I have chosen to write about Business Process Analysis as it is very relevant to my work. I currently support, maintain and develop an Enterprise Resource Planning (ERP) system. The business also uses a Customer Relationship Management (CRM) system. We are also looking at Workflow Management through an ITIL helpdesk package. Currently workflow is managed through a comprehensive set of procedural documents.

We use our CRM system to record all contact with customers and potential customers. Potential customers cannot be recorded in the ERP system as a customer account must exist. The CRM system can be used to record activities such as visitng a customer site, calling the customer on the phone to notify them about a special on a product, receiving a call from a customer regarding pricing, a lead on a potential customer, calling a customer regarding an outstanding invoice and any other contact.

Workflow Management involves the development of procedures relating to the flow of data, documents, and work items from one person or step to the next. Software applications exist to assist in this, and many helpdesk applications now assist organisations in adhering to their procedures.

ERP systems allow a business to enquire on entities such as suppliers, stock, customers, and warehouses. Their main purpose however, is to perform transactions on these entities. Some of the transactions that may be performed are:

Entry, editing, picking, despatching, and invoicing of sales orders.

Entry, editing, sending, receipting, and updating of purchase orders.

Transfer of stock from one warehouse to another.

Place work orders on production to manufacture goods.

Reporting on and following up on all of the above.

Overall these systems can be integrated to make data quickly and easily available to those who need it, and to make internal, customer, supplier and partner transactions easy to manage.

Collaborative agents are autonomous, have social ability, are reactive and proactive. They can interact with other agents to achieve tasks that any single agent could not perform on its own.

Interface agents are autonomous, have social ability, and can learn. They learn to assist the user by imitating them. They can ask other agents for advice. Requirements for an interface agent are that a task is repetitive, and that it is different for different users.

Mobile agents are autonomous, have social ability, are adaptable and mobile. Mobile agents can perform some intelligent operation before sending a more useful request to a server, reducing bandwidth and CPU time on the server.

Information or Internet agents manage, manipulate, or collect information from many sources. They are grouped by what they do rather than what they are, and are more of a social agent.

Reactive agents are very simple and do not have internal models of their environment. Rather, they react to stimuli within then environment in which they exist.

Hybrid agents consist of 2 or more of the above mentioned agents.

Smart agents do not yet exist. They have the properties of all of the previously mentioned agents.

Competitive agents are similar to collaborative agents, but they compete with other agents to get a task done. They may not perform taks for other agents if it could affect the tasks they are performing.

(Agent Behaviours and Types, n.d.).

2. Describe how techniques such as artificial intelligence and statistical techniques are used in software agents.

3. Identify various activities in e-commerce where software agents are currently in use.

Reverse ausctions can use software agents to seek a price from a service or product provider. Software agents can also perform the task of searching the Internet and contacting sellers to obtain pricing information.

4. Computing ethics and bot programming case study: rocky

a. Get an account username and password from the lecturer to LC_MOO at http://ispg.csu.edu.au:7680 and login to the Welcome Lobby.

b. Hold a 5-minute discussion with Rocky on a special topic. Commands act rocky (start bot) hush rocky (stop bot)

I have held a conversation with Rocky.

c. Rocky is an ELIZA-like bot. Report your findings.

Rocky responds with a lot of leading questions, prompting for further information. On doing a quick web search, many websites tell me that ELIZA-bot is a psychotherapist bot, which fits well with its responses.

References

Agent Behaviours and Types (n.d.). Retrieved 19th May 2009 from http://www.comp.nus.edu.sg/~bimlesh/ametrics/findings/Agents%20Behaviors%20and%20Types.pdf

I have chosen this question from the 4 options as it is an area that I have an interest in. Location based services are entertainment or informational services that are delivered based on where you are located. Your location is determined by methods such as GPS on many modern mobile phones, or by triangulation where your approximate location is determined from the 3 mobile phone towers that you have the strongest signal to. Using location based services, you can find the nearest service station, nearby Turkish restaurants, bookshops, or almost any other business or service that you require. You can also receive entertainment and advertising from nearby providers.

A spider is software that follows links on websites to populate it’s database. External links to a website are important to have the website added to a spider’s database (What Exactly IS a Spider, n.d.).

2. What is a meta-search engine? Provide some examples.

a meta-search engine sends request made by users to several other search engines / databases and displays the results as a single list. This could also include the source it was retrieved from. The foundation of meta-search engines is that the web is too big for a single search engine to give the best results. By using multiple search engines, the most relevant sites can be found more easily (Metasearch engine – Wikipedia, n.d.).

3. How can you get your site listed at major search sites; and how could you improve your site ranking?

Some search sites have a function to submit your URL and the search site will visit your site to add it. Some follow links from other sites, and others use databases of registered domains. Depending on which search site you want your site to rate highly on, you can use different methods. One would be to have many other sites link to your site through affiliate programs. Another would be to use keywords throughout your site that you want your site to rate highly against. A further method would be to get a lot of traffic to your site through advertising.

References

What Exactly IS a Spider (n.d.). Retrieved 18th May 2009 from http://www.seologic.com/faq/robots-definition.php

Meta search engine – Wikipedia (n.d.). Retrieved 18th May 2009 from http://en.wikipedia.org/wiki/Meta_search_engine

Customer Relationship Management software is used to keep track of customer relationships by recording contact such as phone calls, customer visits, enquiries, prospects and contacts that are not normally kept in an ERP or accounting package. This includes prospective customers.

Business-to-Business (B2B) software is used to transfer data between businesses such as purchase orders and invoices. This decreases cost of employees to manually enter data, time for goods to ship as orders can be processed almost immediately if stock is available, and time to receive payment, as many businesses using B2B have payments automatically flagged to pay on their due date.

Supply Chain Management (SCM) software is used in a vertical market to increase efficiency, decrease stock on hand, and reduce time to deliver goods. This is done by suppliers making stock and lead times available to customers, and customers using this information to place orders in a timely fashion so as to not run out of stock, but not waste money on excess stock. This can be used on a supply chain for example, where a retail outlet can see the Australian distributors availability, the Australian distributors can see the Asian distributors availability and lead times, and the Asian distributors can see the manufacturers lead times.

These are areas that I have had a lot of experience with through my employment.

A shoppingCart object is associated with only one creditCard and customer and to items in itemToBuy. Persistent customer information such as name, billing address, delivery address, e-mail address, and credit rating is stored in the customer object. The credit card object is associated with a frequentShopper discount object, if the credit rating for the customer is good. The customer can make or cancel orders as well as add and delete items to the shoppingCart product. The credit card object contains the secure method for checking that the charge is authentic.

The class diagram below was created using the open source application “Dia”.

During my search I found that there are 13 types of diagrams used in UML. 8 of the main ones are:

Class Diagrams that describe the static structure of a system

Object Diagrams that describe a system at a particular point in time

Use Case Diagrams that model the functionality of a system

Sequence Diagrams that show the transfer of messages between classes over time

Collaboration Diagrams that combine the static structure and dynamic behaviour of a system by showing the series of sequenced messages

Statechart diagrams that show the reaction of a system to external input

Activity diagrams show the flow of control from one operation on a class to the next

Component diagrams show the organisation of source code, runtime, executable and other physical software components

Deployment diagrams show the physical resources used by a system

(Types of UML Diagrams, n.d.).

References

Types of UML Diagrams (n.d.). Retrieved 17th May 2009 from http://www.smartdraw.com/tutorials/software/uml/tutorial_01.htm

ACID properties is an acronym for Atomicity, Consistency, Isolation, and Durability.

Atomicity means a transaction is either complete or it is not complete. If a transaction did not complete for some reason, it is rolled back to the state it was in before it was commenced.

Consistency is ensuring at the completion of a transaction, that the system is in a valid state. If it is not, the entire transaction will be rolled back.

Isolation means that each transaction thinks it is the only action that the system is performing at the time, and it has exclusive use of the system.

Durability means that once a transaction is complete, all changes to the system are permanent. Safeguards must exist to ensure that even if the system has a hardware failure, transactions will not be lost.

(ACID Properties, 1998).

2. Describe a TP monitor environment. How can a TP monitor stop an operating system being overwhelmed?

In a distributed client-server environment, a Transaction Processing (TP) monitor is a program that monitors and controls the transfer of data between local and remote clients and servers (TP Monitor, n.d.).

This ensures that the amount of data sent between clients and servers does not exceed what the operating system can handle, and prevents the loss or corruption of data.

3. What is the difference in load balancing with traditional and transactional MOM, RPC, and conversations?

4. Why is a two-phase commit protocol better than a one-phase atomic commit protocol?

A two-phase commit protocol (2PC) eliminates the number of failures of committing data. This ensures the integrity of data. The downside however, is that it is slower at committing data (IEEE Xplore, 1998).

References

ACID Properties (1998). Retrieved 17th May 2009 from http://msdn.microsoft.com/en-us/library/aa480356.aspx

TP Monitor (n.d.). Retrieved 17th May 2009 from http://encyclopedia2.thefreedictionary.com/TP+monitor

IEEE Xplore (1998). Retrieved 17th May 2009 from http://ieeexplore.ieee.org/Xplore/login.jsp?url=http%3A%2F%2Fieeexplore.ieee.org%2Fiel4%2F5962%2F15952%2F00741040.pdf&authDecision=-203

Up to 3 of the 10 defined threads can be run at any one time. A variable “running” is defined to track the number of running threads.

Delay is defined as a random variable between 0 and 2, to 12 decimal places. This gives 100 billion possible decimal places times 3 possible integer values, or 300 billion possible numbers. The probability of getting 2 of the 3 running processes starting or finishing at the same time is so minute that it is negligible. This means that a deadlock is extremely unlikely.

The threads run for their random time and when one finishes, the next one starts. This is repeated until all 10 threads have completed running.

www.shoppingsquare.com.au accepts bank cheque, money order, PayPal, and Bpay. I would trust this site and would buy from it as the business has been verified to attain a Bpay number, and with PayPal you do not need to provide your credit card details to the seller. Your card will be charged by PayPal and the funds made available to the seller by PayPal.

2. What measures should e-commerce provide to create trust among their potential customers? What measures can be verified by the customer?

Measures that e-commerce sites can provide include website reviews from trusted sources, digital certificates from a trusted source, encryption, customer reviews (although these cannot be guaranteed), and the provision of telephone numbers, ABN, industry membership, addresses or other details that can be used to verify that the seller is legitimate.

3. Visit the Verisign web site – what solutions does it offer for e-commerce?

 Verisign offer SSL certification of websites. This means that they provide a certificate for the website, which will be recognised by web browsers. It also means that websites using Verisign’s solution provide 128-bit encryption (SSL Information Centre, n.d.).

4. Visit the TRUSTe web site. Describe what services and solutions are offered?

TRUSTe provides certification to inform consumers that an online business has a secure privacy policy in place, and that software it provides is safe. They also provide information about safe web browsing (TRUSTe – Make Privacy Your Choice, n.d.).

5. Get the latest PGP software from http://web.mit.edu/network/pgp.html; install it on two machines and encrypt a message on one machine and decrypt it on the other. Report your findings.

This software is out of date and no longer available. I searched for alternate products, but was not able to find any open source products in a short search. If I have time I will come back to this later.

6. The use of digital certificates and passports are just two examples of many tools for validating legitimate users and avoiding consequences such as identity theft. What others exist?

Some of the validation tools that can be used include:

Wildcard SSL certificate

Authentication logon with a secure USB authentication token

Basic authentication where a username and password is provided for each realm

Digest authentication which is based on challenge – response. This involves the user making a request without any credentials. The server then responds with a WWW-Authenticate header requesting credentials. Instead of sending the clear text credentials, the user sends the Message Digest (MD5) version of their credentials.

NTLM is Microsoft’s own NT LAN Manager authentication algorithm over HTTP. It is similar to MD5 in the way that it works.

Certificate authentication is stronger than other methods. it involves the use of public and private keys to encrypt and decrypt data.

Microsoft passport and other passport methods involve the use of a single set of credentials to sign in to any of the sites that use these passports.

Forms based authentication uses cookies to allow validated users to access parts of a website.

(Web Authentication, 2008). 

References

SSL Information Centre (n.d.). Retrieved 14th May 2009 from http://www.verisign.com/ssl/ssl-information-center/ecommerce-trust-ssl/index.html?sl=hp_banner_ssl_increasetrust1_570x200

TRUSTe – Make Privacy Your Choice (n.d.). Retrieved 14th May 2009 from http://www.truste.org/consumers/index.php

Web Authentication (2008). Retrieved 17th May 2009 from http://www.hackerscenter.com/index.php?/HSC-Guides/Web-App-Security/Web-Authentication.html