There is a lot of internal enthusiasm for open sourcing our tools, but we are heads down at the moment so it’s just a matter of time (or lack of). I think we wi

There is a lot of internal enthusiasm for open sourcing our tools, but we are heads down at the moment so it’s just a matter of time (or lack of). I think we will time it for the next pgAU conference so we can talk about it at the same time. I definitely agree with you that tooling is a big deal. We did spend a lot of time manually proving it out before we spent the time on the tools. It was a bit of a leap of faith but it quickly became obvious that we were onto something. And you’re right, the point of our approach was to move the code to the data. We deal with reasonably large, complex real time data sets Sotheby’s round trips and marshalling become the dominating contributor to total time. Hence our ability to improve some operations by 100x. I can see that if you have a familiar and reliable tool chain and a compatible use case then ORMs could be great. I think our problem was that we had neither, so it was never going to work out too well for us!