I've recently thought about how the open source is kind of doomed nowadays - with the "create new stuff only" approach - do you agree? 👀 Full thoughts: https://www.linkedin.com/posts/marrrcin_lets-talk-about-something-that-too-often-acti[…]084613804032-79BB?utm_source=share&utm_medium=member_desktop

spot on. left a comment there!

I think open source projects are doomed because many packages are being created on a terrible code base. I'd be happy to fix a bug in numpy/pandas/sklearn/scipy because they're great libs with great and well optimized code base. I can't tell you the number of so-called "packages" that are built on copy paste code nowadays -- and yes, llama index also being one of them. It's like they're trying to do too many things. Most of the code pieces I see don't work because env can't be installed. The cutting edge open source is doomed alright but maybe in 5 years there could be something..

totally agree, but I'd also like to add the pain of getting familiar with procedures and wait time to get your change approved/make it live with the change. Recently I've done many minor improvements to kedro-mlflow for example as the need arose on local fork but I'm hesitant to publish this as I'm not sure if I have the time to make it get approved and merged, especially as I need to change things on the fly - I promise to myself that after I'm done with urgent things I'll get back to make a PR with the changes but who knows if I won't forget or be pressured to prioritize other things first...