Re: License unknown

On Monday, June 23, 2014, Jerven Bolleman <me@jerven.eu> wrote:

> Booleans, in this case, are like answers on a math exam without
> showing your work. They might be right or they might be wrong, but no
> one knows how you got there.


Mind your metaphors ;)
How does the *datatype* make a difference on this matter? Does an xsd:float
tell the story of it's history? How do we know any more how a string or a
URI came to be the value of a property than a Boolean?


>
> I think it is critical in the semantic web that you describe what you
> know (you might still be wrong) as in a world where we try to share
> identifiers stating what you don't know often interferes with
> statements made by those who do know.

On this we agree. This motivates my proposal which states something we *do*
know, and seemed to provide the same level of information as the requester
made. I think that on the semantic web we should try to satisfy clients
needs, where that can be done without harming more important principles.

>
> But getting back: to the issue; Warning brain dump following
> The aim of the license statements is two fold. a) to state there is a
> license b) to describe what the license is.
>
> If the publisher does not know the license (e.g. "unknown") if they
> use blank nodes they assert there is a license.
> Which might not be true at all.

There is always a license, in the US, and many other jurisdictions. The
license says , effectively that only fair use is permitted, save a more
explicit license is given.

The publisher, in the sense used above, had no rights to publish if the
content is copyright able and they don't know that they have rights
otherwise.

I think it needs to be thought through what principles are intended with
respect to license. If it is to always respect law, then that is one thing.
If not, a whole set of other principles come into play. Marking the
document as "published without permission, caveat emptor" would see
minimally responsible in the latter case.

-Alan

>
> I think we could split the property: one for licensing organisation
> and another for license. This allows us to say that there is no
> license for this data set per se, but that you can contact a certain
> organization for a different one. It also helps for those cases where
> there is a free to use license with terms that are ok for academic
> purposes but unfriendly to most pharma/business. e.g. in software the
> AGPLv3 applies here. To help those businesses contact a licensing
> organization so that they can negotiate for a different license.
>
> On other option is one license property but model the licenses
> underneath better.
>
> _:kegg a dct:Dataset ;
>            :license [owl:someValueFrom (kegg:licenseForPersonInJapan,
> kegg:licenseForPersonOutOfJapan,
> kegg:licenseForOrganisationOutOfJapan,kegg:licenseForOrganisationInJapan,)]
>
> :swissprotIn1999 a dct:Dataset ;
>            :license [owl:someValueFrom
>              ([ rdfs:Comment "Academic use only" ],
>               [ a :NegotionedLicense ;
>                 :from _:GeneBio ;
>                 :to [ a :end_user_organistation ].
>               ])]
>
>
> We might need to have 3rd property as well to state the copyright
> applicability. e.g.
>
> _:info a dct:Dataset ;
>          :copyrightStatus :publicDomain .
> or
>
> _:info a dct:Dataset ;
>          :copyrightStatus [  owl:onPropery dct:creator .
>                                      :untilDateOfDeath [
>                                          "Some restiction to state 75
> years after creators death"
>                                     ]
>                                   ].
> etc...
>
> Regards,
> Jerven
>
> On Mon, Jun 23, 2014 at 4:30 AM, Alan Ruttenberg
> <alanruttenberg@gmail.com <javascript:;>> wrote:
> > What do you have against booleans? :)
> >
> > That seems like a sort of "too many notes" comment about Mozart's work,
> if
> > you can reach far enough to follow the analogy.
> >
> > -Alan
> >
> >
> > On Fri, Jun 20, 2014 at 3:55 PM, Joachim Baran <joachim.baran@gmail.com
> <javascript:;>>
> > wrote:
> >>
> >>   I am sure we can work out the exact predicate later. The issue I
> raised
> >> was about not using boolean.
> >>
> >>
> >> On 20 June 2014 12:51, Oliver Ruebenacker <curoli@gmail.com
> <javascript:;>> wrote:
> >>>
> >>>
> >>>      Hello,
> >>>
> >>>
> >>> On Fri, Jun 20, 2014 at 3:44 PM, Joachim Baran <
> joachim.baran@gmail.com <javascript:;>>
> >>> wrote:
> >>>>
> >>>>
> >>>> On 20 June 2014 12:41, Oliver Ruebenacker <curoli@gmail.com
> <javascript:;>> wrote:
> >>>>>
> >>>>>   Also, makes me wonder why the EBI has not already been contacted
> and
> >>>>> the license determined? Is that because we didn't have the resources
> to do
> >>>>> so or because different end users might end up being granted
> different
> >>>>> licenses?
> >>>>
> >>>>   That is my point exactly!
> >>>>
> >>>>
> >>>>   Even though the licensing information is not available, there is an
> >>>> indication where to obtain it from (here, in this example, EBI).
> >>>
> >>>
> >>>   Actually, my point is to first answer the question I asked.
> >>>
> >>>   If not only you don't know the license, but you also don't know why
> you
> >>> don't know, it is indeed hard to say anything about the license.
> >>>
> >>>   If all you want to say is who grants the license, you might want to
> >>> consider something like:
> >>>
> >>>   ex:myData   ex:canBeLicensedForEndUseBy   ex:EBI
> >>>
> >>>      Best,
> >>>      Oliver
> >>>
> >>>
> >>> --
> >>> Oliver Ruebenacker
> >>> Founder at Relomics Consulting
> >>> Be always grateful, but never satisfied.
> >>
> >>
> >
>
>
>
> --
> Jerven Bolleman
> me@jerven.eu <javascript:;>
>