Thanks for these further comments. The Editorial Committee has not
considered making environment a separate entity, but I can raise it for
Depending on your implementation, I can imagine that you could indeed
capture specific environment metadata once, as you suggest, and link to
it from the relevant objects. PREMIS doesn't specify how you should
store the metadata within your repository; it just describes what
metadata should be known and, through the schemas, how this metadata can
be expressed in a standard way for exchange or interoperability.
So, you could store specific environment information (or any other
metadata) just once in your database in whatever form is applicable to
your implementation, link it to the relevant objects and only bring it
together with the rest of the object metadata to output to a PREMIS
schema when you need to. I don't believe there is a requirement to store
the metadata in a PREMIS document.
I expect that this could also apply to the proposed 'policy' entity, so
we would need to further consider if there are advantages to separating
National Library of Australia, Canberra ACT 2600 Australia
t: +61 2 6262 1381 | f: +61 2 6273 2545 | e: [log in to unmask]
From: PREMIS Implementors Group Forum [mailto:[log in to unmask]] On Behalf Of
Sent: Friday, 26 January 2007 10:02 PM
To: [log in to unmask]
Subject: Re: [PIG] Revision of the PREMIS Data Dictionary - call for
comments: significantProperties and preservationLevel
If we talk about technical environments, this should be covered by an
advanced (inter)national file format registry. If these environment
information would be within in your archival system, you would have to
put it in every individual object anyhow in case of an exchange with
other archival systems.
You are right about the significantProperties being pretty subjective -
but subjective in context of the policy of an archiving institution! It
should be possible to separate this "subjective" information easily from
the essential object information in case of an exchange with another
archiving institution. E.g. the exact same document could be archived by
a library and by an archive. For the library significant properties may
be totally different and therefore the migration strategy could be
different - although it is the same object.
D-60322 Frankfurt am Main
mailto:[log in to unmask]
> -----Original Message-----
> From: PREMIS Implementors Group Forum [mailto:[log in to unmask]] On
> Behalf Of Susan Thomas
> Sent: Friday, January 26, 2007 10:43 AM
> To: [log in to unmask]
> Subject: Re: [PIG] Revision of the PREMIS Data Dictionary -
> call for comments: significantProperties and preservationLevel
> I think "significantProperties" should remain with the object
> - it can
> be pretty subjective and closely related to the object. I
> would favour
> option 5 with the addition of the wrapper semantic unit of
> "preservationPolicy", as Tobias suggests.
> There is, however, a group of semantic units which I would
> like to see
> presented as an entity: environment. We are working on using
> PREMIS for
> preserving personal digital archives of notable individuals.
> we can record details of the environment(s) used by the
> creator in the
> production of their archive (because we have access to their
> we would love to be able to create this information once and
> link to it
> from the objects derived from that environment, rather than
> add the same
> information to each object. Did the editorial committee
> discuss this as
> a possibility at any stage?
> Gerard Clifton wrote:
> > Thanks for raising this, Tobias.
> > In coming up with the current options, the Editorial
> Committee did discuss
> > whether to create a separate entity for policies or
> business rules, which
> > would include preservationLevel and significantProperties,
> but initially
> > decided that a simple option could be to keep these
> semantic units as
> > 'attributes' of individual objects, rather than link to a
> separate entity.
> > However, we are also interested to hear others' views on this.
> > Also, what other kinds of semantic units might fit under 'policy'?
> > Regards
> > Gerard Clifton
> > Digital Preservation
> > National Library of Australia, Canberra ACT 2600 Australia
> > t: +61 2 6262 1381 e: [log in to unmask]
> > On Thu, 25 Jan 2007 16:55:10 +0100, Steinke, Tobias
> <[log in to unmask]> wrote:
> >> Although I like the structured solution of option 5, there
> should be a
> >> semantic unit "preservationPolicy" for this. The
> preservation level and
> >> the significant properties are heavily related to the policy of the
> >> archiving institution (the agent) and to the associated rights.
> >> Therefore "preservationPolicy" or shorter "Policy" (it's all about
> >> preservation...) could even be an own type of entity in
> PREMIS. Maybe we
> >> should consider to think about other policy-related issues
> before making
> >> a decision.
> >> Best regards,
> >> Tobias Steinke
> >> --
> >> Tobias Steinke
> >> Deutsche Nationalbibliothek
> >> Informationstechnik
> >> Adickesallee 1
> >> D-60322 Frankfurt am Main
> >> Telefon: +49-69-1525-1762
> >> Telefax: +49-69-1525-1799
> >> mailto:[log in to unmask]
> >> http://www.d-nb.de
> >>> -----Original Message-----
> >>> From: PREMIS Implementors Group Forum [mailto:[log in to unmask]] On
> >>> Behalf Of Ruth Duerr
> >>> Sent: Tuesday, January 23, 2007 7:14 PM
> >>> To: [log in to unmask]
> >>> Subject: Re: [PIG] Revision of the PREMIS Data Dictionary -
> >>> call for comments: significantProperties and preservationLevel
> >>> Of the options presented on the wiki, I think Option 5 would
> >>> suit the
> >>> needs of the scientific data archive best....
> >>> On Jan 22, 2007, at 12:37 AM, Gerard Clifton wrote:
> >>>> The PREMIS Editorial Committee has begun revision of the
> PREMIS Data
> >>>> Dictionary for Preservation Metadata and its related XML
> >>>> and is
> >>>> seeking feedback from the digital preservation community.
> >>>> Comments and discussion are invited on proposed changes to the
> >>>> semantic
> >>>> units 'significantProperties' and 'preservationLevel'.
> Notes on a
> >>>> number of
> >>>> issues and suggested options for these semantic units are
> >>> available
> >>>> on the
> >>>> PREMIS Implementors' Group wiki at:
> >>>> http://pigpen.lib.uchicago.edu:8888/pigpen/11
> >>>> Please send comments and suggestions to the PIG-list for
> >>>> discussion.
> >>>> The initial period for comment will end on Friday,
> February 2, 2007.
> >>>> The full list of proposed changes will soon be made
> available via
> >>>> the wiki
> >>>> (see PREMIS Revisions at: http://pigpen.lib.uchicago.edu:8888/
> >>>> pigpen/8).
> >>>> Other topics for discussion will follow in due course.
> >>>> With thanks,
> >>>> The PREMIS Editorial Committee
> Susan Thomas
> Project Manager
> Paradigm & Cairo
> Oxford University Library Services
> Osney One Building
> Osney Mead
> OXFORD OX2 0EW
> Web : http://www.paradigm.ac.uk
> Telephone : +44 (0) 1865 283821