Re: Browser UI & privacy - a discussion with Ben Laurie

Thanks Brandt, for voicing these fears that I think stand mostly hidden
from view and unsaid behind a lot of resistance to WebID, and of course
which are difficult to answer when unsaid.

On 4 Oct 2012, at 18:58, "Brandt Dainow" <bd@thinkmetrics.com> wrote:

> Hi - I'm coming into this discussion late, and though I've tried to catch
> up, please forgive me if you think I've missed something in earlier stages
> of the debate.  However, as a philosopher concerned with online ethics (as
> well as a web analyst), I'm disturbed by the tone of this discussion, so I'm
> throwing in my point:
> 
> The idea that a person can be treated like a computing resource is
> questionable.  It sounds like instrumentalism - treating people as things,
> which is the starting point of most human evil.  The principle that an
> identifier in one system is portable to others refers to computing
> resources, not human beings.  There are no principles in web computing which
> were ever intended to apply to people.  This is why initiatives like WebID
> exist at all - they are trying to compensate for the fact the internet has
> nothing within it pertaining to humans.  

So in web language a URI is a Universal/Uniform Resource Identifier. That
is a bit of a sleight of language, which came from the birth of the web as
a global distributed publishing system, which published *resources*. In logic 
we would just have called those Objects. See for example "Word and Object" 
by Villard Van Orman Quine, where the notion of object I think is used
in a large enough way to contain subjets. Subjects are essentially those
things that require belief/desire vocabulary to describe, and where
substitution of coreferential statements salava veritate does not hold. 
That is the following deduction does not hold:

A. Lois Lane loves Superman.
B. Superman == Clark Kent .
-------------------------
C. Lois Lane loves Clark Kent

C does not follow from A and B because you need to take subjectivity into
account. You can of course reason in subjective positions, but things
become a lot more complicated...

So we are not trying to treat people as resources, nor as objects. 
If you look at my Philosophy of the Social Web you'll find a whole
section dedicated to psychology

  http://bblfish.net/tmp/2010/10/26/

What is going to be possible is in fact quite the contrary of what you
fear. We are going to be able to develop subjectivity aware tools, that
will help make people more aware of different people's beliefs. 

Would that not be great if everybody interacting with their daily tools
could ask questions such as what do my close friends think of X. What do
the others think of X. To the point where people understand that all 
data is to an extent subjective, that it can be merged and unmerged,
that there are an infinite points of view that can be taken. 

In such a world what strength would the idea of a big brother have,
other than just another person, with a bit more experience perhaps,
but also with a partial view of things.

> 
> The concept of a "reputation footprint" is also highly debatable.

yes, because you are thinking of a centralised reputation footprint.

> Personally, I find the idea that I would have a single online profile,
> uniting all my web activities, and traceable back to the real human me, as
> horrifically totalitarian, and a step backward.  I don't have such a
> limitation in the real world.  I can be anonymous when I walk the city,
> enter shops, and pay by cash.  I can conceal my religious or political
> beliefs from my workmates, so as to avoid being judged by them on irrelevant
> criteria, or simply because I want to live privately.  I can decide my life
> has been a mess, then move to a new city, where no one knows me, and start
> afresh, my previous history forgotten.  We must have the same level of
> forgetfulness on the web, the same ability to split our activities and
> present only partial views of ourselves to different groups.  These are
> fundamental aspects of human existence which have remained for thousands of
> years.  They enable us to work and socialise with others who we otherwise
> would be in conflict with.

So yes. We are not speaking of a single online profile at all. You can have
many, link them if you want or not, have short lived ones, or longer lived
ones. It's up to you.

We have been arguing for a web that makes it easier for a user to see what
his relation to the web site he is connected to is. Is he anonymous, or is
he logged in. As who? All this should be transparent.

> 
> Organisations are different.  They are not people.  Any initiative which
> treats organisations, documents and human beings as the same is denying the
> essential dignity of the individual, and their right to chose how openly or
> privately they wish to live.  I can understand why I might want a system
> which enables me to lock my identity to a resource, but that should be a
> voluntary system, and it should enable me to have multiple WebID's (or
> equivalent), and it should permit me to keep my personal identity totally
> anonymous.  

Yes, if we treated humans and documents and organisations as the same we'd be
crazy. But we are not. 

And we do allow multiple WebIDs.

I have 

 http://bblfish.net/#hjs
 http://bblfish.net/people/henry/card#me
 
I really should get working on having a Tor WebID too, with a .onion URL
on a different domain. 

> 
> WebId is a particularly dangerous concept.  It totally depends on the
> unbreakability of the private key.  Does anyone in this group seriously
> believe there's such a thing as unbreakable encryption, or a flawless
> computing system?

I think few people believe that. Security like knowledge is a relative 
concept. You can have more or less of it, but never have totally secure system.

>  If people trust WebID's, what chance do you think anyone
> will have of convincing the world their WebID has been faked or hijacked, or
> their certificate stolen, etc?  

Well it should be relatively easy I hope. What I am not sure is why you think
that people will have more luck with changing their Google+ or Facebook 
accounts. Do you think those institutions really know the millions of people
whose accounts they host? They are using social networks to do the work for 
them. We are just trying to distribute it back. So don't you think that if
you had a bank WebID that your bank would at least know you person to person
better than what a large social network provider does? 

If your Freedom Box ( see Eben Moglen talk about it on CBS here 
http://www.youtube.com/watch?v=SzW25QTVWsE ) gets hacked, then you can at least
reset the software, or look at the source code, or give the source code to a
good friend to find out how the flaw arose. Such systems could potentially
be on the whole a lot more secure because millions more eyes will be able
to look at them.

> If WebID was used for government, financial
> or employment purposes, what harm could fall on someone under such
> circumstances? The same is true of any computing system which seeks to lock
> an IT resource to a real person.  The connection between the two will always
> be problematic and untrustworthy.

IT cannot "always be problematic" or else the web would not function. What
can be done is to make it more trustworthy, more secure, and easier for
people to work together on.

> 
> In terms of online privacy, we cannot possibly imagine what use nasty people
> will make of personal data 10, 20, or 50 years from now.  We simply cannot
> know what technology or business models people will invent.  All we can be
> sure of is that stuff we can't imagine now will dominate the web of the
> future.  This means we can't argue in terms of trying to achieve specific
> effects, because we can't know what the full range of effects will be.  The
> only solution is to focus on avoiding the potential for harm.  This means we
> must take a fantastically conservative attitude to online privacy, and
> resist every attempt to reduce it.  In this light, one has to ask - where
> are the anonymity initiatives?  Where's my IP-rotation plug-in, my user
> agent obfuscation add-on, etc?

The problem with doing nothing is that this is exactly what will lead you 
to the totalitarian nightmare you want to avoid. Because as you make it
difficult to distribute information, and allow people to control their bits
of their data, you end up allowing huge entities to become bigger and bigger
and own everybody's data.

Look at my Philosophy of the Social Web presentation on my home page for 
a detailed explanation.


> 
> The web is a fairly good thing as it is.  Before we seek to "improve" it, we
> need to be absolutely certain we are addressing a genuine problem and that
> the solution won't harm more than it helps.  In the larger context, this
> means "Web-scale verifiable identity" should be no more than a minor item of
> optional technology used by a few people for specific purposes.  It should
> be enacted in a manner which is aware nasty people and governments could
> force it on people as  a means of exploitation and control, which means
> making it hard to manage centrally and avoiding uniform standards.  

Exactly: I like WebID because it can work without central management, and
it uses only existing standards. It could work with Tor URIs so you can
even protect yourself from people knowing where your server is located.

> The emphasis should always be on the avoidance of possible harm, even if this
> means not getting the best technology.

I agree. And the harm is that if we don't allow for distributed, decentralised
authentication where you can be in charge of your data, you will have centralised
authentication, where only a few players are in control of it. 

So it already works. We are using it. :-)

> 
> 
> Regards,
> Brandt Dainow
> bd@thinkmetrics.com
> www.thinkmetrics.com
> PH (UK): (020) 8123 9521
> PH (USA): (801) 938 6808
> PH (IRELAND): (01) 443 3834
> iMedia Articles: www.imediaconnection.com/profiles/brandt.dainow
> 
> This email and any attachments are confidential and may be the subject of
> legal privilege. Any use, copying or disclosure other than by the intended
> recipient is unauthorised. If you have received this message in error,
> please delete this message and any copies from your computer and network.
> 
> Whilst we run anti-virus software on all e-mails the sender does not accept
> any liability for any loss or damage arising in any way from their receipt
> or use. You are advised to run your own anti-virus software in respect of
> this e-mail and any attachments. 
> 
> 
> 
> 
> -----Original Message-----
> From: Kingsley Idehen [mailto:kidehen@openlinksw.com] 
> Sent: 04 October 2012 16:59
> To: Hannes Tschofenig
> Cc: Melvin Carvalho; Henry Story; public-webid@w3.org;
> public-identity@w3.org; public-philoweb@w3.org; Ben Laurie
> Subject: Re: Browser UI & privacy - a discussion with Ben Laurie
> 
> On 10/4/12 11:10 AM, Hannes Tschofenig wrote:
>> Hi Melvin,
>> 
>> On Oct 4, 2012, at 4:49 PM, Melvin Carvalho wrote:
>> 
>>> I think the aim is to have an identity system that is universal.  The web
> is predicated on the principle that an identifier in one system (eg a
> browser) will be portable to any other system (eg a search engine) and vice
> versa.  The same principle applied to identity would allow things to scale
> globally.  This has, for example, the benefit of allowing users to take
> their data, or reputation footprint when them across the web.  I think there
> is a focus on WebID because it is the only identity system to date (although
> yadis/openid 1.0 came close) that easily allows this.  I think many would be
> happy to use another system if it was global like WebID, rather than another
> limited context silo.
>> I think there is a lot of confusion about the difference between
> identifier and identity. You also seem to confuse them.
>> 
>> Here is the difference:
>> 
>>    $ Identifier:   A data object that represents a specific identity of
>>       a protocol entity or individual.  See [RFC4949].
>> 
>>  Example: a NAI is an identifier
> 
> A data object is denoted by an identifier. The representation of a data 
> object is a graph. An data object identifier can resolve to said data 
> objects representation.
> 
> A Web accessible profile document is an example of a data object.
> 
> On the Web a profile document can be denoted by an HTTP URI/URL. In 
> addition, the subject (which can be *anything*) of a profile document 
> can also be denoted by an HTTP URI. Basically, this is what the Linked 
> Data meme [1]  by TimBL is all about. Note, WebID is fundamentally an 
> application of Linked Data principles specifically aimed at solving the 
> problem of Web-scale verifiable identity for people, organizations, 
> software, and other conceivable entities.
> 
>> 
>>    $ Identity:   Any subset of an individual's attributes that
>>       identifies the individual within a given context.  Individuals
>>       usually have multiple identities for use in different contexts.
>> 
>>  Example: the stuff you have at your Facebook account
>> 
>> To illustrate the impact for protocols let me try to explain this with
> OpenID Connect.
>> 
>> OpenID Connect currently uses SWD (Simple Web Discovery) to use a number
> of identifiers to discover the identity provider, see
> http://openid.net/specs/openid-connect-discovery-1_0.html
>> 
>> The identifier will also have a role when the resource owner authenticates
> to the identity provider. The identifier may also be shared with the relying
> party for authorization decisions.
>> 
>> Then, there is the question of how you extract attributes from the
> identity provider and to make them available to the relying party. There,
> very few standards exist (this is the step that follows OAuth). The reason
> for the lack of standards is not that it isn't possible to standardize these
> protocols but there are just too many applications. A social network is
> different from a system that uploads data from a smart meter. Facebook, for
> example, uses their social graph and other services use their own
> proprietary "APIs" as well.
>> 
>> This is the identity issue.
>> 
>> You are mixing all these topics together. This makes it quite difficult to
> figure out what currently deployed systems do not provide.
> 
> Henry isn't mixing up the issues. What might be somewhat unclear to you 
> is the critical role played by Linked Data, and the fact that a WebID is 
> just a cryptographically verifiable denotation mechanism (an identifier) 
> for people, organizations, software agents, and other real world 
> entities that aren't Web realm data objects (or documents).
> 
> Linked Data introduces a power nuance that enables you leverage 
> *indirection* via the use of HTTP URIs to unambiguously denote a Web 
> realm data object (e.g., a profile document) and a real world entity 
> (that's the subject of the profile document) described by said data 
> object. Net effect, either denotation resolves to the same document 
> content (actual data or Web resource). The documents in this context are 
> comprised of RDF data model based structured content i.e., an 
> entity-attribute-value or subject-predicate-object graph.
> 
> Also note that WebID and OpenID bridges already exist in the wild that 
> work, and these serve as powerful demonstrations of the value that WebID 
> brings to bear.
> 
> Links:
> 
> 1. http://www.w3.org/DesignIssues/LinkedData.html -- Linked Data meme
> 2. http://bit.ly/OcbR8w -- WebID+OpenID proxy service showing how 
> password authentication is eliminated from the OpenID flow via WebID
> 3. http://bit.ly/PcQg38 -- screenscast showcasing the combined prowess 
> of OpenID and WebID.
> 
> 
> Kingsley
> 
>> 
>> Ciao
>> Hannes
>> 
>> 
>> 
>> 
> 
> 
> -- 
> 
> Regards,
> 
> Kingsley Idehen	
> Founder & CEO
> OpenLink Software
> Company Web: http://www.openlinksw.com
> Personal Weblog: http://www.openlinksw.com/blog/~kidehen
> Twitter/Identi.ca handle: @kidehen
> Google+ Profile: https://plus.google.com/112399767740508618350/about
> LinkedIn Profile: http://www.linkedin.com/in/kidehen
> 
> 
> 
> 
> 
> 

Social Web Architect
http://bblfish.net/