Basically he suggests publishing a draft of the submission app on August 1st on the Android Market, before submitting it in the contest, in order to get some early feedback to allow for revisions. That would be nice, but I'm just not going to have my app ready by then. Some version of it will be ready by the end of August, but I'm just not going to have enough time to solicit feedback.
He also says to make sure the app isn't buggy. Kind of obvious. And that might be a problem for me. Again, I don't have much time, and this is the most complicated app I've done so far, so there might be bugs. I'm especially worried about out of memory errors, since I'm using a lot of resources.
The four criteria for judging are supposed to be:
- effective use of the platform
I think I'm doing well under each of those criteria. Murphy recommends focusing more on aesthetics, under the premise that judges will have a subjective bias for prettier, slicker apps and unconsciously weight that criteria more.
He also says to try to pick an underrepresented category. Again, probably too late with the advice there. My app most certainly will not fall under anything but a game category.
Another piece of advice is to localize the app, which seems a little strange. The rules state the app must be in English. Murphy thinks multi-language support would score brownie points...but how would you decide which ones to implement?
Finally the best advice is to keep things in perspective. ADC1 had about 1,700 apps and ADC2 will only have 30 winners. I think my concept and design is great, but the implementation will most definitely fall short of my original scope. The hope is that users and judges will enjoy what is implemented and will want to invest in seeing the game fleshed out. I would consider making the final cut (20 apps from each category, or 200 apps) a victory, but even if I don't make it that far, I will still have a workable prototype, will have learned a great deal while making the game, and will still have a viable app to release on the market.