last updated at 2004-05-20 22:15 UTC.

Berners-Lee on the TLD Explosion - Slashdot

AKTiveSpace demo

pshab: demo works in mozilla not ie

C4: Semantic Web Applications

dajobe: Les Carr, Building a companinon website in the semantic web
dajobe: for hypermedia and the web book, next version
dajobe: overview of approaches, starting from the raw textbook text
dajobe: then you get to add a variety of hypertext additions, at your choice
dajobe: tours, gloassaries, cross-refrences, ewxternal links
dajobe: [slides seem to be hypertext, no link atm]
dajobe: resources for students, instructors
dajobe: =CWS sites=companion website
dajobe: a good CWS improves the book reader's levels of understanding
dajobe: techniques used to support the book
dajobe: 1 update the text. 2 integrate with linking
dajobe: 3 use hypertext to underly the book structure 4. exercises for readers/instructors
dajobe: problems - maintenace
dajobe: publisher support does bare bones, but not helping with info management
dajobe: want to deal with, record the impicit knowledge that he authors know about relationships between the content parts
dajobe: so... semweb tech, built an onotology
dajobe: using with some semweb agents that compute and reason over this cotnent
dajobe: goal is to produce something that looks like a textbook but a lot of flexibility
dajobe: tried to avoid things like browsing concept hierarchies, it's not book like
dajobe: start with key concepts ideas in the book
dajobe: tricky part was aexpressing the relationships
dajobe: and this is exactly the value that an author provides
dajobe: 100 domain concepts, 75 is-a arcs, 75 other types of relationships
dajobe: different kinds of content in the book - descs, picturs, summaries examples and also differnet types of reader
dajobe: took about 10% of book content (not doing it all just yet)
dajobe: also added contet from HyperBank textbook and some from WebML
dajobe: stored in AKT triplestore
dajobe: HyberBook agent app, talsk to services, below which talk via RDQL to AKT triplestore
dajobe: web services provided for concepts, tours, provider navigation, provider information
dajobe: result - a web page
dajobe: but contains areas from multiple services and structures
dajobe: presenter Les Carr
dajobe: future work - revision notes - bare bones views, asking about it from some topic
dajobe: --end--
dajobe: Q: why rdf, not xsd?
dajobe: A: this is a generic q, why an ontology?
dajobe: .. trying to codify things that would cross boundaries
dajobe: ... you could compile this down to a use of xsd
dajobe: ... but here we are trying to extend the inferential processing (with rdf, owl)
dajobe: ... looking forward to extensionability, keeps me with semweb rather than compile down to some xml
dajobe: ... instance.

some photos from the conference for annotation perhaps

libby: tuesday, wednesday, thursday -- by libby

Brownsauce RDF Browser

mattb-lap: by Damian Steer

ERH live blogging on WWW2004 May 19

Standardized Uniqueness: Standards as the Key to Personalized Learning (Session P8)

pshab: Panel Session
pshab: no presentations - eclectic collection of people on panel so should have an interesting set of conversations on improving the effectivness of learning...
pshab: room not so full as NY Ballroom... cool with lots of power points :)
pshab: querying audience/panel about areas of interest/foci etc.
pshab: Panel into - Daniel Rehak (CMU), Phillip Dodds (ADLI), Tyde Richards (IEEE, Erik Duval (ARIADNE+), Robert Farrell (IBM), Peter ?, Wayne Hodgins (IEEE, LTSC, Autodesk)
pshab: areas of interest - personalisation, where are bodies moving, intro to SCORM,... invites audiecne to be proactive
pshab: acronym check, IEEE, ADL, AICC (CBT committee), IMS, ARIADNE, ... (we do lots of acronyms...)
pshab: Erik can't remember what ARIADNE stands for... :)
pshab: also LOM...
pshab: question - surely joking personalisation & standards in one sentence - oil and water?
pshab: Tyde - what are you standardising, e.g. if whole course, then no not much personalise, but if small pieces of content ok... granulaity
pshab: [can't keep track of who is talking... and type :)]
pshab: standars for tracking and reporting - part of scorm, have to standardise
pshab: problem is - what does personalisation mean? depends on who talking to... from very specific to very broad
pshab: doing hard stuff now. generative, knowing what from history should know, knowing what rules are
pshab: not want to standardise learning but learning technologies - analogy with clothing industry, clothes sizing not the actual clothes
pshab: :) do have different clothing standards in the world even within the EU.
pshab: need critical mass...
pshab: myths to be dispelled?
pshab: ... ability of issues of what should be mandatory elements of LOM for specific applicaition domains.
pshab: at present there are no mandatory LOM elements
pshab: ... learning objects becomes useless [in medical learnign object] if not tagged properly e.g. for whom, topic, etc. very specific
pshab: from audience - looking at scrom web page dont' get it - panel inline comment, nor do we :)
pshab: is it for agents or humans or ?
pshab: it is about interoperability of systmes nto about who (or what) creates the content.
pshab: standards dont' care about where the knowlede comes from but describes it
pshab: history of computer based larnign, e.g. pre-web interactive, branching etc. kinda lost wen moved to web...
pshab: SCORM aimed to help overcome that flattening...
pshab: SCORM Standards
pshab: ... characterisation of Web being not interacive back then was accurate. Now things have changed with SMIL, etc...
pshab: SCORM focused on making things more learner focused, shift from trainer/Inst. Design view
pshab: question - project in Norway... joining parents/childrne/teachers... crisis in education... need to motivate puipls/teahers ...
pshab: ... Web is there now... but dissapointment. Web access is personalised by defulat(? I think)...
pshab: so challenge... what should be put in the classroom, used in teh curriculum... can work with EU, UNESCO...
pshab: seeme dto ve about waht the real impact is on educaion in practice?
pshab: answer... need resoures to support the National Curriuclum (NC) ... scorm helps there diliver the content to teachers to work they way they want to...
pshab: questioner in audience offers to pay for 50% of casestudy type thing!
pshab: answer... interest in competances wihin curriucum usign scorm infrastructure... meeting the learning needs...
pshab: Question - Scope of work - re:use cases?
pshab: ... focused on use cases to enable web based approaches to do what people wanted to do today...
pshab: ... what we have don't do many things we would like to e.g. modeling user knolwedge, etc.
pshab: But do we know what we need to know to do this or do we need to do mroe reserach - we need to do more reserach to do more generative type interactions
pshab: ... Content Object Repository Discovery and Resolution Architecture (CORDRA)
pshab: CORDRA:http
pshab: allows/facilitates discovery of LOs + somethignj I missed
pshab: from web site "enabling model to bridge the worlds of learning content management and delivery, and content repositories and digital libraries"
pshab: missing thing was resolution of objects
pshab: question... about history of standards and processes?
pshab: history franco/american from aviation industry... manafactures/airlines
pshab: needed to have many cources and log student records. But had multiple systems so non-interoperable and managing it was a problem
pshab: so.. needed to be able to put the pices together, and need object to communicate... SCORM took that model and draw out relevant specs from outside.
pshab: erik makes point that not big bias by american mil i.e. early needs orignaolly form french work
pshab: now fully internationalised... some features of internationalisation that cant' be represented in schemas - working on it
pshab: internationalisaton guy from W3C... interested in the schemas problems [come back to that]... but
pshab: is there a way of describign diffirent 'cultural bias' of content
pshab: need to balance needs for needs of specific countires, etc. and keepign it widely applicable
pshab: work to help people do the localisation themselves
pshab: bindings of LOM to XML and RDF... SCORM is a reference model
pshab: from audience (Vince Wade) is the based on whihc SCORM is based wrong?
pshab: Web is serivce oriented... granuality of LO is really small < screen size, so descibe them and move them around... becomes inractable
pshab: need a serivcde interface - discover me, control me... Not have the LOs on your platform
pshab: ... i.e. not need to move the objecgts around but dilived as services. - I think i correctly paraprhase
pshab: ... problme is with very fine grained and adaptive objects
pshab: .. erik - not sure why think doesn't scale?... documetns must die/elecronic forms must die...
pshab: vincent - lots of objects per min. for really rich interaction. the dilivery would be better as a service?
pshab: ... anser, still need to describe the objects... lots of input from differetn people...
pshab: fine grained ... not necessaryily... [missed hte end]
pshab: little workflow metadata as yet... people werent' really ready for it
pshab: to process 10,000s of service calls on rich systme. Is that scalable? using XML/SOAP
pshab: Question - LO what is one? + [missed first part]
pshab: very general defn. objects ... shift of topic...
pshab: back to service issue - adaptive conent (less static, docs must die related issues) beyond the current scorm...
pshab: different classes of adaptation - as get to individual objects adn adding adaptivity is outside of the scope of hte current standrds
pshab: note - re-use defn. means 'more use' cf. Repurposing i.e. for a new purpose better scale of benefits
pshab: question about community of users... who is the user... what is unique to learnign as opposed or e.g. publising of sciecen related content?
pshab: e.g. learnign managment systme intergration <-- content
pshab: and exchange of that...
pshab: structure of stuff is nto sepecific to learning... but some parts of data model e.g. score of quiz. are
pshab: was tyde, eric. very little specific to learning, pretty generic.
pshab: our focus is on learning compared to entertainment... need to make sure that spec and standars interoperate with external standrds e.g. DC etc.
pshab: now need a consoludation phase ... bring standards togetehr - stack on which others e.g. learnign/entertainment could be built
pshab: question - SCROM only deals with some classes of learning.
pshab: defn. is subjecgtive learning= training, eduction, ?
pshab: question MIT open courseware (OCW) - where is SCORM in that? costs of implimentation of SCORM
pshab: actually OCW moslty pages of content?
pshab: OCW
pshab: SCORM does not fit all applications...
pshab: costs... change as tools devleop - usrs/tutors don't need/shouldn't see the SCORM any more that they see raw HTML
pshab: intergration of diverse conetn form diffrent providers is Much reduced as the standards are implimented.
pshab: missed next Q: but answers about intergration with other systems and interoperation
pshab: ok Q: someting about does scorm do collaboration in learning
pshab: out of time... guy who offered the 50% funding eariler is asking... why not focus on say, chemistry...
pshab: ... can we ... will follow on later
pshab: session ends

Semantic Web, Phase 2: Developments and Deployment by Eric Miller, W3C

libby: including advanced development, swad-e: thesuarus, sem blogging, calendaring
libby: future work - app-speciifc - see similie - devday
DanC: full room here in New York ballroom B
libby: don't forget to corner Dan Connolly and buy him a deli sandwich for info on DAWG
DanC: or some fruit. I got an awesome orange at a street vendor this morning. I already went nutso on NY pizza and deli food, thanks.
libby: BP wg - focus on deployment
libby: among other thiings, transforming some high-visible ontologies/vocabs to rdf
libby: semantic web homepage; semamtic web interest group (friendly :) - good places to start to get involved
libby: chuck Myers - adobe - experience of deployment of Semantic web
libby: all latest adobe products have rdf in them
libby: i.e XMP
libby: wanted easy to create metadata, consistent with using multiple vocabularies
libby: various vocabs supported: incl dublin core, exif...
DanC: dublin core, XMP, rights mgmt, media mgmt, job ticket mgmt, [darn; next slide]
DanC: presentation materials: Chuck Meyers, Adobe (PDF)
libby: next: Frank Careccia - Brandsoft
libby: website is often structured according to company organisation not how you do business
libby: documentation of content that ships with the product as a 'model'
libby: reducing confusion and duplication by automatically-discovered metadata about the creator, workflow, etc
karlcow: It reminds me a question, I had after the QA Talk yesterday. How do you keep your URI alive with a 10 years web site (W3C) and 875,000 Web pages. I replied do not confuse your URI space with the structure of your Web site.
libby: Dennis Quan IBM, bio-haystack
libby: many different public databases full of biological obejcts of interest - oroteins, genes etc, all with own protocols data formats etc
libby: life sciences identifiers (LSID) urns - rdf behind them
libby: naming system and also protocol - OMG sponsored; SOAP-based
areggiori: LSID specifications
libby: enabling non-programmers to use and manipulate this kind of information
libby: biological semnatic web using haystack
FransW: Haystack Home
libby: cool graph interface, drag and drop
libby: collaboration with mygrid (e-science, uk, Manchester, lots of usecases); also creator of LSID/RDF metadata from experiments by the user. using haystack to look at the results
libby: provenance of the experimental data is important - can view the web of provenance using haystack
libby: Jeff Pollock - Network Inference: business case, who description logics are important to customers
libby: SW tech as 'painkillers' for business problems today: lowerign oprational costs, info repurposing; adaptive business capabilities such as inferening, classification; automation of busines activities
libby: metadta prevalent in the business community already
libby: using tools on top of interoperable metadata
libby: business logic - putting it in OWl instead of in java
libby: semnatic data integration - a unified semnatic view of existing data - common query interface for it, non-proprietory way
libby: semantic web services - future use case, no users yet; think will be crucial...grid...
libby: customer: how does the company cope with classification of weirdo devices taht don;t fit neatly into categories, e.g. tivo, camera phones, for reporting sales
libby: autoclassification (using owl somehow(?))
libby: something that uses Xquery
areggiori: federated query using XQuery
libby: heathcare startup - matching care options to customers
libby: customers committed to owl DL rather than other vraiants, because: consistency....soem other stuff
libby: Dave Reynolds, HPL, braod overview of HP efforts to support the SWeb
libby: HP Labs role is to create opportunities for the company, nurture, reduce risk, move it along
libby: SW good fit for HP: broad application, open standards; but risk: network effoect technology
libby: so ...getting the network excited to start with, working with whole community, helping create standards, greater interop, standards very important, HP played large role
libby: also to kickstart needs cheap and easy tools to use. HPs tools: jena (java) rdf toolkit; companion joseki, rdf server
libby: also helping along applications such as damian steer's brownsauce browser :)
libby: Jena has a BSD license - anyone can pick it up, use it, even commercial, don;t need to tell us about it
libby: downside, we dont know who is using it, but we know a lot of people do. lots of apps using it
libby: joseki also very popular, input to DAWG group via W3C submission with ontopia
areggiori: RDF is "horizontal"
libby: no one SW killer app- explore broad range of options
libby: SWAD-E project: application for semantic blogging; also information aggregaion - information portals (see poster)
libby: SIMILE - digital library metadata; also content management
libby: knowledge managemnt and data integration (product catalogues, database integration, connections to KM - the latter an eyeopener, queries accross all that data)
libby: also Semantic web and web services - another EU project SWWS, supply-chain management - integration accrooss enterprises rather than within, hence importamce of open stds
libby: Q: have any of the panelests worked together to create interoperable
libby: network inference have used Jena; Dennis: similie is HP/W3C...uses haystack; eric...missed a bit sorry. symbiotic
libby: Q:relationship to EBXML possibly?
libby: daveR: one of the things on the landscape that need to be integrated together
libby: Q: handling data form such an open world?
libby: DaveR - default is to handle any data; might show some better. Dennis: ditto haystack, see saturday for presentation technology...? for query - you ahve to know what to ask for to ask queries; want to be able to get back unanticipated results
libby: Q:what was the key thing that m,ade the sweb interesting to their customers?
libby: chuck...file browsing..(?), dennis: standardisation of names, assembling fragments of information reduces effort
libby: ericm: add questions to the box for answering at the townhall

a9 search without personalisation and history

As We May Think: Vannevar Bush

A9 Search Engine

edd: fonts too small and screen too low contrast for me

Another clip shown in keynote

FransW: Videoclip "The proud worker" from 3dfx inc.

Smart Personal Object Technology (SPOT)


mattb: a research project of the Social Computing Group at Microsoft Research, exploring how people share media and build conversations in the context of social networks
pshab: there dont' seem to be externally visable demos or even screen dumps?
FransW: They have more info at the Social Computing Group webpage

World-Wide Media eXchange

mattb: ms research project
pshab: just looking at the MS WWME demo - looks cool there is a downloadalbe app
MarkB: looks like w3photo got slashdotted


Digital Convergence

mattb: 'humorous' video made for Bill Gates in 1994
mattb: used in Rick Rashid's keynote
WWW2004 community coverage

What's this?

This web site provides a hub for attendee-generated notes and coverage of the World Wide Web Conference 2004.

The conference is now over. Thanks to all who took part in creating the online coverage, which will remain archived on this site.



Official WWW2004 site
Developer's Day site
WWW2004 on the ESW wiki
W3 Photo and WWW2004 photos
Semantic Web Interest Group scratchpad
Planet RDF

Maintained by Edd Dumbill and Dave Beckett. Chump hosted by XMLhack. Logs hosted by Institute for Learning and Research Technology, University of Bristol. IRC was provided on Freenode, an interactive chat service of the Peer-Directed Projects Center.

Created by the Daily Chump bot.