00:03 <pca006132> kintel: the VBO issue is now gone

00:04 <gbruno> [github] pca006132 closed issue #4858 ( VBO renderer uses excessive RAM on Windows) https://github.com/openscad/openscad/issues/4858

11:41 <pca006132> glad that irc is much more peaceful than the mailing list...

11:43 <juri_> we try. :)

15:25 <pca006132> kintel I probably don't have enough energy to work on the polyset related things, was thinking about manifold triangulator and such...

15:26 <gbruno> [github] pca006132 edited issue #4851 (PolySet needs cleanup) https://github.com/openscad/openscad/issues/4851

15:26 <pca006132> the extrude algorithm should be very similar to the one in manifold, but may need some tweak to use the shortest diagonal

16:47 <Guest46> Does anyone have code for a sieve or many slotted/holed tray or box or something like that?

17:07 <teepee> pca006132: did you know you can now store designs generated with manifold in an application called manyfold? :D

18:53 <InPhase> And we need to start listing all of these on a site called moanyfold.

18:53 <InPhase> We'll fund it all via a non-profit called moneyfold.

19:18 <teepee> ha! I like moneyfold :)

19:53 <J24k59> main money manifold

21:42 <gbruno> [github] dwt opened issue #4952 (Port to iPadOS / iOS) https://github.com/openscad/openscad/issues/4952

21:43 <Scopeuk> poor optomistic soul, I guess atleast we have webasm these days

21:46 <teepee> well, license wise even impossible

21:46 <teepee> we'll see how the latest enforcement in EU goes, not going to hold my breath though

21:50 <Scopeuk> Mozilla didn't seam impressed

21:50 <J24k59> I did not see this coming https://www.reddit.com/r/openscad/comments/1adcw41/i_am_completely_blind_and_this_is_my_first/

21:52 <teepee> nice, it would be interesting to get some feedback on how to make it easier to use

21:53 <Scopeuk> I know screen readers can be funny about how they interact with widgets

21:53 <Scopeuk> I've not tried running one with openscad would be interesting to hear how it sounds

21:53 <teepee> there's probably a number of things we can't easily change, like the editor component itself or Qt, but the might be still some improvements possible

21:54 <teepee> I tried on Linux some time ago, Qt4 was a no-go, Qt5 is slightly better but still not good

21:55 <teepee> but it would be helpful to get that info from someone who knows how to use those tools

21:56 <Scopeuk> yeh, I can comfortably install and run a tool, but I have no idea how someone who needs it and daily drives it will use it

21:56 <Scopeuk> "Thank you very much! :)

21:56 <Scopeuk> When I need to check the output, I take screen shots of every angle, then, I upload the screen shots to ChatGpt, and input in the prompt that I am completely blind, and that I need a detailed description of the model from every angle :)"

21:57 <Scopeuk> I'm not sure of how to better that but it feels like there must be a way

21:58 <teepee> at least: automate it

21:59 <teepee> not sure if local instances are good enough at describing things

22:30 <J24k59> »I am also doing models in openScad, which is 95 % accessible.«

22:59 <InPhase> J24k59: Excellent. I had thought before about the fact that OpenSCAD offers the input side of 3D modeling with vision impairment, but could not figure out how to handle the feedback part. I saw in the comments that this person is using ChatGPT and having it explain the screenshots.

23:00 <InPhase> That's a superb integration of these systems.

23:21 <InPhase> I replied to Mrblindguardian asking for feedback.

23:24 <InPhase> It looks like the overlap rule was missed on one component, but it renders fine, and I suppose the visual artifacts in Preview mode weren't bothering Mrblindguardian.