Discussion List Archives

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

Re: coreCIF.dic-2.4 Discussion List #8

  • To: Distribution list of the IUCr COMCIFS Core Dictionary Maintenance Group <coredmg@iucr.org>
  • Subject: Re: coreCIF.dic-2.4 Discussion List #8
  • From: Brian McMahon <bm@iucr.org>
  • Date: Tue, 20 Jun 2006 12:11:53 +0100
  • In-Reply-To: <444921A5.8060409@mcmaster.ca>
  • References: <444921A5.8060409@mcmaster.ca>
Dear David

Apologies for the delay in commenting on the coreDMG discussion
paper #8. I glanced through this some time ago, and saw nothing
to which I took strong exception; Peter Strickland and I have
now looked at it again.  The comments below reflect mostly our
thoughts on how the new proposals might affect the journals.
Again, we have not much to say.

> ##########################################
> #  5. GEOM_BOND
> ##########################################
> #
> # COMMENT: In high symmetry structures when many bonds are related by 
> # symmetry it is not necessary to list all the bonds in the
> # environment of the first named atom.  Some users may wish to give
> # only the symmetry independent distances and a multiplier to
> # indicate how many such bonds are found in the
> # atomic environment.
> #
> data_geom_bond_multiplicity
>     _name               '_geom_bond_multiplicity'
>     _category            geom_bond       
>     _type                numb       
>     _type_conditions            
>     _list                yes       
>     _list_reference      '_geom_bond_atom_site_label_'   
>     _enumeration_range   0:
>     _definition
> ;       The number of times the given bond appears in the environment of
>         the atoms labelled _geom_bond_atom_site_label_1.          
> ;
> # STATUS: Open for discussion

Journal policy is to encourage the listing of all bonds in the CIF; and
current journal policy is not to print bond distances unless they are
unusual (in which case it's probably unlikely that an n-fold
symmetry-generated bond would be printed anyway). Overall, we're neutral
about including such a new item. It would be helpful to give an example (in
the GEOM_BOND category block) that illustrates such a case, and perhaps
demonstrate that it can co-exist with individual specific bond listings: if
we decide to go ahead with this item I'll see whether we can find useful
examples in the literature.


> ###########################################
> # 6. EXPTL_ABSORPT_CORRECTION_T_max and T_min
> ###########################################
> # The current definition of these items is:
> #     The maximum and minimum transmission factors for the crystal
> #     and radiation. These factors are also referred to as the
> #     absorption correction A or 1/A*.
> # but there have been a variety of different interpretations as to what this
> # means, viz:
> #
> # 1.  These values are the transmission factors that were used in correcting
> # the diffraction intensities for absorption. (Acta Cryst. has adopted this
> # interpretation.)
> #
> # 2. These are the true transmission factors, whether or not any correction
> # was made for absorption.  They are shown to give an indication of the
> # importance of absorption in this specimen.
> #
> # ...
> # These uncertainties should be resolved.  Maybe we need more than one 
> # set of transmission factors.  Your suggestions would be welcome.

It may be helpful to state in a little more detail what current Acta
practice is. The transmission factors are calculated by PLATON and
compared against _exptl_absorpt_correction_T_max and *_min. If they
differ, we store the calculated values as
_iucr_exptl_absorpt_correction_T_min and *_min (not very
expressive, but at least clearly distinguishable tags). In other
words, our practice implies that the existing items are handled as
in your item (1), and calculated transmission factors (2) are also
stored if they are identified as different. It would therefore help
us if the definition (1) were adopted and a new "official" item
introduced for (2). This distinction has however been made only
comparatively recently, so it is possible that older CIFs have used
the data items in the sense of (2).


> ################################
> #  11. _PUBL_SECTION_KEYWORDS
> #################################
> # Doug Boulay (05-06-25):
> # After reading David's report, I notice the the current
> # CIF dictionaries don't seem to have any explicit
> # publication specific keyword metadata definitions...
> # But it could be quite useful for propagating into
> # Dublin-core (RDF) subject metadata for HTML and XML
> # renderings of CIF. It would probably need an official vocabulary of
> # terms and terminology to do it rigorously though.
> data_publ_section_keywords
>     _name               '_publ_section_keywords'
>     _category           publ        
>     _type               char        
>     _list               yes
>     _list_reference     '_publ_section_keywords_id'      
>     _example                       ?
>     _definition
> ;              Keywords associated with the manuscript
> ;

We agree that at present such an item would be useful but
should not be constrained by an enumeration list. The broader
area of keywords specifying scientific topics is one we want
to come back to at a later date.


> ########################
> # 12. DIMENSIONLESS UNITS
> #
> # COMMENT: HDF provides the following information
> #
> # 24.1 Interdivisional Committee on Terminology, Nomenclature and Symbols of
> # the International Union of Pure and Applied Chemistry (IUPAC ICTNS)
> #
> # Another proposal concerns the use of the name 'uno', symbol U, for the 
> # unit one so that dimensionless numbers may be treated in the same way
> # as all other SI units.
> # ...
> #
> # [IDB: adding uno to the list of units would result in the following item.]
>  data_units
>      _definition
> ;              A unique code which identifies the units of the defined data
>                 item. A description of the units is provided in 
> _units_detail.
> ;
>      _name                      '_units'
>      _category                    units
>      _type                        char
>      loop_ _example
>            _example_detail         K    'kelvins'
>                                    C    'degrees Celsius'
>                                    rad  'radians'
>                                    e    'electrons'
>                                    V    'volts'
>                                    Dal  'daltons'
>                                    m    'metres'
>                                    kg   'kilograms'
>                                    s    'seconds'
>                                    U    'uno (dimensionless units)'
> #
> # STATUS: Open for discussion
> ################################

A report of the IUPAP General Assembly 2005
(http://www.iupap.org/commissions/c2/reports/ga-05.html)
suggests that SI adoption is at least some way off:

   The Uno: The introduction of a name, 'uno', for the unit one in SI was
   proposed by the CCU of the BIPM (see René Dybkaer, Metrologi a, 41
   (2004) p.69-73) . It was decided that because of the potential
   inconsistencies that might result from the introduction of such a unit,
   the commission could not support the introduction of this new unit. The
   matter has been dropped for the time being by the CCU.

We are somewhat concerned about the proposal in any case. Proportionality
is indeed dimensionless, but it seems that one is throwing away information
if one does not record that proportionality is by volume, mass, etc.

This matter should be tracked by the Nomenclature Commission, but we
recommend that no action be taken before any formal adoption of the uno
into the SI system - the topic can then be raised again for discussion if
desired.

I hope these belated remarks are of some use. Regarding the other proposals
we are either in favour or neutral.

Best wishes
Brian
_______________________________________________
coreDMG mailing list
coreDMG@iucr.org
http://scripts.iucr.org/mailman/listinfo/coredmg

[Send comment to list secretary]
[Reply to list (subscribers only)]