I believe this makes Amazon S3 behave more similar to Azure blob storage[1] and Google Cloud Storage[2], which is pretty convenient for folks who are abstractin

I believe this makes Amazon S3 behave more similar to Azure blob storage[1] and Google Cloud Storage[2], which is pretty convenient for folks who are abstracting blob stores across different offerings. For what it’s worth, consistency in S3 was usually pretty good anyways, but I ran into issues where it could vary a bit in the past. If you designed your application with this in mind, of course, it shouldn’t be an issue. In my case I believe we just had to add some retry logic. (And of course, that is no longer necessary.) [1]: https://docs.microsoft.com/en-us/azure/storage/common/storage-concurrency#managing-concurrency-in-blob-storage https://docs.microsoft.com/en-us/azure/storage/common/storag... [2]: https://cloud.google.com/storage/docs/consistency https://cloud.google.com/storage/docs/consistency