SocialWebCampW3CTrack
Social Web Camp - W3C Track @ WWW2009
Context
The Social Web Camp, organized by W3C, was held during this this year's 18th International World Wide Web Conference in Madrid, 24 April. See W3C Track @ WWW2009 for a more detailed agenda.
The event featured a mix of structured content (talks and tutorials) and unstructured content. Topics of discussion for the afternoon session were selected at the camp. This Wiki page is intended to collect suggestions. If you're willing to lead a discussion, please add your name to a topic.
Topics discussed at the Camp
- FOAF+SSL hands-on
- Twwweet up sessions: on twitter Group photo
- Social Networks, microblogs and location
- Decentralized Social Networks User Experience - follow-up
- Security, Privacy and Access Control (report missing)
- Mapping between vocabularies (report missing)
Social Networks, microblogs and location
from memory by TonyAbouAssaleh
Participants: @tony_aa, @dirkahlers, @dajobe, @joshsh
- No technological limitations, many tools for detecting, encoding, and displaying location already exist
- How to integrate these tools into interfaces and micro-blogging service?
- How are people going to use them? Diverse needs and usage scenarios.
- Twitter imposes superficial limits -- unjustified
- Future of micro-blogging - how to overcome the size limit?
- encoding (short URLs, tags, abbreviations, location codes, etc)
- extreme case: gzip an encode the message
- encoding (short URLs, tags, abbreviations, location codes, etc)
- relax the limits
- Important that users have control over who sees their location information
- Important that users have control over the granularity of the location information
- point, block, neighbourood, city, region, etc.
- How to uniquely identify location?
- encode/decode by the client in a transparent way
- use natural language processing to disambiguate
Social Web and relations to Mobiliy/Context-awareness
(Notes from DKA)
Let’s not re-invent. Let’s re-use existing ideas of types of context.
Fading of information – value of context information varies over time – when you want to meet up with people and have a discussion you might want to know what room it’s in, but that information is only valuable at that time. Afterwards you only need to know that it was at WWW2009 in Madrid. Is this important?
More precise data kept for a longer time…
Do we need an expires-time for information? It’s important to keep things simple so maybe not.
What is proximity? Proximity refers to something to do with past behaviour of people. Social proximity can be divorced from actual location. Combined with distance-based location… Proximity is defined on underlying topology.
If you’re using a proximity-bases social network. Proximity doesn’t imply relationship. Physical proximity is only important if you’re already a friend of a person (if your social proximity is close)?
Use case – if you have social proximity then you happen to have physical proximity you might want to have a notification of that. Conversely you can infer a context from the physical proximity. Use cases for dating.
Changes of context are also interesting. [implicit] change of context can be used to change behaviour of applications or provide good user experience. A mobile phone sees where you are (home or work) and takes an action on your behalf.
Joined with calendar – you have information that the conference runs for so long, and you have an event after the conference at another venue… so you get some kind of smart notification. But E.g. Dopplr thinks I’m in Madrid but actually I’m in transit to London.
Granularity of groups of friends is an issue...
Events exposed through calendaring.
Context of “I’m in an airplane” or “I’m driving” or “I’m in a train” are important distinctions. “Nearest airport” is useful…
What do most humans care about? They want to publish a lot of detailed information about themselves to people that they know. Could they be triggered to publish more information on a change of context? “We noticed you’re in a pub right now – do you want to inform your nearby friend that you’re in this pub?” Even that might be too disturbing – tough to set up a double-blind system to inform on location/proximity. Also maybe this is too creepy.
Should it be illegal to discriminate based on your facebook profile when you’re hiring? Maybe this has to do with longevity of information…
Pre-camp Topic suggestions
Feel free to edit this section and append your own suggestion to the list or refine an already suggested topic!
Totally distributed social networking
Existing social networks still form closed communities. Could things be opened up?
Existing APIs
geolocation API OpenSocial (See Shindig in particular) vCard at IETF
Social data formats and mechanisms
[DISO Project http://diso-project.org/], XFN Microformat, hCard Microformat FOAF, SIOC, SKOS. OAuth, OpenID, FOAF+SSL. What do we need? SPARQLing social data and needed SPARQL extensions.
Identity and Privacy Technologies
OAuth, OpenID, TLS, SAML
Privacy risks
Machine-readable tags make it easy to link users information. How can users privacy and/or anonymity be preserved?
Reading Grimmelmann: Who's the bigger privacy risk -- the social network provider or your friends? (tlr)
PrivacyAndSecurityInSocialNetworks
Micro-payment solutions
Is the availability of micro-payment payments a key enabler for new business solutions based on social networks?
See ancient abandoned W3C spec on Micropayments.
Suggestion: consider a secure micropayment using a SPARUL INSERT over HTTP POST, and FOAF+SSL for authentication.
Capturing semantic data without bothering end users with it
I'm very interested in techniques which capture better information, especially links to URI's from other peopls systems, without bothering end users. My Mum won't enter a list of her alternate URIs... And I want it to be easy for me to implement.
Lightning Talks
Idea: one slide (optional) and two minutes time (sharp). Anything from announcements, controversial statements, project proposals, observations, etc. is welcome.
Suggested procedure: add your talk/topic, here or let Dom/TLR know at least a bit before the session (just to keep track how many people are interested).
- Michael on SPOT09 - call for demos and participation
- Michael on pushback, RDForms, pac etc. - what are the social implications and privacy concerns that come along with it?
- Harry Halpin on data-driven approach for social network models.
- Would like to see someone talk about OpenSocial
- Antonio Tapiador on web development frameworks (Ruby on Rails, Django, CakePHP..)
- Fabien Gandon on Semantic Social Networks Analysis Slides on Analyzing Global Semantic Social Graph
- Erik and Michael on incubating a W3C 'Location' Incubator Group
- Oshani Seneviratne on Social Web Related Projects at MIT
Who's coming?
If you're planning on participating to the camp, feel free to let other know by adding your name below:
- Dominique Hazael-Massieux, W3C, Mobile Web Initiative, co-Chair of W3C Workshop on Social Networking
- Thomas Roessler, W3C (tlr at w3.org); W3C Security Activity Lead; PrimeLife project
- MichaelHausenblas, DERI, Web of Data and multimedia stuff (still unsure if/how long I can be there)
- Fabien Gandon, INRIA, Fabien.Gandon@sophia.inria.fr
- Daniel Appelquist, Vodafone
- Phil Archer, W3C Mobile Web Initiative (phila at w3.org); i-sieve Technologies, Athens, chair POWDER WG.
- Harry Halpin, University of Edinburgh, co-Chair Social Web XG
- Oshani Seneviratne, MIT.
- Antonio Tapiador, Universidad Politécnica de Madrid
- Håkan Jonsson, Sony Ericsson
- RaphaelTroncy, CWI Amsterdam, co-chair of the W3C Media Fragments WG
- Domenico Gendarmi, University of Bari
- Martin Gaedke, Chemnitz University of Technology, Germany -- Tutorials Chair, @gaedke
- Yiannis Kompatsiaris, CERTH-ITI, WeKnowIt project
- JoshuaShinavier, RPI
- Samuel Santos, Present Technologies
- Tony Abou-Assaleh, GenieKnows.com, Canada, @tony_aa