blog/2024/positron/

[giscus[bot]]

blog/2024/positron/

In May I wrote about my favorite IDE’s.
As a result of that post, I was asked to join Posits private beta-testers for their new IDE, Positron.
Now Positron is out of private beta, so let’s see if it can replace my current favorite IDE’s.

https://drmowinckels.io/blog/2024/positron/

[AlexWeinreb]

Great article, I particularly like the fact you did a "when to use"/"when not to use". Personally I'm still holding out a bit, but this makes me want to give it a try.

When I saw your "the IDEs I use" article just before the Positron announcement, I thought it was an unfortunate timing; turns out it worked out great!

A small note: outside of the Conclusion, external links seem to be broken, as they are relative to your blog. Easy enough to correct, but thought you might want to know.

[drmowinckels]

I think holding out makes sense. It's still pretty early and changing the way one works can take a toll on productivity.

Thanks for letting me know about the links! I did a stupid last minute re-arrange and search replaced a little too vigorously. Fixed them now!

[fkohrt]

Just a small note that Positron is not open source, which unfortunately is a deal breaker for me.

[fkohrt]

Open source is an established term defined by the OSI and the Elastic License 2.0 is not approved by them (and most likely conflicts their definition).

As a consequence:

• If Posit discontinues Positron, nobody will take over
• Altough Positron forked from VSCode, others cannot merge improvements back to VSCode
• Code published under a copyleft license (GPL, among others) cannot become part of Positron
• Positron will not be part of "reproducible pipelines" (e.g., Binder)
• Positron will not be part of any major Linux distribution
• Positron will not be provided by higher education institutions to students as a service

Posit makes the source code available in order to benefit from third party contributions.

[AlexWeinreb]

I see, thank you for the explanation!

[drmowinckels]

It's good you have such dedication, @fkohrt . I'm not working on a Linux, and don't have such strict rules for my own part. But I respect that you do.

I'd like to contest your like point about Positron not being provided to higher education students as a service. In my experience working at a university and being academic, the VAST majority of university provided services are not FOSS. Its mostly very Microsoft and enterprise-centric. At least this is my experience in the academic sector of the Global North.

[fkohrt]

Yeah, I can see why you say that. In Germany it feels like publicly funded institutions, like universities, are increasingly developing and providing FOSS services, but by no means do I have a systematic overview.

Also, I'm not aware of any proprietary alternative to RStudio (or JupyterLab) that universities give their students access to. Meaning: Positron does not compete with Microsoft 365, but with RStudio and JupyterLab, which previously had no contenders (?)

[drmowinckels]

oh yeah! I was just saying the premise as a whole didn't fit. But I agree, given there are free open-source IDEs, its very unlikely Positron will be provided as a service (unless it can really bring something new and competitive, which at the moment it does not).

[rdboyes]

My only current pain point is the restriction in character limit for sending R code to the REPL. I don't hit it that often and it is fixable usually, but it's very annoying when it happens and it isn't THAT high - you can easily hit it with a more complicated ggplot or a long pipe chain

[drmowinckels]

ah yes! I hit this problem for the first time last week. Indeed it's annoying. I did refractor my code to smaller snippets, and it was all-in-all a better experience (both code legibility and this issue), but it for sure is something I'm looking forward to them fixing.