![[spacer]](/schemes/blue/line.gif){kind=small}
|
1. what do you consider the main message of the paper? Code reuse is an important part of software development, but many developers avoid using components they don't know. Codebreaker provides a mechanism for automatically searching components based on the javadoc comments and method signature that a developer provides. 2.1. how is this work applicable to your own work? This work is very applicable to my work. I'm a Java software engineer and Java strongly encourages code reuse. This can be a real time-saver and also allows an inexperienced engineer to develop more complex applications. 2.2. how is this work applicable to your own interests? I'm not really sure how this applies to my interest which is user interface design. I'm not familiar with reuseable interface code, although I'm sure it exists. I don't actually get to work on UIs, so I don't yet have much experience there. 3. do you have ideas how the work in the paper could or should be further developed? No. 4. have you used any reuse libraries 4.2. if yes: which ones? did the libraries have any particular strengths and weaknesses? The product I work on uses several libraries – for some data structures, for xml processing and more. The libraries I'm most familiar with are the data structure libraries. They provide alternative implementations of Arrays and HashMaps; they provide more flexibility than the corresponding Java data structures. I haven't had any trouble with them, but many of the engineers I work with don't like them. They have found some implementation problems. 5. How does Codebroker infer the "task-at-hand"? It takes the javadoc comments and the method signature that the developer provides and forms them into a query. 6. what are the trade-offs between 6.1. letting the systems infer the task-at-hand If the system is wrong, it wastes the developers time by providing irrelevant options. If its inference is correct, it saves the developer alot of time. 6.2. versus that the users specifies the task at hand? The user can provide a more precise search query than an inference. Although is the user doesn't know what she wants and is unable to be precise, the system will fail. |
