The Fractal Web
http://www.w3.org/2008/Talks/0617-fractal-tbl/
Tim Berners-Lee
MIT Computer Science & Artificial Intelligence Laboratory (CSAIL)
University of Southampton School of Electronics and Computer Science
This talk
- Fractals and Zipf on the Web
- Engineering for a scale-free systems
Shapes of data
- lines: tape, cards
- matrix: databases
- Trees: SGML, XML, Top-down structured design, OO
- Net: Internet (?), WWW (?)
Cultures, Groups and boundaries
- Groups form their own language
- This makes them work
- This forms a barrier
- This is a constant trade-off
- Work with a larger or smaller group?
Cultures, Groups and boundaries
Extreme 1: Monoculture
- Culture: MacDonalds
- Engineering: One huge DB. Cyc
- Many people feared for web
- Can't meet the requirements of diversity
- Develeopment slows to a crawl
Extreme 2: Extreme diversity
- No overall coordination
- Culture: Cults, e.g Heaven's Gate
- Engineering: Individual OO systems, DB systems
- Lack of general communication
- Rapid deevlopment
The shape of the web
Society includes communities on many scales
Universal WWW must include communities on many scales
Applications connected by concepts
For example in biopax
[Diagram:
Joanne Luciano,
Predictive Medicine Drug discovery demo using RDF, Siderian
Seamark and
Oracle
10g]
- Communities will be of many sizes.
- There will be very many small ones (6.10^9 of size 10^0)
and a few global ones (e.g. W3C Recommendations)
- Kleinberg shows
that fractal (1/f) distribution is optimal under some
assumptions
- Swoogle
results for example (right)
- We have less experience when fractal is not constrained
to a 2D surface.
Total Cost of Ontologies (TCO)
Assume :-) ontologies
evenly spread across orders of magnitude; committee size
as log(community), time as committee^2, cost shared across
community.
| Scale |
Eg |
Committee size |
Cost per ontology (weeks) |
My share of cost |
| 0 |
Me |
1 |
1 |
1 |
| 10 |
My team |
4 |
16 |
1.6 |
| 100 |
Group |
7 |
49 |
0.49 |
| 1000 |
|
10 |
100 |
0.10 |
| 10k |
Enterprise |
13 |
169 |
0.017 |
| 100k |
Business area |
16 |
256 |
0.0026 |
| 1M |
|
19 |
361 |
0.00036 |
| 10M |
|
22 |
484 |
0.000048 |
| 100M |
National, State |
25 |
625 |
0.000006 |
| 1G |
EU, US |
28 |
784 |
0.000001 |
| 10G |
Planet |
31 |
961 |
0.000000 |
Total cost of 10 ontologies: 3.2 weeks. Serious project: 30
ontologies, TCO = 10 weeks.
Lesson: Do your bit. Others
will do theirs.
Thank those who do working groups!
Engineering for scale-free systems
Allow groups to form
- DNS allows new URIS to be owned
- RDF, RDFS, OWL, allow anyone to create terms
- Terms interconnect
- Patchwork connected at the edges
User interface for adding data
- Prompt for well-known terms
- Allows discovery and selection of local group terms
- Allow creation of new terms
Message mixes vocabulary from many cultures
Data mixing: Term by term
| dc:title | Data Integration and Transparency |
| cc:license |
<http://creativecommons.org/licenses/by-nc/3.0/> |
| dc:creator |
| foaf:name | Tim Berners-Lee |
| foaf:homepage | <http://ww.w3.org/People/Berners-Lee> |
| foaf:email | <mailto:timbl@w3.org> |
|
| tk:event |
| dt:start | 2007-06-12T09:00 |
| dt:end | 2007-06-12T10:00 |
| dt:summary | W3C-WSRI eGovernment workshop |
| geo:lat | 38.9 |
| geo:long | -77 |
|
| tk:slides | <http://www.w3.org/2007/Talks/0618-egov-tbl> |
| tim:slideCount | 12 |
One item may involve data from many ontologies
The tradeoff
| Local | Wider |
|---|
| Local reuse only | Wider reuse |
| Local terms | Global or shared terms |
| Fast | Takes effort |
| |
Semantic Web optimizes the tradeoff
- Mixing of ontologies in each application
- Tradeoff choice term by term not project by project
- Retrospective connections between ontologies
- Widest interoperability with limited subset of data
- Understands overlapping communities
Data owners should
- Take inventory
- Decide priorities, most likely benefits
- Look for existing ontologies
- Don't change the way data is crrently managed
- Set up standard (RDF, SPARQL) portals onto existing data
- Where necessary, adapt or write new ontology bits
Thank You
Thank you for your attention
http://www.w3.org/2008/Talks/0617-fractal-tbl/
