W3C home > Mailing lists > Public > www-ws-desc@w3.org > June 2004

Issue 210: component equivalence

From: Mark Nottingham <mnot@mnot.net>
Date: Mon, 21 Jun 2004 16:25:30 -0700
Message-Id: <49165ACA-C3DA-11D8-AB5E-000A95BD86C0@mnot.net>
To: www-ws-desc@w3.org

I have an AI to start discussion of issue 210 with a straw-man proposal.

Section 2.15 of part 1 says:

>  Two components of the same type are considered equivalent if,  for 
> each property, the value in the first component is the same  as the 
> value in the second component.

I propose replacing this with:

-->8--
Two component instances of the same type are considered equivalent if, 
for each property of the first component, there is a corresponding 
property with an equivalent value on the second component, and the 
second component has no additional properties.

Instances of properties of the same type are considered equivalent if 
their values are equivalent. For string values, this means that they 
contain the same sequence of Unicode characters. Values which are 
references to other components are considered equivalent when they 
refer to equivalent components (as determined above). Finally, et-based 
values are considered equivalent if they contain corresponding 
equivalent values, without regard to order.

Extension properties which are not string values, references or sets of 
strings or references MUST describe their values' equivalence rules.
--8<--

For string equivalence, we might also consider referencing Unicode 
technical note #5: <http://www.unicode.org/notes/tn5/>, as the text 
above doesn't cover some scenarios.

Section 2.15 goes on to say:
>  With respect to top-level components (Interfaces, Bindings and  
> Services) this effectively translates to name-based equivalence  given 
> the constraints on names. That is, given two top-level  components of 
> the same type, if their {name} properties have the  same value and 
> their {target namespace} properties have the same  values then the two 
> components are in fact, the same component.

I don't know what to make of this, as it seems to contradict the 
statement above it; we're first told that equivalence is determined 
across all properties, and then just across {name} and {target 
namespace} for an ill-defined subset. What was the intent here?


--
Mark Nottingham     http://www.mnot.net/
Received on Monday, 21 June 2004 19:25:32 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:58:31 GMT