[odf-discuss] A bit more detailed explanation of the AFNORposition

[Lars Noodén]

Daniel Carrera wrote:
> marbux wrote:
>> ... I've never said that *all* foreign elements and attributes 
>> need to be preserved...
> 
> Only ones needed for interop? ...

1) how it is proposed to sort foreign elements (aka garbage) needed for
interoperability from those designed to enforce vendor lock-in from
those that are truly optional from those that are errors?

>> You continue to claim it is impossible for apps to preserve foreign
>> elements and attributes needed for interop purposes despite the fact
>> that: [i] the ODF Conformance section says that conforming apps *may*
>> preserve them;

It is not impossible, simply not necessary, and it is up to the
application to do it or not.  Before proposing a change to the
specification, explain how to overcome the problem in #1 above or
provide a reference to a short explanation.

> That does not say that applications shall or should preserve foreign
> elements. It says that the content should be processed and that the
> document shall be valid ODF if the unknown elements are replaced with
> their content only. ...

That is very good, since it requires that applications to not choke on
broken documents.

In addition to the theoretical risk (see #1 above), we have a great many
major real-life examples of abuse in other specifications as a result of
added garbage (aka foreign elements) used by MS to prevent
interoperability with standard compliant applications and systems:

	Kerberos
	Java
	HTML
	HTTP

2) how it is proposed that requiring preservation of foreign elements
(aka garbage) would be engineered into the spec in such a way that abuse
is prevented?

Regards,
-Lars