Not very satisfying that he didn't actually design a flatpack bed.
There is an interesting optimization problem in this that the author didn't touch on: create a design that maximizes the useful material use.
So you might start with a design that is really spindly, and has too many joints. Then run that through the optimizer, so that you can see the proposed layout. This design will have lots of waste, but you will also be able to see where you could add strength or functionality by consuming the offcuts.
In the author's design[1], the overlaps can simply be cut out of the other members without reducing strength too much. The central portion can then be cut into cantilevered tables, and a few more offcuts can provide bracing or legs, which gives you free integrated nightstands (as wings off the existing bed).
>what does it feel like to have 90% code/abstraction with 10% of visual sketch input
I was playing around with automating some aspects of architectural model making a few years back. I was using grasshopper (because I am an architect and can't really code). I found it interesting at the time, that the program encourages you to play with the ratio of visual-to-programmatic metric, but that ratio also reinforces how you interact with the project.
It was pretty simple stuff.. automatically unfolding meshes for paper cutting, adding tabs and rebates to sheet material for more precise joins, tagging elements for correct assembly, sorting elements by size and shape and using some basic genetic algorithms to optimise orientation and position on sheets to minimise time and material wastage.
But it really got me that when you push the programmatic elements to be more precise and repeatable, you really close down the problem space. When you increase the visual feedback, you tend to do it at the expense of precision and repeatability - your process becomes more tightly coupled to the specific project at hand.
Anyway, I am def going to have to make some time to try some of these out.
I hate to say this, but it seems like a pretty clear case of using the wrong tool for the job.
There's no conceivable reason to cut something this simple on a large-format CNC mill. It's literally just a couple of straight cuts. It's not going to be faster (not with standard endmills), not going to be easier, and it's not going to be cheaper unless you're making them by the thousands.
You can likely buy S4S lumber for less than an oversized sheet of furniture-grade 1" plywood.
If the machine is available and it’s a hobby project it’s ok, but you’re right this is not a cheap or easy way to do it.
But as a mechanical engineer, the whole project is such a “software” approach to me, starting from turning the selected tool into a requirement to use. Then, rather than just powering through in solidworks for a one-off design, the author spends a lot of time looking for automation tools like they are a library, points out this approach feels like coding in rust, gives up, and even blogs about it.
We grew with overbuilt furniture, trailers, radio masts, gas fittings, plumbing, etc. that my father built. A whole lot of knitted jumpers, wool socks, bed covers, etc. from my mother and the aunts with a freezer full of home cooked meals scaled for shearing teams and put away for later.
Most of that carries on, and we've got good relations with a slew of people that are solid craftspeople who put their best work in galleries for the extra big bucks.
The good stuff is as accessible to us normies as good software is to the HN crowd .. it costs less or nothing if you're in the maker circles swapping food, yarns, and other goods.
FWiW I was a silent partner and occasional assistant in a glass studio / wood shop for two decades, that helps - that was some money (way back when you could get a lot of land for a lot less than today) and a lot of sweat equity modifying buildings and landscaping, etc.
The bit about Mozaik hits home. I use that software every day. It's terrible and I hate it and it's so bad that you have to quadruple check everything before cutting anything... but it does work, and it's cheap. The custom cabinetry company I work for has nested thousands of sheets of material using it, which has allowed the company to grow well beyond the manual cutting they used to do.
But the software is horrendously bad, and I struggle to understand how so many serious faults that have existed since the very beginning have never been fixed. Faults that hinder usability to the point you want to bash your head in. The fact that Mozaik is still in business (and has even been acquired!) is simply because there is no competition in the space—alternatives are either incredibly expensive (tens of thousands of dollars per year), or not suited to production use.
I did, and by many accounts CV is just as bad as Mozaik, although significantly more capable.
Mozaik is bad because it tries to do everything itself, badly, instead of adding functionality to established CAD software. That's the approach Microvellum takes with AutoCAD, and I expect it's probably a lot better software, but like everyone in this space except Mozaik, they do the whole "contact us for a quote" thing. In other words, it won't be cheap.
Yes we've had this discussion with management before when discussing CV short comings. Pathetic offering of a program, and simply would not pass even the most basic usability tests in other fields. Microvellum afaik is only a better program "in theory." I have not heard much positivity from actual users.
One thing that wasn't mentioned is that its important to measure the plywood you're using. 18mm ply can vary from 17 to 19mm so when you're making tight finger joints its important to know the thickness of the sheets that the parts will be made from.
Once it comes to actual cutting, the tool (in my case, a water jet) will often have an auto-nester to lay out the parts, but being able to prototype the part nesting in CAD means its faster to optimise the design to fit the sheets you've got.
I have also just moved into a new apartment in Copenhagen and needed to find new furniture, including a bed. The cheapest and easiest option was to “buy” a used bed (buy is in quotation marks because many beds are either given away or cost a modest amount, as long as you pick them up yourself). The first bed I bought was too wide to fit through the stairwell, so I sold it the next day without any problems. The next bed (a futon in perfect condition) was easy to take apart and fit through the stairwell without any problems.
Compared to buying new, it was much cheaper, and compared to building it myself, it was both much cheaper and easier.
Maybe I'm crazy (or it is an accidental illusion most likely) but when scrolling it seems like the content of the image below the 'flatpack cad' heading almost moves (i know it doesn't actually). I'm curious what kind of illusion is causing this or if it has something to do with vsync in the browser?
I think I can see what you mean when I scroll fast enough. It looks like horizontal movement either to the left or right depending if you're scrolling up or down.
Author could spend a lot more time learning how to use Autodesk Inventor sketch environment. You can setup sketch dimensions as ‘driven’ and use equations to comply to whatever exterior box dimensions one chooses. His creating assembly from parts and arranging is silly. All parts /solids should come from same master sketch.
And then also how to use Autodesk CAM, you can use sketch geo to set paths, define offsets, etc.
Perhaps this might fill the gap if you are able to access this resource. The following I have not verified, and the information therein is provided for commentary and educational purposes. I was able to find what appears to be two versions of the same or interchangeable content:
If only there were a company that produced flat pack bed frames with minimalist design for little money
I have a bed frame that's around 24" above the floor. I took parts of Ikea bed frames I found on the street that were thrown out around the time students change schools. I then bought some 2x4s, cut them down to size, and used a mallet and chisel to cut out mortises. The other parts of the bed frame were the tenons, and support the normal Ikea bed slats, with a couple 2x4s added underneath for extra support, and some diagonal 1x4s for leg reinforcement. Probably cost me $20 (including screws) and it packs down to a couple long boards and two rolls of bed slats.
Not very satisfying that he didn't actually design a flatpack bed.
There is an interesting optimization problem in this that the author didn't touch on: create a design that maximizes the useful material use.
So you might start with a design that is really spindly, and has too many joints. Then run that through the optimizer, so that you can see the proposed layout. This design will have lots of waste, but you will also be able to see where you could add strength or functionality by consuming the offcuts.
In the author's design[1], the overlaps can simply be cut out of the other members without reducing strength too much. The central portion can then be cut into cantilevered tables, and a few more offcuts can provide bracing or legs, which gives you free integrated nightstands (as wings off the existing bed).
[1] https://kevinlynagh.com/newsletter/2025_07_flatpack/cad_nest...
There is a whole book based on this idea, but for clothes making https://www.birgittahelmersson.com/ “zero waste patterns”
>what does it feel like to have 90% code/abstraction with 10% of visual sketch input
I was playing around with automating some aspects of architectural model making a few years back. I was using grasshopper (because I am an architect and can't really code). I found it interesting at the time, that the program encourages you to play with the ratio of visual-to-programmatic metric, but that ratio also reinforces how you interact with the project.
It was pretty simple stuff.. automatically unfolding meshes for paper cutting, adding tabs and rebates to sheet material for more precise joins, tagging elements for correct assembly, sorting elements by size and shape and using some basic genetic algorithms to optimise orientation and position on sheets to minimise time and material wastage.
But it really got me that when you push the programmatic elements to be more precise and repeatable, you really close down the problem space. When you increase the visual feedback, you tend to do it at the expense of precision and repeatability - your process becomes more tightly coupled to the specific project at hand.
Anyway, I am def going to have to make some time to try some of these out.
I hate to say this, but it seems like a pretty clear case of using the wrong tool for the job.
There's no conceivable reason to cut something this simple on a large-format CNC mill. It's literally just a couple of straight cuts. It's not going to be faster (not with standard endmills), not going to be easier, and it's not going to be cheaper unless you're making them by the thousands.
You can likely buy S4S lumber for less than an oversized sheet of furniture-grade 1" plywood.
If the machine is available and it’s a hobby project it’s ok, but you’re right this is not a cheap or easy way to do it.
But as a mechanical engineer, the whole project is such a “software” approach to me, starting from turning the selected tool into a requirement to use. Then, rather than just powering through in solidworks for a one-off design, the author spends a lot of time looking for automation tools like they are a library, points out this approach feels like coding in rust, gives up, and even blogs about it.
Craftsmanship has been replaced by 3d printing.
Not at all, if anything good craftsmanship is commanding higher prices to those that appreciate it as it becomes relatively scarce.
Some guy makes a wooden bed: https://www.youtube.com/watch?v=sL96mw1uCmA
Lino Tagliapietra still makes a packet at 90 just for orchestrating others making non functional glassware.
I guess I should add the qualifier: for the middle class.
I grew up in homes that were well built with well made appliances etc, but it wasn’t fancy luxury, it was just well built.
Now it’s either junk or unaffordable luxury priced (and maybe not even craftsman made!)
We grew with overbuilt furniture, trailers, radio masts, gas fittings, plumbing, etc. that my father built. A whole lot of knitted jumpers, wool socks, bed covers, etc. from my mother and the aunts with a freezer full of home cooked meals scaled for shearing teams and put away for later.
Most of that carries on, and we've got good relations with a slew of people that are solid craftspeople who put their best work in galleries for the extra big bucks.
The good stuff is as accessible to us normies as good software is to the HN crowd .. it costs less or nothing if you're in the maker circles swapping food, yarns, and other goods.
FWiW I was a silent partner and occasional assistant in a glass studio / wood shop for two decades, that helps - that was some money (way back when you could get a lot of land for a lot less than today) and a lot of sweat equity modifying buildings and landscaping, etc.
The bit about Mozaik hits home. I use that software every day. It's terrible and I hate it and it's so bad that you have to quadruple check everything before cutting anything... but it does work, and it's cheap. The custom cabinetry company I work for has nested thousands of sheets of material using it, which has allowed the company to grow well beyond the manual cutting they used to do.
But the software is horrendously bad, and I struggle to understand how so many serious faults that have existed since the very beginning have never been fixed. Faults that hinder usability to the point you want to bash your head in. The fact that Mozaik is still in business (and has even been acquired!) is simply because there is no competition in the space—alternatives are either incredibly expensive (tens of thousands of dollars per year), or not suited to production use.
I am in the same boat, daily user of Cabinet Vision. Did you know the same founder made both and sold them off within a few years of one another?
I did, and by many accounts CV is just as bad as Mozaik, although significantly more capable.
Mozaik is bad because it tries to do everything itself, badly, instead of adding functionality to established CAD software. That's the approach Microvellum takes with AutoCAD, and I expect it's probably a lot better software, but like everyone in this space except Mozaik, they do the whole "contact us for a quote" thing. In other words, it won't be cheap.
Yes we've had this discussion with management before when discussing CV short comings. Pathetic offering of a program, and simply would not pass even the most basic usability tests in other fields. Microvellum afaik is only a better program "in theory." I have not heard much positivity from actual users.
I've found the OnShape flow pretty good for this, using the Laser Joint and Auto Layout scripts he referenced - thread here: https://forum.onshape.com/discussion/14166/designing-laser-c...
One thing that wasn't mentioned is that its important to measure the plywood you're using. 18mm ply can vary from 17 to 19mm so when you're making tight finger joints its important to know the thickness of the sheets that the parts will be made from.
Once it comes to actual cutting, the tool (in my case, a water jet) will often have an auto-nester to lay out the parts, but being able to prototype the part nesting in CAD means its faster to optimise the design to fit the sheets you've got.
I have also just moved into a new apartment in Copenhagen and needed to find new furniture, including a bed. The cheapest and easiest option was to “buy” a used bed (buy is in quotation marks because many beds are either given away or cost a modest amount, as long as you pick them up yourself). The first bed I bought was too wide to fit through the stairwell, so I sold it the next day without any problems. The next bed (a futon in perfect condition) was easy to take apart and fit through the stairwell without any problems.
Compared to buying new, it was much cheaper, and compared to building it myself, it was both much cheaper and easier.
Maybe I'm crazy (or it is an accidental illusion most likely) but when scrolling it seems like the content of the image below the 'flatpack cad' heading almost moves (i know it doesn't actually). I'm curious what kind of illusion is causing this or if it has something to do with vsync in the browser?
I think I can see what you mean when I scroll fast enough. It looks like horizontal movement either to the left or right depending if you're scrolling up or down.
Closest thing I could find is something called the "serpentine illusion". https://www.frontiersin.org/journals/neuroscience/articles/1...
Author could spend a lot more time learning how to use Autodesk Inventor sketch environment. You can setup sketch dimensions as ‘driven’ and use equations to comply to whatever exterior box dimensions one chooses. His creating assembly from parts and arranging is silly. All parts /solids should come from same master sketch.
And then also how to use Autodesk CAM, you can use sketch geo to set paths, define offsets, etc.
Immediately made me think of Victor Papanek
https://en.m.wikipedia.org/wiki/Victor_Papanek
I made several of his designs from the 70s onwards and still use a modification of his bookcase design.
The reference most relevant would be "Nomadic Furniture" which I believe is a Dover reprint.
Nomadic Furniture => https://archive.org/details/nomadicfurniture00henn/
Looks like a cool 70's book. There's a sequel too — but not that I could find on archive.
Perhaps this might fill the gap if you are able to access this resource. The following I have not verified, and the information therein is provided for commentary and educational purposes. I was able to find what appears to be two versions of the same or interchangeable content:
https://annas-archive[.]org/md5/d6bcfd2e04361e5eba8358a31ada...
https://annas-archive[.]org/md5/21c997325f84137e2d3381aa4f7d...
If only there were a company that produced flat pack bed frames with minimalist design for little money
I have a bed frame that's around 24" above the floor. I took parts of Ikea bed frames I found on the street that were thrown out around the time students change schools. I then bought some 2x4s, cut them down to size, and used a mallet and chisel to cut out mortises. The other parts of the bed frame were the tenons, and support the normal Ikea bed slats, with a couple 2x4s added underneath for extra support, and some diagonal 1x4s for leg reinforcement. Probably cost me $20 (including screws) and it packs down to a couple long boards and two rolls of bed slats.