[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]
Re: [xml-dev] Separate data from rules ... is the XML Schema 1.1 <assert> element a step backwards?
- From: bryan rasmussen <rasmussen.bryan@gmail.com>
- To: Michael Kay <mike@saxonica.com>
- Date: Tue, 7 Jul 2009 15:30:06 +0200
hmm, after sending off my last message I noticed - as is generally the
case - that Michael Kay already touched on the issue..
on the types of validation..
> (a) laws of nature - children have two parents, planes take off before they
> land
>
> (b) rules decided by the business (or more generally, by the user community)
> - every customer belongs to one branch office, every employee is over 16,
> ISBNs are ten digits long
>
I would split the (b) into (b) rules decided by the community
exchanging the data -
ISBNs are ten digits long, contracts must be signed, currency must be
in a recognized format..These and (a) are the actual validity rules,
the rules that allow exchange of documents defined as valid by a group
of those doing the exchanging..
and
(c) rules decided by the business for their processing of valid
documents - workflow related - actual category of 'business rules'
that is to say something can fail these but still be 'valid':
payment due in more than 2 months, payment greater than this amount
needs two signatures, every customer belongs to one branch office.
> (c) rules decided by IT system designers - every customer has at most three
> phone numbers, tables cannot be nested within tables, messages are limited
> to 140 characters
>
And this would be
(d) the workflow to allow data that is valid to be processed to be
placed within various IT systems - generally automated.
I would say that any workflow that can be automated and is of type d
is not a business rule though...
rest of the message was better phrased than what I could do of course.
Cheers,
Bryan Rasmussen
[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]