Re: DIN 5007, Swiss Sorting

From: Tex Texin (texin@progress.com)
Date: Sun Mar 12 2000 - 01:15:22 EST


Thanks for the information Alain.

I have a user that is making claims about an old Swiss collation.
I think he is fine with our algorithms for sorting, but wants
some accented characters treated differently. I could just
adjust tables to give him what he wants, but if it is close to some
standard
I prefer to give it a proper label and synchronize all the
characters exactly, instead of just the few he expressed concerns
about. Also, it wouldn't be the first time a user completely
fantasized what the sort ordering should be, so I like to seek
other evidence.

Ever hear of an old Swiss collation? Or is there a recent
change in Swiss collation?

tex

Alain wrote:
>
> À 12:36 2000-03-10 -0800, Tex Texin a écrit:
> >I am looking into conventions for Swiss collation.
> >
> >I was pointed at DIN 5007. Is this on the web anywhere?
> >
> >Are there old and new Swiss collations, and are the
> >collation conventions different depending on the language
> >and/or dialect being used, since the Swiss use multiple
> >languages.
>
> [Alain] DIN 5007 is for German. It is pretty similar to CAN/CSA Z243.4.1
> which harmonized ordering for at least French, English, German, Italian,
> Portuguese and Dutch at once. The only real difference is level-2 sorting
> (Canadian standard -- based on French dictionaries sorting practice
> discriminates quasi-homographs on multiple accents differences starting at
> the end of words -- DIN 5007 does not do that, it starts at the beginning,
> but for German dictionaries it practically does not matter [except for
> words of French origin!], or at least much less than for French dictionaries).
>
> 3rd level discrimination (on case) is done the same in DIN 5007 as in the
> Canadian standard. Actually we took German as a model for sorting lower
> case before upper case as there is no firm practice for this in French
> (except in encyclopedias which seem at first glance to do the reverse but
> there are too many exceptions to deduce a rule) and we did not find firm
> practice in English either while there was one in German...
>
> Michael Everson disagrees that for English lower case should be sorted
> first in case of quasi homographs (ex. : august before August), based on
> what he deduces from the short version of the OED. But the full OED has no
> preference, it is easy to demonstrate (same kind of exceptions as for
> French encyclopedias -- German case discrimination is systematic, whence
> the reason for Canadian preference, which we find compatible with French
> and English).
>
> For Switzerland, I would recommend the Canadian standard practice as it
> includes German, French, Italian [and Romansch] at once, compatible
> languages for sorting (unlike Spanish which would require tailoring for ñ
> [and for ch and ll if traditional sorting is to be used]).
>
> Alain LaBonté
> Charlesbourg

-- 
Progress is a proud sponsor of the 16th International Unicode Conference
March 27-30, 2000 in Amsterdam, Holland
http://www.unicode.org/iuc/iuc16/index.html
See our panel on Open Source Approaches to Unicode Libraries
http://www.unicode.org/iuc/iuc16/a206.html
------------------------------------------------------------------------------------------------
Tex Texin                     Director, International Products
                                 
Progress Software Corp.       +1-781-280-4271
14 Oak Park                   +1-781-280-4655 (Fax)
Bedford, MA 01730  USA        texin@bedford.progress.com

http://www.progress.com The #1 Embedded Database http://www.SonicMQ.com JMS Compliant Messaging- Best Middleware Award http://www.aspconnections.com Leading provider in the ASP marketplace

Progress Globalization Program http://www.progress.com/services/partners/globalization/index.htm ------------------------------------------------------------------------------------------------ Spanish Proverb: Don't speak unless you can improve on the silence. Tex's Proverb: Don't email unless you can improve on the screen saver.



This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:59 EDT