Re: one-way sameAs and friendOf links

On 7/19/12 7:54 AM, Michiel de Jong wrote:
> Hi Melvin,
>
> On Thu, Jul 19, 2012 at 1:05 PM, Melvin Carvalho
> <melvincarvalho@gmail.com> wrote:
>> +1 sameAs is going to have to become a fact of life, especially for anyone
>> that chooses to use webfinger ... you should read it as :  entity A is the
>> same as entity B
> I think you may want to rethink that one. You shouldn't just believe
> anything you read. I mean, people obviously sometimes say things that
> aren't true. You would get into situations like this one:
>
> http://www.imdb.com/title/tt0079470/quotes?qt=qt0471978
>
> Therefore, a sameAs claim, or any claim for that matter, should only
> be trusted in the /outgoing/ direction, never in the reverse
> direction, unless cryptographically signed by the other party.

It depends on the usage context. That's the critical point here. The 
good, bad, and ugly of owl:sameAs relationships have always been with us 
and aren't going anywhere soon. Thus, we have to deal with it as 
explained in my initial response re. conditional reasoning.

>
> It seems quite fundamental to me. In fact, if your bank would follow
> your advise, then i could easily go there, tell them "#me sameAs
> Melvin Carvalho", and they would hand me all your money, telling you
> "ah, but we just learned from Michiel that entity Michiel is the same
> as entity Melvin." :)
You are over simplifying Melvin's point. He knows enough about 
equivalence to not be implying a scenario with such a simplistic 
interpretation.

Kingsley
>
> Cheers,
> Michiel
>
>
>


-- 

Regards,

Kingsley Idehen	
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca handle: @kidehen
Google+ Profile: https://plus.google.com/112399767740508618350/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen