00:01 <InPhase> joseph_: From talking to you I have pretty good confidence, but I like to see some documented clarity that people have a sense for what they're getting themselves into when seemingly simple development steps on larger projects explode a bit in interaction complexity.

00:01 <InPhase> That's the sort of thing that tends to catch a lot of people off guard. :)

00:35 <joseph_> InPhase: Thanks, I will add sections that address those areas. The likelihood of expansion during development makes me inclined to not add anything from #56 to the scope, and tell Google it's a 175 hour project (but quite likely I'll spend more time than that). Do you agree?

00:36 <InPhase> You'd only get paid for the 175 hours, and the final progress submission is due at the end of it.

00:36 <joseph_> InPhase: To reiterate, my schedule could support up to 350 hours if you think that's what is necessary for the project to be successful. Are you saying that if I choose 350 hours, I get a later deadline?

00:37 <InPhase> I'm not sure if that's true actually.

00:37 <InPhase> Last year it was more compacted of a timeline and only the smaller time periods were used.

00:37 <InPhase> This year I don't know how it's arranged with the two options.

00:38 <InPhase> I guess I only see the one timeline specified.

00:40 <joseph_> Google says "The standard timeline accounts for 12 week coding projects. Mentors and GSoC Contributors can agree to extend the coding period up to 22 weeks. " This appears to be independent of the project size that is chosen.

00:42 <InPhase> So in some ways this functions like a standard job. There's an application, there's a work expection, you get evaluations, google sends you money. But there's the other aspect in that it's also an open source project that anyone is free to volunteer on. We've had people in the past do extra programming on the project topics after the end of the GSoC because they wanted it working nicer, completed,

00:42 <InPhase> and merged. This differs from a normal job. :) It's considered a good side-effect of GSoC when contributers get hooked on the project and like working on it afterward. (But I think it's inappropriate to ask for such at application time. It's worth just seeing what people come to enjoy.)

00:47 <joseph_> InPhase: Thanks. Fortunately I would not mind if I end up working more hours than I'm being paid. After all, this is FOSS and I use OpenSCAD for my own projects. My goals with the scope are 1) I want you to have confidence in my proposal and 2) I don't want to overpromise

01:10 <InPhase> Sounds good. :)

02:58 <Joel> It's been a bit since I've worked on my openscad drawing project... One thing I've been curious about, is can I some how transform something called in a module? So for example, in the module, I want it to call itself, and orient on the X axis, but when the module is included in another openscad file, and the method is called, I want to use transform to orient it on another axis

02:58 <Joel> I presume I could pass another argument to orient it another way... was wondering if there was something more elegant

02:59 <Joel> https://github.com/OpenSourceHouses/TheTiny/blob/main/house.scad

03:54 <retrosenator> call itself?

03:55 <retrosenator> you are using recusion?

05:29 <J22> Joel you can use the parent module calling it to determine the action - but only works if the parent is not a build in module

05:40 <J22> Joel https://bpa.st/HPIA

09:45 <Scopeuk> peepsalot does running a reformat over your branch ahead of the merge make git happier? My understanding is the reformat was automated

10:16 <Scopeuk> Joel it is possible for a model to get access to it's children and manipulate them but I'm not sure that's what you want, passing an orientation parameter to the module is probably best

10:21 <Guest9022> I've got version 2021.1.0 on Gentoo. Dump files for animate seems to work, and unchecks the box when it thinks everything is stored. However, the doc says the files are stored where the thisproject.scad file is. I don't see them there and can't find them elsewhere. Any ideas?

10:23 <J22> have you opened or saved any other project after

10:25 <Guest9022> yes, i've opened example projects. But there should have been a ton of files.

10:26 <J22> maybe have a look in the example folder  ( or search for frame00000.png)

10:28 <Guest9022> i just did an updatedb and locate frame00000.png... no dice.

10:30 <J22> you can try the latest scad snapshot version ..  normally those files are on the latest open/saved project location

10:32 <Guest9022> rofl... i thot i'd reloaded openscad before... apparently not.

10:32 <Guest9022> I just restarted and it worked. :P thx for the help tho! :)

10:34 <J22> glad you solved it

11:39 <Joel> J22 I'll have to digest that a bit

11:39 <Joel> thank you!

11:41 <J22> Joel if you tell us what you try to accomplish we might help you better

11:42 <Joel> J22 Yeah, I need to build a small demo, share it, and ask if there's a way to simplify :)

13:15 <Guest63> Hello

13:16 <Guest63> I am trying to access the website opencad but it not letting me its saying 403 error code

13:35 <J22> http://openscad.org/

13:39 <J22> i am trying to reach Guest63 but  i get error 1168

13:45 <teepee> yeah, lost the s in all that haste it seems

14:00 <J22> https://www.opencad.io/

14:01 <J22> computer aided dispatch    Ü

14:01 <teepee> that's an awesome idea to overload names :)

14:02 <J22> https://www.abbreviations.com/CAD

14:03 <J22> always check that your counterpart  in a discussion is from the same field

14:03 <J22> i like "Complicated And Difficult"

14:04 <J22> Come And Die

14:50 <peepsalot> everyone knows CAD stands for Ctrl-Alt-Delete

14:52 <peepsalot> Scopeuk: that's a good idea that I suspect will help a lot, dunno why it didn't occur to me. I will try that on my next attempt

16:24 <joseph_> InPhase teepee peepsalot: I submitted the first draft of my proposal to the Google website. Please let me know of further suggestions. I also made a public repository with the key parts of my GSoC proposal. It's here: https://github.com/jbinvnt/gsoc-2022-proposal-public

17:02 <InPhase> joseph_: Excellent. Now that that's all completed, I have a totally unrelated question for you. From your past exploration of Ethereum contracts, do they actually have enough flexibility and reliability to have widespread and flexible applications, or do you think they are fairly limited to niche applications set by the available inputs to the contracts, the expenses of adding more, and the challenges

17:02 <InPhase> of establishing trust for inputs.

17:04 <InPhase> joseph_: I've been trying to sort out the real merit of that particular item, but from a critical-thinking consideration of technical merits, rather than the common available perspective of true-believer investors.

17:47 <gbruno> [github] MichaelPFrey edited issue #3854 (Freezes if main-window position is negative with size being bigger than the screen.) https://github.com/openscad/openscad/issues/3854

17:48 <InPhase> That again?

17:52 <teepee> *edited*

17:54 <InPhase> I missed the edited, but also the 2021 bug is a flashback to a 2020 bug.

17:54 <InPhase> I posted the link back.

17:54 <teepee> yeah, might indeed be a duplicate

17:55 <teepee> and I still think that's a Qt bug, not that it helps

17:55 <teepee> well, or windows+qt

17:55 <InPhase> We needed more data initially, so probably a good thing. It was sounding in 2020 like probably a Windows only problem that arises from Qt only under dual monitor setups.

17:56 <InPhase> And that's definitely holding true for everyone reporting it.

17:57 <InPhase> As I don't have Windows or two monitors going, that's what capped my exploration of it 2 years ago.

18:15 <teepee> those platform issues are quite annoying

18:15 <teepee> same with that build change that drops Win7 without any change on our side

18:16 <teepee> it's very obsolete, but I guess people still cling to the old windows versions like gold nuggets

18:37 <InPhase> Yeah. The oldest systems I cling to are all work systems, like the 2014 Linux server with 2013 kernel version 3.10.0, the 2014 macbooks not allowed to upgrade past macos 10.12, and the important Windows 7 computers.

18:38 <InPhase> Although it's not so much clinging as nobody has established a viable plan for upgrading and replacing them and all the associated consequences.

18:42 <InPhase> At least we got all of the Python 2 out of circulation though, except for the one Python 2 program that was deprecated, taken out of commission, replaced by an entirely different system, and then resuscitated from the dead and then blocking off the upgrade of those macbooks.

19:04 <gbruno> [github] rcolyer closed issue #4178 (Performance degradation caused by fast-csg) https://github.com/openscad/openscad/issues/4178

19:07 <joseph_> InPhase: That's a great question. My limited experience in this area was during freshman year of college when I did some smart contract prototyping with a few other guys. The goal was a micro-lending application, but it failed because we saw transaction costs were too high

19:07 <joseph_> InPhase: I think all decentralized systems have to fight against a competitive disadvantage because robustness and consensus are by nature slower and more expensive than a centralized authority. If people want these benefits they must accept the cost

19:11 <InPhase> It does seem like every cryptocurrency is fighting a constant battle with transaction scarcity making them too expensive to be useful. Ethereum seems to be trying to fix this with layer 2 add-ons, but I do not understand if these should have the same trust, or if any have the same capability. I also do not understand why none are designed with true scalability where the full chain only needs to be

19:11 <InPhase> grabbed if you need it for some purpose, opening up more transaction space.

19:12 <InPhase> I keep finding myself with the conclusion that while many are making mall fortunes, it's wise to stay away. I'm just watching for when and where this might not be true.

19:12 <teepee> also fighting against one of the inherent problems of software... bugs

19:14 <InPhase> But, thanks. I just wanted to hear an experience from someone who actually sat down and wrote something for contracts. :)

19:15 <teepee> bitcoin has none of those right? otherwise you could visit el salvador :)

19:16 <teepee> I just saw a report they adopted it as official currency

19:16 <teepee> crazy stuff :)

19:16 <InPhase> joseph_: On-topic, now or soon would be a good time to pop up a PR for some sort of small change to demonstrate grasp of the basic mechanics.

19:16 <InPhase> joseph_: I was poking around the issues, and didn't see one yet.

19:17 <teepee> I said it's more important to have the propsal in as that's deadline in 2 days

19:18 <teepee> while ideally the PR would go along, it's ok to have it while we are in the selection/decision period

19:18 <teepee> like half way in, not last day ;-)

19:20 <teepee> I always think the application period should be a bit longer

19:20 <InPhase> Yep. I think that's the right ordering of priorities.

19:22 <teepee> ah, next up at revision: Shader Jam feat. Wayfinder

19:23 <InPhase> Looks like there's a full 3 weeks to submit slot requests after receiving applications, which is excessively long. That could have been 1-2 weeks.

19:23 <teepee> joseph_: one inspiration regarding handling of shaders could be RetroArch, not sure if you have heard about it. it's a frontend for various retro system emulators and other applications

19:24 <InPhase> joseph_: Let's say a good timeline is 1-2 weeks from now for some sort of demo PR. But seriously, feel free to ask questions about it. It's not an exam. Communication is an asset in open source work.

19:24 <teepee> the focus is quite different as their shaders are mostly for effects based on bitmap, but it might still give some ideas

19:41 <joseph_> InPhase: I'll check out retroarch, recognizing teepee's point that it serves a different purpose. I've started investigating #3706 but there's no PR bc I don't have a working fix yet. My findings: https://github.com/openscad/openscad/issues/3706#issuecomment-1098561884

19:48 <InPhase> Sounds like a logical assessment.

20:19 <crazy_imp> with a spaceball and "SpaceNav" selected in the Axes configuration - is the axis mapping fixed / needs to be configure via spnavcfg?

20:19 <peepsalot> Scopeuk: somehow formatting them exactly the same doesn't seem to help. I actually have no idea at this point what git is even doing for "find-renames"

20:20 <teepee> crazy_imp: I believe it still needs to be configured, not 100% sure though, I have not tried that for a long time

20:21 <teepee> uh, print fits first time, awesome \o/

20:22 <teepee> now the question is how to clamp things together :)

20:29 <Scopeuk> peepsalot that's really annoying

20:31 <crazy_imp> teepee: if an axis isn't mapped in openscad, it does nothing. once mapped i can manipulate it from spnavcfg

20:48 <peepsalot> i guess I need to recreate the file moves/renames in every branch that requires merging/rebasing, since rename detection does fuckall

20:55 <InPhase> peepsalot: https://github.com/git/git/commit/9dd29dbef01e39fe9df81ad9e5e193128d8c5ad5

20:56 <InPhase> peepsalot: 2021-07-15 edit to git master that sets up an unlimited number of renames able to be detected. Which then leads me to, do you know what your rename detection limit is set to? You have a good number of them all at once.

20:56 <InPhase> Maybe you smacked into a performance "feature".

20:59 <peepsalot> i set merge.renamelimit=0 for unlimited

20:59 <InPhase> With what git version?

20:59 <peepsalot> 2.35.3

21:00 <InPhase> I guess that's recent.

21:00 <InPhase> Well, that's all I had for you. Recreate all the moves by scripting was my previous idea.

21:02 <InPhase> Honestly I don't understand why it has to find them anyway. They should be part of the git history being merged...

21:05 <peepsalot> i have to set find-renames=03 to avoid false deletions. that means it detects a file as a rename if their "similarity index" is at least 3%

21:05 <peepsalot> this is like the point where file license info counts though

21:07 <peepsalot> and it makes no sense because eg my branch has src/VetexArray.cc vs master src/glview/VertexArray.cc the file is ~470 lines long, and only 3 lines differ between them. but this fails to match at 4%

21:11 <InPhase> I see no logic behind even attempting the similarity test if git mv was used in the git history to rename the file. That's history. Git shouldn't need to validate it.

21:12 <peepsalot> i'm pretty sure git mv is just syntactic sugar for delete/create ... because that's the only thing git understands

21:12 <InPhase> Really?

21:12 <InPhase> Well that's the problem then.

21:12 <peepsalot> it literrally has no concept of moves

21:12 <InPhase> That's a deep design flaw.

21:13 <peepsalot> but linus says its a feature

21:13 <peepsalot> somehow

21:13 <InPhase> And tell us peepsalot, how's that excellent feature working out? ;)

21:17 <Scopeuk> peepsalot if you do make that script might be worth publishing it in the repo to mop up

21:19 <InPhase> peepsalot: Git tips based on this move as delete/create seem to be, "Never ever move and change at the same time".

21:19 <InPhase> peepsalot: Which of course is exactly what the thing you're attempting is like, even though that's not what you did.

21:21 <peepsalot> right, the tip should be never move and change any file as long as unmerged PRs exist

21:21 <InPhase> Which should be a good reason to not use git, provided there were anything suitable to replace it.

21:22 <InPhase> I guess on the one hand, I don't really see a way around it with any combination of merging or picking intermediate points, or anything like that. But on the flip side, it might actually be okay to just script redoing it, since there was never a move stored in the first place.

21:22 <InPhase> I'd say script it, change one, and then test out a merge of that with master and see if it actually can handle it.

21:23 <InPhase> (Locally, of course.)

22:11 <joseph_> teepee InPhase: Now that my proposal is submitted, do you know if there is any specific coordination I need to do with BRL-CAD or OpenCAx? It seems like they are the endpoint with Google. Also I will follow up about my PR progress, after I finish a big school project due Tue

22:20 <crazy_imp> https://youtu.be/sTmTEuZUf_4 - did some greenhouse construction, still wip

22:20 <othx> crazy_imp linked to YouTube video "greenhouse" => 1 IRC mentions

22:38 <teepee> joseph_: I don't think there is at this point, only in case the project is accepted, but I can check with Sean again if there's anything new this year

23:02 <teepee> crazy_imp: looks nice, even the plants are growing