Why the Semantic Web will nerver work
1 like672 views
The document discusses the successes and challenges of achieving the original vision for the Semantic Web. It summarizes that: 1) The Semantic Web has seen more usage and success than initially expected, with technologies like semantic search, linked open data, and RDFa being widely adopted. 2) However, the vision of a unified network of linked, machine-readable ontologies has not yet been fully realized. The ontology and vocabulary story remains unclear, and there is a gap between linked data practices and formal ontology languages. 3) Bridging this gap will require new formalisms that can model the behavior of incomplete, decentralized linked data at web-scale, as well as research to unify competing models
Why the Semantic Web will nerver work
- 1. Tetherless World Constellation “Why the Semantic Web will Never Work” (note the quote marks!) Jim Hendler RPI http://www.cs.rpi.edu/~hendler @jahendler (sorry, not in rhyme)
- 2. Friends, Romans (& Greeks), Countrymen Lend me your ears Tetherless World Constellation • I have come to bury the semantic web, not to praise it
- 3. Does it mean Tetherless World Constellation • Why our critics were wrong when they said “The Semantic Web will Never Work” or • Why the Semantic Web will Never Achieve the Vision we had for it (at least if we don’t fix things)
- 4. Yes (not Xor) Tetherless World Constellation • Outline – Some current Semantic Web Successes – Revisit the Semantic Web vision • What did we say we would do – Review successes and failures • What has worked as well (or better) than we expected • What hasn’t – What are some challenges to overcome to achieve the latter?
- 5. Revisiting the Vision… Tetherless World Constellation
- 6. History Tetherless World Constellation >200 Semantic Web talks since 2000
- 7. Pre-History Who first conceived of the Semantic Web? Tim Berners-Lee (WWW Geneva, 1994) Tetherless World Constellation • "This is a pity, as in fact documents on the web describe real objects and imaginary concepts, and give particular relationships between them... For example, a document might describe a person. The title document to a house describes a house and also the ownership relation with a person. ... This means that machines, as well as people operating on the web of information, can do real things. For example, a program could search for a house and negotiate transfer of ownership of the house to a new owner. The land registry guarantees that the title actually represents reality.” – Tim Berners-Lee plenary presentation at WWW Geneva, 1994
- 8. Prehistory: 1st funding talk Oct. 1999 Beyond XML:Agent Semantics • DARPA will lead the way with the development of Agent markup Language (DAML) – a “semantic” language that ties the information on a page to machine readable semantics (ontology) • Currently being explored at University level – SHOE (Maryland), Ontobroker(Karlsruhe),OWL(Washington Univ) – Largely grows from past DARPA programs (I3, ARPI) • But not transitioning – W3C focused on short-term gain:HTML/XML <ONTOLOGY ID=”powerpoint-ontology" VERSION="1.0" <Title> Beyond XML DESCRIPTION=”formal model for powerpoint presentations"> <subtitle> agent semantics </subtitle> </title> <USE-ONTOLOGY ID=”PPT-ontology" VERSION="1.0" <DEF-CATEGORY NAME=”Title" ISA=”Pres-Feature" > PREFIX=”PP" URL= "http://iwp.darpa.mil/ppt..html"> <DEF-CATEGORY NAME=”Subtitle" ISA=”Pres-Feature" > <CATEGORY NAME=”pp.presentation” <DEF-RELATION NAME=”title-of" FOR="http://iwp.darpa.mil/jhendler/agents.html"> SHORT="was written by"> <RELATION-VALUE POS1 = “Agents” POS2 = “/jhendler”> <DEF-ARG POS=1 TYPE=”presentation"> <DEF-ARG POS=2 TYPE=”presenter" >
- 9. Berners-Lee et al, 2001 Tetherless World Constellation (May 21, 2001)
- 10. This leads to a radically new view of interoperation uses uses uses uses uses uses uses uses Distributed,partially mapped, inconsistent -- but very flexible! Brighton, Mar 2002 10
- 11. This leads to a radically new view of interoperation uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses uses Distributed,partially mapped, inconsistent -- but very flexible! uses uses Brighton, Mar 2002 uses 11
- 12. But, like the web… Brighton, Mar 2002 12
- 13. DAML Notional Schedule Tetherless World Constellation Now Later 2001: We will change the world!
- 14. Web “travel agents” How many cows are there in Texas? Query processed: 73 answers found Google document search finds 235,312 possible page hits. Http://www…/CowTexas.html claims the answer is 289,921,836 A database entitled “Texas Cattle Association” can be queried for the answer, but you will need “authorization as a state employee.” A computer program that can compute that number is offered by the State of Texas Cattleman’s Cooperative, click here to run program. ... The “sex network” can answer anything that troubles you, click here for relief... The “UFO network” claims the “all cows in Texas have been replaced by aliens “Agent” Markup Language Southhampton, 1/03 14 www.mindswap.org 14
- 15. Making Markup Easier Brighton, Mar 2002 15
- 16. Animal ontology Brighton, Mar 2002 16
- 17. Use that markup in query/portal interfaces Brighton, Mar 2002 17
- 18. Services need Web Logics 2001: Semantic Web Services Southhampton, 1/03 www.mindswap.org 18 18
- 19. 2003: Semantic Web Services Services off the desktop Southhampton, 1/03 19 www.mindswap.org 19
- 20. 2003: Semantic Web Services Services off the desktop Southhampton, 1/03 20 www.mindswap.org 20
- 21. 2003: Semantic Web Services Services off the desktop Southhampton, 1/03 21 www.mindswap.org 21
- 22. 2003: Semantic Web Services Services off the desktop Southhampton, 1/03 22 www.mindswap.org 22
- 23. So where have we got to Tetherless World Constellation • Semantic Web technology use has exceeded even my wildest expectations – What is different now? • Semantic Search – All the big kids are playing! • Advertising drives Web markets – “Markets are created by disaggregating the producer and the consumer” • “Buzz” around data on the Web – esp. Open Government Data
- 24. Example: OGP use growing quickly Facebook incentivizing use of RDFa like buttons Tetherless World Constellation 15,178 sites of top 1,000,000 as of 3/3/11 Oct 2010: FB reports RDFa is ~ 10-15% of > 3,000,000 likes per day! Facebook is encouraging developers to use the RDFa version
- 25. Because they want the links! Tetherless World Constellation The network is where their money is made! (predicted >$5B of advertising in next two years)
- 26. Creates a platform for SW-powered apps Tetherless World Constellation
- 27. They said it couldn’t be done Tetherless World Constellation • Common Criticisms
- 28. The Shirky fallacy Tetherless World Constellation • Folksonomy will win – Tagging the technology of choice • Tagging has largely failed to meet its promise – Tagging doesn’t achieve goals without “social context” • Example: Flickr tag “James”; Amazon tag “My-…” The Network effect requires links (Hendler & Golbeck, JWS, 2008)
- 29. The database community fallacy Tetherless World Constellation • The semantic web will never scale,1,000,000 triples and things go to heck Winner of the 2009 Billion Triples Challenge Just plain wrong!!
- 30. “ad hoc” data integration example: Linked Open Govt Data Tetherless World Constellation More than 50 of these at http://logd.tw.rpi.edu See also http://data.gov and http://data.gov.uk
- 31. And we do things the DB community struggles with Tetherless World Constellation
- 32. Another Shirky criticism Tetherless World Constellation • This is just a make-work program to keep AI scientists busy doing what they’ve always done • Cannot create an ontology at Web Scale • AI never works so it won’t this time • Logic and reasoning will not work on the Web because people disagree and because logic isn’t powerful enough for what is needed – (ok, he called it syllogism, but we know what he meant)
- 33. Sem Web 2010 Tetherless World Constellation April 2010
- 34. Semantic Web 2010 Tetherless World Constellation July 2010
- 35. Sem Web 2010 Tetherless World Constellation August 2010
- 36. Sem Web 2010 Tetherless World Constellation July 2010
- 37. Sem Web 2010 Tetherless World Constellation August 2010
- 38. Enterprise Semantic Web Tetherless World Constellation
- 39. The “bottom” of the Semantic Web Tetherless World Constellation • What is seeing the most use?? RDFa
- 40. The success of “Linked Data” Tetherless World Constellation • Maturation of RDF technologies – SPARQL endpoints • Fits Web development models – RDFa • Works well with current search paradigms – A little semantics goes a long way • BUT WHAT IS STUNNING IS JUST HOW LITTLE! – Equality via same URI – RDFa mostly w/DBMS not triple store – Not only no reasoning, but hardly any “principled” inferencing!
- 41. The bad news… Tetherless World Constellation • The ontology story is still confused
- 42. Ontology: the OWL DL view • Ontology as Barad- Dur (Sauron's tower): – Extremely powerful! Decidable Logic basis – Patrolled by Orcs inconsistency • Let one little hobbit in, and the whole thing could come crashing down
- 43. ontology: the linked-data view • ontology and the tower of Babel – We will build a tower to reach the sky – We only need a little ontological agreement • Who cares if we all speak different languages?
- 44. ontology: the linked-data view • ontology and the tower of Babel – We will build a tower to reach the sky – We only need a little ontological agreement • Who cares if we all speak different languages?
- 45. OWL has had successes Tetherless World Constellation • Examples from Clark and Parsia (2011) – Decision-support tool for sales people to automate policy driven cross-selling recommendations at very large US bank built out of RDF integrated data, OWL reasoning, and Pellet – At global 25 company (another bank) OWL and Pellet form the core of a bank-wide Entitlements service to represent, analyze, and query every access control policy for the entire bank, globally, in 50+ legal jurisdictions • And many other companies could claim similar – But most of these sorts of systems are still just coming out of prototype phase – And most are still more “expert” system than Web app
- 46. The tough love stuff Tetherless World Constellation • OWL is succeeding to a large degree as a KR standard – Building “expert systems” as a business has never gone away; OWL improves tooling • But it is largely failing in bringing representation to the WWW – cf. “misuse” of owl:sameAs >> “proper” use – cf. rdf:class >> owl:class – cf. it is rare that ontologies link to others
- 47. The gap is growing Tetherless World Constellation • Linked-Data-based applications are growing in size, number and importance on the Web – But the “vocabulary” story is still unclear • Ontology research is turning OWL into a usable KR standard, – But the linking story is still unclear No linking without vocabularies No network effect without links
- 48. What I think we MUST do Tetherless World Constellation • Bridging the gap between the linked- data and ontology views requires some key research challenges to be addressed – DL (and FOL) are useful formalisms for KR&R, but do not address the needs of the Web! – Empirical comparisons are useful in scaling systems, but do not address the needs of an academic community!
- 49. My Challenge to you Tetherless World Constellation • A sufficient formalism for Semantic Web applications must – Provide a model that accounts for linked data • What is the equivalent of a DB calculus? – Provide a means for evaluating incomplete reasoners • In practice we must be able to model A-box effects as formally as T-box technologies
- 50. Be bold! Tetherless World Constellation • A sufficient formalism for Semantic Web applications must also – Define what an ontology is ontologies really are • Including external referents linking between terms • Including ontology alignment partial mapping • Including non-expressive formalisms real-world “errors”
- 51. Be bold! Tetherless World Constellation • A sufficient formalism for Semantic Web applications must also – Define what an ontology is ontologies really are • Including external referents linking between terms • Including ontology alignment partial mapping • Including non-expressive formalisms real-world “errors”
- 52. It just might work… Tetherless World Constellation • One idea on how to get there – Define common problems that offer features of interest to both communities – Compare approaches with respect to performance – Develop hybrids that have best features of both as necessary – Repeat (thanks Bettina!)
- 53. IADIS-2008 Summary • The infrastructure needs of intelligent systems are now being met by a combination of Semantic Web, Linked Data, Web Services and Rule-based systems – Knowledge engineering can be jumpstarted from existing terminologies/ontologies, semi-structured systems, and other Web resources – Web Services (esp WSDL, SAWSDL) provide "wrappers" and other methods to let "legacy" systems play with agents – Reasoners and rule-based systems are scaling in new ways, and receiving some standardization • So where are all the agents???
- 54. Conclusion: “Why the Semantic Web will never work”? Tetherless World Constellation • No reason at all – The Semantic Web is here, it is working, and it will continue to do so • But, for it to move to the next level and be all that we as a community have aspired for – We must revisit and update the early visions for the modern web – We must unify the “competing” models of linked-data and machine-readable vocabularies – We must step up to some critical research challenges
- 55. Appendix Tetherless World Constellation • Research Challenges (ca. 2008)
- 56. Research Challenges • What is the Web culture? – Design/use/analysis are connected to "cultural stereotypes" (Think HSBC ads) • What are the cultural stereotypes in the emerging online community? • What level of "knowledge" is needed by Web users? – Is this dependent on application? User community? – Is expressivity a plus, minus, non-issue? • Especially in an open system (previous AI systems were "closed"
- 57. Research Challenges • Computational challenges as "end user" support – Scaling – Semantic Web HCI (What do we show "real users"?) • What are the trade-offs in use – Virtually all AI literature assumes a high-cost, high-value model – The Semantic Web is showing us alternative models • What are the trade-offs, analyses • If more and more of what we see includes integrated data from multiple sources, will that change the trust models – Do we need to expose provenance? Will "provider" model be changed?
- 58. Research Challenges • Who are the "experts" – What level of expertise is needed to become "dangerous" with this new technology? • What is the "ecosystem" (what is the equivalent of Web developer/web master/web user?) • If more and more of what we see includes integrated data from multiple sources, will that change the trust models – Do we need to expose provenance? Will "provider" model be changed? • Formal vs. informal models of ontology – I didn't discuss "folksonomy" but a key aspect is "social context" (Hendler & Golbeck, 08) • Can social contexts use
- 59. Research Challenges The Biggie