[open-bibliography] Orphan data

[Tom Morris]

On Thu, Mar 1, 2012 at 8:16 PM, Karen Coyle <kcoyle at kcoyle.net> wrote:
> I have run into a case of "orphan data" that, in part, arises from a desire
> to be "open." In this case, an organization hosts -- but does not create --
> bibliographic data in a large database. There are no stated "owners" of any
> of the bibliographic records; they are being shared by a community with no
> one claiming any rights.
>
> When asked to let someone use the data under PDDL (or even CC0), the hosting
> organization declined because they consider themselves to have no rights to
> grant a license.
>
> Thus, openness has led to an "orphan data" problem. Potential users cannot
> obtain a license or even a statement of public domain because there is no
> declared owner.
>
> Any ideas for a solution?

There seem to be lots of people encouraging someone else to take a
risk.  A better approach would be to offer to indemnify them from any
and all damages if they take that risk.  That would show the depth of
your convictions.

This data didn't just appear out of nowhere.  Where did they get it?
If they provide the downstream consumers with the provenance
information, they can make their own risk assessment, in conjunction
with their lawyers, and decide what they want to do.

What does it consist of? If it's all unarguably data which is not
subject to copyright or any other type of IP protection (e.g. contract
law binding on some who received the data as part of service), then
that information can be provided to downstream consumers.

To my mind it's a little bit disingenuous of the hosting organization
to refuse to provide any information about the data.  The cynic in me
says that they know there's a problem, but don't want to 'fess up.

Situations like this are why organizations which deal in intellectual
property should be rigorous about tracking the provenance of data that
they get, mash up, redistribute, etc.  It's very difficult to
impossible to sort things out after the fact.

The reason the lawyers involved are skittish is because a) they're
paid to be paranoid and b) they know that people play fast and loose
with this stuff.  Having said that, it's not the lawyer's call at the
end of the day.  There will be some business type who takes the
lawyers risk assessment and makes a business decision as to whether or
not the risk is worth the reward.

Years ago I was involved in putting together a public domain software
compilation CD (back when "public domain" was the predominant flavor
of open source software).  Finding something on the net labelled
"public domain" wasn't good enough because it might have been some
proprietary piece of code that someone had rebadged.  For each program
we had to track down the author and get a letter stating that a) they
were the author and owned all rights to the program and b) that they
were releasing it into the public domain.  No author, no letter, no
inclusion in the compilation.

It's all well and good to tell someone else "go ahead in and take the
risk," but in the real world they're just going to ignore you (and
think you're naive/nuts to boot).

Tom