Up here in Portland (well really the western suburbs) we had a rather ambitious TechShop start up about a decade and a half ago, but they didn't last a year. I made a decent sized donation at the start because I liked what they were doing, and took some great classes there on woodworking and CNC. The woodworking instructor in particular was really interesting as he was an old school pattern maker (patterns for casting) which involves different techniques than finish or furniture carpentry. He could build to 1/64" tolerances by using various combinations of sanders, including a stationary disc sander with a 1 yard diameter. That thing was a pretty cool piece of equipment.

Unfortunately I'm sad to say besides the finances they had some cultural issues. A few of the instructors went pretty heavy on the "boys club" style humor in a way that would clearly alienate a lot of women.

https://doubleunion.org/

At first, you have an influx of excited members, so it feels like you have a shot at it - but within a couple of years, most of them drop out. They either get discouraged, or they find their groove and really want the convenience and flexibility of working on their own schedule with their own tools that aren't abused by strangers.

If you want to run a Makerspace in Portland, Seattle, or in the SF Bay Area, you better have an endowment from a wealthy donor, a long-term government grant, or something of that sort.

Of course, that market has other problems - cost and neighbors who don't want 24/7 power tool use - but it's better than the alternative where anyone serious about their hobby can easily build their own shop.

Things like milling machines are expensive, and spend 99%+ of hours unused in home shops.

On the flipside, accountability and safety are expensive in shared spaces.

The best version of this that I've usually seen is where 10 people who want to work on cars rent a shared space and provide shared tooling.

It would be neat to see votech programs/community colleges fulfill some of this role, particularly as a "meeting space for makers."

I find it heartbreaking but it seems inevitable. Not a lot of people want to keep an old building available for rent when they can sell the property for redevelopment into "luxury" apartments or condos.

I totally understand though. People don't realize how expensive precision metalworking stuff is. And that's before you start to insure the place.

For something like a really rigid CNC (not a desktop sized one or for wood), someone needs to come up with a reasonable cost machine that can hold thous and has a spindle that isn't stupidly overpriced.

A while back I looked into building my own, and ended up with a Tormach because I couldn't compete on price. There are a couple near that price point now, but going transformatively cheaper seems very hard.

Lathes are easier. The C4 class is both quite cheap and impressively capable (much more than the cheaper C3 mini lathe class).

Similar competitions (solar car, concrete canoe, pinewood derby, hackathons, yearly DEFCON attendance, 4-H robotics, Zero robotics, INTEL/GOOGLE/Microsoft/ThermoFischer science fairs, math olympiads, science olympiads) might be reasonable draws for other maker shops.

And then there was American Steel…

You’re not imagining things. It’s hard to sustain places like this when private equity firms are buying up all the land to build high end apartments.

It sounds like there were some differences in the membership structure, but American Steel was bought by a private equity firm, the existing tenants were kicked out, and now as you say the new VC funded firms bring in far more income than poor artists ever could. That sounds like exactly the kind of loss to our community that we suffered when Techshop went under.

But as others noted, it is not an attractive business. High fixed costs, you're significantly more optional than gym subscriptions to many people, and your best customers naturally tend to graduate to owning their own.

I wonder if it's possible to also have a series resistor between the battery and microcontroller to bring it down to its lower end of acceptable voltage range for longer. Just a thought.

That said, I'm always sad to get shot in the eyeball by bare LEDs.

After giving this idea a little thought, I'd love to see an iteration that diffuses the LED light through the skin...

Wow - subtly glowing ears, noses, lips, etc?!

I got his satellite clock kit for the office and love it!

I wonder how hard it would be to make a set of large-gauge lobe jewelry with a screen in it. Perhaps 1/2" or 3/4" would be doable?

A thought re sales and sizing, though: industrials are usually sold in quarter-inch increments and most people buy 1-2.5" from what I understand, so it should be doable to make them to sell.

At best you could probably do a small low-res (8x8?) LED matrix with some programmed patterns but even that would require more batteries etc than this and get exponentially heavier.

the one-finger orbit and 2-finger pan and zoom was fantastic and intuitively obvious. with a mouse it's left-button orbit, right-button pan, and mousewheel zoom, which is not as immediately obvious but still pretty usable

https://mitxela.com/other/3d/loader.js?2

view-source:https://mitxela.com/img/uploads/metal/scaffold/scaffold-case...

incidentally meshlab knows how to open .wrl files, and the loader on this page also supports .stl

surprisingly, this is not a hack of modelviewer.dev, though it uses the same tag name and underlying 3-d library. it supports different model formats and is an independent effort. https://mitxela.com/projects/model-viewer is a page from last month by jacobs about it

no license visible tho, which is a shame because it's awesome and i'd like to be able to use it. still, not that hard to reimplement i guess? now that jacobs has done all the hard and non-copyrightable work of figuring out what kind of user interface and programming interface works best for this

I feel like it may have been like some XML format (VRML?), but I may be misremembering and it may have just been Java applets...

With a strong enough magnet, instead, just its torque might be enough to train yourself to.

I’ve often wondered why smart watches don’t have battery link bands… seems you could add a ton of capacity in a fashionable way.

It seemed to result in a fairly thick and inflexible strap, with limited length adjustment. It was generally panned by reviewers for its bulk, although a microsoft-loving large-wristed friend of mine reported it was just fine.

It will be an interesting world when Sneaker Nets are the safest smuggle.

And then years down the road you know how to make most things from scratch from the electronics design, PCB gerbers, and the enclosure both in metal (both machined and fabricated), laser cut, and 3D printed. But just conjuring up any cool thing you want to make summons a mental spreadsheet of the time it will take to complete the project, possibly including making a small-run of extras to sell for much less than your time is worth, and then you set your drink back down and shake your head because ain't no one got time for that any more. And then you buy the Etsy/Kickstarter version instead because you realize they have woefully underpriced their super cool widget.

I took engineering tech and welding classes at my local community college as a non-matriculated student.

Once you've got the hang of it, more complex PCBs aren't really that much different.

I started with double-sized PCBs, but 4-layer PCBs are now almost equally cheap, which makes routing a lot easier too.

[1] https://tomverbeure.github.io/2019/02/05/VGA-I2C-My-First-PC...

The hardest part by far is figuring out what you want to do. Even if you think you want to do a hundred different things and you can't do them all, you still have to figure out what you want to do.