[MMSEM-UC] Usecases, Interoperability issues and discussion

Dear all,

 

I would like to remind that I am expecting from all use cases to provide a
summary of the interoperability issues. All summaries will be included in
the “Interoperability issues” section of the deliverable. Also, based on
these issues, we will proceed on the “common framework” section.

 

Regarding the usecases, there is a little progress from our last meeting in
Athens. I know that everybody is very busy, but we must all try to find some
time to finilise the very good work that we’ve done so far. The
interoperability issue is becoming more and more a key issue in the metadata
standards area. I attended last week’s info day for FP7 in Luxembourg, and
the most important issue discussed in the Digital Libraries Unit was the
issue of metadata interoperability. There have been set up already many
proposal dealing only with this aspect and most of them refer to the work of
SKOS of the SWD WG and to the work that started in the MMSEM XG.  

 

I believe that the bottleneck in the completion of the usecases is the
possible solutions section. For this reason, my suggestion is to divide the
work in two parts. The first part is to report on the issues through the
motivation example (some of the usecases have already done this) and the
second part is to propose a possible solution. In this way, we can have a
first draft of the deliverable with all the interoperability issues.
Afterwards, if we manage to have possible solutions for each use case we
include them in the second draft. Therefore, I propose all the usecase to
finilise the first part as soon as possible.

 

Regarding the common framework section, I propose to include all the ideas
we discussed in Athens meeting including a report on possible solutions to
enable interoperability. From what we’ve discussed and studied in various
papers so far, I can distinguish three levels of interoperability. These
levels are 1) the schema level, 2) the record level, and 3) the repository
level. At the schema level, interoperability action usually takes place
before the operational metadata records are created. Methods used to achieve
interoperability at this stage mainly includes: derivation (from an existing
schema), application profiles, crosswalks, switching-across and other. At
the record level, the applications have  adopted a particular metadata
schema without being aware of other applications metadata schemas. As a
result these two applications cannot be interoperable. Activities at the
record level focus on integrating or converting data values associated with
specific elements. At the repository level, when multiple sources are
searched through a single search engine, one of the major problems is that
the retrieved results are rarely presented in a consistent systematic,
reliable format. The reason is that the source provider may have used
different metadata schemas and/or applied them differently. Interesting
processes related to ensuring interoperability at the repository level
include metadata harvesting, supporting multiple formats, aggregation,
thesauri and control vocabularies. Another level that can be distinguished
is the interoperability in the semantic level that is not presented before.
In this level, interesting ideas include upper harmonising ontologies and
ontology alignments tools.

 

We will further discuss these issues in tomorrows telecon.

 

Regards,

Vassilis