[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [ubl-dev] Re: Code list extensibility andsubstitutiongroups
An afterthought: if there's no way then to say what *isn't* valid then surely there is no meaning to what *is* valid and one actually has virtually no validation at all How is this different to not having a Schema for a codelist? There are codelists for which UBL 1.0 has no Schema (many) but following this reasoning isn't it better to have no Schema than to have one which seems to validate but doesn't? Steve >>> "Stephen Green" <stephen_green@bristol-city.gov.uk> 23/02/05 10:15:40 >>> Thanks Chee-Kai Is there any way to then specify what 'is' being used for validation when substitutionGroups are involved? My fear is that there isn't so there is no way to say what *isn't* valid. Steve >>> Chin Chee-Kai <cheekai@softml.net> 23/02/05 06:45:36 >>> The presence of substitutionGroups can also create a sense of "the element is not the type which it's present manifest type suggests" -- in other words, what type you see that's associated with an element at present may not be what is being used for validation. The situation would be worsened if the actual substitution definitions come in from another <include>d file stored physically elsewhere, and might also introduce possible authenthicity issues (eg. allowing validation using bogus subsituted groups when invalid codes exist in instances). ..... Best Regards, Chin Chee-Kai SoftML Tel: +65-6820-2979 Fax: +65-6743-7875 Email: cheekai@SoftML.Net http://SoftML.Net/
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]