I have stated several times that Massimo should be very careful in advocating for Web2py. I'm still holding this opinion. But I won't elaborate on this.
About the facebook clone. I do not see anything offensive about this video. I thought it's pretty good. But it can be improved. How? Well, Facebook, Reddit, etc. are quite trivial applications in terms of functionality and features. But Facebook doesn't work perfectly and Reddit fails once every other day. Why? The answer is Scalability!!!! These applications are hard to build to withstand a tremendously large user base. Here's my suggestion. I remember Massimo did a Reddit clone once. Let's redo this. This will be a hit if done right (the political climate is right :-)). The Reddit clone should show these: 1. Expressiveness and conciseness of Web2py. 2. Exhibition of included batteries: form, authentication, permission, DAL and CRUD. 3. Scalability. There are many frameworks that will take 2 of these 3 items. A good microframework will take 1 and 3 easily. I think Web2py can leverage itself to take all 3 items. Items 1+2 can be easily shown. But item 3 will take a thoughtful experimental design to show. And it might cost some money (but probability not much). Item 3 encapsulates many practical aspects, including deployment. Possibly, choose a fast server like nginx, fine tune the set up to produce optimal results. And if possible, set up a scenario multiple replicated databases. I believe this demonstration will be far better than the mere words of advocacy for web2py. The benefits are multi-fold and will affect new comers and current users (many of whom, I think, are still struggling or at least tweaking with optimal, scalable deployment of web2py apps). Nobody expects that this Reddit clone will replace Reddit (for instance). Fortunately (or unfortunately), not many real apps will have the "problem" of breaking Reddit frequently is having. But, many people will jump in if they see that Web2py is productive, easy and able to build decently scalable apps.
