Comment by joachimhs | Hacker News Reader

Comment by joachimhs | Hacker News Reader

Parent Full thread

I run a small makerspace for kids and teens in Norway, and we kept hitting the same wall: the 3D tools we tried were either too fiddly for a 9-year-old or locked behind installs and accounts. So I built Akse — a browser-based 3D modeller for beginners, where the path from idea to a printable STL is as short as I could make it.

It's deliberately limited, but with a rather powerfull 2D Blueprint mode where you draw an outline on millimetre paper and extrude it to 3D. You build by placing and combining primitive shapes (box, cylinder, sphere, cone, pyramid, wedge, torus), set any shape to "hole" mode to cut it out of another. Everything is in real millimetres, so what's on screen matches what comes off the printer. Output is a single STL. That's most of it — no parametric constraints, no assemblies, no fillets. For teaching beginners that's intentional, not a gap.

The obvious comparison is Tinkercad — same space (primitive-based, browser, education-oriented), and I'm not claiming Akse is better. The differences are that it's open source, embeddable as a Svelte component, works in Norwegian as well as English, and is even more stripped down. It mostly exists because I wanted something I could shape around how our workshop actually runs, and put in front of Norwegian-speaking kids without an account or install.

Under the hood it's a Svelte 5 component using Three.js for rendering and three-bvh-csg for the boolean operations; storage goes through a small port interface so it's backend-agnostic, and the standalone version just uses localStorage. It's early (v0.1) and has rough edges. I'd really value feedback on where it trips up first-time users, since that's the entire point of the thing.

Source (AGPL-3.0, with a commercial option): github.com/joachimhs/akse3d

I have similar hopes for Solvespace - that every middle school student can pick it up and design things. We have a couple issues that keep me from recommending it too strongly though - bugs in the boolean code are IMHO the biggest blocker for kids.

Any chance you could have the kids make comparisons between the two? Solvespace is completely constraint-based, so it may be a bit harder to learn but also more flexible.

It's a single exe, but there is also an experimental web version: https://solvespace.com/webver.pl

> that every middle school student can pick it up and design things.

The Solvespace UI is a long, long way from being the sort of UI a contemporary kid has any kind of comfort with, I'm afraid, and will be obtuse even to teachers (many of whom, with subjects that concern technology, do not have time to develop expertise in an obtuse UI and may indeed only be confident they understand the meaning of all the lessons they are teaching and not much more).

I don't think bugs in your booleans are your biggest problem at all.

I think Tinkercad has weaknesses as a classic CAD package, and there are things I would like to see done better, but as a package to teach younger people how core concepts in 3D modelling (rather than the ontologies of bRep) actually work, it is the standard you are working against.

A UI which seems a bit more in-line w/ contemporary expectations is Dune 3D:

https://dune3d.org/

which I find a little less confusing than the traditional 3D CAD packages I've tried (and failed to learn) --- at least for Dune 3D I've made it through the tutorial successfully.

Dune 3D is OK — and is after all dependent on Solvespace.

As noted on the footnote for the project on Github:

>I ended up directly using solvespace's solver instead of the suggested wrapper code since it didn't expose all of the features I needed. I also had to patch the solver to make it sufficiently fast for the kinds of equations I was generating by symbolically solving equations where applicable. ↩

>> The Solvespace UI is a long, long way from being the sort of UI a contemporary kid has any kind of comfort with...

Respectfully, I disagree. Even adults have used the word "fun" to describe using solvespace. But I don't actually have feedback from kids, hence the question to OP.

Don't get me wrong, I think Solvespace is really important and I am glad you are keeping it ticking over, but it is the least fun UI I have ever encountered, I'm afraid (kind of obtuse on a trackpad too)

Not saying any of this is easy — Shapr3D on an iPad, expensive and marketed on its extraordinary usability, is just utterly perplexing!

>> kind of obtuse on a trackpad too

Oh, I actually hate trying to use solvespace with anything other than a 3-button mouse. Unfortunate because the web version can actually run on my phone but is unusable in practice.

I guess this is, sort of, the filter through which I experience the frustrations; it may be magnifying them even.

I've not used a three button mouse in thirty years. Probably only used a mouse for a dozen hours in the last 15 years.

Not the person you were responding to, but genuine question: How are you interacting with computers without practically ever using a mouse? And how do you expect to use CAD software without a mouse?

Trackpad?

I have used Macbooks with good trackpads since 2003. The models from the 2008 unibody Macbook and onwards are so very obviously superior to mice, that I simply have never wanted to use a mouse since.

My preference for mice was two-button Microsoft Mouse models (I had the magnificent early "dove bar" Microsoft mouse, which I regretfully had to replace with a 2.0 at some point). On the Mac, you only had one-button mice anyway.

For Linux I did try a nicely made Logitech three-button mouse, but ended up giving it away because some quirk of my neurology or physiology simply will not let me consistently independently address the middle button while holding or moving the mouse. I will almost always end up pressing one or other finger either side as well (sometimes enough to click it).

Scroll wheel mice are slightly better because the middle button is physically different (and often slightly raised). I do have one stored away in case I need it.

On a trackpad I do have individual fingertip tap control, for some reason.

I also use a wacom pen for photo editing, but those are not really much use for CAD. (Honestly, I don't get how people think Shapr3D is usable on the iPad with the Pencil; it is not)

For CAD: FreeCAD's "Gesture" control mode is superb! The "touchpad" control mode not bad. I've toyed with the idea of getting a Spacemouse for my left hand (or making one of the DIY ones) but I have never really felt enough need, because the Gesture mode is so fluid and instinctive.

>> Scroll wheel mice are slightly better because the middle button is physically different (and often slightly raised). I do have one stored away in case I need it.

Solvespace does great with a scroll-wheel mouse. Scroll = zoom.

> That's most of it — no parametric constraints, no assemblies, no fillets. For teaching beginners that's intentional, not a gap.

Fillets seems like an omission. For 3d printing, fillets are pretty important for getting good results.

Every other limitation makes sense to me, tho it’d be cool to see parametric constraints eventually, if you could find a way to add them without confusing beginners.

Thanks for making this! I will be recommending it over tinkercad going forward!

Group transform is out of order. It does not transform the group but the elements. This leads to the suspicion that position and rotation are not transformation chains on object trees but attributes. That would be the wrong architecture.

I am very sorry, but please explain. Why is this a nice looking Svelte / Three / CSG app, but the basics are wrong?

Someone built something useful and you complain that it's "nice looking"? Consider reframing your feedback.

No, they are complaining that a basic functionality is not working and hinting at not having the right architecture to deliver correct results in more complex scenarios.

but this is clearly not meant to be onshape, its for kids who want to 3d print something. the 'hole' mode does not seem to subtract like I would expect.

Yes, sure. I will try to be more blunt, as I was not trying to complain about anything.

I think you have invested with agreeable results, but you missed fundamentals. Adjust course to "architecture first" and I expect a great product.

I think that's a lot better. "Please explain your wrongness" really had me scratching my head. Hopefully they can make improvements to the 3D architecture or find someone who can help.

Reply on news.ycombinator.com