Back in December, I published a post entitled, “How I spot not-yet-documented features in SQL Server CTPs.” In that post, I revealed a few of the ways that I get an early jump on what’s changed between CTP and/or RC builds of SQL Server. You can do those same things if you want to see what new objects or columns have been created, or which system modules have changed, between – say – SQL Server 2014 and SQL Server 2016. This will likely be a bigger list than any of the individual sets of items I’ve posted about in our SQL Server 2016 builds post, but the same concepts apply – create a linked server to the older instance, optionally create some synonyms for easy adaptation, and go to town.
There are a few other things I check as we get closer to the final release, and they can be quite revealing about what features have made it into the product. We can also get some insight into things they tried to get in but couldn’t (for example, there are error messages and Intellisense verbiage for
STRING_AGG()
, which does not seem to be in the cards for RTM). I’m going to point out a few, but I’m not going to iterate through all of the things I’ve learned – this is more to serve as as a starting point so you can experiment on your own.
If you’re interested in this kind of spelunking, you can learn a lot without having to reverse engineer binaries.