[foaf-dev] Re: Privacy When Aggregating Public Data?

Story Henry henry.story at bblfish.net
Thu Aug 23 14:02:18 BST 2007


Hey, I think this Bloom Filter is pretty neat. This could be useful  
in foaf.
In cases where we don't want to publish information about who we  
know, we could just publish
the bloom filter of our address book email addresses, and people  
could work out if we probably know someone.

<http://bblfish.net/people/henry/card#me> foaf:bloomMboxFilter  
"sdfsdf;j;k123413iifszflkn431" .

One could probably do the same of other identifying features such as  
aim skype or other ids.

I just thought I'd highlight this.

Henry

On 23 Aug 2007, at 06:09, NickLothian wrote:
>
> However, I think there might be a better way to do it. If an identity
> is packaged along with a bloom filter (http://en.wikipedia.org/wiki/
> Bloom_filter) of their contacts then when that person joins a new
> network it's possible to check all existing members of the network
> against that filter to see what links should be created.
>
> A similar scheme could be implemented for the identities themselves.
> Instead of sending all profiles to any system a bloom filter generated
> from the identities is sent. The new system can then ask "Someone here
> claims to know http://twitter.com/daveman692 - are you that person?
>
> Obviously this isn't as efficient as a foreign key-type lookup, but it
> has the big advantage that the new system will never know about
> contacts who aren't already on the system, nor will it be able to find
> identity profiles which it doesn't already have data on.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 2429 bytes
Desc: not available
Url : http://lists.usefulinc.com/pipermail/foaf-dev/attachments/20070823/eb904c87/smime.bin


More information about the foaf-dev mailing list