Following are observations and recommendations regarding the project flow for the translator project for CSCI3081. To date we have completed iteration One "The Lexical Scanner," which consists of packaging component programs that were already developed with programs that needed to be designed and developed. In the iterative process we work to split a project into smaller chunks or pieces which permit us to more effectively manage the over all completion. We are able to take advantage of the entire design process from modeling the specification, constructing the tests, to creating the object oriented architecture. A very positive component of this exercise was the emphasis on testing. Although I have seen quality assurance efforts that have tested a completed program I have not previously thought much about an organized testing construct that would permit the developer to think of the testing piece as an integral part of the development process at a module level. This approach not only permits bugs to be eliminated much earlier in the process but encourages a new design metaphor in which one is able to see a design for the programing phase through the construction of tests prior to laying down code. The use of the SubVersion control system for managing the versions of files proved itself over and over again as a welcome and viable solution for ensuring that work was stored and archived in an organized manner. SubVersion proved to be a valuable tool for facilitating team contributions to the source code creation and modification process. In the first iteration we were able to touch the very surface of the human communication processes required to enable an efficient design and implementation. Since the team size was limited to two people our need for communication would seem to be small. However, unforeseen complications immediately emphasized the requirement that communication be emphasized as the highest priority goal. I was not able to participate in a timely manner for much of iteration One because of a personal family matter that could not be avoided or scheduled around. This issue frequently occurs on teams as a normal experience. Even for two person teams a expectation of the management structure of the project must be worked out at the very beginning. It is also very important to consider the over all contributions and experience. THus since we have only completed iteration One judgement must be reserved on the over all process. From a personal perspective, the jury is still out as to wether the lexical scanner developed for iteration One is extensible and will integrate with the remaining phases of the project without modifications. This is partly due to the lack of confirmation of the over all design. I.E., we are relying on a design that is given to us in pieces. There is nothing wrong with this approach and I am in no way questioning the designer's methods. Thus we have to move forward based on the trust that this effort has been completed successfully. on the Micro level it very well may be that the design of our piece I.E., the lexical scanner could be more modularized or defined differently to accommodate changes.