Re: gap analysis (input regarding PML)

Hi James,

I think you've identified an import gap, namely, verifiability of the 
provenance. I don't know whether this belongs in the News Aggregator 
Scenario or another scenario but it should definitely be listed somewhere.

Technically, I know that the p-structure and the Semantic Web Publishing 
vocabulary are as you described: they represent claims about  data 
sources/derivations. Importantly, these claims can then be digital 
signed by the entity making the claim. In the p-structure, we had the 
notion of checking agreement between two participants in some message 
passing interaction, which allowed you to "verify" that they agreed that 
a message was passed. Whether you can verify whether a provenance trace 
actually corresponds to the real world beyond checking that yes x signed 
that it occurred, is a bit of an open question, I think.

Anyway, it's a good gap.

Paul

James Cheney wrote:
> On Aug 6, 2010, at 10:13 AM, Paulo Pinheiro da Silva wrote:
>
> <snip>
>
>> As a response to my message, Paul cited a European provenance project 
>> that according to Luc is based on p-streucture, which pre-dates OPM. 
>> The information that p-structure pre-dates OPM, however, is not 
>> enough for me to know whether p-structure provides a solution for 
>> connecting derivation traces to information sources that I cannot see 
>> in OPM of if it does, why the connection was not propagated to OPM. 
>> Also, it does not clarify the relation between p-structure/OPM and 
>> the technical issues in the original gap analysis. 
>
> Hi all,
>
> Being not intimately familiar with either PML or OPM, I'm confused by 
> the statements above that suggests that Paulo believes that PML has a 
> property that OPM might not have:
>
> "a solution for connecting derivation traces to information sources".
>
> I'm actually not convinced that either OPM or PML has this property.  
> To me, both appear to be data formats in which one can represent 
> information about *claims* made about derivations and information 
> sources, but these claims may or may not be verifiable against reality.
>
> Paulo seems to be referring to aspects of the "infrastructure" 
> developed around PML that provide stronger verifiability for 
> provenance claims.  These are (I think by design) not specified by 
> OPM, but I guess they were considered in other earlier work.
>
> Paulo, can you define this problem (and how PML / surrounding 
> infrastructure solves it) more precisely, or point to where it is 
> defined in one of the papers mentioned earlier in this discussion (for 
> those of us without time to read all of them)?
>
> I think it is an important observation that just having a data format 
> for provenance is not enough; there needs to be some infrastructure 
> that supports it to provide verifiability.
>
> --James

Received on Friday, 6 August 2010 12:36:05 UTC