Hi everyone,
Many thanks to Greg for raising a really interesting question. When I
was learning my museum informatics, I was always told to remember the
'cow vs cheeseburger' principle - that you can turn a cow into a
cheeseburger, but you can't go the other way. In other words, you can
use a rich schema to render data according to multiple simpler schemata,
but you can't go from a simple schema to a rich one.
In this instance, the richness of museum information which includes
information about the object, related to the object (and its other
relations) and describing processes involved in the care and management
of the object would be the cow. LIDO would be the cheeseburger.
Our preference at Collections Trust has been to work towards a data
standard which simplifies the process of capturing, creating and
managing object-oriented information in management systems but which
preserves the full richness of the relations that make it a useful
heritage asset, and then to use crosswalks to support the rendition of
this richness to other formats, such as LIDO/EDM, DC et al. LIDO stands
for 'Lightweight Information Describing Objects' and is therefore
fine-tuned to that specific task. In this sense, it is not so very
different from ObjectID or the other task-specific schemata that
preceded it.
The point we're stuck on and on which we've been soliciting views from
the sector is whether to try and reach this data standard by developing
a 'canonical' CIDOC CRM recipe for museum information (object, event,
process etc) using the approach developed by the BM with ResearchSpace
and other efforts. We're also considering the implications of trying to
roll out such a model sector-wide when the CRM barrier to entry remains
quite high. The primary vehicle for doing this, given that it is used
all over the world, would be to de-couple the SPECTRUM Units of
Information from the procedural standard and adapt them towards a
SPECTRUM Data Standard.
Whatever it's called, a non-reductive industry-led standard for
capturing, managing and expressing the full richness of museum
information would enable us to support rendition to LIDO and whatever
comes next. This is the underlying idea of COPE - not just to develop
data that is shareable using current-gen metadata standards but also to
be resilient to future development in standards. In my short career,
I've seen us go from Z to OAI to SRU (and many others in between) and
I'm sure there will be many more to come!
Interesting thread, and thanks again for raising it.
Nick
Nick Poole
Chief Executive Officer
Collections Trust
Join Collections Trust's Collections Management Group
Follow us on Twitter
Like us on Facebook
www.collectionstrust.org.uk. Company Registration No: 1300565 Registered
Charity No: 273984
Registered Office: Collections Trust, WC 209, Natural History Museum,
Cromwell Road, London, SW7 5BD
-----Original Message-----
From: Museums Computer Group [mailto:[log in to unmask]] On Behalf Of
Richard Light
Sent: 24 November 2014 23:13
To: [log in to unmask]
Subject: Re: [MCG] To LIDO or not to LIDO
Greg,
I would say that an event-based approach is both a worthwhile
aspiration, and increasingly a working reality. The Europeana
Connection Kit (ECK) provides a standard LIDO mapping as being "the most
popular exchange format for museums" [1]. Major CMS vendors are
announcing support for the ECK - last month it was the turn of Adlib -
so I expect the pressure will be on the remainder to offer this
functionality in their own offerings.
The extent to which offering support for LIDO reflects a truly
event-based model in the underlying system is something you would have
to ask individual vendors about.
Whatever conclusions you draw about LIDO would apply equally, of course,
to the CIDOC CRM, which is also event-based. My own view is that you
can say very little of value about cultural heritage, using simplistic
dbpedia-style binary links between objects and "other stuff".
Best wishes,
Richard
[1] http://www.europeana-inside.eu/eck-workflow.html
On 24/11/2014 16:52, Reser, Gregory wrote:
> I'm weighing the pros and cons of object-based (VRA
Core<http://www.loc.gov/standards/vracore/>) vs. event-based
(LIDO<http://network.icom.museum/cidoc/working-groups/data-harvesting-an
d-interchange/lido-overview/lidos-background/>) data models. I can see
the benefits of LIDO, like the potential to capture related information
about what has happened to an object over time and the flexibility it
affords in the search and display of this data, but I wonder if it is a
beautiful theory rather than a workable solution. My very frank
question: is LIDO really advantageous and do institutions plan on
utilizing event relationships in their discovery systems?
>
>
> In the LIDO records I have come across, the most heavily used events
are creation and acquisition. Other events, like printing and
distributing a print, are handled in a note or edition display
statement, leaving the structured event properties unused. This suggests
that data is being mapped to LIDO from an object-based CMS.
>
> http://dac-collection.wesleyan.edu/Obj3351 (LIDO download (in really
> small font at the
> bottom)<https://mail.ucsd.edu/owa/redir.aspx?C=LmRoDFIN1kmtjzHqHOJ__nG
> uk02L2dEIwtiAxIFZnmiLZwhl8wouqPAUnQx98EVtF8EEVp26gyU.&URL=http%3a%2f%2
> fdac-collection.wesleyan.edu%2fObj3351>
>
>
> Is it a matter of time before CMSs move to object-based models? I
understand that LIDO is intended for delivering metadata not as the
basis for a content management system. Still, it seems that if one
intends to deliver data structured in the event conceptual model, at
even a medium level of specificity, one must collect the data in a
schema that allows mapping into this model. In other words, it will be
much easier to export rich LIDO metadata from a CMS that captures the
relationships of event information.
>
>
> While current object-based input forms can capture information related
to creation and acquisition, do any of them capture structured data for
things like excavation, modification, or restoration? I'm wondering if
data input tools have changed to accommodate LIDO's event based approach
- do they allow multiple events complete with creator, date, location,
materials, etc.? Do most systems still use the conceptual model, "here
are all the dates, here are all the locations, here are all the
materials" or are they changing to, "an event happened to this object on
this date, at this location and these materials were used, another event
happened to this object on this date..."? I made a mockup of a simple
event-based input
form<https://docs.google.com/drawings/d/1ywI_B6-rTIxvt0vo63ienkYHu3sbvmd
XpcncpkWEnFo/edit?usp=sharing>, just to see how it might look.
>
>
> I see the benefits of LIDO - the theory sounds wonderful, I just need
some real examples that demonstrate the benefits of event-based over
object-based metadata.
>
>
> Greg Reser
> UC San Diego Library
> 9500 Gilman Drive, 0175K
> La Jolla, CA 92093-0175
>
> Phone: 858.246.0998
> Skype: gregreser
>
>
>
> ****************************************************************
> website: http://museumscomputergroup.org.uk/
> Twitter: http://www.twitter.com/ukmcg
> Facebook: http://www.facebook.com/museumscomputergroup
> [un]subscribe: http://museumscomputergroup.org.uk/email-list/
> ****************************************************************
> .
>
--
*Richard Light*
****************************************************************
website: http://museumscomputergroup.org.uk/
Twitter: http://www.twitter.com/ukmcg
Facebook: http://www.facebook.com/museumscomputergroup
[un]subscribe: http://museumscomputergroup.org.uk/email-list/
****************************************************************
****************************************************************
website: http://museumscomputergroup.org.uk/
Twitter: http://www.twitter.com/ukmcg
Facebook: http://www.facebook.com/museumscomputergroup
[un]subscribe: http://museumscomputergroup.org.uk/email-list/
****************************************************************
|