Positron and Julia: a short history of a long-awaited integration effort

This post may have gone off-topic and lacks a clear point of view.

Are you speaking of the Positron project?

The amount of AI/generative expansion here makes it really hard to understand what you’re actually trying to express! It’s ok to use AI for translation, but please try to use your own framings when discussing things here.

Great historical account. As the economist C. Northcote Parkinson, of Parkinson’s Law remarked

Officials multiply subordinates, not rivals.

Unless posit.co moves into the business of supporting Julia in the same way it supports R, I doubt there will be much more accommodation of Julia, either in Positron or in Quarto than there is already.

Wall of slop.

What is the point of this post?

I found it very interesting

For what it may be worth, in just a few lines, my interpretation of the OP is that its author was once excited by the promising possibility of having Julia as well integrated in Positron as R and Python, but now is disappointed by how the efforts in that direction have progressed - with a nod to the Flexible Julia project for JetBrains as a counterexample that they might consider a “happier” story.

Although I’m not a native English speaker, I found the message fairly understandable (assuming that my interpretation given above is correct, my apologies to the author if that’s not the case). So I am genuinely curious about what kind of details in the original message indicate that AI has been overused in order to create it, as a couple of people have suggested. I suppose that it is not the length of the message itself, because that could also be the product of human effort. Maybe my lack of fluency in English made me overlook some hints in how the language is used?

I originally didn’t want to give a clear viewpoint; I just wanted to record this story. But I feel that people don’t understand this viewpoint. So alright, I’ll submit a flag to the moderators, please delete this.

I found the post very interesting and informative. I’m sad it has been removed for no obvious reason.

Wait… they liked my project (Flexible Julia) - AI or not, that is awesome :slight_smile: :clap:

A shame I can’t read the post now. Always enjoy getting feedback

Although the author edited the post and deleted the original content, you should be still able to read it if you look into the edit history, isn’t it like that? (Click on the pencil icon at the top-right corner of the first post)

Thanks for that.

I thought it raised an interesting question: why are some tools paid and is it better or worse for the community. I did not understand his own take on that, however.

My own take: open source is awesome. I am a huge proponent, and this community is living proof of how far it can go. I have contributed to a few projects over the years, most notably Apache OFBiz, and I would not be the developer I am without the ones I learned from.

But open source has a built-in tension. Direction can be murky, because you are usually building on the backs of a few willing people and bound to their goodwill. Projects still tend to need strong leadership, and that leadership faces a hard choice. Accept every contribution and slowly lose a clear direction, or hold the line with a handful of core committers who only have so many hours in the day. You cannot thank those people enough. But unless a larger company is backing the project, the odds are it stalls out eventually.

The reason I moved away from that model for my own work is commitment. I have bills to pay and limited time, and I have bootstrapped my companies my whole life. Not every project has to turn a profit. Some of it is just passion, and Flexible Julia is one of those that does not come close to covering its own keep. But if I want to commit to something for the long haul, I have not found a way to do that without asking for some form of compensation, even if it only covers part of the bill. I do get criticism for that from some of the hardcore open-source contributors though and I guess that’s fair, but life’s about choices and I’d rather do something right or not at all and getting paid even a little bit helps me keep on track.

That said, I still contribute to open source whenever I can, especially where the maintenance stays low. And I try to keep my own tools within reach of the people who need them most, which is why most downloads are free for students and open-source contributors, and discounted for startups.

None of that changes the simple fact: open source is the backbone of countless communities and technologies, and we should thank every single person who keeps it standing.


I also do not understand why the tooling here gets so much criticism. I think they are rather good, and everyone has options to choose from (unlike with systems like ABAP or Delphi or Swift). But the feedback itself is good and necessary, of course…

Excellent point of view.

My post takes a neutral stance (if I had to say, there are pros and cons). My main goal is to present the facts for discussion and reference for the community, because I found this “story” itself interesting and informative.

Given that the initial post is now deleted, I’ll bring this to a close. That really wasn’t my intention and I’m sorry I derailed things here, but it was quite unclear how much was your own thought process — and what that thought process itself was! By the content alone, this would be the sort of post that we would flag and moderate if it came from a brand new user because it’s likely AI-driven account astroturfing. You’re someone I know, though, so I went with a softer reminder of our guidance instead… but that ended up just derailing this further. I should’ve tried to more clearly steer this towards a discussion instead.

The reason we have that AI guidance is for two reasons: we have a longstanding “be consise” community norm where we even suggest using blogs for posts like these, and this is a discussion board for humans to talk to each other… and it’s on the second point that I myself failed!