Re: Digital assets

Hi and welcome Mark,

It sounds like a both interesting, and very important for many, task you
are embarked upon.

The point we are at in our efforts at the moment is as follows:

   - As per your preference we have constructed a proposal to Schema.org to
   introduce the second proposal.
   - This was submitted via the Schema.org Github as Pull Request 1784
   <https://github.com/schemaorg/schemaorg/pull/1784>
   - Hopefully this will be accepted into the next release of Schema.org -
   which should be soon-ish.
   - That will establish the ability [in Schema.org] to describe any
   thing/entity, that can be otherwise be described using Schema.org, as being
   part of an archive — a major step forward.
   - An archive [collection] and the holding/responsible organisation will
   also be describable.

As referenced in the wiki “[the] *model attempts to separate out the
question of ‘what’ something is, from the fact it is in an archive and
therefore has a set of archive specific properties related to it*.”, which
brings me to some of your thoughts.  These relate to issues that are
relevant right across the many domains addressed by Schema.org, independent
of if a described item resides within an archive or not.  My initial
thoughts being:


   - Digital Access Rights — Schema.org’s core approach is to provide
   methods of publicly describing resources for publicly sharing on the web to
   aid discovery.  This area I believe could benefit from some Schema.org
   enhancement in being able to describe the type of access required to a
   resource (both digital as in access tokens etc, and physical e.g. ’photo id
   needed’).  Delving deeper into actual mechanisms and token values etc., is
   probably beyond the scope of a generalist vocabulary such as Schema.

   - authority — As with the web in general ‘authority’ stems from the
   reputation and authoritative nature of the Person/Organisation asserting
   something.  Authority in the library world, where much of my work has been,
   comes from stating that my [Individual library] description of a
   Book/Article/ImageObject is ’sameAs’ the description provided by The
   Library of Congress, WorldCat, Europeana, WikiData, etc.  Thoughts in this
   area, and for example the nature of trust, are being discussed in a few
   areas of Schema.org, especially with respect to fake news and fact
   checking, and 'ClaimReview' - see Issue #1828
   <https://github.com/schemaorg/schemaorg/issues/1828>.

   - related content archives  — The general use of ’sameAs’, ‘about’,
   ’subjectOf’ when applied to individual items or the collections they are
   ‘partOf’ often are sufficient to derive what you are looking for here.
   There have been been suggestions for a ‘related’ property, especially for
   Event and CreativeWork but have not been successful as the term meaning is
   so broad and difficult to define in a meaningful and useful way.

   - trustworthiness — this is all part of the same scope as ‘authority’.

   - Urls — Not sure what you are getting at here.  Every Thing in
   Schema.org implicitly has a url, and explicitly with the ’url’ property can
   have several.  By use of ’sameAs’ any Thing can be related to a description
   of the same thing elsewhere.

   - tags — String based tags are discouraged (although, as a hangover from
   basic html markup, CreativeWork has the ‘keywords’ property) in favour of
   supplying one or more Thing references (preferably as URLs) in the ‘about’
   property which has the potential for creating a graph of useful
   relationships.


As we move on beyond the initial implementation of our proposal in
Schema.org, and hopefully the appearance of Schema.org on sites describing
archives and their resources; we I am sure will discuss the practical
aspects of describing resources based upon experience.  Most if not all the
issues you raise will be part of those discussion, that archivists can
bring to their experience to, but in most cases they are/will be relevant
to much broader domains.

~Richard.

Richard Wallis
Founder, Data Liberate
http://dataliberate.com
Linkedin: http://www.linkedin.com/in/richardwallis
Twitter: @rjw

On 29 March 2018 at 01:10, Mark Chipman <markchipman@gmail.com> wrote:

> I'm a new member of the working group.   I joined this group as I'm
> working on solutions for Alzheimers disease and seeking to address the
> needs for providing useful memory queues to help with coping.  The data
> represented in this regard, would serve to rekindle memories (IE provide a
> way to help associate and remember events, facts, images, recordings, etc)
> for Alzheimers sufferers.
>
> The ultimate goal would be provide a means to catalog any kind of digital
> data as cursory steps to the development of all sorts of technologies and
> memory aid solutions that could be developed; provided that archives of
> digital assets could yield such information in a searchable semantic way.
>
> So simply, I would like to develop a way to offer schema.org friendly
> digital assets stored in repositories on the web.  All of the content needs
> to be crawl-able and index-able.
>
> I have had a chance to review the proposals presented to date.  I do like
> the second proposal as it lends nicely to adoption into the schema.org
> graph.
>
> After reading the proposals, its seems clear that the requirements for
> digital assets have not been addressed as far as I can tell in regards to:
>
>    - Digital Access Rights (authentication and authorization)...
>    basically the same requirements necessary for access permission of web
>    resources for securing any other type of content... think acess tokens
>    (OAuth, SAML, or OpenID) as well as claims to control access.
>    - authority -  the establishment of the archived content's nature as
>    authoritative (content owners/official record bearer repositories) versus
>    unauthoritative (available collections that serve as alternate repositories
>    of the same content)
>    - related content archives - collection of one or more other locations
>    of similar or related content
>    - trustworthiness (the accuracy of the content) - perhaps where the
>    content originally came from
>    - urls - collection of one or more locations of the online repos (CDNs)
>    - tags - useful keyword identifiers that would promote filtering
>    abilities  and allow data ETL into searchable data stores
>
> These are just some initial thoughts, but I would like to bring them up
> for discussion.  Perhaps the architypes schema may not be the best fit for
> this kind of need... but it appears to be the most closely matched in terms
> of relevance.
>
> I appreciate all feedback.
>
> Regards,
>
> Mark Chipman
>