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

Re: Question: representation of uncertainties in scientific

• Subject: Re: Question: representation of uncertainties in scientific
• From: "Richard G. Ball" <Richard_Ball@xxxxxxxxx>
• Date: Mon, 5 Aug 2002 20:57:32 +0100 (BST)

On [2002-Aug-02]  Brian McMahon <bm@iucr.org> wrote:
> 
> The *intention* is to specify one particular representation, namely
> -1.2345e1(2) and *not* -1.2345(2)e1. This is making concrete the usage
> implied in paragraph 5. on page 657 of Hall, Allen & Brown (1991),
> Acta Cryst. A47, 655-685.

Good.

> 
> As Nick's response indicates, there are grounds for arguing in favour of
> alternative representations, but it adds a little more burden to parsers.

An error in the exponent? I would hope not otherwise I'd have to say there wasn't much point in doing the structure.

> So
> it is certainly interesting to see whether Brian T.'s question raises anyone
> who is using the other version in practice. If not, we intend to stick with
> the current single allowed format.

Please; my CIF-life is complicated enough :-)

> 
> However, I draw your attention to the fact that the current production for
> <Exponent> in the cifsyntax document generalises the Hall, Allen & Brown  
> paragraph 5 to explicitly permit the following representations to be
> considered as valid:
> 
>       -1.2345e1(2)
>       -1.2345E1(2)
>       -1.2345d1(2)
>       -1.2345D1(2)
>       -1.2345+1(2)
>       -1.2345e+1(2)  (etc)

My highly technical, jargonish, response would be: yuck!

Can the production not be un-generalized?

> 
> Again,
>    (a) is anyone currently using the 'd' or '+' forms;

Not here.

>    (b) are there any strong views on this generalisation?

Yes. 

After all the "d" form is a holdover from ancient computing rituals and isn't of much relevance in these days of 64-bit native word formats. Is there any real use for extended precision (beyond 64-bit) in crystallography that would be appearing at a CIF level?

Richard
-- 

richard_ball@merck.com
(I regret the presence of the legal disclaimer but I have no control over it)

------------------------------------------------------------------------------
Notice: This e-mail message, together with any attachments, contains information of Merck & Co., Inc. (Whitehouse Station, New Jersey, USA) that may be confidential, proprietary copyrighted and/or legally privileged, and is intended solely for the use of the individual or entity named on this message.  If you are not the intended recipient, and have received this message in error, please immediately return this by e-mail and then delete it.

==============================================================================

Reply to: [list | sender only]

• Prev by Date: Re: CIF-DEVELOPERS digest 36
• Next by Date: Re: Question: representation of uncertainties in scientific notation
• Prev by thread: Re: Revised draft of CIF 1.1 syntax document
• Next by thread: Re: CIF-DEVELOPERS digest 36
• Index(es):
  • Date
  • Thread