- From: Bjoern Hoehrmann <derhoermi@gmx.net>
- Date: Sun, 07 Sep 2003 06:16:28 +0200
- To: Karl Dubost <karl@w3.org>
- Cc: www-qa@w3.org
* Karl Dubost wrote:
>> information". How would you "test" for example
>>
>> <address
>> xml:lang='x-bjoernhoehrmannish'>##39(ö)64..werhq%__</address>
>>
>> whether it meets my hypothetical requirements?
("The <address> element MUST contain contact information")
>It could be exact. If it meets an address scheme. Yes. The XHTML Spec
>doesn't say what must be your address scheme, you are free to choose.
>If you think that address is not testable, it should not be in the spec
>at all.
Well, I am not concernd about XHTML, my question was how to understand
"testable" as used in the Specification Guidelines and I've provided an
example to illustrate that my reading of "testable" is not necessarily
the one intended by the QA WG. If I understand you correctly, "testable"
means that someone shall be able to tell whether the requirement has
been met. If that's the case, everything that could be a requirement of
a specification is by definition testable and so it does not make sense
to me to use that term at all.
>It's exactly the same problem when you have an image and you have an
>"alt" attribute. The content of the alt attribute must describe the
>image, but the semiotics is something which is very subjective, even
>more than the one for address.
The alt attribute must actually contain /alternate/ text /if needed/ to
understand the document... However, yes, that's similar to my example.
Received on Sunday, 7 September 2003 00:16:50 UTC