After my post on Google App Engine and its lock-in factor I was surprised to see this on the top of Techmeme...
One of the biggest criticisms of Google's App Engine have been cries of lock-in, that the applications developed for the platform won't be portable to any other service. This weekend, Chris Anderson, the Portland-based cofounder of the Grabb.it MP3 blog service, just released AppDrop — an elegant hack proving that's not true.
My first thought: "Wow, That's incredible"
My second thought: "Wait a minute, what now?"
Here's the thing, the information part of an API is a way to get information to and from somewhere. It is a gateway to an information repository such as a database or a membership directory. So while this is an impressive feat by Chris Anderson it doesn't change the fact that the information repository on the other end of Google's API is proprietary and using that repository locks you in.
Having a copy of the APIs is great but you are still faced with the fact that your data is stored in a proprietary directory (bigtable), your users are using the Google log-in accounts, and so on. So to make the transfer to another server you are still going to have to recreate all that.
Which is exactly what I and every other Google App Engine critic has been saying from the start. No one ever claimed it was impossible to move your Google App Engine application to another server, just that it was difficult. From my previous post (bold added for emphasis)...
In fact, the end result of all these libraries is to lock you into Google's proprietary system and force you to use their language of choice (at least until they expand beyond python). So if you do use those features you really can't move your site to another host without a massive rewrite.
That's still true. As is my original point which is that using a standard apache web host with mySQL and your own membership provider is still a better bet. It allows you to move your application anywhere and it allows you to do so seamlessly.
Again, this isn't meant as an anti-Google App Engine post as much as its meant to say "lets be realistic about this and actually look at what's going on here". A good part of the blogosphere seems infatuated with this and that infatuation is causing them to spread a lot of mis-information. The idea that people would see this port and say "that's it, lock-in problem solved!" just shows how blinded some of them are.