[okfn-discuss] Right to be forgotten ruling

Tue May 20 10:47:36 UTC 2014

On 20 May 2014 07:22, Tony Bowden <tony.bowden at gmail.com> wrote:

> On 20 May 2014 02:06, Anna Daniel <a.daniel at griffith.edu.au> wrote:
> >  My position is that a fact is a fact and should remain so.
>
> Most facts don't fall into the "universally true for all time"
> category. This isn't just pedantry — a lot of the discussion in this
> area comes down to this sort of framing. We naturally treat some facts
> as having an obvious time component, but generally aren't as good at
> doing so with classes of information with a longer half life.

Good point Tony.

> > If it's been released into the public sphere it should stay there.
>
> Many countries already have laws that recognise a value in allowing
> information to decay — e.g. expunged records or spent convictions
> under Rehabilitation of Offenders laws. There are certainly
> interesting technical challenges to achieving things like this in
> practice, but that doesn't mean that the underlying goal isn't a
> worthwhile one.

Excellent point here - I was also going to mention this re convictions. The
interesting questions are the interaction between those technical
challenges (and the differential burdens they may impose) and those
worthwhile goals.

> > Secondly, putting the
> > liability onto intermediaries rather than content owner (person/entity
> who
> > made the decision to make the data public) is a dangerous precedent
>
> Anyone republishing information — even just in a passing reference to
> it — is already open to lots of legal challenges in most jurisdictions
> anyway (e.g. libel). I don't believe anyone is claiming that the
> intermediaries must proactively decide what they can and can't publish
> — simply that (as with lots of other areas), once their attention is
> drawn to problematic material, they have to remove it

Very good points too. It is quite correct that one has legal (and moral)
obligations to correct "libellous" and generally incorrect information. I
think the issue in this case is that one was essentially working with
official or semi-official info which one would normally assume to be
reliable plus the processing in an automated way (no editorial function).

As such, the potential scope of the ruling is both very broad and could
potentially have a significant negative impact on many projects (including
open data ones) that prepare databases by collecting and processing data
from various sources.

Rufus
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/okfn-discuss/attachments/20140520/a059190b/attachment-0002.html>