...
1) We were, perhaps, overzealous in latching on to the video chat/sewalong idea during the paper prototyping idea. It is easy to focus on non essential but interesting features and to believe that they are essential (because they are interesting)! Our paper prototyping/testing sessions were perhaps not as helpful as they could have been in the long run, since we ended up changing so much that the information that we gathered was not as relevant to our later designs. This was one thing that we would definitely have done differently-- paper prototyped a more modest design and therefore get more useful feedback earlier on.
2) We should not try to make our low fidelity computer prototypes overly complete . or get too attached to it! In our first computer prototype we implemented project search and sewing help search using js, and we never got around to updating them to use a database, potentially because we wanted to still use the code we wrote and not waste it (or to save work). Also, the project pages were all done with static web pages. This made moving them into a database a bit tricky.
3) Because it It was difficult to design something meant to instruct for a user group in something that we were not a part of, we may have benefited from having someone with experience in the area serve as a sort of advisory role. Perhaps someone who was already skilled at sewing but could remember what hurdles they had to overcomehave no expertise in. This made making a representative project difficult. Also, features such as Project Search and Sewing Basics may not be as useful or efficient to users. For example, the criteria by which we chose to filter projects may not have been correct or useful ones.