AFRICA NAZARENE UNIVERSITY

Name: Eric Wahinya – 11JCS005

Unit: Software Engineering

Code: CSC320.

Title: Assignment 2 Lecturer: Mr. Njuki Due Date: 26th June 2013

1. Using the waterfall model; model the March 4th 2013 voting process in Kenya.
The software process models in software engineering are the workflow model which shows the sequence of activities in a process along with the inputs, outputs and dependencies, the dataflow or activity model that represents a process as a set of activities each of which curves out some data showing how the input to a process is transformed to an output and finally the role or otherwise referred to as the action model which represents the roles of people involved in the software and the activities for which they are responsible. The role or action model is divided into a number of categories and the waterfall approach is one of these categories aside from the others which are the evolutionary development and the CASE scenario.
The waterfall approach takes the fundamental process activities of specification, development, validation and evolution and represents them as separate process phases such as requirements specification, software design, implementation, testing and so on. (Sommerville, pg 65). The figure below shows the software lifecycle as explained by the waterfall approach;

Figure 1. The Software Lifecycle. (Sommerville, pg 66)

Before the March 4th elections that took place in Kenya a number of preparations were made before the voting process took place to ensure that events on that day would take place swiftly without any complaints of rigging by the respective candidates and their fellow supporters; Kenyans were given an opportunity by the Independent Electoral Boundaries Commission to apply for clerk and registration officer jobs so as to coordinate and count the votes during the election day, IEBC recruited their desired applicants and contacted them to be interviewed on week before the election date, they were summoned to attend five days of training where they were taught how to use the Biometric Voting Registration machine (BVR), they were taught how they would coordinate people during the election who would show up in large numbers and finally they were taught how to set up the polling stations i.e. preparing the ballots with different colours where the colours represented the different ranks of candidates being voted for e.g. president, governor etc. After the training the trained applicants were sworn in and made an oath to ensure that they would not violate the election laws during the voting process. On March 3rd in the evening they were allocated to their respective polling stations to prepare for the Election Day. In my opinion the events that took place while preparing for the elections would not be modeled well using the waterfall approach but the usage of the Biometric Voting Registration machine would be explained very well because it a system and the waterfall model breaks down the different steps that a system undergoes during its development until it is complete. * Requirements Definition Phase – The systems services constraints and goals are established by consultation with the system user and they are then defined in detail and serve as a system specification. The electoral commission applied this phase when they performed a research of how the BVR kits worked in other countries i.e. India and were prepared to give a 3.9 billion BVR tender to a number of companies i.e.4G Identity Solutions, Symphony, Face Technology and On Track Innovations who they got in touch with to see of the features they provided would meet their standards which was the supply of 9700 voter registration kits. The system was expected to make rigging of elections difficult by eliminating impersonation and multiple voting. (http://www.africog.org/news/fresh_blow_to_IEBC) * System Software and Design Phase – The system design partitions the requirements to either hardware or software systems. It establishes overall system architecture. Software design involves identifying and describing the fundamental software system abstractions and their relationships. After the electoral commission had gone through the requirements definition phase and found a favourable company to give the tender of supplying the BVR kits, they were brought to Kenya and the developers configured them in the manner IEBC desired them to be and immediately it was announced that voters would get rid of their previous voter cards and were given a time period of one month to visit their nearest polling stations to be registered again as voters using the Biometric Voting Registration machine. The machine was able to scan and save the fingerprints of all the voters and it also had a webcam that took the voters’ photos. Information of a specific voter could only be altered when the voter’s fingerprints were scanned meaning the system was very secure that even the clerks could not alter any voter information. * Implementation and Unit Testing Phase – During this stage, the software design is realized as a set of programs or program units. Unit testing involves verifying that each unit meets its specification. The Independent Electoral Commission decided to test the BVR machine once they were through with registration of voters by using some of the machines in a number of by-elections in the country and the machine proved to be quite useful because the voting processes were faster and more efficient and the results were announced with a short time. * Integration and System Testing Phase – The individual program units or programs are integrated and tested as a complete system to ensure that the software requirements have been met. After testing the software testing is delivered to the customer. In the March 4th election many polling stations countrywide used the BVR system to cast their votes and the machines were also used at the Bomas of Kenya which was the IEBC headquarters then during the tallying of the candidates’ votes.

* Operation and Maintenance Phase – Normally (although not necessarily) this is the longest lifecycle phase. The system is installed and put into practical use. Maintenance involves correcting errors which were not discovered in earlier stages of the lifecycle improving the implementation of system units and enhancing the system’s services as new requirements are discovered. As IEBC tallied the votes on March 4th they realized that the number of rejected votes was quite high and that some of the figures being analyzed were inconsistent. As they reviews the results more critically the IEBC Servers and the IEBC vote tallying computer system disc-space completely malfunctioned and collapsed failing to respond in correct downtime. The British High Commissioner to Kenya, Christian Turner through proxies particularly Maina Kiai began lobbying the inclusion of spoilt votes in final tally besides attempting to delay the presidential total tally announcement. The National Security Service quickly intervened and stopped the IEBC from making more use of the Electronic vote tallying system. Strategic Intelligence Service reported that the Red-October virus was used to attack Government of Kenya computer systems by Western spy-agents to mine crucial data. This clearly shows the challenges experienced by the BVR machine in the operations and maintenance phase where the system was used and it experienced some errors when hackers tried to alter the data. To solve this issue so as to avoid complaints which would have later caused violence in the country, IEBC began recounting the votes again manually to ensure data consistency. The number of rejected votes when the IEBC was using the electronic tallying system was nearly 300,000 at the 5million total votes cast mark, a very significant number that makes a case in the total vote cast. When IEBC took up the manual tallying system the total number of rejected votes was 39000 at the 4.6million total votes cast mark, this is a disturbing significant decline in the electronic figures compared to the manual figures.( http://www.kenyan-post.com/2013/03/revealed-what-really-happened-to-iebcs.html)

REFERENCES
Sommerville, I.(2007).Software Engineering.(8th Ed),Edinburgh Gate, England.Pearson Education Limited.
Lewela, M & Kisiangani,E.(29 October 2012).Kenya’s Biometric Voter Registration: New Solution, New Problems.http://www.issafrica.org/iss-today/kenyas-biometric-voter-registration-new-solution-new-problems
Goldman, D.(29 July 2012).Fresh Blow to IEBC in 3.9bn Kit Scandal.http://www.africog.org/ news/fresh blow to IEBC
Goldman, D.(7 March 2013).IEBC SYSTEM was HACKED by a VIRUS called the RED-OCTOBER VIRUS. http://www.kenyan-post.com/2013/03/revealed-what-really-happened-to-iebcs.html http://elog.or.ke/index.php/partners/10-ied/69-statement-on-biometric-voter-registration-bvr-processes-issued-by-the-institute-for-education-in-democracy-ied-and-the-elections-observation-group-elog