Can you explain what it means to "check the data"? The point here is that, for the last year, Twitter has NOT been generating sequential IDs. Just look at the s

Can you explain what it means to "check the data"? The point here is that, for the last year, Twitter has NOT been generating sequential IDs. Just look at the source code[1]. They save a few low bits for the worker ID, data center, and worker-specific sequence. The highest bits are the timestamp bits. Maybe you can correct me if my assumptions based on the code are wrong, but my guess is you'll magically find that Twitter's highest ID grows linearly with time. Then you're just randomly sampling within this maximum timestamp. At this point you're assuming that Twitter sees a constant rate of signups wrt time, which I highly doubt. [1] https://github.com/twitter/snowflake/blob/master/src/main/scala/com/twitter/service/snowflake/IdWorker.scala https://github.com/twitter/snowflake/blob/master/src/main/sc...