[rfc-i] issue: canonical formats

[duerst at it.aoyama.ac.jp ( "Martin J. Dürst" )]

On 2012/06/02 0:58, Peter Saint-Andre wrote:
> On 6/1/12 9:55 AM, Paul Hoffman wrote:
>> On Jun 1, 2012, at 8:12 AM, Joe Hildebrand wrote:
>>
>>> I do think we need a canonical output to support some of the legal use
>>> cases.
>>
>>
>> What the heck is a "legal use case"? Do you have any evidence of a law anywhere that says "the canonical format for publishing must be displayable"?
>
> Patent disputes and the like. One version says XYZ and another version
> says XZ because of some formatting glitch. What the exact definition of
> the technology?

We (both the list and Heather with input from IETF legal) have come to 
the conclusion that exact formatting isn't an issue as long as the 
content is the same (which is a technical problem and so we should be 
able to deal with it).

Just for a change, and some entertainment, I'd like to tell you about a 
court case that I was involved with that indeed at one point dealt with 
a formatting issue. If you look at the encoding table at 
http://tools.ietf.org/html/draft-duerst-dns-i18n-02#section-3.2, you'll 
notice that the column headings and the columns don't match. The same 
problem appears in draft-duerst-dns-i18n-01, but the table at 
http://tools.ietf.org/html/draft-duerst-dns-i18n-00#section-3.2 has it 
right.

During a deposition, this layout issue (which I somehow never was aware 
of, otherwise of course I'd have fixed it) and the fact that the patent 
involved used the same layout as draft -02 were shown to me. I was asked 
what I'd think about it, and I explained that to anybody understanding 
what's going on it was immediately clear that the layout in -00 was 
correct, and the others were weird. I think I also said that it was 
strange that in the patent, this problem wasn't detected and fixed.

As far as I am aware, that case never went to court, and so I'm not sure 
how much of a contribution to the case this formatting problem was. But 
if anything, it certainly didn't give a good impression of the patent.

Regards,   Martin.