W3C home > Mailing lists > Public > www-international@w3.org > October to December 2012

Re: Proposal to NOT address I18N-ISSUE-183: Use of types questionable?

From: Steve Harris <steve.harris@garlik.com>
Date: Mon, 22 Oct 2012 11:42:33 +0100
Cc: Eric Prud'hommeaux <eric@w3.org>, RDF-WG WG <public-rdf-wg@w3.org>, Internationalization Core Working Group <www-international@w3.org>, Gavin Carothers <gavin@carothers.name>
Message-Id: <082E5BDD-42D7-4A85-9F91-F46AA5E8CF56@garlik.com>
To: Richard Cyganiak <richard@cyganiak.de>
On 2012-10-21, at 20:20, Richard Cyganiak wrote:

> Eric,
> 
> I have to disagree. Neither untyped integers nor untyped doubles are hard to find; examples for integers include world rankings (e.g., 34th in the world) and unit multipliers (e.g., 9 for billions); examples for doubles include ratios (e.g., change compared to previous year) and, indeed, structured typed values:
> 
>  :landArea [ :unit :km2; :value 9.827E6 ];
> 
> The current example does not just lack rigour, but shows two examples of poor practice:
> 
> 1. the use of integers where a more appropriate XSD type (gYear) exists; don't do that!

I'm not 100% convinced that this one is bad.

> 2. the use of floating point values for currencies; don't do that!

Yeah, that's not cool.

- Steve

> Also, a census does not produce GDP figures. Call it </stat-facts2007> or whatever.
> 
> Best,
> Richard
> (who often hangs out with statistics people)
> 
> 
> 
> 
> On 17 Oct 2012, at 17:53, Eric Prud'hommeaux wrote:
> 
>> Proposal to NOT address I18N-ISSUE-183: Use of types questionable?
>> ===============================================================
>> 
>> Issue: In the example [[
>> @prefix : <http://example.org/stats> .
>> <http://somecountry.example/census2007>
>>     :censusYear 2007 ;              # xsd:integer
>>     :birthsPerPerson .0135 ;        # xsd:decimal
>>     :gdpDollars 14074.2E9 ;         # xsd:double
>> ]], 2007 and 14074.2E9 should be typed.
>> 
>> Discussion: Untyped integers and doubles are hard to come by. The datatypes for some, like gene sequence location, don't spring as readily to the average reader's mind, but they are typed. While the current example doesn't demonstrate the most rigorous data modelling, it illustrates the parsing of integer, decimal and double barewords.
>> 
>> Proposal: no change, noting that proposals from I18N (or anyone else) are welcome and easily incorporated after LC.
>> 
>> Please indicate whether this addresses the stated issue.
>> -- 
>> -ericP
>> 
> 
> 

-- 
Steve Harris, CTO
Garlik, a part of Experian
+44 7854 417 874  http://www.garlik.com/
Registered in England and Wales 653331 VAT # 887 1335 93
80 Victoria Street, London, SW1E 5JL
Received on Monday, 22 October 2012 10:43:03 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 22 October 2012 10:43:04 GMT