[foaf-protocols] Poll on WebID Protocol Ontology Changes
nathan at webr3.org
Tue Oct 5 22:24:12 CEST 2010
Henry Story wrote:
> On 5 Oct 2010, at 21:47, Melvin Carvalho wrote:
>> I've attempted to add a poll summarizing some of the key topics that
>> have come up in the last week or so.
>> I've added my preferences, perhaps it will give greater clarity if we
>> could work out where we're in agreement, and where there's room for
> Thanks for putting that together.
> I voted, but thought I'd like to explain my vote a little bit first, so that it is not
Likewise, and after taking everything in to consideration, I put a MUST
Interested to here if somebody can say why both shouldn't be supported?
> As far as cert ontology and foaf go, both are RDF and can work well together. To move
> something from cert to foaf is not really going to gain us much traction, because
> in the RDF world, people don't have that much trouble with using different namespaces.
> (Though we could really move RSA into cert, if that was felt to make things a bit
> lighter, and if this is felt to be a problem. Initially I had them separate, because
> I did not know enough about the crypto space to understand enough how to do this)
> The people we need to convince are not at all in the RDF space. The big group
> we want to "embrace" won't usually care a jot about RDF - and could even still
> (though this is less likely, and becoming less so every year) hold this against us.
> I mean here people like the Portable Documents group, and such. They will want something
> that fits into an XML or JSON format. Thankfully we have GRDDL and GRDDL-JSON there
> to help us. If we convince that group we will gain real serious number of users. I
> don't see that we gain much by convincing the foaf group - they are on board already.
> So my feeling is that we are not using our resources that well by altering something that
> already works well.
> By convincing the XML people we will find out what they want. And that could be pretty
> much anything: inclusion of DER, PEM, mod/exp, .... we don't know yet. All that we know
> is that in the long term the only thing that counts is mod/exp - mathematically. Given
> that, we can integrate anything else.
> The issue of having an inverse for cert:identity is a bit minor, but it was useful
> for PHP people to have the inverse. Merging the rsa ontology into cert is even
> more minor. Moving a relation to foaf is I think a minuscule gain, and also quite
> restrictive as it essentially forces only one notation with binary components.
> Though we could also do a mod:exp notation, which is also restrictive for future
> But well, perhaps moving to the foaf namespace at the price of dealing with ASN.1 will
> make a huge difference. I just find that difficult to believe.
> Ok, I probably have been writing way too much here. We'll see what comes out in the vote.
> foaf-protocols mailing list
> foaf-protocols at lists.foaf-project.org
More information about the foaf-protocols