Re: Towards a Context TF Charter (was Re: Context in the XG?)

Hi Phil and all,

The idea of having a TF focused on a small-scale goal sounds great. I'm
working on a vocabulary of Online Presence and would like to contribute to
this work as much as time permits me. For a start I would have some comments
on the Charter text:

The Context Task Force is concerned with identifying and codifying the
> factors that are likely to affect which identity is active at any given time
> such that a user can easily switch between profiles.

This is very important. However, my perspective on this is slightly
different. I don't see a user as a profile-switching thing that can easily
draw a line between his different profiles (I find it difficult even to
define those different profiles). Let's imagine that I am in some kind of a
business mode (where business identity is active), I'm on work, on my chat
client I am available only for the people I work with, etc... Yet, it's
Friday, I'm dreaming of fun, so I want to broadcast a status message to my
personal friends to invite them on a beer. Should I then pass to some kind
of leisure mode to post this message? But then if in leisure mode I am
normally available for my personal friends, their chat messages could
distract me, and I don't want that.

So my concern is about the use of words ''active identity'' and ''switch
between profiles'', since I think that users are who they are, they do what
they do. Different identities arise in perception by others. Therefore I
prefer to say that different presence/identity data have their particular
audience for which they are intended to, rather than say that the user has
an active profile/identity.

I'm not sure how clear this is, so maybe my blog post about faceted online
presence [1] could help clarify it further.

Regarding the factors that affect the intended audience of presence data (or
the active profile if we call it that way), I have done a user study about
the intended audiences of status messages, and I believe it might be
valuable for this work (unfortunately it is not published anywhere yet, so
I'll make it available to you as soon as it is published, or prepare a
reduced version if needed). One of my goals was to determine what context
elements influence a status message to be dedicated to personal friends or
to business contacts or whatever audience. I assumed that user's location
(whether she is at work or at home), device being used (laptop, desktop,
mobile phone) can determine who the status message is for. To my surprise,
it seamed to have very little to do with those contextual elements. Content
type (like whether there is a picture link in a status message, or a link to
a song, etc.) was much stronger influence. I'll try to extract user stories
from the studies and put them on the wiki in the following days.

[1] http://www.milanstankovic.org/blog/?p=55

Best,

Milan Stankovic

e-mail: milan@milstan.net
homepage: http://milstan.net


On Tue, Jun 23, 2009 at 1:11 PM, Phil Archer <phil@philarcher.org> wrote:

> Hi all,
>
> I took an action last week to work with Claudio on developing a charter for
> a context task force. Reading through the various posts on this thread I
> _think_ I've got a clear idea of what the TF could do and how it could
> remain manageable.
>
> I'm sure everyone agrees with Alex that privacy and context are very
> closely related and I quite see why he's suggesting that the 'two task
> forces' are really treated as one. However, breaking down the work into
> manageable chunks, I think we can and should set a small-scale goal for the
> context TF, knowing, of course, that interaction with the putative privacy
> TF is essential. Actually, I think the context TF should get its work done
> quickly so as not to hold up the privacy TF (or anyone else).
>
> So what 's the work of the context TF? I think it should concentrate on
> delivering a vocabulary that describes user context. The text below is my
> first attempt at writing a charter. Claudio - you'll recognise some of your
> original post in here. Ronald - let's work on that vocabulary. SKOS may be
> relevant to us there.
>
> Comments?
>
> Phil.
>
> === begins ===
>
> Social Web XG Context Task Force Charter
>
> Background
>
> Human beings maintain multiple identities all the time. Although there is
> clearly an interaction and interdependence between them, people will assume
> and project different identities at work, at home, among friends, among
> family, among special interest groups and so on. An individual's use of
> social networking tools will vary according to which identity is in force at
> a given time, as exemplified in the Multiple Identities Use case [1].
>
> If context is the cause, the obvious effect is on privacy. Referring to the
> Multiple Identities use case for example, Kavita wishes to keep her
> membership of the Society for Creative Anachronism private from her family
> members. A documentary broadcast in the USA in 2008 called Growing Up Online
>  [2] details some very poignant (real) use cases in this area too. For
> example, it tells the story of what happened when a mother disregarded the
> privacy wishes of her teenage son and how for one girl it was her online
> persona that was the 'real her' – it was the conforming, moody teenager that
> was the fake.
>
> The Context Task Force is concerned with identifying and codifying the
> factors that are likely to affect which identity is active at any given time
> such that a user can easily switch between profiles. These include:
> location symbolic names (e.g. home, office, transportation)
> social activity (e.g. business meeting, with friends)
> related information (e.g. temperature, brightness, remaining battery &
> more),
>
> These factors provide input to systems that can, for example, protect
> privacy, control whether a given message is delivered immediately or later,
> whether particular events trigger alerts or are simply logged and so on.
>
> The Task Force Deliverables
>
> 1. A social context vocabulary that captured relevant aspects that could
> either be detected by the device automatically or provided by the user.
> 2. Examples of the kind of APIs that would be desirable to extract such
> data.
>
> Get Involved
>
> All members of the Context Task Force are members of the Social Web
> Incubator Group [3]. Join the discussion through the @@@publicly archived@@@
> mailing list at public-swxg-context@w3.org
>
> Documents
> ...
>
> [1]
> http://www.w3.org/2005/Incubator/socialweb/wiki/UserStories#Multiple_Identities
> [2] http://www.pbs.org/wgbh/pages/frontline/kidsonline/
> [3] http://www.w3.org/2005/Incubator/socialweb/
>
>
> Alex Korth wrote:
>
>> Hi all,
>>
>> I totally agree on that. It it absolutely neccessary that there is an
>> explicit context that is controllable by the user, e.g. the user's active
>> persona (private vs. business, see the profile switcher on your nokia as a
>> starting point), mood, and so forth.
>> But: do not let us draw a border to the privacy TF because these TFs'
>> domains are not disjoint. Let me give you a quote which is my favorite
>> definition of privacy:
>>
>> "Privacy is the claim of individuals, groups, or institutions to determine
>> for themselves when, how, and to what extent information about them is
>> communicated to others."
>> - Alan Westin, Privacy and Freedom, 1967
>>
>> As I mentioned in a past Email thread, privacy is not about secracy and
>> thus, massively overlapping with explicit context. I agree that privacy has
>> not much to do with the technical availability and standardization of
>> implicit context, e.g. geo-location, time and so on, but it does with its
>> accessability. Splitting up the team to TFs that are overlapping is no good.
>> I propose having privacy and context in one TF.
>>
>> Cheers,
>> Alex
>>
>>
>>  -----Original Message-----
>>>> From: public-xg-socialweb-request@w3.org [mailto:public-xg-socialweb-
>>>> request@w3.org] On Behalf Of Venezia Claudio
>>>> Sent: den 3 juni 2009 15:44
>>>> To: public-xg-socialweb@w3.org
>>>> Subject: Context in the XG?
>>>>
>>>> I've not received any comment to the mail below so far....
>>>> Is there any?
>>>>
>>>> Regards
>>>> Claudio
>>>>
>>>> -----Original Message-----
>>>> From: public-social-web-talk-request@w3.org [mailto:public-social-web-
>>>> talk-request@w3.org] On Behalf Of Venezia Claudio
>>>> Sent: Friday, February 27, 2009 10:56 AM
>>>> To: 'Harry Halpin'; Renato Iannella
>>>> Cc: public-social-web-talk@w3.org
>>>> Subject: RE: Proposal: Keep Group Unified, Don't Divide into Taskforces
>>>>
>>>> Hi Harry, all,
>>>>
>>>> as said during the workshop operators would be interested in getting to
>>>> a standard definition of social context, as a set of information which
>>>> might range from location symbolic names (e.g. home, office,
>>>> transportation) and social activity (e.g. business meeting, with
>>>> friends) to sensors' related information (e.g. temperature, brightness
>>>> and whatever),
>>>> this work has been partially carried out within W3C UWA but just with
>>>> respect to the delivery context,
>>>>
>>>> the aim would be collecting use cases, generate requirements (e.g.
>>>> functional, security etc) and collect the critical mass to
>>>> propose/achieve:
>>>>
>>>> 1) a social context vocabulary
>>>> 2) a definition of Web social context which would extend the web
>>>> browsing context
>>>> 3) the specification of APIs to get/set social contextual information
>>>> 4) a (privacy) security model
>>>>
>>>> We believe that contextual information will be more and more important
>>>> in social networking (especially in mobility) and we'd like to prevent
>>>> a massive market fragmentation,
>>>>
>>>> Regards
>>>> Claudio
>>>>
>>>> Questo messaggio e i suoi allegati sono indirizzati esclusivamente alle
>>>> persone indicate. La diffusione, copia o qualsiasi altra azione
>>>> derivante dalla conoscenza di queste informazioni sono rigorosamente
>>>> vietate. Qualora abbiate ricevuto questo documento per errore siete
>>>> cortesemente pregati di darne immediata comunicazione al mittente e di
>>>> provvedere alla sua distruzione, Grazie.
>>>>
>>>> This e-mail and any attachments is confidential and may contain
>>>> privileged information intended for the addressee(s) only.
>>>> Dissemination, copying, printing or use by anybody else is unauthorised.
>>>> If you are not the intended recipient, please delete this message and
>>>> any attachments and advise the sender by return e-mail, Thanks.
>>>>
>>>>
>>>
>>>
>>>
>>
>
> --
>
> Phil Archer
> http://philarcher.org/
>
> i-sieve technologies                |      W3C Mobile Web Initiative
> Beyond Impressions                  |      www.w3.org/Mobile
>
>