Обсуждение: Re: [GENERAL] Map of Postgresql Users (OT)

Поиск
Список
Период
Сортировка

Re: [GENERAL] Map of Postgresql Users (OT)

От
"Magnus Hagander"
Дата:
Well, I can speak for Sweden. In which case it's basically that you
cannot put up *any* personally identifyable information that was not
already *explicitly made public* or that you have written permission
from the person in question, on any computer medium that can be used
outside the EU (which pretty much means the internet, it's kinda hard to
restrict it to the EU).

I have also read on other lists that there are similar considerations,
and eve nworse in some cases, in other countries. I'm afraid I don't
have any actual references to that.


Putting up pins with locations are fine, because that's not personal
information. When you put names on them, you are getting into a gray
zone.


Anyway, to my point. If there was a big gain by doing this, it'd
*probably* be ok, but I'm not sure. But as I don't see it adding all
that much, I just don't think it's worth risking it.

IP based locations are going to be far enough off that in most cases it
won't really be useful to know who is there. And there's going to be
loads and loads of them. But it may still be enough in *some* cases that
ppl might complain. Therefor, doing the email based list without names
makes sense.

Then having a map with manually registered people that *does* put out
names won't have as many pins on it, and the pins that are there will be
more accurate (because they are actually submitted by the person who's
there). As the information is provided by the person in question,
there's also no question of wether it's legal or not.

Bottom line: I think we should have two maps, fulfilling the different
needs. That'll also buy us out of the possible legal gray-zone.

(If we want to do the register-to-get-a-pin, it would be nice to
integrate it with the upcoming login system for the website, so ppl don'
thave to register more than once. Assuming this system is accepted once
it's done, of course :-)

//Magnus


> -----Original Message-----
>
> Well, we certainly do not want to do anything against the
> law. What countries would these be?
>
> On Dec 7, 2005, at 9:38 AM, Magnus Hagander wrote:
> >
> > It is an issue with the law in several countries. The two sets of
> > information may be public, but the combined info is still sensitive.
> > It's law, it doesn't have to make sense :-) I'm not sure if
> it applies
> > exactly in this case, but it could very well do, and it's just
> > unnecessary to risk stepping on it.
> >
> >
> > It's a different thing if we have people who want to be
> listed by name
> > register on a webpage. Then they volounteer the combined
> information,
> > and it's perfectly ok to use it that way (especially since we'd say
> > that's why).
> >
> > //Magnus
> >
> > ---------------------------(end of
> > broadcast)---------------------------
> > TIP 9: In versions below 8.0, the planner will ignore your desire to
> >        choose an index scan if your joining column's
> datatypes do not
> >        match
>
>
> ---------------------------(end of
> broadcast)---------------------------
> TIP 4: Have you searched our list archives?
>
>                http://archives.postgresql.org
>

Re: [GENERAL] Map of Postgresql Users (OT)

От
Robert Treat
Дата:
On Wednesday 07 December 2005 17:35, Magnus Hagander wrote:
> Well, I can speak for Sweden. In which case it's basically that you
> cannot put up *any* personally identifyable information that was not
> already *explicitly made public* or that you have written permission
> from the person in question, on any computer medium that can be used
> outside the EU (which pretty much means the internet, it's kinda hard to
> restrict it to the EU).
>

Well, it would pretty hard to argue that the information was not explicitly
made public since the bullet points were generated from a completly
autononomous 3rd party source.

> Putting up pins with locations are fine, because that's not personal
> information. When you put names on them, you are getting into a gray
> zone.
>

I odn't agree, but can't argue that you're opening up to more trouble by
including names.

>
> Anyway, to my point. If there was a big gain by doing this, it'd
> *probably* be ok, but I'm not sure. But as I don't see it adding all
> that much, I just don't think it's worth risking it.
>
> IP based locations are going to be far enough off that in most cases it
> won't really be useful to know who is there. And there's going to be
> loads and loads of them. But it may still be enough in *some* cases that
> ppl might complain. Therefor, doing the email based list without names
> makes sense.
>

I'd +1 this line of thinking. I don't see much argument against it.

> Then having a map with manually registered people that *does* put out
> names won't have as many pins on it, and the pins that are there will be
> more accurate (because they are actually submitted by the person who's
> there). As the information is provided by the person in question,
> there's also no question of wether it's legal or not.
>
> Bottom line: I think we should have two maps, fulfilling the different
> needs. That'll also buy us out of the possible legal gray-zone.
>

Well, I suspect we might actually have even more maps anyways, so it's no
problem to assume this out of the gate afaics

--
Robert Treat
Build A Brighter Lamp :: Linux Apache {middleware} PostgreSQL