00:14 <kintel> teepee oh, sorry, I didn't realize that had made it. We should author an example :)

04:27 <pca006132> peeps[work]: manifold is not calling StackCheck, so it should be fine

04:28 <pca006132> but to be honest I am not sure if StackCheck is using defined behavior in C++, iirc C++ disallows comparing pointers that are not in the same allocation

04:30 <peepsalot> yeah that was basically my point. but i think StackCheck would give incorrect results if ever called from multiple threads

04:31 <pca006132> yes

04:36 <peepsalot> i know openscad depends on manifold, not the other way around, not sure why i worded it so weirdly. i guess I meant as long as multithreading is *entirely contained* within manifold then we are fine

05:41 <pca006132> no, I think minkowski (for manifold backend, but implemented in openscad) can use multithreading

05:41 <pca006132> it should be fine cuz that code does not invoke StackCheck

05:42 <pca006132> iirc it is only used in the interpreter part

05:42 <pca006132> not in the backend glue code

07:38 <J24k42> anybody heard about bambu lab testing openSCAD?  just was contacted regarding an "develop openSCAD as an experimental project" on  https://makerworld.com/en/makerlab.

07:40 <J24k42> teepee?

09:09 <teepee_> J24k42: no, sounds interesting, if they are asking for contact, the info@openscad.org mail address works and reaches marius and me

09:10 <J24k42> great i ll tell them (if they respond again - they were asking to use my model for advertising or something .. told them it is public domain)

09:12 <teepee> thingiverse did express some interest regarding the WASM stuff too a while ago but nothing happened so far

09:13 <teepee> but then, they are probably still busy with taking over all the stuff, so it's not too surprising

09:14 <J24k42> i told them (bambu) that wasm was a great opportunity to do so  however that link i got requires a login and that failed thrice to create

09:16 <teepee> oops, not so good, but yeah logins are getting hugely complex nowadays

09:18 <teepee> in some other good news we are at least back to *some* working builds https://build.opensuse.org/package/show/home:t-paul/OpenSCAD

09:19 <teepee> not sure what to do about OpenSUSE and Fedora, I don't even know if they are used by more than 2 people or at all

09:23 <J24k42> seems that bambu makerlab is something like a customizer

11:05 <J24k42> teepee hope you will get an email from Candice (collab management - marketing team at Bambu lab) soon.

11:11 <J24k42> or their  OpenSCAD project's chief developer

11:12 <J24k42> but this really sounds like the wanting to include openSCAD into their universe

13:57 <InPhase> teepee, J24k42: Looks like the Bambu Labs CEO is an engineering mentality type. Might be worth getting involved with them if they're interested, as vested interest from a company run by someone who actually "gets it" could be valuable in the long run. https://www.reddit.com/r/3Dprinting/comments/zmneu9/i_am_dr_ye_tao_ceo_of_bambu_lab_ask_me_anything/

13:58 <J24k42> i guess that links to the interview with cnc kitchen .. worth to watch

14:58 <J24k42> InPhase https://www.youtube.com/watch?v=7pFtbybLlk0

14:58 <othx> J24k42 linked to YouTube video "Bambu Lab's NEW A1 Mini and why they didn't release an X1 XL!" => 1 IRC mentions

14:59 <J24k42> or in text https://3dprinteracademy.com/blogs/news-1/exclusive-interview-with-bambu-lab-ceo

15:08 <a6-webm> (obligatory I'm thinking about GSOC) what's the state of making an official LSP? I was thinking "why doesn't OpenSCAD use an LSP under the hood?" and then found out what Scintilla is 😔, so has there been much thought around here abt that?

15:17 <a6-webm> (will be reading logs, so dw about me leaving)

15:20 <pca006132> I think no active dev is thinking about that for now? there are many important issues that require fixing...

15:21 <pca006132> btw, curious what is the typical number of applications openscad get for GSOC?

15:28 <Scopeuk> I believe someone did prototype a language server for openscad a while back but I don't think it went very far

15:29 <Scopeuk> https://github.com/openscad/openscad/issues/3648 and it's children

15:44 <InPhase> pca006132: Binomial distribution with values ranging from 0 through twice the number of slots available.

15:54 <a6-webm> yeah doesn't seem like there's an easy way to make an LSP the standard, but my other project idea had something to do with showing line numbers and file names if you hover your mouse over geometry

15:55 <a6-webm> which in my head doesn't sound too hard, just a setting that means each vertex gets an extra field that says which file made it

16:16 <Scopeuk> I think there is a prototype for that as well somewhere

16:17 <Scopeuk> things get fuzzy with boolean operations

16:19 <Scopeuk> hmm it's in main line, right click on part of the preview, select the component to jump to

16:30 <a6-webm> Scopeuk: loooooool, my ideas are that good they're already main

17:18 <pca006132> InPhase: manifold only has one slot this year, and some how there are more than 5 people asking us now

17:18 <pca006132> thinking about how to turn some of them down

17:27 <InPhase> "After carefully filtering your applications through ChatGPT, we have chosen the applicant that our supreme overlord has selected."

17:30 <pca006132> sounds fair

17:30 <pca006132> and are we qualified as AI project now?

17:31 <InPhase> pca006132: After you've made your choices, there's no real way to turn no into a positive. But it's worth pointing out the unexpectedly high number of applicants, that it was a difficult choice between the project proposals, and express your encouragement that they continue to be part of the community and maybe also try again in another session.

17:32 <pca006132> yeah I'm thinking if they are expecting a high probability for selecting them

17:32 <InPhase> pca006132: Which is pretty boilerplate, really, so a person used to receiving "thank you for your application" letters will glaze their eyes over it. Still worth doing the polite form though. :)

17:33 <InPhase> pca006132: And, you can point out your hands were tied by only having the one slot.

17:33 <pca006132> yeah definitely

17:33 <pca006132> and one of them is already implementing the project now

17:33 <pca006132> so fast

17:35 <InPhase> We had some early starters. I don't remember any of them sticking around after the pay period ended though, except a small bit to try to push a final merge through.

17:36 <InPhase> I think GSoC achieves a valuable goal, but I haven't seen it increasing open source hobbyist contributer counts for us, other than as a byproduct of the work itself maybe nudging in some other interested people.

17:37 <InPhase> This is to be expected though. When you pay people for something, then stop paying them, their brain is in a different place from the people who were never paid. :)

17:37 <pca006132> indeed

17:51 <teepee> did you see actual proposals yet? asking is easy, writing up a proposal with a clear timeline needs some work

18:17 <InPhase> Oh, "asking", yeah. Don't turn anybody away until proposals are in the hand. At that stage, only guide on proposal content.

18:21 <teepee> well, turn people away is not allowed anyway as the rules don't allow to give indication on acceptance before google announces the official lists

18:21 <teepee> but I guess tuning the effort based on what people bring in works just as fine