Making Exception Information Understandable for Persons without Computer Science Education : Improving the Software by Analyzing Exceptions Data

• Main Author: Edeljung, Johan
• Format: Others
• Language: English
• Published: Umeå universitet, Institutionen för datavetenskap 2016
• Online Access: http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-128574

This thesis consider exceptions from a different (maybe unique) perspective. Not talking about how and if they should be handled, instead it purposes ideas how to analyze exception data. Both in an analytical way, by plotting large amount of exception information with different parameters, and also by seeing if the information that each exception object carries can be made available for others than developers. These analyses are made available because of a centralized database, filled with exception data, that the software company Vitec has. The method used to evaluate these analyses, was to develop an application, that could perform such analysis. A model to translate the exception information was also developed, and the resulting translations was integrated with the application. Tests to evaluate how useful the outcome could be for the target users, developers and support personnel of Vitec, was conducted. The conclusion of this work was that in fact both types of analyses can help improve the software. The translations by allowing the support personnel to faster resolve the issues at hand, and taking some of the burden of the developers. The charts plots can assist the developers to see how a system is working, and where the most errors occurs. The work answered the question if there would be any use of such analysis and translation of exception, however, in answering these questions opened the door to a new big question. How can the analyses and translations be done in the best way? The thesis discuss this matter as well, but this work has just started and as but the investigations, strongly suggest ,further work would be beneficial.