As I see it, there are two elephants in the room that need addressing.
- Do you really want to get involved with anything Google given their links to NSA?
- Do you really want to place company or personal resources into this product given their track history of shutting down experimental, and not even experimental technologies (as this is)?
(or Maybe it is time to take off my tin-foil hat. I don't know about that, I just know I am deeply disappointed with Google)
For question #2, this doesn't seem like a heavy duty product, just a tool for experiments and spikes. If you're doing real development, you wouldn't be using it (not unless it grows a lot in power and scope).
Question #1 isn't really relevant for serious projects.
- Statistical analysis of Google closures shows that they deprecate products at below industry pace, so your impression about that is also wrong: http://www.gwern.net/Google%20shutdowns
I mean, I guess you could boycott their services and products if you're deeply disappointed with Google. These concerns probably aren't relevant to this because similarly to the OAuth playground, it's simply a developer usability tool. Like the webdev tools and other web-related Google things.
I think the App Engine is perfect for developers who don't want to have to make decisions about servers and infrastructure. This can be beneficial. However, your design must fit their infrastructure. Depending on your application, this may or may not be a good thing.
If you ever need/want/have to migrate your application elsewhere, you'd have to replicate their environment or rewrite the application.
For the kinds of applications I work on, those are deal breakers so I use AWS (which also has its limitations) or some other server hosting service instead.
I run a number of apps on Python/App Engine specifically because I don't want to deal with servers/infrastructure/scaling. Sometimes the limitations aren't ideal, but it's still a trade-off I've been happy with overall.
Right, this proves my point.
You can run Java apps on JBoss. That's it.
You are still limited to using their infrastructure.
Don't get me wrong, that's not necessarily a bad thing.
For many, it's wonderful. I'm just saying that one needs to weigh needs, capabilities and goals carefully before deciding to use Google Apps (or any other "platform") because your application has to work in their platform and moving to another platform could be very costly.
It seems obvious to say, but many technical decisions involve trade offs of one kind or the other. The ones that bind early and have high switching costs are worth thinking through at some depth.
- Do you really want to get involved with anything Google given their links to NSA?
- Do you really want to place company or personal resources into this product given their track history of shutting down experimental, and not even experimental technologies (as this is)?
(or Maybe it is time to take off my tin-foil hat. I don't know about that, I just know I am deeply disappointed with Google)