[dsdl-discuss] Re: Patent problem?

Alex Brown wrote:

>Dear all,
>
>After the plenary yesterday, the observer from Systemwire suggested that any
>DSDL functionality for multi-document validation might conflict with
>Systemwire's patent in this area.
>
>see http://www.systemwire.com/xlinkit/validator.html for the product
>
>IANAL, but it seems to be hyperlink generation is an essential feature of
>their claim (though I can't find the patent online).
>
>
>Rick - I know you've looked into this, what's your conclusion?
>
>
>
See http://www.geocrawler.com/archives/3/6317/2000/10/0/4455310/
for my earlier take. I have had quite good relations with them: they paid
for me to go and speak at an ACM Software Engineering conference
in Toronto 3 years ago, and are very nice people.

For details on XLinkIt see www.systemwire.com/whitepapers/xlinkit.pdf
and www.cs.ucl.ac.uk/staff/c.nentwich/publications/ RN/ht2001-xlinkit.pdf
both of which mention Schematron. See their application at

In my discussions with them, it came out that XLinkIt is more concerned
with ensuring
consistency within sets of distributed documents rather than linking
from schema to instance
errors. This is pretty clear in their patent application
http://l2.espacenet.com/espacenet/viewer?PN=AU5547300&CY=gb&LG=en&DB=EPD
where you send a set of documents for checking, rather than a single
document.

In other words, Schematron differs because:
 1) you validate only a single document as your starting point
 2) when you locate other documents, these are located using a URL
created from data
 in that document or in hardcoded the schema or passed in as a
parameter, or concatenated
 from those sources.

I don't believe, based on my earlier discussions with them, that their
patent prevents us
from generating validation result documents in the form of links from a
schema to an instance.
That is not "monitoring and maintaining the consistency of distributed
documents".

In other words, I gather that this kind of link is caught by the patent:

<error num="26">
  <data xlink:href="doc1.xml/#id(i26)" />
 <data xlink:href="doc2.xml/#id(ii26)" />
 <message>These two pieces of data are inconsistent</message>
</error>

but not

<error num="26">
  <data xlink:href="doc1.xml/#id(i26)" />
 <message>This piece of data is incorrect</message>
</error>

and not

<error num="26">
 <schema xlink:href="schema.xml/#id(rule64)" />
  <instance xlink:href="doc1.xml/#id(i26)" />
 <message>This piece of data is incorrect</message>
</error>

(I note that my company has put out a validator which automatically
generates links
between errors in a document and descriptions of the errors, and the
XLinkIt people
are aware of it and have never raised any objection. )

> I suppose we may need look into this more thoroughly ...
 
Yes.

Cheers
Rick

--
DSDL members discussion list
To unsubscribe, please send a message with the
command  "unsubscribe" to dsdl-discuss-request@dsdl.org
(mailto:dsdl-discuss-request@dsdl.org?Subject=unsubscribe)