[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [ddlm-group] A dictionary-writing workshop in Hyderabad?

I agree with John W that the considerations in writing dictionaries are largely technology-agnostic, and indeed my proposal would be that a large component of the workshop would be separate from any particular ontological language and indeed applicable to e.g. NeXus-style efforts.  I think, however, that the workshop has to have a concrete (perhaps stretch) goal, which is that actual draft CIF dictionaries/definitions are started (or even finished) as a result of the workshop, which means that towards the end of the workshop an introduction to DDL2 and DDLm would be needed.

I've had some offline encouragement so I plan to move this idea along.  I will probably take discussions offline (with perhaps some progress updates to this group), so if you would like to be included in further discussions please let me know.

James.


On 6 November 2016 at 06:02, john.westbrook@rcsb.org <john.westbrook@rcsb.org> wrote:
Hi all,

We continue to actively engage our producer and consumer communities in the macromolecular
domain in developing content extensions that are compatible with our DDL2 infrastructure.
Increasingly this involves developing compatible content extensions for domains outside
of the traditional diffraction methods.

As David Brown has observed, there is a significant intellectual component in carefully
defining a domain dictionary content and these are fairly distinct from the particular
choice for encoding this information.

I would suggest to focus any workshop more specifically on the fundamental aspects of
data definition and organization.   From a technology perspective I would prefer that
the discussion also address how to contribute content for macromolecular applications as
well.

Regards,

John




On 11/2/16 10:29 PM, James Hester wrote:
Dear DDLm group,

From time to time I get enquiries on creating CIF dictionaries for various disciplines (pressure, reflectometry, electron
diffraction are the most recent).  It occurs to me that there might be sufficient interest in the general and technical aspects of
CIF dictionaries that a one-day workshop at Hyderabad would be worthwhile.  At the conclusion of the workshop, participants should
be able to produce a syntactically-correct (DDLm) dictionary with consistent and correct semantics, and the attendees would be
invited to come prepared with a list of concepts that they would like to put into a dictionary.  In a broader sense, the workshop
would kick-start new dictionary efforts and may interact nicely with those commissions that are being pushed to produce some
metadata for their fields by the DDDWG/IUCr executive.  In an ideal world, the content of dictionaries begun at the workshop could
be finalised in consultation with colleagues on the spot during the IUCr conference, whether at commission meetings or in ad-hoc
discussions, and some of us could make ourselves available for immediate consultation throughout the meeting.

Suggested topics:
(1) Introduction: dictionary layout, role the dictionary plays, what is a category?
(2) First steps: collecting and classifying concepts (single-valued, multi-valued, assigned, measured, derived,...)
(3) Sorting concepts into categories
(4) Writing the human-readable definitions
(5) Assigning datanames
(6) Assigning attributes; overview of available attributes
(7) Describing interaction with other CIF dictionaries (import)
(8) How expansion works (the _audit.schema system)
(9) Adding to existing dictionaries

I would not propose covering dREL in detail due to time constraints. Note also that much of the material is not CIF specific, but
whether we want to sell it that way I'm not sure.  There would be plenty of "practical" sessions where participants actually worked
on their own dictionary - we might choose a useful mini-topic as an example for those who haven't come with any particular field in
mind.

It would also be useful to have some DDLm dictionary-writing tools (checkers/GUIs) available, if anybody has something they could
polish up in time.

What does the rest of the group think of this idea?  While I'm happy to take a lead role in putting it together, we would need some
volunteers to help prepare materials, present, and wander around the room helping.

all the best,
James.
--
T +61 (02) 9717 9907
F +61 (02) 9717 3145
M +61 (04) 0249 4148


_______________________________________________
ddlm-group mailing list
ddlm-group@iucr.org
http://mailman.iucr.org/cgi-bin/mailman/listinfo/ddlm-group


--
John Westbrook, Ph.D.
RCSB, Protein Data Bank
Rutgers, The State University of New Jersey
Department of Chemistry and Chemical Biology
174 Frelinghuysen Rd
Piscataway, NJ 08854-8087
e-mail: john.westbrook@rcsb.org
Ph: (848) 445-4290 Fax: (732) 445-4320
_______________________________________________
ddlm-group mailing list
ddlm-group@iucr.org
http://mailman.iucr.org/cgi-bin/mailman/listinfo/ddlm-group



--
T +61 (02) 9717 9907
F +61 (02) 9717 3145
M +61 (04) 0249 4148
_______________________________________________
ddlm-group mailing list
ddlm-group@iucr.org
http://mailman.iucr.org/cgi-bin/mailman/listinfo/ddlm-group

Reply to: [list | sender only]