BA 572 - J. Galván1 THE INFRASTRUCTURE FOR E-COMMERCE Web technologies.

BA 572 - J. Galván 1

THE INFRASTRUCTURE FOR E-COMMERCE

Web technologies


WWW- THE TRIUMPH OF ANARCHY

Perhaps the most important human technological artifact that evolved more or less ad-hoc

Limited original vision of the WWW has very little to do with today’s impressive reality

Web Users have consistently innovated in figuring out new ways of leveraging this powerful medium

Web architects then try to catch up by extending (read “patching”) the Web infrastructure to support these new uses


HOW IT ALL STARTED

The web as a static document repository PROBLEM:

“You have documents in your computer that I would like to see” How can I reach you? OR “I would like to send you a text message”

Tim Berners-Lee’s original vision for the WWW (circa 1989) An easy way to access cross-linked static documents stored in a

variety of servers around the world Initial specification defined:

A language for formatting such documents (HTML) A simple protocol for communicating between browsers

and servers (HTTP)


THE TURNING POINT

1993 -Marc Andreesen (student at UIUC) writes Mosaic – first graphical WWW browser (precursor of Netscape)


WWW IS A CLIENT SERVER SYSTEM

Web Clients Use HTTP protocol to connect to servers Request and display Web pages stored in

servers Typical clients: Web browsers

Web Servers Listen for incoming connections from clients Use HTTP protocol to converse with clients Store and transmit Web pages to clients


WWW THE ORIGINAL CONCEPT (1989-1995)

Human Readers INTERACT with Interconnected Static Documents

Main advantage: Universality

Main disadvantage Lack of interactivity Yet another mass broadcast medium

Main business use Awareness building


HTML

Hypertext Markup Language Used to place texts and images in web

pages Understood by all web browsers However, there are different versions

of HTML and different browsers may display different pages


HTML EXAMPLE

HTML EXAMPLE CODE <html> <head> <meta http-equiv=Content-Type content="text/html; charset=windows-1252"> <link rel=Edit-Time-Data href="webpage_archivos/editdata.mso"> <title>HEADING</title> <style> </style> </head> <body bgcolor="#FFFFFF" link=blue vlink=purple class="Normal" lang=ES> <div class=Section1> <p>HEADING</p> <p><span class=SpellE>Subheading</span></p> <table class=MsoTableGrid border=1 cellspacing=0 cellpadding=0> <tr> <td width=192 valign=top class="Normal"> <p><span class=SpellE>Text</span></p></td> <td width=192 valign=top class="Normal"> </td> <td width=192 valign=top class="Normal"> </td> </tr> <tr> <td width=192 valign=top class="Normal"> </td> <td width=192 valign=top class="Normal"> <p> <img width=192 height=164 src="webpage_archivos/image002.gif" v:shapes="_x0000_i1025"> </p></td> <td width=192 valign=top class="Normal"> </td> </tr> <tr> <td width=192 valign=top class="Normal"> </td> <td width=192 valign=top class="Normal"> </td> <td width=192 valign=top class="Normal"> <p><a href="http://www.google.es/">link</a></p></td> </tr> </table> </div> </body> </html>


LAYERS IN THE WEB

Web serverBrowser

World Wide Web

Internet

Data communication network

Web application


The World Wide Web

Client/server application Layered on top of Internet Web page – basic information unit Web site – set of related pages Hyperlink – logical page-to-page link Home page – Web site’s starting point


Web Software

Browser Client-side application program

Internet Explorer, Netscape/AOL, Mozilla, Opera,...

Requests and displays (views of) Web pages Web server

Server-side application program Where a Web site resides Manages access to Web site contents Fills page requests – “serves” pages


STRUCTURE OF A WEBSITE

Home page

Content page 1 Content page 2 Content page 3

HyperlinkHyperlink Hyperlink


CONNECTIONS TO A WEBSITE

Browser Browser Browser

Web server

Client workstation Client workstation Client workstation

Host computer

Databaseof

web pages


HOW THE ORIGINAL WEB WORKS


ANATOMY OF AN URL

URL = Uniform Resource Locator

http://web.mit.edu/sloan/www/index.html

Name of protocolfor communicating with server Domain name of

Web server wherepage is stored

Pathname of pagewithin server localFile system

Name ofwebpage

Domain levels


HOW THE WEB WORKS (I)


HOW THE WEB WORKS (II)


HOW THE WEB WORKS (III)


Web Servers

Server hardware The host computer

Server software Application program

Web server (Apache, IIS) Database server E-mail server

Runs on server hardware or host


MULTIPLE SERVERS CAN RUN ON THE SAME HOST.

Server is software

Public webserver

Subscriberweb server

Employeeweb server

File server E-mail server

Host computer


MULTIPLE SERVERS CAN RUN ON

MULTIPLE INTERCONNECTED HOSTS.

A Web or Application Server is software!!

File server

Host computer

Public webserver

Host computer

Subscriberweb server

Host computer

Employeeweb server

Host computer

E-mail server

Host computer

Router

Internet


SEPARATE FIREWALLS AT EACH LAYER

Firewall protects Internal network from public network

Runs on Router (via Network

Address Translation)

Host Server Workstation

Public webserver

Host computer

File server

Host computer Host computer

E-mail server

Router

Internet


Interactivity

Early browsers limited to static pages Business need for

Animation, sound, flashing logo Access to global markets Client side data input

Need for client side interactivity


Client-side Interactivity

Scripts Macro-like instructions JavaScript, Jscript, VBScript Embedded in HTML text

Applet Small program executed within browser Java, VisualBasic

Plug-ins and helper applications


SOME POPULAR PLUG-INS.

Plug-in is “integrated” software application Processes registered file types (e.g., proprietary file formats) Often available as free download

Plug-in Supplied by Function

Acrobat Reader Adobe Display PDF filesRealAudio/RealPlayer RealAudio Play real-time audioShockwave Macromedia Display animationQuicktime Apple Animations, audio


Web Information Systems

Client-side interactivity Server-side interactivity

Personalized marketing JIT inventory Inter- and intra-corporate communication


Web Applications

Applications that support a Web information system Client/server Browser on client side Server software on server side

Middleware translates between HTML Non-Web server format


Middleware connects two applications/servers.

Server-side middleware links Web server Database server Other applications

HTMLrequest

HTMLdata

Browser

Client host(manufacturer)

Supplier's Database

Database server

Middleware

Web server

Server host(supplier)

Internet

Data

Data

HTMLdata

HTMLrequest

Databasequery

Databasequery

HTMLdata

HTMLrequest


Maintaining State

The Web is stateless Each transaction is independent No problem for Web surfing

Web information system is stateful Series of related transactions Must “remember” previous steps

Cookies provide one solution; another is registration, login, and tracking via DB


EVOLUTION OF THE WWW

Business Drivers Enable transactions Allow interactivity between

browser and server Facilitate personalization Support multiple browsing

devices Better organize and retrieve

Web content Support Business-to-Business

applications


THE WEB AS A TRANSACTION FACILITATOR

Business Motivation: Low-cost front-end for allowing customers to connect to

corporate computers Customer registration/Address changes Order tracking/Customer support Online Transactions: eCommerce

Problems of original Web concept Static web pages No interactivity Stateless protocol: no support for multi-step transactions Insecure communications


WEB FORMS - PRIMITIVE INTERACTIVITY

Pages which contain fields to be filled by user Usually contain a “Submit” button When user presses “Submit”, server responds by

sending a page containing information specific to the user-supplied parameters

• Examples: Web search tools Order forms in commercial web sites See e.g. http://www.google.com


WEB FORMS


BUSINESS USES OF WEB FORMS

Server sends original html page containing input fields User types info into fields and presses submit button Client establishes connection with handler script at server side

(script filename contained in web page) Client collects user input into a long string and sends it along

with an HTTP command back to server POST customer=John+Doe&cardno=1234567890&expires=6/98&product=123&quantity=5

Handler script at server reads parameter string and processes it, usually producing a new page as a result


CGI

Common Gateway Interface Set of standards for writing handler scripts How it works

All URLs that refer to a special directory (e.g. /cgi) cause the execution of a corresponding script at the server (for example http://web.mit.edu/cgi/test)

Scripts typically translate parameters into SQL statements for a database and translate the query results into an HTML page


HOW IT WORKS - FEDEX


MICROSOFT ACTIVE SERVER PAGES (ASP)

Competing technology to CGI Scripting Language is similar to Visual

Basic MS Access can automatically convert

database tables, queries and forms into ASP pages

Requires Microsoft web server



Business Drivers Enable transactions Allow interactivity

between browser and server

Facilitate personalization Support multiple browsing



applications


THE INTERACTIVE WEB

Business Motivation: Allow complex interaction between user browser and

corporate server Web becomes an extension of the user’s PC Browser becomes a window to a variety of corporate

applications Problems with Web Forms/CGI/ASP

All processing done at server side Rapid user interaction with Web page not possible Need local processing to create highly interactive Web

pages


ENTER JAVA APPLETS

Programming language to enable interactive Web pages

Developed by Sun Microsystems originally for programming intelligent microwave

ovens!!! Java programs are called applets Applets are platform-independent

They can run equally well on Windows, Macs, Unix, etc.

Require special browsers that can support Java though


HOW DOES JAVA WORK?







applications


PERSONALIZED INTERACTION

Business motivation: Low cost medium for gathering

information from customers to allow Personalized service Targeted advertising

Problems with current model Does not allow easy identification of

distinct customers


COOKIES

A method for identifying web users and delivering customized web sites First time user connects to a web site, s/he is asked to fill

in personal information form Server packages information into a “cookie” file and sends

cookie to browser Browser stores cookie in local file system Each subsequent time browser visits site, it sends cookie

back to server Server uses information stored in cookie to identify user

and possibly customize the supplied web pages Privacy implications?


COOKIE APPLICATIONS


SUMMARY: WWW THE CURRENT CONCEPT (1995-TODAY)

Human Users INTERACT with Documents and Applications

Main business use B2C Transactions Customer Support




browser and server Facilitate personalization Support multiple

browsing devices Better organize and retrieve


applications


MULTIPLE DELIVERY DEVICES

Business motivation: Allow users to access web content from a variety

of devices PC Browsers PDAs (e.g. Palm Pilots) Mobile Phones Telephones (via voice interface) …

Problems of current Web model Each access device has different look-and-feel

requirements HTML specifies formatting for PC browsers only





devices Better organize and

retrieve Web content Support Business-to-Business

applications


ORGANIZE AND INDEX WEB CONTENT

Web is useless unless we can easily locate relevant resources

Current solution: Search Engines Index the Web by automatically

“discovering” web pages and organizing them around keywords found in their text


HOW DOES GOOGLE WORK?

Before you ever enter a query: Programs (called “web crawlers” or “spiders”)follow links

from one page to another all over the web. The programs construct indexes of which words appear

on which pages and save the indexes (and often copies of the pages) on massive “server farms” maintained by Google.

Each page is also assigned a “page rank” based on the number of other pages that link to it. Links from pages that, in turn, have lots of other pages linking to them are weighted more heavily.


GOOGLE’S PAGE RANK FORMULA

PR(A) = (1-d) + d(PR(t1)/C(t1) + ... + PR(tn)/C(tn)) Where

t1 -tn are pages linking to page A, C is the number of outbound links that a page has d is a damping factor, usually set to 0.85.

In words: a page's PageRank = 0.15 + 0.85 * (a "share" of the

PageRank of every page that links to it) "share" = the linking page's PageRank divided by the

number of outbound links on the page.


HOW DOES GOOGLE WORK? (CONT.)

After you enter a query: Programs check the indexes to determine

which pages contain the combination of words you entered.

Google provides a list of these pages in order of page rank (order is probably affected by other factors, too)


GOOGLE NUMBERS AS OF DECEMBER 2004

Over four billion Web pages, each an average of 10KB, all fully indexed.

Up to 2,000 PCs in a cluster. Over 30 clusters. One petabyte of data in a cluster --so much that hard disk

error rates of 10-15 begin to be a real issue. Sustained transfer rates of 2Gbps in a cluster. An expectation that two machines will fail every day in each

of the larger clusters. No complete system failure since February 2000.

Source: ZDNet – UK <http://www.zdnet.com.au/insight/software/0,39023769,39168647,00.htm>


PROBLEMS WITH TODAY’S SEARCHES…

Text keywords are misleading… HTML does not give any clues as to

the true meaning of the data Example:

Desperately seeking Wendy Cook






Web content Support Business-to-

Business applications


SUPPORT FOR B2B APPLICATIONS

Original Web was conceived as a communication medium between computers and humans

Amazing new applications will become possible if computers can automatically read and understand Web pages Electronic purchasing Intelligence gathering ….

Problem: HTML pages are unstructured HTML only provides information about presentation, not

meaning


WHAT IS THE UNDERLYING ISSUE?

When storing documents on the web, specify not only their appearance, but also their semantics (i.e. their meaning!)


WHAT IS THE MEANING OF WEBSITE ADDRESSES? BEST RADIO IN TOWN

http://www.wwoz.org/ SEARCH OF THE WORD RETREIVE IN GOOGLE

http://www.google.es/search?hl=es&q=RETREIVE&btnG=B%C3%BAsqueda+en+Google&meta=

CORRECTION GOOGLEhttp://www.google.es/search?

hl=es&sa=X&oi=spell&resnum=0&ct=result&cd=1&q=RETRIEVE&spell=1 ACCESS TO WELCOME PAGE OF A BANK

https://www.sabadellatlantico.com/es/HERRAMIENTAS/INDEX/?url=/es/PAGINA_INICIAL/BANCO_ATLANTICO/BIENVENIDOS/

INTRANEThttp://portal/default.aspx

INTRANEThttp://10.5.162.254/

LOGIN PAGE IN A BANK (Access to the Intranet)https://oi.cajamadrid.es/CajaMadrid/oi/pt_oi/Login/login

ARTICLE SELECTED IN A WEBSITEhttp://www.baquia.com/noticias.php?id=11667


BEST RADIO IN TOWNAn ordinary address

http://www.wwoz.org/

Hypertext protocol

Addressing the whole web

2nd level domain, owned by the organization

1st level domain. Identifies activity


SEARCHES

Placing “RETREIVE” in Google http://www.google.es/search?hl=es&q=RETREIVE&btnG=B%C3%BAsqueda+en+Google&meta=

After the address comes “search?” and all the data of the query

CORRECTION GOOGLE (RETRIEVE)http://www.google.es/search?hl=es&sa=X&oi=spell&resnum=0&ct=result&cd=1&q=RETRIEVE&spell=1

Includes changes in the original query after answering that the word seeked was really RETRiEVE

This type of query changes with time. It cannot be assured to recover the same document executing the same query at different times


ACCESS TO WELCOME PAGE OF A BANK

https://www.sabadellatlantico.com/es/HERRAMIENTAS/INDEX/?url=/es/PAGINA_INICIAL/BANCO_ATLANTICO/BIENVENIDOS/

Protocol – web – secondary + primary domains – Folders +

?url – Spanish + welcome page There are language options in previous pages


INTRANETS

Portal supplying asp dynamic pageshttp://portal/default.aspx

INTRANET through IP addresses within the local networkhttp://10.5.162.254/

LOGIN PAGE IN A BANK (Within the Intranet)https://oi.cajamadrid.es/CajaMadrid/oi/

pt_oi/Login/login


RECOVERABLE DOCUMENTS

Adobe document from National Science Foundation

http://www.nsf.gov/pubs/2007/nsf07534/nsf07534.pdf Article selected in a website

http://www.baquia.com/noticias.php?id=11667 Dynamic webpage selecting articles from a

database. The server is programmed in PHP and articles are indexed (?id=11667)


THE SEMANTIC WEB

The “Next Generation Web” with well-established infrastructure for expressing information in a Precise, Human-readable, and Machine-interpretable form.

Enable syntactic and semantic interoperability among independently-developed Web applications, allowing them to efficiently perform sophisticated tasks for humans.

Enable Web resources to be accessible by their semantics rather than by keywords and syntactic forms.

Enable inferencing: Chris is an associate professor at MIT. Associate professors are permanent employees. Chris is a permanent employee of MIT.


EVOLUTION


Please, beware of technical problems


NEW BUSINESS MODELS


HUMAN INFRASTRUCTURE

BA 572 - J. Galván1 THE INFRASTRUCTURE FOR E-COMMERCE Web technologies.

Documents

Transcript of BA 572 - J. Galván1 THE INFRASTRUCTURE FOR E-COMMERCE Web technologies.