What bugs me about this change is that after three years of preview with one billing model, GAE comes out of preview with a very different billing model, and gives developers virtually NO time to adjust their apps for it. The difference of the new billing model to developers cannot be overstated--it is huge.
I admit I cannot be considered a prime customer for Google: my app is a backend for a small Android app called "Rate This Spot", which just implements a rating system for geographic areas. I developed it in my spare time. I spent A LOT of time optimizing the CPU load of it. In the end, I came up with a system that aggregates ratings in memcache and a cron job that commits them to the data store. Very low CPU usage, and even though volume is low, it was designed to scale and still cost very little if volume increased. Now the billing model changes and it turns out my app is designed completely wrong for the new model. My cron job keeps the instance alive too much and I will end up having to pay for an app that hardly gets any traffic and makes me no money. Turning down the cron rate will increase the risk of data loss of memcached values before I can store them. The ridiculous minimum 15 min billing time forces me to really crank my cron rate down to make any difference. When I designed this thing I had assumed that by the time I'd actually have to start paying, I would have enough users to be able to monetize this in some way. But now I actually have to hope that the app doesn't become popular, because if it does, I will likely have to shut it down because it will cost me too much before it reaches any critical mass. My optimizations have been pointless and very counterproductive. I would have to rewrite it for the new model and I don't have any spare time to do that right now. Google has been evaluating over the past 3 years how people use GAE so they can make money. That's fair. But at the same time, during these 3 years developers have been pouring a lot of effort in learning how the system works, which knobs to turn to make efficient use of the environment, and yes, how to design their app so as to minimize cost and maximize profit. Now the rules are turned on their heads. Early adopters are punished because by now they have deployed countless large apps that are mis-architected and mis-optimized and will suddenly cost them much more than anticipated. Developers feel trapped and betrayed--they signed up for one thing, and now they are offered something else instead. They made their decisions based on data: cost of CPU usage, API calls, etc. I know I spend many hours benchmarking, calculating, extrapolating. It seemed like something I could do with minimal risk. If it didn't take off, I would not loose any money. If it did, I would have enough breathing room before I'd need to find a way to make it pay. Now I don't know what will happen. I'll probably end up paying money without making any. That's just what I didn't want to happen. I for one am abandoning GAE, unless something drastically changes with the new billing. I should have known buying into a proprietary system was a bad idea. It always is. My Rate This Spot app will likely stay on GAE, unless it ends up costing me too much money. I was working on a different project on GAE, but I am going to do it on a proper open source system instead, using node.js and CouchDB. Much safer. It's sad: I used to be really excited about GAE. Here was a system that seemed to be aimed at people like me: people with ideas that can build something in their spare time, and maybe see it grow into something big. That doesn't seem to be the case anymore. Maybe I don't feel like I'm abandoning GAE. Instead, GAE seems to have abandoned me. -- You received this message because you are subscribed to the Google Groups "Google App Engine" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/google-appengine?hl=en.
