“The Once and Future Web”

observancecookieΑσφάλεια

5 Νοε 2013 (πριν από 3 χρόνια και 9 μήνες)

48 εμφανίσεις

“Lessons From

the Web”

Tsinghua University, Beijing

April 2008


Bebo White

bebo@slac.stanford.edu


The Web Will Be 20 Years


Old Next Year

The Chinese Web Will Be 14 Years
Old Next Week

No One Predicted the Web


Science fiction is sometimes a good predictor of the
future:


It wasn’t in “Star Trek” or “2001: A Space Odyssey”


The Web was not the first hypertext system


Vannevar Bush’s Memex


Ted Nelson’s Xanadu


Apple’s Knowledge Navigator


None were really glimpses of the Web

The Web Was Unpredictable

The Web Was Invented to Solve a
Problem


To facilitate the sharing of documents and services within
international high
-
energy physics collaborations


To “make life easier” for computer
-
phobic physicists

The Invention of the Web Was a
Convergence


The demands of the physics/scientific communities


State of technology at CERN


Availability of the Internet


Popularity of client
-
server systems


Adoption of SGML for document processing


Interest in the NeXT and OOP


Tim B
-
L’s interests


Hypertext systems


Open source software

The Web Was NOT an Immediate
Success


An interesting computer science exercise


Another documentation system (and no one likes to write
documentation)


It took a great deal of “evangelism”

The First Web T
-
Shirt

Today’s Web T
-
Shirt

Today’s Web T
-
Shirt


155,230,051 Web sites

(news.netcraft.com


12/07



45,045.045 sites/square inch



212.23 sites/inch


The Web Took Off Not For the
Reasons It Was Invented (1/3)


February, 1993


the Mosaic browser










April, 1993


the NCSA Httpd server


Netscape

The Web Took Off Not For the
Reasons It Was Invented (2/3)


February, 1993


the Mosaic browser










April, 1993


the NCSA Httpd server


Netscape


IPO announced 11/95


No profitable quarter


Priced at $28 (typical tech was $15)


Opened at $71, peaked at $75, closed at $58


The Internet “big bang” had begun!

The Web Took Off Not For the
Reasons It Was Invented (3/3)


March, 2000


Yahoo! Hits a market value of $104 billion


greater than the entire US auto industry, parts suppliers
included










The browser wars

It Took Off Because


It was unpredicted


There are no expectations as to form or function


It is unpredictable


There are no rules guiding its evolution


It is dynamic (unlike a traditional data collection)


It Works Because (1/2)


It is huge, but scaleable


It is unconstrained in its scope


URIs work


It is hyperlinked


Easy to add content to


Searching is possible


Discovery/surfing is fun

It Works Because (2/2)


Web technologies are intended to be interoperable


The Web is based on a large collection of technologies


No technology can pretend to cover all needs on the Web


Hence the interoperability of technologies is necessary


Web standards should be open and not proprietary


The Web should be accessible to all

It is Successful Because (1/2)


Simple architecture


HTTP, URI, HTML


Networked


value grows with data, services, users


Extensible


from Web of documents to…


Tolerant


works with imperfect markup, data, links,
software


Universal


independent of hardware, operating system,
software, language, ability

It is Successful Because (2/2)


Free/cheap


browsers, information, services


Simple (and fun) for users


text, multimedia, links


Powerful


for people and machines


Built upon open standards

Learned Perspectives of the Web
(1/2)


It is both a technical and a social phenomenon


It has grown into a huge and complex organism

Learned Perspectives of the Web
(2/2)


The computer science perspective: infrastructure and
intelligent systems


The information science and knowledge management
perspectives: data, information, knowledge, wisdom
hierarchy


The social intelligence perspectives: connectivity, social
network intelligence


Application perspectives: e
-
commerce

5 Laws of the Web


These laws have evolved from the evolution of the Web:

1.
Web resources are for use

2.
Every user his or her Web resource

3.
Every Web resource its user

4.
Save the time of the user

5.
The Web is a growing organism

(Ref: Noruzi)

Is It Possible to Use These Lessons
to Predict the Future of the Web?


No one predicted the “dot
-
com” bust


No one predicted Web 2.0


Is it a lesson from the past that the future of the Web is
unpredictable?

Diffusion Models


Tools for
Forecasting


Models can be


Quantitative


based on mathematics/statistics


Rogers’ S
-
Curve


Quasi
-
qualitative


conceptually founded


Gartner Group
Hype cycle

The Rogers’ S
-
Curve of
Technology Adoption

S
-
Curve Examples

Ref: Paul Saffo

What Can the S
-
Curve Tell Us
About the Web?


It depends upon where we are on the curve


are we at the plateau?


What is our perception of diffusion? Acceptance? Ubiquity? How are
people using the Web? How do they want/expect to use the Web?


Is the Web successful?


What are the intra
-
technology relationships/dependencies?


Web and
PC, Web and TV, Web and mobile phone; how do the diffusions
correlate?


If the S
-
curve is fractal, have we only reached one step of the Web’s
future?


We Tend to Predict Linearly

Ref: Paul Saffo

Underestimation and
Overestimation

Ref: Paul Saffo

S
-
Curves and New Eras of
Technology

Ref: Paul Saffo

Lessons from the Information
Revolution


We won!


The Web was our secret weapon


Information is no longer scarce but ubiquitous


It has become media


But it can’t (IMHO) be explained merely by the S
-
curve


Where is the Web on the S
-
curve? It has not become
invisible

Gartner Group “Hype Cycle”

How Much of the Web is Hype?
(1/2)

How Much of the Web is Hype?
(2/2)

Again
-

Is It Possible to Predict the Future
of a Technology Like the Web?


No one predicted the “dot
-
com” bust


No one predicted Web 2.0


Is it a lesson from the past that the future of the Web is
unpredictable?

Before the “Dot
-
Com” Bust


Many people thought that they had figured out what Web
technology was good for


The development of the technology was focused on the validity
of that belief


To many the Web was synonymous e
-
commerce


The Web was data
-
centric and application
-
centric


The governance of the Web was “top
-
down”


The Web had to show that it is unpredictable

After the “Dot
-
Com” Bust


A re
-
evaluation of the Web took place



Web 2.0 is the result of that re
-
evaluation


The Web becomes more user
-
centric


Web governance becomes “bottom
-
up”

Applications Lessons


It’s the applications that make the Web


It’s the killer applications that make the Web diffuse


It’s also the killer applications that generate the hype


What was the killer app of Web 1.0?

Killer Apps


It’s the killer applications that make the Web diffuse


It’s also the killer applications that generate the hype


What was the killer app of Web 1.0?


I think forms, CGI, and SSL


They drove e
-
commerce


Browsers were certainly killer apps

What’s the Killer App of the
Future Web?


Maybe there won’t be one


The technology will “stand on it’s on”


“The death of the browser”

What’s the Killer App of the
Future Web?


Maybe there won’t be one


The technology will “stand on it’s on”


“The death of the browser”


My guess is that semantics + mobility + personalization
will lead to numerous killer apps


Think searching, education, entertainment, science

What’s the Killer App of the
Future Web?


Maybe there won’t be one


The technology will “stand on it’s on”


“The death of the browser”


My guess is that semantics + mobility + personalization will lead
to numerous killer apps


Think searching, education, entertainment, science


Tim B
-
L once said that the days of the Web are numbered


It will disappear into the background


The Web becomes as OS; the network is the computer; the world’s
largest database


This would be the “ultimate diffusion”

Lessons From the Past Will Define
the Web of the Future


The “dot.com bust”


what is the Web good for?


User reaction/involvement


e.g., Web 2.0


Standards vs. Growth (W3C vs. Designers)


The “Media Revolution” has succeeded the “Information
Revolution”


Portable devices with rich interfaces must be a part of the
Web’s future


Are we headed for another bust?

46

The Big Ideas of Web 2.0


Fresh, useful data is the core


The ability for other parties to manipulate that data


“Living” applications that can be easily adapted


Harnessing the collective experience


“The Web as a platform,” independent of user platform


Primary focus of participation, rather than publishing


Trusting of users to provide reliable content

The Big Ideas of the Semantic
Web (1/3)


Information has “machine processable” and “machine
-
understandable” semantics


Can be built upon the framework of the existing Web
technology


Ontologies are the basic building block of a semantic Web

WWW

URI, HTML, HTTP

Bringing the computer back
as a device for computation

Semantic Web

RDF, RDF(S), OWL

Dynamic

Web Services

UDDI, WSDL, SOAP

Static

The Big Ideas of the Semantic
Web (2/3)

WWW

URI, HTML, HTTP

Bringing the Web to its full potential

Semantic Web

RDF, RDF(S), OWL

Dynamic

Web Services

UDDI, WSDL, SOAP

Static

Semantic Web

Services

The Big Ideas of the Semantic
Web (3/3)

Are We Ignoring the Lessons From
the Past?


By trying to shape the future of the Web with technologies
that seem to be relevant?


By losing track of the Web as an evolving natural system?

Thank You!


Questions? Comments?



bebo@slac.stanford.edu