I almost skipped this article because it's long and I've never used Google Cloud, and I assumed it would be complaining about Google deprecating a bunch of cloud services I don't know anything...
I almost skipped this article because it's long and I've never used Google Cloud, and I assumed it would be complaining about Google deprecating a bunch of cloud services I don't know anything about. But it ended up being a great essay about how different development and company cultures affect their willingness to support a platform over the long term, and whether the burden for maintaining compatibility is the responsibility of the platform or the developers. I thought it was especially interesting to look at the conflict between trying to "clean things up" and the damage that ends up doing to users, including lowering their interest in using future platforms/offerings after being bit by one of your previous ones.
The sad thing is that Google App Engine mostly had this right for a long time. I have a little app that has been running there since 2008. There are not many places that will run a little hobby...
The sad thing is that Google App Engine mostly had this right for a long time. I have a little app that has been running there since 2008. There are not many places that will run a little hobby server for free and for so many years. I've long considered App Engine's stability to be a refreshing contrast compared to the constant churn inside Google.
However, I was lucky that I only used the core parts of App Engine. Many of the original API's have been deprecated and disappeared, and a business that relied on App Engine's ability to send email (for example) would have had to migrate to a different email-sending service.
I almost skipped this article because it's long and I've never used Google Cloud, and I assumed it would be complaining about Google deprecating a bunch of cloud services I don't know anything about. But it ended up being a great essay about how different development and company cultures affect their willingness to support a platform over the long term, and whether the burden for maintaining compatibility is the responsibility of the platform or the developers. I thought it was especially interesting to look at the conflict between trying to "clean things up" and the damage that ends up doing to users, including lowering their interest in using future platforms/offerings after being bit by one of your previous ones.
The sad thing is that Google App Engine mostly had this right for a long time. I have a little app that has been running there since 2008. There are not many places that will run a little hobby server for free and for so many years. I've long considered App Engine's stability to be a refreshing contrast compared to the constant churn inside Google.
However, I was lucky that I only used the core parts of App Engine. Many of the original API's have been deprecated and disappeared, and a business that relied on App Engine's ability to send email (for example) would have had to migrate to a different email-sending service.