[IUCr Home Page] [CIF Home Page]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Permitting new physical units?



	Not being involved in programming I certainly defer to Syd's
chastisement for suggesting the use of modified data names.  However, I
suspect he will be putting out this brush fire for many years yet to come
because it is not obvious to a non-programmer that this is a poor
solution. 

	I am not repentant however on the question of dimensions versus
units.  We may have had this discussion before, and I believe Syd when he
says that when we talk about 'units' we are really talking about
'phys-dimensions'.  However, for those of us without Syd's abilities to do
these mental gymnastics and who tend to take the simplistic view that the
word 'units' means 'units' rather than 'dimensions' there is a real danger
that we will get confused, as clearly has been happening in the
discussion.  I prefer to call a spade a spade rather than a shovel, which,
while apparently similar, is in fact a different tool with a different
function.  We have already run into this problem with the data name
_crystal_setting which should be _crystal_system and which therefore is
causing considerable problems for users who appreciate that the two are
not the same thing.  I wish now that I had really caught that one before
it got into the dictionary, but we were being bullied into getting a
functional dictionary as quickly as possible and that fell through the
cracks.  I recall that we were told at that time that the data names were
not changeable and that we were responsible only for the definitions.

	If, for good reasons, it is necessary to stay with the term
'units' in the DDL, the definition there should make it quite clear that
this field is used to specify the dimensions (by giving the units) and
that it is not permissible to define different units if the
phys-dimensionality is not changed.  Comcifs of course has no control over
the definitions of the DDL, but Syd might avoid setting more brush fires
in the future by using words that mean what they say and that do not send
people off on the wrong track.

	Finally, can I appeal to Syd to be careful in the way he assigns
the subject to his email messages?  If the discussion system that Brian
has set up in Chester is going to work, we need discipline in assigning
the subject line.  Specifically, every time a new subject header is used,
a new thread is set up on the web page.  If the message is a continuation
of an existing discussion then it should have the SAME subject heading so
the computer can recognise where it belongs.  The way things are running
at the moment, most of Syd's comments are being detached from the
discussion to which they belong because they start new threads.  Thus,
Syd's pertinent explanation of why name extensions will not work will not
be read by someone following the thread in which my suggestion appears. 
The result is that Syd's attempt to control the brush fire may be lost on
someone following the discussion thread on the web.  'David' has made many
contribution to the discussion (some under the less transparent name of
'I.D.Brown') and which of these has given rise to the thread at the head
of this message will not be obvious.  The result is that the reader may go
away thinking that attributes encoded in the data names are the best thing
since ice cream.  It does not take much effort to keep the thread running: 
just keep any email you want to comment on, and use the Reply feature,
rather than sending off a new message.  Have a look at the coreDMG web
page to see what I mean.  Please, Syd, can we keep you on track? 

			David


*****************************************************
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
*****************************************************



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


Copyright © International Union of Crystallography

IUCr Webmaster