on a third server within their control. 
This server will provide the general tools for browsing the repository 
and linking the images to conferences, and conference concepts.

___b) Seed photos, data__

We are working only with JPEGs through www2004. It is fine for 
vocabularies to be inclusive of multimedia, but we are staying just 
with JPEGS to start.

Images should be stored in a higher resolution and a web resolution. 
The primary reference to an image should probably be for the web 
resolution.

We've identified a few links to existing photos online, some with 
associated RDFs. Everyone is welcome to contribute some photos to give 
us an initial set of 100+ photos. (It is expected these photos will 
eventually be contributed under the license, but we will also be 
prepared to delete them and/or re-contribute them.)

Because we may want to change a few things in the near future, let's 
keep this alpha collection smallish.

___c) Default Vocabularies and Namespace__

Default vocabularies (ontology definitions) have been drafted. Effort 
was made for compatibility with existing vocabularies.  When faced 
between defining something abstractly or specifically to the context, I 
think we've been decided to be specific to this exercise.

Vocabulary for an image:
  	-  vocab for entire image has been described (Libby, is there a URL?)
  	- vocab for part/region of image (describing region + what's 
depicted) - 
http://www.wasab.dk/		morten/2004/01/image-regions-schema.rdf and sample
		 http://www.wasab.dk/morten/2004/01/image-regions-sample.rdf (Morten 
Frederiksen)

Vocabulary for conference:
  	-  vocab for conference - http://www.mindswap.org/2004/www04photo.owl
		(Jennifer Golbeck)
  	- vocab for W3 conference materials, concepts (coming from 
Southampton)

(Discussions around vocabularies were very interesting with active 
participation from all.  Extra thanks to Jennifer Golbeck for putting 
the generalized conference vocab online; the Southampton team for 
tackling the hairball of W3 conferences; and Jim Ley, Morten 
Frederiksen, and Libby Miller for gathering and sorting out the photo 
vocab discussions; Libby has been a special glue in these discussions 
and put together example RDF.)

___d) Default Policies__

http://creativecommons.org/licenses/by-nc-sa/1.0/   (summary)
http://creativecommons.org/licenses/by-nc-sa/1.0/legalcode (full legal 
code)

Libby's email summarizing policy discussions put forth four suggestions 
which seem like a good alpha version of policy:
	1. 	one license for the core set of photos held
		on server http://creativecommons.org/licenses/by-nc-sa/1.0/
	2. 	remove any photos from the core collection that people don't
		wish to be there, without quibble, as long as either they created
		the photo or are depicted in it
	3. 	undertake	 to correct any errors in the metadata from the core
		collection, and remove identification of people in the metadata if
		requested by them
	4. 	reserve the right to remove photos and metadata from the core
		collection for any reason (but in practice don't do this unless 
there's
		a really good reason)

We are proposing a single license for photos and annotations 
contributed to any of the "official" servers.  The proposed Creative 
Commons license is royalty-free, requires attribution, limits use to 
non-commercial, and share-alike. We are putting this forth as an 
*alpha* license. We need some legal guidance on this (and the privacy 
policy that follows below). There are open issues here.  Please *read* 
the full legal code before raising any comments on the license.

It is important to note that the "core" collection of images are images 
that are contributed by others. Since the images accepted into the core 
agree to the license, which includes attribution, it should be possible 
for people interested in using the images in ways beyond those 
described in the license to contact the contributing party directly.

___e) Default User Interface__

AKTive Space is expected to be the default UI for browsing the 
repository by the time we get to www2004. A variation of Fotonotes will 
be the default UI for regions of images. A simple form will generate 
RDF for the image. (The server code will write data appropriately both 
into the JPEG via and to the annotation server.)  A description of UI 
issues can be found at http://esw.w3.org/topic/UserInterface

(I've got up some Fotonotes stuff at http://w3photo.org/photostest ; 
however it should be much better Monday; just threw some things up 
their today...)

Greg