<td colspan="4"><strong>Implementation of Delta Debugging & DriverTest Case(s)</strong> ([http://www.stmozilla.cs.uni-sb.de/papers/tse2002org/ Simplifying and Isolating Failure-Inducing Input, Ziller and Hildebrandt, 2002tinderbox.html Tindexbox])<br />

<td>General AlgorithmCreation / Extrapolation of Test Case(s)</td> <td>The general delta debugging algorithm implementationWe need test cases that can return whether or not the test passes or fails. For details, see [http://wwwTinderbox has a couple of tests that are executed after the source is built.st.cs.uni-sbExtrapolate those tests from the Tinderbox source code so that we can use them in this project.deWe also need a test case that can pass/papers/tse2002/ http://www.st.cs.uni-sb.de/papers/tse2002/]fail consistently so that we can test the delta debugger.</td> <td>[[User:dwwoodsiAnkuswan|Dean WoodsideAditya Nanda Kuswanto]]</td> <td>Work in progress, check SVN repository from time-to-time.</td>

<tdcolspan="4">Minimizing <strong>Implementation of Delta Debugging Algorithm</tdstrong> <td>The minimizing delta debugging algorithm implementation. For details, see ([http://www.stinfosun.csfmi.uni-sbpassau.de/papers/tse2002/ http://www.st.cs.uni-sb.de/papers/tse2002/].</td> <td>[[User:dwwoodsi|Dean Woodside]]</td> <td>Not started.</td> </tr> <tr> <td>Schema Defintion of Driver Data<-99-01/td> <td>A simple XSD which defines the driving test type (e.g. user interactionYesterday, my program input)worked. Today, the minimal set of circumstance (scenario) to reproduce the failure, and the expected outcome after automating the circumstancesit does not.</td> <td>[[User:dwwoodsi|Dean Woodside]Why?])<br /td> <td>Work in progress, check SVN repository from time-to-time.</td>

<td>Choosing a Record/Replay FacilityThe Algorithm</td> <td>In The delta debugging algorithm. Drives the framework to retrieve change sets, apply changes, build source tree, run test case of scenarios that require user interaction (namely, mouse actionss), to find the minimal set of failure inducing changes. The intersection of all other parts of the framework will require a record/replay facility that will record user interaction the first time through and then replay it later during the automationto make them work together. Ideally, should be abstract enough for easy extensibility with little impact.</td>

<td>Found some Windows tools Work in James Whittaker's famed [http://wwwprogress.howtobreaksoftware.com/ How to Break Software]. Checking these out for their appropriateness. Tending to think we might have Check the SVN repository from time to roll our own (can't script existing ones [/well/])time.</td>