2010YOUY01 commented on issue #26225: URL: https://github.com/apache/datafusion/issues/26225#issuecomment-6158217163
> 8\. What I care about is that people are spending their effort on the work that matters most (at least in my view). As for how we get there, whichever approach we can agree on and move faster with works for me. This is also how I decide whether to refactor now or clean up later. I try to estimate the total amount of work needed to reach the final, stable version (including all the optimizations and features I currently have in mind), and see which path requires less effort: refactoring first or cleaning up later. For major operators like Aggregate and HJ, it's quite obvious to me that refactoring first (and it's cheap to do with AI) with a better problem model is easier. But I probably hid my chain of thoughts, making it confusing to others. I'll try to explain that in the writeup as well. I'm more tolerant of localized complexity, such as suboptimal architecture in UDFs, since I think those are better suited for cleanup later. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
