W3C home > Mailing lists > Public > public-privacy@w3.org > October to December 2010

Re: UK Government is suggesting a data deletion systems

From: Rigo Wenning <rigo@w3.org>
Date: Mon, 8 Nov 2010 17:25:08 +0100
To: Caspar Bowden <casparb@microsoft.com>
Cc: David Singer <singer@apple.com>, "public-privacy@w3.org" <public-privacy@w3.org>
Message-Id: <201011081725.08642.rigo@w3.org>

to delete _on-the-Web_ you have to know all resources in order to know what to 
delete. Knowing all the resources about me is the first privacy thing. But 
nobody knows all the resources on the web, so how could you assert that this 
is just nothing new? When building the web architecture, they forgot the 
delete button. Deletion is simply not in the architecture as we have no bi-
directional links. Having a 404 doesn't mean that it doesn't exist anymore. 

The legal deletion pretension for deletion already exist in most 
jurisdictions. Those are rather complex legal constructs needing a valid 
reason to ask for deletion in most cases. 

Consequently is a technical challenge and a privacy challenge at the same 
time. The privacy challenge is that in order to delete everything about you on 
the web, I have to know everything about you. The technical challenge is that 
the architecture of the web makes an "open world assumption" that nobody can 
possibly know "everything about you". All search engines united get to 
somewhat over 40% of the available information on the Web. There may be more 
information about you remaining in the 60% that is not indexed. And it can 
become indexed at any moment in time.

So if you can point me to a comp.sci article that has a solution, that would 
be great!

If this sounds provocative, you've read it the right way. :) I really want to 
challenge your mind here!



On Monday 08 November 2010 16:44:14 Caspar Bowden wrote:
> > request@w3.org] On Behalf Of Rigo Wenning
> ...
> > The delete button is one of the ideas floating around in Brussels at the
> > moment. The technical savvy people know that this is a research topic.
> Why? What is the engineering or comp.sci novelty?
> Caspar
Received on Monday, 8 November 2010 16:25:41 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:23:52 UTC