IS 524 CORPORATE INFORMATION SYSTEMSfaculty.wiu.edu/C-Amaravadi/is524/notes/is524_notes2.pdf · Chandra S. Amaravadi March 8, ... Rapid access to information is a life-or-death issue

©Copyright 2016 by Chandra S. Amaravadi. All rights reserved. Electronic reproduction or distribution is strictly prohibited.

IS 524 CORPORATE INFORMATION SYSTEMS

SUPPLEMENTARY NOTES

Chandra S. Amaravadi

March 8, 2017

Part II

Managing data resources (database I)

Database planning and analysis (database II)

E-Commerce

1

MANAGING DATA RESOURCES

Need for Databases Rapid access to information is a life-or-death issue for many organizations. For example a physician may want to know if an incoming patient is allergic to a particular drug or an airline company manager may want to know the forecasted flight occupancy on a given flight. Organization is key to accessing the data in a timely fashion. Real World Concepts: In the real world there are objects such as cars, animals, basketball games etc. Entities- Are individual examples of objects and things in the real world. A computer with a particular serial number, Bob’s car, the city of ‘Colorado Springs’ are examples of entities. Eclasses- Are collections of related entities. Students in Maryanne’s course, a collection of car parts sold in a store, vendors in the Midwest are examples of entity classes. Is the weather an entity class? Attributes – Are properties of eclasses (which are ultimately recorded in the database). A person’s hair color , a credit card#, processor speed and storage capacity are examples of attributes. Data Models

A data model is an abstract view of the data. It captures relationships between two or more entity For example, students take courses, ships have captains and products are placed on order. Here students and courses, ships and captains, products and orders are eclasses respectively. Three types of relationships are possible between two eclasses, a concept that is referred to as cardinality: 1:1 – Each instance of eclass A is associated with only one instance of eclass B and vice versa. E.g. ship: captain; company: CEO; country: president. A ship has only one captain and a captain is assigned only one ship. 1:M – Each instance of A is associated with many instances of eclass B. Each instance of B is associated with one instance of eclass A. Example, hospital: wards, department: courses, department: employees. A department has several employees but an employee belongs to only one department. M:N -- Each instance of A is associated with many instances of eclass B and vice versa. Example, students: courses; hospital: special treatments; advertisement: TV programs.

Fig 1. Types of cardinality relationships among eclasses.

Cardinality is indicated by ‘crows foot’ notation. A crow’s foot indicates ‘many entities’ of an eclass. Note

2

that the only basis for judging cardinality are the rules given above not whether the eclass is singular or plural! The Database Approach The overall concept here is that objects in the real world have a representation in the database. Organization deals with entity classes such as customers, employees, products, vendors, benefits etc. Not only that, eclasses have relationships among themselves (‘customers buy products’, ‘vendors supply raw materials’). Entity classes and relationships are depicted via data models (also known as ER models) as shown below. Once a data model is developed, the database can be designed and developed. In fact there is something of the software engineering life cycle that is present in the diagram below – planning, analysis, design and implementation. Which phases correspond to which here?

Fig 2. The database approach

Basic Organization Concepts: Data – Consists of raw facts e.g. a scoring average of 24 points per game, 30 items ordered, a building has ten floors. These are not meaningful unless we add more details. Information – Raw facts are not by themselves meaningful. Information can be thought of variously as data presented in a context or as a collection of raw facts. For example, we may have the following collection of facts about a person. Name: Chris Nash DOB: 2-4-1965 Profession: Engineer Address: 415 Walnut street. In database terminology this would be called a record and the individual facts would be attributes (consisting of attribute name and a value). This collection of facts makes the information meaningful. Attribute – Property of an eclass e.g. hair color, part#, parcel weight, expense amount etc. Primary key- Is a particular field or combination of fields that is used to uniquely identify any one given record within the database. For example part# identifies a part and account# identifies an account uniquely. When tables are listed as below, the primary key is generally underlined and is the left most attribute. Note that a given primary key value can occur only once in a table – that is what uniqueness means. Each table should have a primary key by database rules. The primary key can consist of multiple values as in WIU ID#

3

and Course# (in a grade table). File/table- Traditionally, a computer file was a group of characters, but here it is simply a collection of records. File and table are used interchangeably. Record -- A grouping of attribute values -- row within a table. The product table below has two records. Schema - The schema of a database describes the structure of the file. Overall organization – the database is organized into files which are organized into records and then into attribute values. The product file below has sku# as a primary key, a description, price and quantity. Description, price and qty (as well as SKU#) are attributes. The table has two records corresponding to the two rows.

Product table/file

sku# description price qty

11967-4 crankshaft $250.00 400

11978-5 Cylinder ring $28.00 15

Fig 3. Basic organization of a database table

Summary of Basic Organization Concepts: A database is a collection of files or tables. Each row describes one entity – it could describe a student named Jack. The row is called a record and describes an entity as stated. Attributes are columns in the table. Attributes store values that contain the information about a given attribute of an entity. Thus a record can also be thought of as a collection of attribute values. The primary key is used to uniquely identify each record in the database. Secondary keys are attributes other than primary key e.g. ‘Descr’, ‘Price’, ‘Quantity’ in the table below. The schema/structure of the database shows a list of tables and their attributes. The schema of the product table above consists of sku#, description, price and qty.

Fig 4. Organization of a database table

4

DBMS DBMS is a software program such as Access and Oracle, to manage the data. A DBMS allows queries such as how many products were sold yesterday. DBMS Activities The DBMS is a software program that allows users to physically manage the data in an organized fashion.

This includes defining the tables, entering the data, getting reports and carrying out queries.

Data definition This is used to define the structure or schema of the table. Essentially it is the process of identifying the attributes of a table and defining their data type i.e. whether it is numeric, currency, text etc. The primary key as well as allowable length for each attribute are also defined during this stage.

Fig 5. Defining the structure or schema of the table

Data entry A form is created to enter the data. The form has a title or heading, labels which are used to guide the data entry and fields where data is actually entered. A data entry form facilitates data entry. It can be used to check for valid data.

Fig 6. Data entry form

Modification Data can be added, deleted or changed. For e.g. a new customer may be added, an inactive customer deleted or a customer’s address may be updated. Query Query refers to getting information from the database. There are two methods of querying data. QBE – QBE stands for Query By Example. This is a form used to retrieve the data. The form has the

5

attributes from a table (e.g. a product table). The user can select attributes as needed (e..g product description and price) and additionally specify criteria for e.g. they can type “chair” under description column, to retrieve all products where the description is “chair”. Here the query is for products costing < $200, no fields are selected for output yet.

Fig 7. QBE for finding products costing < $200

SQL – SQL is a scripting language (4thGL) used to retrieve data from the database. The SQL statement has three parts:

SELECT <attr1, attr2,….>

FROM <table1, table2…>

WHERE <condition1 AND/OR condition2 AND/OR …..> The ‘select’ part has a list of attributes desired in the output, separated by commas. The ‘From’ part has a list of tables also separated by commas, corresponding to the source of attributes. The ‘where’ part has a list of conditions. If there are multiple conditions they are linked by ‘AND’ or ‘OR’. The example below shows a reservation table with Flt#, Pname and Confirm# all as combined primary keys and it shows three records. The SQL query below the table is selecting Flt# and Confirm# for output with the condition “where Pname = Smith”.

Fig 8. Query on reservation table and query result

Multi-table retrieval – When there are multiple tables, there are three changes to the SQL format. In the ‘select’ part, attributes are preceded by name of table e..g prod.prod#. (what are the three parts of an SQL query?) The ‘From’ part will have a list of all the tables separated by commas. Thirdly in the ‘where’ part values of the common key are equated. For e.g. if ‘cust#’ is a common key between the ‘customer’ and ‘order’

6

tables, the ‘where’ part of a query involving ‘cust’ and ‘order’ tables will have …Where cust.cust# = ord.cust#. See if you can use these principles to write a query to LIST Employee name, title, salary, and department name for the table below. What is the result of the query?

dept

emp e_ssn e_name e_title e_salary

956-34-8895 Smith Developer 35000

959-66-6785 Johnson Analyst 27000

967-89-8898 Weintraub Manager 60000

Fig 9. Employee and department tables

Reports – Data is commonly obtained from the database in the form of reports. Reports enable the data to

be formatted according the needs of the user (this is information). A query produces raw unformatted data. A

report is specified with a report specification. This consists of report and page headers, detail line (this has

rows from the database), page and report footers.

Fig 10. A report specification.

Importance of DBMSs: Like many information systems, DBMSs have both an operational and strategic usage. Operationally, DBMSs are useful in retrieving stored facts. ‘Was check# 1013 cashed?’ ‘How many units of product with SKU#112393 are remaining in the store?’ ‘Is a white refrigerator manufactured by whirlpool available for less than $1000?’ But their strategic usage outweighs the advantages of such information. Strategically, information in databases can be mined to yield information such as ‘which brand and style of jeans are best sellers?’ ‘what type of employees make the best managers?’ ‘What size carpets have the highest volume of sales?’ etc. This information is useful in making better decisions for production, hiring, retail etc.

d_no d_name d_mgr_ssn d_phone

142 Manugistics 967-89-8898 845-9878

230 IMS 979-99-0045 989-0087

345 Pilot 978-64-8005 787-9934

467 InfoSec 998-98-0967 884-5768

7

DATABASE PLANNING

The first step in the database development process is database planning. There are two main goals here, first to get an overview of the organization’s data requirements (but it is not to identify their data needs, this is done at a later stage). The second is to define the resource requirements for the overall development process. Resource requirements can include personnel, hardware, software and schedule. What hardware/software is required to ultimately implement the database? How many personnel? How long will the development process take? In addition, the scope of the project is also defined at this stage. The process starts with getting some background information on the organization first. This can include the number of employees, number of products, items in inventory etc. What is the purpose of getting this background information? The method used to carry out planning is called Business Systems Planning (BSP) or sometimes Enterprise Planning. The method is akin to planning for a library. Suppose you are asked to build a library, what information do we need to know? viz. the number of expected books and the areas they cover. These can be used to determine the size and layout of the building. BSP relies on the principle that activities generate or use data. For e.g. ordering raw materials requires checking available inventory. In BSP, the organization’s functions are first identified. Functions consist of processes and activities so these are also identified. Next the information used or generated by these activities are tabulated against the functions and processes. E.g. determining land use for a county administration uses information on tracts in the county. See table below that is a result of this exercise. This is called an Enterprise Analysis or a Planning Matrix. From the information classes (in the top row) an enterprise data model is also developed. See the second diagram below.

Fig 1. An Enterprise Analysis Matrix (Planning Matrix) for a county admin.

8

Fig 2. An enterprise data model from the information classes

The Enterprise Data model illustrates the following:

1. County has many tracts 2. County runs many conservation programs 3. A conservation program can be run on several tracts 4. A tract has many parcels 5. A parcel can be owned by more than one person 6. A tract has a single land use.

The EA matrix along with the Enterprise Data Model are used to obtain an overview of the data requirements of the organization. The results of planning (what are these?) are used to determine the size and scope of the database as well as the resource requirements.

9

DATABASE ANALYSIS

During the analysis stage of database development, (or requirements definition) the actual data requirements are identified. Generally this is done by examining “views” of the data. “Views” can include forms, reports etc. that exist within an organization. McDonald’s restaurants for e.g. have ‘waste sheets’ that list items and quantities that have to be discarded that day. So it is an example of a view. ER modeling This diagramming method is used to depict relationships existing between the different entity classes in the real world that are of interest to the organization. This is same as a data model (and crudely similar to an Enterprise data model). The fundamental concepts are: Eclasses- Are collections of related entities. Student’s in Maryanne’s course, a collection of car parts sold in a store are examples of entity classes. Is the weather an entity class? Eclasses are modeled with a rectangle with the name of the eclass inside it.

Fig 3. Modelling an entity class

Attributes – Are properties of eclasses (which are ultimately recorded in the database). A person’s hair color , a credit card#, a gpa are examples. Attributes are modeled with ovals and labeled appropriately. The primary key is drawn as the left most attribute and the label is underlined. Attributes are shown only once. They are not duplicated except for the primary key. If a pkey is not given, it can be assumed. One challenge in modeling attributes is properly associating an attribute with an eclass – this is based on commonsense as there are no set rules. Does ‘fare’ belong with ‘flights’ or with ‘booking’?

Fig 4. Modeling attributes with ovals

Multi-valued attributes- attributes that can have more than one set of values per entity instance. For e.g. an employee can have multiple skills, a classic car can have two colors. What are the entity classes in each of these cases? Multi-valued attributes are generally represented with a double oval. Fare is a multi-valued attribute in the diagram below. Why is it multi-valued in this case?

10

Fig 5. A multi-valued attribute in an eclass

Relationships – are logical and meaningful connections between two or more entity classes. E.g. students take courses or physicians treat patients. Relationships are modeled with the diamond symbol and labeled with the name of the relationships. Relationships can be two way (binary), three way or four way (more than that is also possible). The diagram is read left to right or top to bottom. Relationship attributes – sometimes an attribute does not belong to eclasses in a relationship, but with the relationship. For e.g. a distributor can place order with a book publisher. If distributor and publisher are taken to be the eclasses, the quantity of books ordered does not belong with the distributer or with the publisher. It is treated as a relationship attribute. When there are relationship attributes, a dotted rectangle is placed around the relationship symbol (diamond) and the attribute is shown sprouting from the relationship. See the diagram below. Here ‘date’ is a relationship attribute.

Fig 6. Modeling relationships with relationship attributes

Cardinality- Cardinality shows the number of entities that participate in a relationship, i.e. how many instances of an entity class correspond with how many instances of another entity class? For 1:1 cardinality each instance of the first entity class corresponds to one instance of the other entity class. For 1:M one instance of the first entity class can correspond to many instances of the second entity class. M:N cardinality means that each instance of either entity can correspond to many instances of the other entity class. You should recall this from earlier discussion. Max Cardinality Rule: When we have more than two entity classes in a relationship, maximum cardinality is used. Class/subclass relationship -- Eclasses can be grouped into classes and subclasses based on generalization/specialization of properties. For e.g. a sports car is a specialization of a car that seats two, has an aerodynamic shape etc. Here ‘car’ is a class and ‘sports car’ is a subclass of it. Class subclass relationships are modeled by an elongated hexagon. The label for this hexagon is always “is-a”. Cardinality is not shown for subclasses. Attributes specific to the subclass may be shown, otherwise they are shown on the superclass. Only horizontal and vertical lines are used.

11

Fig 7. Classes/Subclasses relationship

Drawing the ER chart First step is to list all eclasses and attributes. The following format needs to be used: Prod: prod#, descr., price, qty

Cust: cust#, name, dt. Joined

Then relationships between eclasses need to be mapped. Is there a relationship between prod and cust?

between ord and products? Between orders and salespeople? Map all relevant relationships among the

eclasses. Here relationships between the eclasses students, courses, departments, faculty, majors (Note that

these have been arbitrarily selected for illustration) are shown. Also note that attributes are not shown. Note

that the relationships are mapped as follows (see diagram following):

1) Students take courses (with faculty).

2) Faculty offer courses (to students).

3) Courses are offered by majors (and departments).

4) Majors are offered by departments (could also show mapping between courses and departments).

5) Departments have faculty

6) Faculty teach students (in courses)

Fig 8. Map the relationships between classes

12

Start by selecting one of the more important eclasses (those that have lots of relationships with others). Draw

a rectangle corresponding to the eclass. Then gradually add other entity classes one at a time. Make sure that

between two entity classes there needs to be a diamond symbol. Since there can be more than two entity

classes in a relationship, watch out for three way and four way relationships. These can be determined via

mapping as illustrated or by articulating “Students take courses with faculty,” “Salesperson sells cars to

Customers.” Here we start with students:

Fig 9. An eclass named students

Students are involved in relationships with both faculty and courses. This is a three way relationship. So we draw

a diamond symbol to represent the relationships and connect it with all three entity classes. Note the use of

horizontal and vertical lines and maximum cardinality.

Fig 10. Modeling relationships with students, courses & faculty

In the remainder of the model we add departments and majors. Since we mapped courses to majors (in our

mapping diagram) the logical place to draw majors in the diagram is beneath the courses. It’s an m:n

relationship.

Fig 11. Add majors in the relationship mapping diagram

13

Then we add the departments eclass as shown below. Both courses and majors are offered by departments.

There are relationships between courses and majors, courses and departments and departments and majors. It is a three

way relationship and is shown also with maximum cardinality.

Fig 12. Add departments in the relationship mapping diagram

Finally, we add the relationship between faculty and departments. A department has multiple faculty but each

faculty member works only for one department. Note that linking faculty and departments results in a cycle

which is indicated with the multi-sided arrow. Cycles are considered undesirable because they may indicate

error in modeling. Here it cannot be helped since eclasses such as courses, departments and faculty are part

of more than one relationship. Cycles need to be recognized by the bent arrow as seen below.

Fig 13. Final model

At this stage attributes are added. These are not shown here in this illustration.

14

E-COMMERCE

What is E-Commerce?

E-commerce is business carried out using the internet. An E-commerce transaction has a buyer and a seller

carrying out commerce on the internet. The buyer selects goods/services and makes a payment and the seller

responds by providing the product/service.

Fig 1. E-commerce

Types of E-Commerce

E-commerce can occur in several modes as follows.

Customer to Business (B2C or C2B): In B2C or C2B, the buyer is a retail customer and the seller is an online

retailer such as Amazon.com. This is the type most commonly encountered on the web.

Business to Business (B2B): In B2B, both buyers and sellers are businesses.

Customer to Customer (C2C): In C2C, customers buy and sell to one another as in Ebay. Some computer

companies such as Dell allow their customers to buy and sell to each other.

Intra-organizational: This is a microscopic portion of the overall e-commerce and is mentioned here for

interest. It occurs if buyers and sellers are within the same organization. Many organizations have internal

websites or marketplaces where employees can buy and sell from one another.

Motivators for C2C online purchases

Motivators include low prices, 24/7 convenience, free shipping (in some cases), ease of carrying out the

transaction (example, ordering flowers from a cell phone at a bus stop), range of products and ease of

product comparison. These can be conveniently grouped into low prices, convenience and choice.

Motivator for Businesses

The acquisition cost per customer is the cost of advertisements and promotions divided by the total number of

customers. So if a company has one million customers and spends ten million on marketing, its cost per

customer is $10. The acquisition costs per customer are quite high in conventional retail businesses. In

15

consumer electronics it is $56/customer, in groceries it is $84/customer, in apparel it is $53. In the cell

phone industry, some companies have spent as much as $250 acquiring each new customer. In internet based

businesses, in theory, the acquisition cost/customer should be $0 since customers should be able to find the

retail store through a search engine. In practice, except for the major retailers, the smaller retailer has to

advertise in order to have customers browsing their site.

Net Shoppers key demographics

Net shoppers are upwardly mobile and young with an average age of 42 and an average income of $75,000.

The distribution of the sexes is equal (50-50) between males and females. Males predominantly shop for

electronic goods while women tend to buy clothes and accessories online. 90% of the U.S. population has

shopped online at one time or the other. The B2C market is only $304 billion while the B2B market size is in

the order of $780 billion (the U.S. GDP is $16.8 trillion). It is expected to grow to $1 Trillion annually by

2020.

Most commonly purchased items

The most commonly purchased small ticket items (inexpensive) are:

Apparel

Toys/video games

Books

Software

Music

Health and Beauty

Office supplies

The most commonly purchased large ticket items (expensive) are:

Plane tickets

Computer hardware

Hotel reservations

Consumer electronics

Car rental

What characteristics do these items share in common?

An Extended View of E-Commerce

E-commerce has been defined as a buyer and seller carrying out a transaction via the web. To enable this it is

necessary to have a physical infrastructure in the form of internet and e-commerce system as well as a method

of processing payments. Thus an Extended View of E-Commerce includes the Buyer and Seller, the

infrastructure on which e-commerce transactions take place as well as financial entities that enable the

payments. The buyer can pay the seller via banks, intermediaries and other merchants. Since the data travels

through public networks, there is a need for security. There are secure protocols for carrying out internet

transactions. SSL and SET are two such protocols.

16

Fig 2. An Extended View of E-Commerce

Infrastructure -- Internet

There are two components to the infrastructure: the internet and the e-commerce system. The internet is the

physical medium through which e-commerce transactions take place. It is important to realize that there are

no direct wired connections between the buyer and seller (in this case, it is a client and a web server). When a

web client requests a web page, the request is broken down into packets (small units of data). The packets are

addressed by the host (see diagram below) with the destination IP address (e.g. 143.43.78.9) and sent to their

destination. Routers at the junctions of networks route the packets to their destinations using routing rules.

They perform the function of traffic lights at street corners – to guide traffic. Sometimes the host may not

know the destination address. In such cases it sends an enquiry to its neighbors. If its neighbors are unable

to answer the query, the query goes to a DNS server (previously known as root server). The DNS servers

also take periodic ‘attendance roll calls’ to maintain their list of IP addresses.

Fig 3. Infrastructure -- Internet

17

Components of the Infrastructure

Web clients – Computers that access the webservers usually using browsers as software. The web client is

often the ‘front end’ of an e-commerce system. It could be a mobile phone or a personal computer or other

computing device.

Host/node – This is an addressable location on the internet with an IP address. It is a computer that is often

dedicated to internet services such as caching, packet collection etc.

E-commerce system – The set of hardware and software that supports e-commerce transactions. The front

end of an e-commerce system is the web client while the back end is the web server. Together with e-commerce

web site and e-commerce application, the client and server constitute an e-commerce system.

DNS server – The ultimate reference for all web site addresses. DNS servers poll web sites in their area i.e.

take ‘attendance’ to maintain the DNS lists. Unknown web site addresses are resolved this way.

Routers – Components that examine packet addresses and forward data at network branches.

Service Providers – (ISPs) provide internet access.

E-Commerce System Purpose

An e-commerce system can fulfill many different functions related to the business. The important ones are:

a) Attract and interact with visitors – In E-Commerce, the web site represents the company and only point of

contact between the customer and company. Therefore the site has to be attractive and easy to navigate,

b) Display products/services – An e-commerce site exists to sell products/services or information (see

discussion on business models). One of the important functions of a web site is to display these attractively,

c) Process payments – In the case of companies selling products or services, collecting payments is an

important function. Payments are handled usually through credit cards or electronic accounts (discussed

later).

d) Shipment/payment confirmation – When payments are made the order has to be confirmed and

subsequently when the item is shipped, the customer has to be notified.

e) Update databases – Databases are updated at various stages in the process. Specific updates include user

profiles, orders, invoices, inventory, shipments etc. (such activities have been discussed as part of

‘organization concepts’ as well as in ‘intro to corporate systems’).

Conversion Ratio

The ratio of buyers to visitors is called the conversion ratio. It is in the vicinity of 5% for small companies and

75% for the better known sites. The attractiveness and features of the web site is one of the factors that

influences the conversion ratio. Can you think of others?

E-Commerce System

The set of hardware and software that supports e-commerce transactions is called an e-commerce system. The

front end of an e-commerce system is mostly the web client while the back end is the web server. The client

and server are often separated by geography. Together with e-commerce web site and e-commerce

18

application, these components constitute an e-commerce system. (e-commerce application is discussed

further on). Thus an e-commerce system consists of:

Web clients

Web Servers (or simply ‘servers’)

E-commerce site (Web site )

E-commerce application

Interface

Business logic

Database backend

Web Client

Any device that can access a web site is called a web client. Web clients range from conventional computers

to tablets and smart phones. Even the devices carried by tradespeople (e.g. fedex delivery person) may be

classified as a web client.

Fig 4. The Web Client

Web Server Functions

Web server is the device (and software) that hosts the web site. The functions of the web server go beyond

simply providing the home page. Web servers sometimes provide authentication i.e. verification of identity

usually through username and password. They respond to DNS protocols and capture visitor information

(site traffic) for advertising purposes.

Server Architecture

Most organizations (i.e. large organizations) do not have a single web server. Instead, they have a hierarchy of

web servers carrying out specialized functions. The architecture shown in the diagram below is a typical three

tier architecture used by many large organizations. The three tiers separate the functionality (purpose) and

protect web resources:

a) Webserver layer – this is the set of web servers that respond to incoming requests for web sites. These

servers simply furnish the home page.

b) Application server layer – this is also known as the e-commerce layer. The user is switched to this layer

whenever he/she wants to view products or to place an order.

c) Back end layer – This layer connects the application servers to the back end systems consisting of the

organization’s information systems – ERP, inventory, accounting etc. the user is not directly in contact

with this layer. For the organization, this layer handles the order processing, accounting and inventory.

19

Fig 5. Server Architecture

E-commerce site

A simple e-commerce site is shown below. The top part of the site is called a banner. The banner has left

and right panels for images. The bottom left has a side bar with menu options. The main part of the web

page has content.

Fig 6. A simple e-commerce site

Web Site Content

One of the major components of an E-commerce system is the web site (what are other major components?).

The purpose of web site content is to convert browsers to buyers. Web site content is mainly developed in

HTML which is a language to format text and images on a web page. Data in a webserver could sometimes

20

be in XML format. The code can perform various functions such as loading a form, loading a web page,

verifying credit card information, verifying form data etc. The code could reside on the client side or on the

server side. In the old days code resided on client side, but it proved to be a boon for hackers, so nowadays

code/applets reside on the server for greater protection.

The web site content consists of:

Text and graphics

HTML

XML (data)

code -- (part of E-commerce application)

client side -- Java script

Server side – Applets, servlets (PHP, JSP/ASP)

Web site Content – HTML and XML

HTML – Hypertext markup language is a language used to ‘mark up’ text for display on the web. Note that it

is not a programming language. In HTML text is “tagged” with different types of tags to create a variety of

effects such as bold, color etc. Most tags have a start, marked by angled brackets (e.g. <start tag>) and an

end marked with a forward slash and angled brackets (e.g. </end tag>). Thus ‘<b> Bolded text </b>’

creates bolding for Bolded text. Also note that all html documents start with <html> and </html> tags. In

the example below the html code and how it appears when displayed in a browser are shown. As an exercise

see if you can copy and paste it into notepad (its an “Accessory” in windows), save it as “testhtml.htm” and

open it using “file”, “open” in your browser.

Fig 7. A example of the HTML code

XML – XML or Extended Markup Language is used to ‘mark up’ configuration (e.g. ‘date’ to be in

mm/dd/yyyy format) files, product files or other data files for transfer of information on the internet. It has

become a universal format for data exchange on the internet. XML uses user defined tags, but it needs a

21

schema to define the syntax of the tags following it. In the example below, ‘xs’ stands for XML schema. A

‘note’ element type is described as a complex type consisting of the elements, ‘to’, ‘from’, ‘heading’ and

‘body’. Note that the part listed directly below is the XML schema so that a browser or application could

‘understand’ the data. and the second part is an instance of the schema.

<xs:element name="note"> <xs:complexType> <xs:sequence> <xs:element name="to" type="xs:string"/> <xs:element name="from" type="xs:string"/> <xs:element name="heading" type="xs:string"/> <xs:element name="body" type="xs:string"/> </xs:sequence> </xs:complexType> </xs:element>

Following is an example of the <note> data type, that is defined above.

<note> <to>IS524</to> <from>Amaravadi</from> <heading>Proficiency with IT</heading> <body>Proficiency with every aspect of IT is vital in this digital age</body> </note> Another example is shown below. Here a ‘bookstore consists of various books. Each ‘book’ has the

attributes (what are these called in XML?) title, author, year and price. Could you write the XML schema for

this data?

<bookstore> <book category=“YOUNG ADULT"> <title>The Maze Runner </title> <author>James Dashner</author> <year>2010</year> <price>6.35</price> </book> <book category="WEB"> <title>Beginning XML</title> <author>Jill Fawcett</author> <year>2012</year> <price>24.87</price> </book> </bookstore>

Web site Content – Code

As we have learned previously, the content of a web site includes code as well. The complete code will

constitute an e-commerce application (discussed next). The code can perform various functions such as

loading a form, loading a web page, verifying credit card information, verifying form data etc. There are a

number of technologies to perform these actions including Java and JavaScript. As stated earlier, the code

could reside on the client side or on the server side. In the early days of the internet code resided on client

22

side, but it proved to be a boon for hackers, so nowadays code/servlets reside on the server for greater

protection.

Code – Client Side

In JavaScript high level functions are called within the html document (is HTML a programming language?

Confusingly HTML is also referred to as ‘code’). In the script below, the function is ‘write’ (document is the

class, a concept similar to entity class). ‘document’ refers to current document, so the code will write

whatever is in the quotes i.e. Hello World! to the current document. You could copy and paste this in the

notepad, save it as a ‘.htm’ file (test.htm) and open it as a file in your browser. See if you can change it to

print “Hello IS524 Class!” two times!

<!DOCTYPE html> <html> <body> <p>Click the button to greet class</p> <button onclick="greetClass()">Greet Class</button> <script> function greetClass() { alert("Hello IS524 Class!"); } </script> </body> </html> Code – Server Side

Scripts are usually hidden in the server side and called from the client page, as the example below illustrates.

<html>

<head>

<title>Test form for yor own php</title>

<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">

</head>

<body bgcolor="#FFFFFF" text="#000000">

<font size="-1" face="Arial,Helvetica"></font>

<h2>Try it yourself - a test form to try your own php</h2>

<form method="GET" action="http://www.faculty.wiu.edu/C-Amaravadi/xxx/xxx/myServer.php">

<p>Name:

<input type="text" name="yourname" size="50" maxlength="50">

</p>

<p>Email:

<input type="text" name="email" size="50" maxlength="100">

</p>

<p>

<input type="submit" name="Submit" value="Submit">

</p>

</form>

</body>

</html>

23

In this example, the code above opens a form (see form above) that takes a user’s name and email and then

passes these (i.e. ‘yourname’ and ‘email’) to a script called myServer.php. This script is shown below:

<html>

<head><title>PHP EXAMPLE</title>

</head>

<body bgcolor="#ffffff">

<font size="-1" face="Arial,Helvetica"> </font>

<h2>PHP SERVER SIDE SCRIPT</h2>

<p>You said your name is:<?php echo $_GET['yourname'];?>

<p>and your email is:<?php echo "<a href=\"mailto:" . $_GET['email'] . "\">" . $_GET['email'] . "</a>";?>

<p>That's it folks!

<p> Don't forget to have fun!

</body>

</html>

Note that the script is mixed with the html code so that the output appears formatted as:

An e-commerce application

We learned earlier that a customized application has three major components:

1) the interface or front end of the application,

2) the business logic or the component that does the actual processing (e.g. adding up items at a retail

counter) and

3) the back end or database.

An e-commerce application consists of scripts that usually are written in Java/JSP that run in the application

server layer. The higher level functions are illustrated in the diagram below and provide the needed web

server functionality. In this example, the application is a cellular provider that allows clients to log in and get

24

information on cellular packages (such as ‘1GB’, ‘2GB’ etc.). ‘GetClientInfo()’ can get the user name and

password and ‘SelectPkg()’ allows the client to select one from all the choices displayed. Unfortunately, the

web site code is not transparent being a mix of HTML, scripts and java code. Higher level functionality is

difficult to perceive unless one looks at pages and pages of webpage source.

Fig 8. A customized application has three major components

Fig 9. The application in action

Security and Security Problems

Since buyer and seller cannot see each other, web transactions require a secure environment. Both physical

security and transaction level security is required. The following figure shows that web site vandalism is the

most common security problem for web sites followed by denial of service and distributed denial of service.

25

Fig 10. Common Security Problems

Physical security

Physical security is safe storage for information and is ensured via firewalls and electronic accounts. A

firewall is a set of hardware and software to protect the web server. Electronic accounts hide the customer’s

balances.

Transaction security

Transaction security is the safe exchange of payment information. In the internet data can travel through many

nodes before reaching its destination. The data can be potentially intercepted within the network or en-route,

so it is important to encode the information so it is visible only to the parties in the transaction. Transaction

security is ensured via encryption and authentication schemes.

Encryption

Encryption is encoding the transaction. There are two major methods of encryption: SSL and SET. Both of

these work on the ‘public key’-‘private key’ scheme. The private key consists of the factors of the public key. For

example the factors of 1650 are 55 x 30 so the public key is 1650 and the private key is 55 and 30. The

scheme is based on the idea that there is no algorithm to identify the correct factors of a very long number.

The public key is is used to encrypt information before sending and the private key is used to decrypt it.

SSL – SSL stands for Secure Sockets Layer and is used widely in e-commerce transactions. In SSL, as in SET

keys are exchanged at the start of the transaction. In SSL the key exchange protocol is called S-HTTP. Note

that SSL is an encryption method.

SET – SET is Secure Electronic Transaction is an encryption method used for financial information such as

currency exchanges, credit cards etc. In SET every transaction is ‘witnessed’ by a third party.

26

Fig 11. Encryption

Authentication

Authentication is a method of verifying identities of the parties in a transaction. When one writes a check,

store clerks ask for a drivers license or other identification. This ensures that the person writing the check is

the same person who is on the driver’s license. Similarly on the internet it is important to verify identities of

the buyer and seller, so digital certificates are used.

Digital Certificates – Because parties in an e-commerce transaction will not typically see one another, it is

essential to have verification of identity. Digital certificates are similarly exchanged between buyers and

sellers before the start of a transaction. A digital certificate is a document issued by a certificate authority that

has information about the owner, owner’s public key, validity date, serial number etc.

Fig 12. Digital Certificates

Electronic Payments

Payments are obviously an important part of e-commerce. Payments need to be made when purchasing a

product or service. Choices in payment methods are credit/debit cards, smart cards, electronic checks and

electronic accounts. Payments can be made as in the physical world by providing the card#/account#.

When a shopper makes a payment, the merchant sends the transaction to the customer’s bank and receives

payment. If it is a credit card, then the customer’s credit card company is billed and the customer pays the

company at the end of the billing cycle. Sometimes there is a third party who acts as an intermediary between

the customer’s bank and the merchant’s bank (e.g. paypal).

27

Fig 13. Payments in E-commerce

Problems with credit card, electronic payments

People typically do not want to expose their credit card to the web, besides there are some categories of

people like elderly, school kids and travelers who may not have credit cards. The solution is supposed to be

digital cash or electronic accounts. In the early days of the internet, vendors tried to introduce digital cash

that mimicked paper cash, but that became problematic as it needed a machine (just like a credit card

processing terminal). Now electronic payments are carried out via electronic accounts. Leading vendors

include Paypal, Verisign, RBS etc. In order to use the electronic account, shoppers simply select this is a

payment option and log into their account and approve the order amount.

E-Commerce Business Models/E-commerce Strategies

An E-Commerce business model is a method of operation or primary reason for existing or what the business

does. It includes, being an E-tailer, a transaction broker, a market creator, content provider, community

provider, portal and service provider.

CATEGORY DESCRIPTION EXAMPLES

E-tailer Sells physical products directly to

either other businesses or

consumers

Amazon

Wal-mart.com

Transaction broker Processes online sales

transactions for users, charging a

fee for each transaction in order

to save those users both time and

money

E-trade.com

Expedia

28

Market place Sets up a digital environment

where buyers and sellers can meet

to browse and exchange products

allowing for the setting of the

prices on these products. They

make profit off transaction fees

e-Bay

Priceline.com

Content provider Generates revenue by providing

digital content over the web.

This revenue is gained either

from selling access to this content

or through providing advertising

space.

WSJ.com

GettyImages.com

Itunes.com

Social Networking Enables the sharing of

information between people with

similar interests

Facebook.com

Portal Serves as an initial point of entry

to the web providing additional

services as well.

Yahoo

Google

Service Provider Provides web 2.0 services such as

photo sharing and online backup

to users

Photobucket.com

Dropbox

E-Commerce Revenue Models

Model Description

Advertising Revenue Model Revenue is generated by gathering a large audience and then

subjecting that audience to a large volume of advertisements

Sales Revenue Model Revenue is derived from selling: goods, information, or services

to customers.

Subscription Revenue Model Revenue is generated from the charging of a subscription fee in

exchange for access to the provided content.

29

Free/Freemium Revenue Model Revenue is generated from a premium that is charged for

access to additional services above and beyond the basic

services which are provided for free.

Transaction Fee Revenue Model Revenue is generated from charging a transaction fee for each

transaction that a company generates.

Affiliate Revenue Model Revenue is generated from by web sites that send visitors to

other web sites in exchange for a referral fee.