Re: Usage Scenarios for Versioning and Extensibility

[Bill de hÓra <bill@xxxxxxxxxx>]

Joe Gregorio wrote:

> On Mon, 12 Jul 2004 17:14:54 -0700, Mark Nottingham <mnot@xxxxxxxx> wrote:

> Tim's policy seems to work, but with the following caveat/addition:
>
> 1. Newer versions of the format are kept backward 
> compatible with older versions. 

Yes, I think that is entailed by rules 3 and 4 (they seem to 
conflict/fail without it).

The idea is of adding new element names or QNames rather than 
altering existing ones approaches what Dan Brickley described as 
"monotonic" elsewhere... and on that, I would be happy to see a 
versioning rule/policy that new elements (from any namespace) added 
to new version will not affect the expected behaviour over previous 
elements and versions, ie there's no versioning approach that 
involves/encourages co-occurence constraints between elements; but I 
have no idea how to test for it (except maybe trying to write it 
down in xsd).

cheers
Bill