Discussion List Archives

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

Annual Comcifs report

Dear Colleagues,

	I have been asked by Mike Dacombe to submit an annual report for
Comcifs for 1999.  I enclose below my draft of the report for your
comments in case I have missed out something important or misrepresented
an issue.

	Please let me have any comments as soon as possible.  The deadline
for submission of the report is the end of February.

				David

keep on scrolling - the report is below.
*****************************************************
Dr.I.David Brown,  Professor Emeritus
Brockhouse Institute for Materials Research, 
McMaster University, Hamilton, Ontario, Canada
Tel: 1-(905)-525-9140 ext 24710
Fax: 1-(905)-521-2773
idbrown@mcmaster.ca
*****************************************************


Annual Report of Comcifs to the Executive Committee of the IUCr

                     Annual Report for 1999
                     Submitted by I.D.Brown

1. Introduction
     1999 saw important developments in the operation of Comcifs.
The Glasgow congress of the IUCr not only provided the first
opportunity for the triennial review of membership under the
terms of reference approved in 1998, it was the occasion for the
first formal meeting of Comcifs since its appointment in 1993.

     The Comcifs meeting at Glasgow had two components, an open
meeting which outlined some of the problems facing CIF, and a
closed meeting in which we developed a strategy to deal with
these problems.  Details are given below.

2. Membership
     The membership of Comcifs consists of a small number of
voting members appointed by the Executive Committee of the IUCr
and an unlimited number of non-voting members added at the
discretion of the chair.  The non-voting members comprise all
those with an interest in the development of CIF who request to
be placed on the Comcifs discussion list.  Both kinds of members
are fully involved in the work of Comcifs, but approval of new
CIF dictionaries and policies is restricted to the voting
members.  The voting membership approved by the Executive
Committee consists of:

               David Brown (chair)
               Brian McMahon (secretary)
               Helen Berman
               Herbert Bernstein
               Paul Edgington
               Syd Hall
               Gotzon Madariaga

This is a reduction of one in the number of voting members from
the previous Comcifs.

3. Meeting in Glasgow
     This meeting provided an opportunity for direct discussion
of the issues facing CIF.  To address the problems (described
individually in the sections below) four subcommittees were
established.  These are:

     Dictionary Definition Language (DDL) Development Committee
     Publicity and Outreach Committee
     Software Development Committee
     Dictionary Review Committee

Details of their terms  of reference and membership are given
below.

4. CIF Dictionaries
     Approval was given on 1999.3.22 to version 2.1 of the Core
CIF dictionary.

     By the end of the year several dictionaries were close to
being submitted for approval.  These are:

     Macromolecular CIF dictionary Version 2
     Modulated Structure CIF
     Image CIF
     Symmetry CIF

In addition, dictionaries are in preparation for:

     Diffuse Scattering
     Electron Densities
     Small Angle Scattering
     Magnetic Structures
     Graphics

Each of the approved dictionaries (core, macromolecular and
powder dictionaries) has a Dictionary Maintenance Group whose job
is to update the dictionary and present the revisions to Comcifs
for approval.

     Following the meeting in Glasgow a Dictionary Review
Committee (membership: Brown, McMahon, John Westbrook) was
appointed to ensure that each dictionary submitted to Comcifs for
approval received a detailed review.  When satisfied that the
dictionary complies with Comcifs policy this committee
recommends to the voting members of Comcifs that the dictionary
be approved.

5. The future of the Dictionary Definition Language
     The CIF dictionaries, like the CIFs themselves, are written
as STAR files whose semantics are defined by a Dictionary
Definition Language (DDL).  Currently two different versions are
in use:  DDL1 is used for the core and powder diffraction CIF
dictionaries and DDL2 is used for the macromolecular dictionary.
DDL2 is more prescriptive and less permissive than DDL1, but this
has been deemed necessary for the projected applications of the
macromolecular CIF.  Dictionaries written in different DDLs (and
hence CIFs written using these dictionaries) are not compatible,
an arrangement that is not viable in the long term.  Further, if
CIF is to remain competitive with other file structures being
developed for the web (e.g., XML, ASN.1), it will need increased
functionality.  The possibilities were demonstrated by Hall
who showed how one can code algebraic relations between
different items into a CIF dictionary.  This development can be
likened to the addition of verbs to a CIF dictionary that
currently only contains nouns.  Such a development would greatly
increase the power of CIF editors and simplify crystallographic
programming.

     Following the Glasgow meeting a DDL Development Committee
was appointed (Westbrook, Hall) and charged with making
recommendations for a version of DDL that is upwardly compatible
with DDL1 and DDL2 and which will provide for the inclusion of
algebraic expressions and other algorithmic methods.

6. Concatenation of Dictionaries
     McMahon, Bernstein and Westbrook presented a proposal to
Comcifs for a protocol that  allows different CIF dictionaries to
be concatenated.  This  simplifies the structure of specialised
dictionaries since they do not need to duplicate the definitions
of items that appear in other dictionaries.  It also makes it
easier for users to define any local items included in their
CIFs.  An early approval of this protocol is anticipated.

7. Software
     CIF is a powerful crystallographic language, but an
electronic language is of little use without the software to
manipulate it.  Comcifs has identified the lack of software as a
major impediment in the adoption of CIF causing many potential
users to turn to less suitable languages for which there is good
software support.  Most users continue to see CIF as merely a
format for submission of experimental results rather than as a
language for manipulating crystallographic ideas.  An urgent need
is a CIF editor and browser to simplify the process of reading,
constructing and modifying CIFs.  Two editors are under
development: an editor based on DDL2 is being developed by the
Protein Databank (PDB) to simplify the preparation of submissions
to the PDB, and an editor based on DDL1 is being developed at the
Cambridge Crystallographic Data Centre to help in the
preparations of submissions to the Cambridge Structural Database
and to crystallographic journals.  Both editors will be useful
for preparing submissions to Acta Cryst. B. C and D and will be a
pivotal tool in the mooted all-electronic structure journal.  
Lack of software is currently a serious impediment to the full
exploitation of the potential of CIF.

     Following the Glasgow Congress a Software Development
Committee (Bernstein, Westbrook and Edgington) was appointed to
review the current state of CIF software and make recommendations
for its improvement.

8. Publicity
     CIF needs to be  advertised and promoted at a time when the
internet is expanding and standards are rapidly changing,
otherwise we may lose our advantage through ignorance of the
potential of the language.  It is essential that all branches of
the crystallographic community move to a common, or at least
compatible, electronic language.  CIF has been tailored to the
community and provides the flexibility and functionality that the
community needs.  It is essential that this message be heard and
understood across the crystallographic world.

     Following the Glasgow Congress a Publicity and Outreach
Committee (Berman, Hall, McMahon) was appointed and charged with
ensuring that the virtues of CIF are widely disseminated and that
software developers are encouraged to incorporate CIF into their
programs.

9. Interoperability
     The rise of the World Wide Web and other Internet protocols
has fuelled new developments in information interchange.
Abstraction of information about information (so-called metadata)
now allows for richer and more structured exchange mechanisms
between and across disciplines. Some of the initiatives in areas
related to crystallography include Chemical Markup Language (CML
- a DTD for describing chemical structures within SGML or XML
documents), macromolecular structure descriptions in terms of
CORBA objects and RDF (Resource Description Framework) schemas.
In addition, the embedding in CIFs of external objects tagged by
MIME type (an Internet standard for defining types of data files)
and the provision of appropriate MIME types for CIF documents
allow easier manipulation of crystallographic data within the
current Web protocols. There are also arguments for formalising
CIF within the ANSI or ISO standards frameworks.

     In addition to these very general extensions, a number of
disciplines related to various areas of crystallography are
developing their own data vocabularies; often it is mutually
helpful to work towards common data structures, expressed in STAR
of CIF formalisms or at least in terms of related database
schemas.

     Members and consultants of COMCIFS are actively involved in
all these areas, and their work is an important element of
ensuring that CIF is relevant in the wider context of information
provision. 
 
10. Intellectual Property
     Ownership of CIF by the IUCr is essential to prevent the
development of incompatible CIF dialects.  That ownership needs
to be publicised and the standards enforced.  On the other hand,
the IUCr wishes to see the standard widely used and does not wish
to discourage the use of CIF by implied threats of legal action
for software that inadvertently fails to follow the standard.
Various methods of protecting our interest have been suggested
such as registering CIF as a service mark, providing a set of
legal guidelines for the use of CIF, and providing software to
certify that a document or program claiming to be CIF compliant
is actually CIF compliant.

     The Executive Secretary of the IUCr was requested to proceed
with registering appropriate service marks, and Bernstein and
Phil Bourne were appointed as an ad hoc committee to lead a
further discussion on this topic.

11. Acknowledgement
     I would like to thank the members of Comcifs for their
dedication and hard work.  In particular I would like to thank
the secretary, Brian McMahon, for maintaining the internet
connections that allow Comcifs to function.