UbiWeb Workshop Day 1, 9 March 2006 -- 8 Mar 2006

<scribe> Scribe: Kazuyuki

<scribe> ScribeNick: Kazuyuki

<olivier> should the record be public?

shall i scribe for karl?

for morning session?

<olivier> I will do it

thanks

<olivier> Dave Raggett starts introduction session

<olivier> going round the room for self introductions

yes, public

<olivier> thanks, kazuyuki

you're welcome

<olivier> ScribeNick: olivier

<scribe> Agenda: http://www.w3.org/2006/02/ubiwebws-agenda.html

chair introduces the workshop, organization

does anyone have a link to Dave's slides? can't see the URI from back of room

<Don> http://www.w3.org/2006/03/ubiweb-workshop-intro.html

thanks

Dave: Introducing sessions

<chaals-> [I guess I should have noted that I am also proposed (by the group) co-chair for WebAPI group and working in a couple of other areas in W3C]

Dave: Introducung Don Retallack, Boeing

Position Paper for Don Retallack, Boeing: http://www.w3.org/2006/02/Boeing.txt

tidbits from the talk (feel free to add to the log)

Don R: less concerned about people stealing our data than we are about people changing it (on Security)

<chaals> Don R: It is important to be able to rely on rendering in scenarios like factory floor use, telemedicine

scribe: we need things to talk to things, not people to things, things to talk to people

<chaals> Don R: useful to have a mechanic crawling around the inside of an aeroplane have access where they are to a drawing, not climb down to the floor, look at it, and climb back up, taking maybe 15 minutes or so.

<chaals> Don R: Need to be able to take a long-term view of this. The life span of an aircraft from conception to retirement from use is generally more than 50 years

Don R: all of our information (pay check, personal info) is now web based - huge investment

Don R: taking questions

Dave R: how are seemingly opposed goals of resilience against drop out and thin client satisfied?

Don R: we're working on such systems to have clients communicate even in loss of connection. p2p networks

Kazuyuki Ashimura, W3C: voice browser working group considered voice print for identification

scribe: you may be interested in such technology for security

Don R: we had contracts with US Gov for such technologies, actually

scribe: and much more
... voice input increasingly important
... time saving, in factories for instance, instead of badging

<chaals> "People in factorys like to input information with a wrench" (Actually, this is a serious use case - also applies to people working in dangerous environments with clumsy protective equipment, etc)

Dave R: what makes the Web more attractive than other frameworks?

Don R: it's for everybody

scribe: simplicity and universality of using the web
... and the sophistication of the protocols
... make it more interesting

[thanks, applause]

Shinichi Matsui, Panasonic, takes the floor

Panasonic position paper: http://www.w3.org/2006/02/panasonic-paper.html

use case oriented presentation

S. Matsui: presenting 2 business visions

central concept: "ubiquitous" and "global environment" from the home

scribe: expecting ubiquitous network society by 2010
... appliances digitized and networked (by 2010?)
... we are now in the end of phase of broadcasting dogitization, digitization of mobile
... "3D value chain"
... SD card used on multiple devices
... second D is "DVD", third is "DTV"
... evolved from HD (full high definition), CD (content distrib), UD (universal design)
... will talk about content distribution a little more
... importance of secure 3D platform
... alliance between appliance builder and content distributors

<chaals> [/me thinks "the ideal user interface is widely implemented for fridges"...]

scribe: House-wide broadband - no extra wiring should be needed
... use e.g power cabling, or coax, or wireless
... 1 button connection, 1 button diagnosis
... introducing HDPLC High Speed Power Line Communication
... (for High Definition pictures and Internet)
... Ubiquitous Home Appliances - disassemble and combine functions through digitization

S.Matsui: future is "organic combination by networking" of digital home appliances functions
... input, processing and output by network of appliances
... -> Ubiquitous Home Appliances
... Picture Monitoring "every/anywhere door phone"

<chaals> [So I could let people into my house if I am in another country, but could see and be sure it was them. That's actually a really useful thing for a group of people I know]

S.Matsui: skips lifestyle 2015 slides
... conclusion
... Requirements: secure and safe, extensible, works with some (any?) device, open and RF based

Q&A

J Hjelm: [missed q]

<chaals> Johan Hjelm: What do you think is a reasonable timeframe to expect a ubiquitous web?

S.Matsui: expecting ubi society 2010
... hope by 2008 or 9

[thanks, applause]

Nobuo Saito, Keio U taking the floor

paper: http://www.w3.org/2006/02/saito-paper.pdf

PUCC: name of consortium on p2p and ubi comp

N.Saito: Role of universities: neutral, innovative - can act for consolidation among industries,univs,countries
... PUCC technologies: p2p network, mobile integration, overlay network protocols, app/service integration
... shows architecture of PUCC
... (available in paper http://www.w3.org/2006/02/saito-paper.pdf )
... PUCC bridging between p2p home network and web/application
... listing members
... Web has an important integrating role to play

Q&A

Q: on PUCC protocol

scribe: why define a new protocol instead of using web services?
... SOAP application protocol

A: SOAP is a t higher level

scribe: it may work, but we are not sure
... we will start experimenting from this summer, try existing protocols

Q: you are proposing an overlay to interact with p2p network

scribe: are you proposing new naming mechanism/service
... to deal with resolving p2p resources ?

[please input names of commenters, if possible]

A: we're looking into solutions

Commenter 1 was Kapil Sachdeva, Axalto

Commenter 2 was F. Reynolds, Nokia

Tahar Cherif, Sharp - Toward Ubiquitous Web

scribe: presenting SHARP's view on Ubi web, quite close to what S.Matsui of Matsushita presented
... still see a huge challenge for accessibility
... defining access terminals

Sharp Position Paper: http://www.w3.org/2006/02/sharp-paper.pdf

T. Cherif: Shared terminals (ST), Personal Terminals (PT)

scribe: Use case 1: Input interface
... text input handicap, lack of secrecy for password input

Use Case 2: Personal Information management

scribe: Toward ubiweb: raising importance of ST
... personal mobile terminals
... (mobile phones, pdas)
... comparing ST and PMT
... collaborative web access
... linkage of ST and PMTs - securely
... linkage of shared and personal page
... and synchronization
... proposed solutions
... solution 1 input interface - shared page is updated based on the interaction with personal page
... effective text input, web navigation by PMT

solution 2 personal information management: storage in pnt, update in ST according to provided info. risk of leak is very low

scribe: Conclusion

Q&A

F. Reynolds: [missed q]

(discussion on sharing in parallel)

Don R: shows badge with chip including a lot of information (passwords included)

scribe: would like to be able to use this and get personal environment in shared kiosk

T.C: agree with approach

D.R: badge has a lot of technologies: bar code, magnetic stripe, chip
... how do we get all these systems to work together

Dave R: discussion session later on security

Charles MCN: personalized setup with a card exist, in libraries for instance

scribe: what is needed is standardization and harmonization
... key point identified in this presentation is what kind of info do you want to give
... give to whom, when, and where do you keep it?s example of google offering to host all your info

J. Hjelm: seems you are thinking of web as data display readable by humans

scribe: but it's much more than that
... all the data and information, how does that fit into your ideas

tx chaals

T.C: we have only identified some areas, are willing to participate. Agreed that this presentation focuses on display and usage by humans
... this is our approach for this new work

J.H: how would you see the sharing/split in the phone world?
... how do you envision avoiding leaking of personal information into shared environment

T.C: these are questions that the new work should answer

Kapil Sachdeva: wanted to answer Don's remark - as smartcard maker. will be the focus of my presentation

F.R: usability concern with having both ST and PMT - how does the user share attention between the two

T.C: depends on culture/generation. think of nintendo DS, multiple screen and multimodal input

[scribe reminds speakers that this is an international audience. speak slowly and in the microphone]

[thanks, applause]

<Kazuyuki> I agree with you ;-)

Franklin Reynolds, Nokia - http://www.w3.org/2006/02/reynolds-paper.pdf

F.R: ubiquitous web, upnpn and smart homes
... smart home: environment saturated with computing and networking
... smart spaces
... types of homes (many types - complicate connectivity and security)
... upnp forum (universal plug and play)
... upnp technology ipv4, xml, soap, ssdp, gena, presentation service device and service profile
... digital living metwork alliance (DLNA) consortium
... vision of a wired and wireless network of computers and consumer electronics
... CE companies are driving - they have different business models than communication or PC companies
... hate interdependence on other brands
... deployed for a long time - backward compat important

<karl>

F.R: they also need to sell replacement
... future proofing is important
... important: consumers are not trained

<karl>

F.R: they make bad network/security administrators
... hard to get everything to actually work - it has to work by itself
... consumer will buy dumb over smart if dumb more likely to "work"
... security - who owns, is responsible for all the devices? user change, too, people come and go
... data integrity compromission could mean lives in danger
... Model Mis-Match
... often there is no browser - machine to machine
... not all interactions are request-response
... some communication is one to many
... real world not easily modeled as decentralised data store of pages
... not stateless
... may have side effects, may move
... service discovery should be ubiquitous
... propose to integrate service discovery with URIs
... HTTP and SOAP
... slow and complicated, but works in most cases
... problem is in low and high end of performance requirements
... User Interface... web browsers not the best UI for control
... e.g page-at-a-time not natural for all devices

Q&A

Q1: have you tried multipass DNS?

multicast?

A: yes, but not an IETF spec

scribe: but still are problems when moving devices from one network to another
... adding key is part of the solution

Kapil Sachdeva: [acromyms overload, sorry, missed that]

WS-discovery for service discovery

Tahar Cherif: p2p paradigm - will it be used mostly for service discovery?

A: if I can relax what peer to peer means, then think it should be used everywhere

break out session: Security || Use Case

[reconvening]

Charles McCathieNeville summarises discussion of Security subgroup

(minutes to come)

-> http://lists.w3.org/Archives/Public/www-archive/2006Mar/0014.html

<chaals> CMN: (what I didn't say...) We looked at a couple of problems and use cases. Making things that are useful in factories and in home fridges is tricky - what works for securing an aeroplane might be unreasonable for securing a milk carton.

<chaals> ... how do you know that the printer you see in the airport is reliable enough to print a sensitive document? Can you prove it promised something then didn't do it - so you can go to court afterwards? Can you technically ensure that this is the printer? Can you find some information that you trust that says the printer is trusted by someone?

<Kazuyuki> welcome ;-)

<Kazuyuki> ScribeNick: Kazuyuki

Session on device coordination

- Kapil Sachdeva: Device coordination with Web applications

<Kazuyuki_> (back)

<Kazuyuki_> wonders why sometimes I'm dropped...

<Kazuyuki_> Kapil: explains Smard card and its applications

<Kazuyuki_> ... it includes client authentication based on two way SSL

<Kazuyuki_> ... in addition one time passwords

scribe: explains: Web applications and Smart cards
... architecture of Smart card security for web applications
... using Javascript
... mentions: Issues
... browser extension problem
... browser setting, device dependency, etc.
... explains: Network Addressable Devices

<karl> Component Extension API Requirements Version 1.0 - December 2001

thanks

Kapil: Position: standardization of web API

Q&A:

<karl> s/Member has/W3C Members have/

Franklin: how is the power of smart card?

A: additional reader using USB (e-card technology)

Charles: The WebAPI group is looking at APIs primarily for documents and applications, although they are looking to some extent at devices. But they are not handling the more complex question of security of device connections at the moment, and so browsers probably will not open up authentication and security via connected devices given the current web security

Michelle: What if somebody steal my smart card?

A: smart card is protected by your own PIN

Melinda Grant: The Ubiquitous Web...

<chaals> framework.../

thanks so much, charles

Gerrie: explains: Problem of Today's Web
... Usecases: Photo Archiving, Photo Shareing, Enterprise Print
... user actions and standard technology required
... HP's primary solution interests

HP's position: various standards must be integrated seamlessly

Q&A:

Johan: how to integrate?

A: don't jump to answers. people must @@@

DonR: @@@
... How can web server and browser access to the printer for example in Kinko's. using URL?

A: we need protocol to support server and client connection

Dave: (missed the Q)

A: scalable protocols for various devices and environments are required

Don Wright: The Ubiquity of Print

DonW: Today's Printing and the Web
... Ubiquitous Web for Print
... What functionality is needed?
... Printing Device Model?
... Tools available: XHTML-Print, CSS Profile for Print, PWG's work, UPNP, etc.
... How to discover Printers?: range, distinction, sub-net, geographic information
... How to delively job?: protocols (IPP, SOAP, HTTP PUT, others?), drivers, intermidiate
... Capabilities of Devices: PWG?, protocol/format
... Security issues: how to determine security needs?, What security is needed for printers?
... Path Forward: use cases and needs for devices

Q&A:

Charles: In Kinko's use case, how can we get the printer that couldn't print the job to say "well, there is another printer over there that can do it - is that OK by you?

thanks

Johan: how to determine possible sets?

<chaals> [kinko's in Singapore works for me if I don't mind the result being faxed... and trust Kinko's with information but not Chucko's...]

Gerrie: It can be done, depends on who has a motivation to implement this

thanks

<Reto> still, if I can detect Kinko's there is no reason why my printer cannot, is there?

Hide Tokuda: Ubiquity Japan - some video clips

scribe: A Love Triangle
... - combination of technologies already available in Japan at the moment
... - future technology: context-aware display
... - purchasing using ubiquitous device
... - location + presense service
... not only protocol but our life is important
... A First Love
... - Future elementary school
... - the child would like to extend his stay at Okinawa...
... - magic mirror: shows the past
... Problem of Uniquitous connection:
... - How to find services
... - how to distinguish devices?
... Resulution:
... - u-Photo: snapshot based approch: use information included in image file, using some special viewer
... - superinposing geographic information into picture

Discussion

thanks so much. see you tomorrow :)

Dave: 2 topics for discussion: (1) Printing and the Ubiquitous Web, (2) Device Coordination

Gerrie: Printing issue is interesting. But it's not easy.
... not exciting :)

[discussion by two groups]

Breakout reports and discussion

Dave: Printing and the Ubiquitous Web: @@@

<hallvors> "device coordination": whatever it takes to make your things work together?

oops

thanks

Toshihiko: Summarizes: Device Coordination
... drove three questions
... (1) Scalable Discovery
... - architecture on higher levels
... - vocabulary (ontology, semantic web) for attributes, functions, etc.
... (2) Values beyond existing technology?
... - scalable descovory is one of keys
... (3) Markup and scripts
... - what's scripts?: ECMAScript and/or Java
... (4) Misc
... - What's "device coordination"?
... - and its standardization?
... - current technology landscape in W3C: choreography
... - OASIS
... - request to W3C: too short slot!, many requirements!

Franklin: vocabulary?

Dave: @@@

Johan: light weight translation is needed.

Dave: mentioned tomorrows schedule and restaulant for dinner

<Reto> 12http://www.swuma.org/2006/ could be a target for vocabulary questions

thanks

ok, good boy

[day1 adjournd]

- DRAFT -

UbiWeb Workshop Day 1, 9 March 2006

8 Mar 2006

Attendees

Contents

Session on device coordination

Discussion

[discussion by two groups]

Breakout reports and discussion

Summary of Action Items

Scribe.perl diagnostic output