<M6piz7wk[m]>
which is object in the center and then translated x/2 ?
<M6piz7wk[m]>
so the center is by 0.1 offset for x axis?
<Jack2123>
teepee there is no local config folder on windows ?!
<teepee>
it might not exist, check the library-info window, it should give the full path
<M6piz7wk[m]>
hmm is there a way to get variable storing the OpenSCAD version?
* M6piz7wk[m]
wants to apply the workaronds only for that version
<Jack2123>
teepee yes i have made that folder but it isn't used there is only backup and libraries
<teepee>
create the color-schemes/editor in there
<Jack2123>
what in libraries ? .. ok *kopfschüttel*
<teepee>
no
<teepee>
I have: /home/tp/.config/OpenSCAD/color-schemes/render
<teepee>
with: User Config Path: /home/tp/.config/OpenSCAD
<teepee>
same should work for the editor color schemes (instead of "render")
<Jack2123>
that was what i did .. but it will not show the file
<teepee>
did you change the name inside the file?
<Jack2123>
oh
<Jack2123>
*facepalm*
<teepee>
I think there's even an issue for that, I guess the expectation would be the custom one should overwrite the global file
<Jack2123>
nah still not there
<teepee>
hmpf
* M6piz7wk[m]
found `version()`
<Jack2123>
teepee works fine under the program folder though
<teepee>
strange, it's fine on Linux, just copied the visualstudio one and changed: "name" : "Visual Studio 2",
<teepee>
OpenSCAD needs to be restarted though
<teepee>
(I think)
<teepee>
M6piz7wk[m]: there's also version_num()
<Jack2123>
yes as i said doing the same on the program folder works but not with the file in "documents/oscad"
<teepee>
what path does it show as "User Config Path" ?
<M6piz7wk[m]>
if (version_num() == "2.02101e+7") echo("WHEEEEEEEEEEEE");
<M6piz7wk[m]>
why is this not true?
<M6piz7wk[m]>
it's what `echo((version_num())` returns
<teepee>
it's not a string, it's a number
<M6piz7wk[m]>
oh
<teepee>
== 20210117
<M6piz7wk[m]>
works thanks
<ccox>
The logic for highlighting lines with # is in scintilla, and not all the options are exposed by Qscintilla.
<ccox>
We can make it better by allowing highlighting of everything after the first word, but the first word on the line will remain unhighlighted.
<teepee>
yes, the problem is that there's no way to completely hook into the lexer logic without putting code into the QScintilla build itself
<teepee>
which is not possible for all the builds
<Jack2123>
teepee C:\Users\Jack\AppData\Local/OpenSCAD but that folder is not existing
<ccox>
worse, the logic we need to change is in scintilla, NOT Qscintilla
<teepee>
oh? ok, that's unexpected, but I did not look at all the details how they interact
<teepee>
I thought the QScintilla parsers are pretty much an alternative to the Scintialla logic
<ccox>
I've been drilling down into the source, and it looks like the assumption that is causing this is in the scintilla line processor
<ccox>
QScintilla links in much of scintilla
<teepee>
Jack2123: create that with subfolders color-schemes/editor and put the file in there?
<ccox>
yep, Qscintilla/scintilla/lexers/LexCPP.cpp LexerCPP::Lex() parses the text, and assumes anything that starts with # is a preprocessor instruction
<Jack2123>
ok that worked .. also with templates ( but not with examples)
<teepee>
yeah, examples are not meant to be customizable
<Jack2123>
Ü well -
<teepee>
maybe at some point when we have some integrated library support, but just user configurable seems strange
<ccox>
lexer eats the # and sets state to proprocessor, then looks for specific keywords. Sadly it is very non-customizable in that part of the code.
<teepee>
oh, talking templates, the cc0 seems like a good idea, but maybe it should use the official wording?
* teepee
goes looking if that lexer library still exists...
<Jack2123>
thought this is official wording .. or ccZERO
<teepee>
oh, oops :) This branch is 55 commits ahead, 3987 commits behind master.
<teepee>
it gives that text: " To the extent possible under law, <Your name> has waived all copyright and related or neighboring rights to <Title of work>. "
<M6piz7wk[m]>
JakeSays: does that have anything useful to made OpenSCAD designs?
<M6piz7wk[m]>
*make
ScadUser has quit [Client Quit]
<JakeSays>
M6piz7wk[m]: it has raspberry pi boards and accessories
<teepee>
M6piz7wk[m]: I'm not sure if that's the exact values you are after, but aligning the things a bit differently seems to make it much easier, e.g. https://bpa.st/P66A
<M6piz7wk[m]>
Got this.. i did that originally, but i want to add additional logic and geometry later so it has to be exact with projected tolerance 1mm
<M6piz7wk[m]>
Also the design is not the issue i think that openSCAD renders the shape wrong by 1.0
<M6piz7wk[m]>
bcs in my design the math to me adds up for the 1.0 to not be there
<Jack2123>
M6piz7wk[m]: do you know you can use orthogonal view .. and view from x / y / z to see things clearly ?
<M6piz7wk[m]>
so i think that the OpenSCAD centers everything on 0.1,0,0 (x,y,z) instead of 0,0,0
<M6piz7wk[m]>
Jack2123: i believe so, it looks like it's clearly there
<Jack2123>
and for sure openSCAD is not doing that .. i am sure i would have noticed
<Jack2123>
the code you posted doesn't match the picture .. so i don't know what your problem is and how to replicate
<M6piz7wk[m]>
this is how it looks on my end.. the highlighted is unexpected shape
<dTal>
to clarify - I guarantee OpenSCAD isn't doing that
<M6piz7wk[m]>
from GNU Guix
<M6piz7wk[m]>
oh
<M6piz7wk[m]>
what's doing that?
<Jack2123>
check if the version code you used really works
<dTal>
with overwhelming probability, a mistake in your math
<M6piz7wk[m]>
the shape is 4.2 lenght and it's placed in the center so 4.2/2=2.1 which should result the shape being on the exact edge without the 0.1 geometry ?
<dTal>
can you make a minimal example of openscad doing the wrong thing?
<M6piz7wk[m]>
i try
<dTal>
if it doesn't show the behaviour in the simple case, gradually make it more and more like the broken case
<Jack2123>
M6 you are not using the cube with +.1 but the one without
<teepee>
oh, oops, does it work? I missed the time to check :)
<teepee>
but now we should have seen all errors I guess. unless there's a typo in index.js
<teepee>
having a nicer, maybe even customizable indicator would be cool. I wonder if that could be done via a shader, but that probably would need shader stacking...
<Jack2123>
is 19 also some kind of prefix?
<teepee>
prefix?
<Jack2123>
like the oct with 8 08 .
<Jack2123>
or why should it fail today?
<teepee>
it's computer stuff, this always finds creative ways to fail
<teepee>
like I just got 4 fedora build warnings after adding the cc0 text file :)
<teepee>
which obviously would not have changed anything in the actual build
<Jack2123>
but all 10 checks have passed or?
<teepee>
sure, whatever happened on OBS again, maybe disk full or something like that
<teepee>
I'm sure it has nothing to do with the change itself
<Jack2123>
i meant i didn't see any warning did you see that on git ?
<M6piz7wk[m]>
Where is the code that is defining the symbol of tab indentation?
<M6piz7wk[m]>
The default one is `->` which is pain to work with so i want to submit patch to use `|` and can't find the relevant codeblock x.x
<teepee>
and it looks like there's some issue with all the ppc64le builds this time, seems to be some config issue actually, nothing we can do, the'll fix it soon enough
<teepee>
M6piz7wk[m]: that would be probably code from the QScintilla editor component (or maybe Scintilla which is the base for that)
<Jack2123>
everytime i fetch my fork i get "run failed" emails "no job were run" i think it is dangerous if one get used to ignore messages
<teepee>
yeah, I think that's from the MacOS builds kintel prepared and disabled as we use the CircleCI version
<M6piz7wk[m]>
teepee: so outside of the main repo?
<teepee>
If I find some time I'll try to quiet that, it's not a problem
<Jack2123>
no worry i can ignore that .. i meant in general we have this that we get used to ignore things because we don't understand or know they are not importand.. know a case where someone blocked all warning light in a car with tape because they distracted him o_O
<teepee>
M6piz7wk[m]: they are pretty responsive via mailing list, I just recently discussed an issue and there's a fix now in the newer releases
<teepee>
they might accept a patch making this configurable (at least I think it currently is not changable)
<teepee>
Jack2123: yeah, constant useless warning messages to ignore are a very bad training, eventually there's one that's important
<teepee>
I've seen that at work from time to time. luckily I'm not working in an area where it would be dangerous to people
<InPhase>
Lot of geometry in this year's calendar.
<Jack2123>
who knows .. maybe some one getting very upset if scad is not working proper - Ü
<Jack2123>
inPhase geometry was easy to fill the days with .. and i am not good in making balloon animals
<InPhase>
:)
<InPhase>
Balloon animals would be a fun one. I bet we could make a nice module system for that.
<Jack2123>
the moment i wrote that .. i was thinking .. wait can't we use .... for that
<InPhase>
I figured. :)
<teepee>
haha, topic for next year \o/
<teepee>
24 balloon animals
* M6piz7wk[m]
can't find the codeblock so he submitted proposal
<InPhase>
Jack2123: While falling to bits on legos.
<teepee>
I'm not seeing why SDF would be of much benefit here
<teepee>
it's just positioning differently sized rounded tubes, or Pille() as Jack2123 would say :-)
<Jack2123>
thought curv is a sausage thing
<InPhase>
teepee: A whole straight stretch of balloon can be one equation, while a cube is 3-6 things unioned together. It's all backwards. :)
<Jack2123>
i think unity used the term capsule
<teepee>
yes, SDF can do rounded unions easily, or blend shapes, but that does not seem needed for balloon animals
<InPhase>
s/unioned/intersected/
<teepee>
balloons tend to blend rarely
Jack2123 has quit [Quit: Client closed]
<teepee>
also intersection is something ballons do for very short time only ;-)
<InPhase>
On the whole I'd say OpenSCAD is scaling better than curv for now. But SDF approaches done right can do Inigo Quilez magic that is way out of scope for what we could consider doing.
Jack21 has joined #openscad
<teepee>
true, but also that magic is very cool for visuals, which is something I do not care about for any of my designs so far
<Jack21>
was thinking about the bending of balloons and when they squish were they meet
<InPhase>
teepee: Yeah, I'm probably not going to have a career as an artist at this point. :) But the way he can get elaborate video rate temples with ornate columns and things like that is very impressive.
<InPhase>
teepee: This is indicative of things I could make serious use of.
<InPhase>
Jack21: I made a bent santa hat last year.
<InPhase>
Jack21: Your connection is a bit wonky today. Entry 23 from 2019, grab the santa scad and navigate around to see the bend in the hat. https://openscad.org/advent-calendar-2019/
<Jack2133>
i bluescreened as usally every day cuz some nvida update is not working with ms
<InPhase>
That whole santa design is just cheap alignment hacks for visual effect.
<teepee>
InPhase: it's just a matter of someone having the time to code that stuff, mapping CSG to a shader SDF is possible and has been done already (although based on the Implictcad code)
<Jack2133>
right click still blacks out the window ..
<teepee>
yes, does not happen on Linux, so I have no idea how to debug that
<Jack2133>
the bend in the hat looks more like an attached part
<teepee>
it's probably some OpenGL state getting messed up by the right-click renderer or something like that
<Jack2133>
hmm one bit at a time .. i tell you when it is gone Ü
<InPhase>
Jack2133: Yep. Well, every bit looks like an attached part on that Santa when you look real close. Because they are. ;)
<M6piz7wk[m]>
teepee: Sorry i missunderstood what you meant by it (though that openscad has some kind of config file that can change this to avoid asking QS.. upstream)
<InPhase>
Jack2133: Those whiskers also cut clear through Santa's back molars.
<teepee>
M6piz7wk[m]: no, we can provide option to configure what QScintilla gives us, but at this point this is only enable/enable-after-first-non-white-space/disable
<M6piz7wk[m]>
i see
<teepee>
if QScintilla would implement an option to change the display, we can add that too
<M6piz7wk[m]>
noted
<Jack2133>
well CGI .. i new someone was every time surprised that things happen when he was there .. so everything is smoke an mirror
<teepee>
what I meant was they seem open and responsive to suggestions and feedback, at least I can report a nice experience talking to (I assume) one of the developers
<InPhase>
I wouldn't change a thing about that Santa. Because if I change anything about that Santa everything else will be out of alignment. ;) It's probably my least parametric publicly released design.
<Jack2133>
noooo santa don't wave your .aaaahhh
<Jack2133>
but he is looking cute
<teepee>
omg! we don't need santa out of alignment, there's too much out of alignment already :)
<M6piz7wk[m]>
Can 3D Printers print STL files or do i have to slice that first?
* M6piz7wk[m]
got Ender 3 and is trying to make it to print his design
<M6piz7wk[m]>
.. using openscad
<Jack2133>
i think i had a ragdoll anywhere with parameters for every joint to move
<M6piz7wk[m]>
or like what is the non-octoprint way on openscad to sent the object for manufacturing?
<Jack2133>
if you use octoprint .. the slicing can be done there
<teepee>
normal workflow is slice first, there might be printers with more powerful systems in there that have a slicer running
<Jack2133>
but anyone has to make a gcode - what makes the printer tick
<teepee>
yeah, or built-in octoprint maybe :)
<M6piz7wk[m]>
Jack2133: i don't have octoprint set up.. i built the printer yesterday x.x
<teepee>
it's awesome to have, but should work fine without
<Jack2133>
then you have to slice it yourself (or online)
<M6piz7wk[m]>
oke O.o
<teepee>
and if the slicer allows 3MF, it's a better option than STL
<M6piz7wk[m]>
is there any slicer that works with openscad? I tried to import it in FreeCAD, but it has distro-related issue on my system
<Jack2133>
.oO( soooo much better )
<Jack2133>
you need to export rendered geometry
<Jack2133>
no slicer (i know) can interpret scad code
<teepee>
directly no, but Cura notifies on file save, so I just open the file once, and just click refresh in Cura after exporting a new version in OpenSCAD
<M6piz7wk[m]>
Jack2133: so like the STL file?
<Jack2133>
or 3mf
<teepee>
Cura does have an OpenSCAD plugin, but I've never used it
<Jack2133>
oh what does it do
<teepee>
but it certainly just runs openscad to convert the file
<M6piz7wk[m]>
is cura even libre? i always used freecad for the manufacturing as i can do stress testing and simulations there x.x
<teepee>
same as FreeCAD does too
<teepee>
yes, the slicer core is AGPL and the UI is GPL I think?
<M6piz7wk[m]>
gut gut
<teepee>
oh, wikipedia says it was changed to all LGPL
<teepee>
interesting
<Jack2133>
allmann is taking over
<Jack2133>
and it seems the plugin works
<M6piz7wk[m]>
cura is not packaged on guix.. does prusa slicer works?
<teepee>
yes, BOSL2, dotSCAD and nopscadlib are huge
<M6piz7wk[m]>
hmm is there something like relation finder in OpenSCAD that allows me to find the relations inbetween constrained elements to avoid making a quadratic function to hardcode it?
<InPhase>
Those dotSCAD dragons are pretty impressive.
<InPhase>
I haven't dug into how he did that, but, clearly impressive.
<InPhase>
We should get that guy to make us some reindeer for next year. :)
<JakeSays>
the nop lib is impressive just in the quantity of things it provides
<teepee>
yes, did you check his thingiverse account, i think all the stuff is also in the examples folder
<othx>
Jack2133 linked to "Cat corner protector by JustinSDK" on thingiverse => 1 IRC mentions
<teepee>
heh, yes, flexible filament, otherwise it's just 2 ears instead of 1 table corner :)
<JakeSays>
Jack2133: LOL those are cool
<Jack2133>
replaces a 90 ° corner with 2 very pointy ears ..
<Jack2133>
means it protects the corner .. not you
<teepee>
now print very fluffy ear protectors
<M6piz7wk[m]>
teepee: are those a thing?
<M6piz7wk[m]>
🌟_🌟
<teepee>
ear protectors for cat corner protectors? probably not. but it would look funny
<teepee>
the suggested flexible material is certainly the better option, I want to try that stuff at some point
<InPhase>
TPU is amazing stuff. It prints like PLA, and at only about 1mm thick it's so durable you basically cannot destroy it without tools.
<Jack2133>
and with heat treatment it gets glossy like acetone ABS
<InPhase>
I used TPU to print up group label place cards for student groups, and after years of students bending those things out of boredom, they remained in perfect condition.
<InPhase>
When I bend one of those 1mm sheets at 90 degrees, it snaps back to 45 degrees in a sudden springing motion, then it slowly returns the rest of the way to flat over a couple seconds. So the flex dynamics are a little different than rubber, and less bouncy. But it might overall be more durable than rubber.
<Jack2133>
you can apply an offset (-1) to the 2D from the projection - and `linear_extrude()` it to use in a difference with the object .. using `children()` makes this more convenient in a module. Especially if needed for all 3 axis
TheAssass1n has quit [Remote host closed the connection]
TheAssassin has joined #openscad
TheAssassin has quit [Ping timeout: 276 seconds]
TheAssassin has joined #openscad
blink is now known as fling
Guest95 has joined #openscad
Guest95 is now known as jmatson
<jmatson>
hey anyone alive in here?
<Jack2133>
sure
<Jack2133>
teepee: is there a way to get echo("x") in a template.. or how do i escape the ""
<jmatson>
so i downloaded a file for Christmas ornaments. i can't get opencad to recognize any font that i install.
<Jack2133>
jmatson have you checked font list?
<jmatson>
yep it's not listed
<jmatson>
it's listed in the font folder.. Works in illustrator
<Jack2133>
if the list string is longer or the 2D array build after patten for row and pattern for columns it should be something like that hitomezashi pattern
arebil has quit [Quit: My keyboard has gone to sleep. ZZZzzz…]
<teepee>
looks nice, I don't know if it's the same
<Jack2133>
think when the matrix is build with an algorithm it should work
<Jack2133>
but it may only be similar as we have this 45° rotation which offsets every %2 line
<M6piz7wk[m]>
<Jack2133> "https://i.redd.it/pa3d25g1fi681..."; <- cool o.o i was wondering about similar design to make the 3d printing faster by avoiding lifting the extruder
<Jack2133>
Alicia thanks thought its probably something i posted - Ü
<Alicia>
yeah, looked like a quote. that's how I found it
<Jack2133>
M6 the normal bottom/top pattern are meandered (zigzag) so it can be done in one go
<teepee>
yes, that's how the matrix bridge quotes replies
<M6piz7wk[m]>
<Jack2133> "M6 the normal bottom/top..." <- ye cool O.o i would probably want to add some angles to it though to better utilize the support material and less complicated shape for the CNC to interpret
<M6piz7wk[m]>
Can i use the projection to cut the Z axis ?
<M6piz7wk[m]>
seems like it only works for X which complicates things for me
<Jack2133>
it always projects along Z
<Jack2133>
onto the XY plane .. you can decide if it cut at 0 or make a full projection
<M6piz7wk[m]>
ehh O.o
<Jack2133>
like you take a slice at Z=0
<Jack2133>
z is the blue axis
<M6piz7wk[m]>
<Jack2133> "you can apply an offset (-1..." <- i am trying to figure out this tbh to hollow out the shape
<M6piz7wk[m]>
so i have this and trying to make a projection from it which projects only on the X/Y and i need to project on Z/X
<M6piz7wk[m]>
(the orientation here is important)
<M6piz7wk[m]>
.. for material optimization bcs it doesn't need to be solid
<Scopeuk>
M6piz7wk[m] projection from it as in to extend it in that direction, or projection, to take the 2d form of it?
<Jack2133>
rotate 90 so Z going through the parallel sides .. projection and offset ⇒ offset(-1) projection() rotate([90,0]) Object();
<M6piz7wk[m]>
that seems ugly x.x
<M6piz7wk[m]>
hmm i can just make a function that takes the shape and then makes another that is smaller to determine the cutout.. that seems like more workable O.o
<InPhase>
Jack2133: Not sure exactly where that appeared, but somewhere... WARNING: Unable to convert translate([0, 0, true]) parameter to a vec3 or vec2 of numbers
<InPhase>
true is leaking into h somehow
<Jack2133>
it was working here ..
<InPhase>
Hmm. Works in the master branch.
<InPhase>
Maybe that comma change?
<InPhase>
My eyes can't spot it...
<Jack2133>
line 20 has the cut not explicitly
<Jack2133>
sorry 19
<InPhase>
Oh right, there was a change in how named variables in the middle of the sequence were assigned.
<InPhase>
peeps[win] fixed that. It was doing some absurd stuff.
<InPhase>
This looks like some of those test cases.
<InPhase>
So your code isn't "wrong", just slamming into a buggy feature in all the previous releases for explicit followed by implicit.
<InPhase>
I guess assign the cut for now, or don't assign any, until we get a fresh release out there. :)
<Jack2133>
minimal effort Ü
<Jack2133>
InPhase if i explicitly assign a value to a variable this should have priority .. and shouldn't be overwritten by any implicit value
<InPhase>
Yes, that was the conclusion of our discussions on the topic, which is why the master branch handles this sensibly now.
<InPhase>
I think it had just gone unnoticed for a while.
<InPhase>
We tried discussing smarter choices than this, but the best we could come up with is "stop it, implementer..."
<InPhase>
The warning there is for the output line following.
<InPhase>
function is f(a="a", b="b", c="c") [a,b,c];
<InPhase>
So that implicit 2 fills into a, the 1 overrides it, and then a warning is thrown because it's not clear.
<JakeSays>
i'm not a fan of wokibooks for documentation
<JakeSays>
that was my random thought for the day
<Jack2133>
InPhase hmm i think it is clear , a is given so the value is moved to the next free slot
<Jack2133>
so first all assigned are occupied and then the remaining as they are sequenced
<Jack2133>
a bonus would be if datatypes are differently treated so a boolean can only assigned to a the right slot .. however every mix up is not a good idea and will lead to confusion
<Jack2133>
btw is there a way to break errors and warnings in the output"console" .. it is annoying if you have 10000 points and due to a typo you have to wait until all warnings are shown
<InPhase>
Jack2133: So actually I misspoke earlier, remembering an intermediate option we had discussed. The one which was ultimately chosen was to assign the values left to right, and complain when a value gets assigned twice.
<InPhase>
Jack2133: This works with almost all sensible usages, and will warn people when they most likely made a mistake.
<Jack2133>
.oO(i remember when i switched from 2015 .. i saw a lot of warnings because i got away with a lot bad code Ü )
<InPhase>
Jack2133: Dif([1,0],true,h=4) This for example, throws a warning.
<Jack2133>
InPhase fair enough - but not because it is not clear - more as someone doing this may not be aware what he is doing
<InPhase>
Yeah. One could construct an example where that middle value was actually supposed to be the 2nd parameter, but it got to the 3rd parameter because of a value later in a long list, making debugging confusing.
<Jack2133>
InPhase what is about Stopping the output?
<InPhase>
lol. Can of worms.
<Jack2133>
flummm
<InPhase>
OpenSCAD needs a proper threading architecture. I know how to do it, and do this all the time for work, but haven't had the time and energy to dive into it.
<Jack2133>
so ctrl s ⇒ alt F4
<InPhase>
With a proper threading architecture the gui will never lock up, and most operations will be abortable in some manner.
<JakeSays>
and then the preview can still be active while rendering is taking place!
<InPhase>
Sure. There's no reason at all this should be prevented.
<Jack2133>
sounds like a wonderfull dream vision parallel universe
<M6piz7wk[m]>
What's the abstract of openscad? Is it just an interpreter making calls to API that does the rendering?
<M6piz7wk[m]>
or like could i define a lisp library that makes the call to use lisp to make these designs?
<InPhase>
Jack2133: It's not really that hard. All my work does this on equally complicated programs.
<InPhase>
s/my work/my work code/
<JakeSays>
M6piz7wk[m]: iirc something like that exists
<M6piz7wk[m]>
like lisp for OpenSCAD ?
<JakeSays>
lisp based csg
<InPhase>
M6piz7wk[m]: There are multiple libraries doing computation that are glued together for OpenSCAD.
<InPhase>
M6piz7wk[m]: Plus editor libraries and gui libraries. So there's no single OpenSCAD engine really, although that is on the "well wouldn't that be nice" list.
<JakeSays>
lol
<Jack2133>
InPhase - paid open source programmer is one thing - but imagine you wouldn't need to earn money UBI
<InPhase>
M6piz7wk[m]: Broadly speaking geometric computation is done by CGAL, OpenCSG, Clipper, and custom routines. The code computation parts for evaluating functions and module recursions and such are all done in-house.
<JakeSays>
i wouldn't mind working on an open source code base for a living, but not an open project.
<InPhase>
Jack2133: Yeah. Although I think it's conceivable paid positions for OpenSCAD could arise at some point. Thingiverse would have been the first logical source for that, but Makerbot abandoned investment in even its own site, so that was not going to manifest.
<M6piz7wk[m]>
i was searching and my searX-fu is not on point today :c
<JakeSays>
InPhase: i guess that explains why the configurator doesn't work
<InPhase>
Utilization for OpenSCAD is reasonably high. It's just about some money source of a company or collective of companies ending up with a vested profit interest in it, which will motivate some funding toward development.
<JakeSays>
InPhase: well, if log4j can't get any love..
<InPhase>
I never even heard of log4j before it showed up in the news the other week. :)
<JakeSays>
lol
<InPhase>
And then suddenly it's "bugs will be all over the place!" And when you look to see WHERE these bugs are supposed to be showing up, nobody can say.
* InPhase
shrugs.
<Jack2133>
problem is that goverments should ensured that - but they are 20years behind IT development
<JakeSays>
oh yeah. big time panic
<JakeSays>
Jack2133: governments?
<Jack2133>
yes they need to ensure that dangerous situations can not arise ..
<JakeSays>
in opensource?
<Jack2133>
they should monitor potential risks so if opensource software is used in many critical areas ( like a big bank is system critical) they should step in and ensure that maintainance and monitoring is in place
GNUmoon has joined #openscad
<JakeSays>
yean. no. not at all.
<JakeSays>
the LAST thing i want is governments getting involved in open source
<Jack2133>
well not the US governments Ü
<JakeSays>
any government
<M6piz7wk[m]>
JakeSays: already happening in the EU and FSFE is recognized as authority on it
<JakeSays>
yeah i figured it'd start with the EU
<M6piz7wk[m]>
great progress done on replacing government's use of proprietary software with free software so far ^-^
<Jack2133>
i think it is fine if the state using tax money to give people proper free software
<JakeSays>
M6piz7wk[m]: that's not the same thing though
<JakeSays>
Jack2133: as long as the state is just a consumer, like any other joe.
<M6piz7wk[m]>
states are participating on development e.g. German government investing in conduit.rs to use the matrix protocol for their internal communication and medical field
<InPhase>
JakeSays: Government funding for open source software is actually extremely reasonable. It's governments that are the right source of funds for things that broadly benefit society as a collective rather than specific commercial interests.
<Jack2133>
sure they just giving resources not taking over decisions
<M6piz7wk[m]>
EU itself giving major funding to FLOSS e.g. protonmail.com
<JakeSays>
InPhase: that's not what Jack2133's suggesting. he wants governments to be the opensource overseers
<InPhase>
JakeSays: The logic for government funding of open source infrastructure is identical to the logic of government funding for scientific research.
<Jack2133>
but if EU putting 10 fulltime programmer on openSCAD .. they sure will influence the software
<JakeSays>
but that's not what you suggested
<InPhase>
JakeSays: I think Jack2133 is more advocating governments looking at it as a responsibility that governments have to make sure security critical things are not left buggy.
<M6piz7wk[m]>
Jack2133: well did you asked them?
<M6piz7wk[m]>
germans would probably give you a region with a castle for this kind of thing
<JakeSays>
InPhase: the only way for a govt to ensure that is to regulate it.
<Jack2133>
JakeSays not what i said .. monitoring doesn't mean that .
<InPhase>
JakeSays: And I agree. Heartbleed should have never happened... Government funding should have existed for openssl in a properly functioning world.
<JakeSays>
Jack2133: that's exactly what monitoring means, at least in the real world.
<JakeSays>
InPhase: govt funding is not what this is about.
<InPhase>
JakeSays: That doesn't require regulating. It requires cash grants through some agencies.
<JakeSays>
M6piz7wk[m]: hmm. i think libfive is the lisp thing
<InPhase>
OpenSCAD is not first in the lists of projects needing such an influx, but it would also be a reasonable recipient. We have entire GDP producing markets that are being propped up by open source 3D printing software.
<InPhase>
And these software efforts are not adequately supported by anything remotely on par with the amount of value being generated.
<JakeSays>
that's definitely true
<JakeSays>
but as long as people are willing to work for free nothing will change
<M6piz7wk[m]>
JakeSays: what's libfive other then some weird python library that generates roots of 5
<InPhase>
teepee: $20/month is pocket change. I sure don't mind expanding my DO account to include a build server at that scale. $hundreds / month is a different scenario. :)
<JakeSays>
Jack21: interesting
<Jack21>
JakeSays btw i can monitor the stock market - without having any influence
<teepee>
I guess it needs some closer look to get some more reasonable estimate for opencollective, I guess a safe range would be somewhere between $20 and $50 monthly
<JakeSays>
Jack21: you're not a government
<Jack21>
the government also can monitor the stock market without influence