Comic Sans Typeball Designed to Work with the Ibm Selectric Typewriters

It’s not far fetched though. With careful design, prints often do work on the first try.

Yeah, feels a bit like "I'm sharing this code I've written here. I haven't tried to compile it yet, but it should work."

Looks like the same person made a typeball for Cherokee, and that comes with a picture of the printed output:

https://www.printables.com/model/441262-cherokee-typeball-fo...

So maybe Comic Sans (and Wingdings) will work as well.

> they don’t have Y to test it but “it should work”.

SW development process 101. What are users for ? /s

Works for shipping commercial software!

(Slightly /s)

Amazingly, I think you’re right, but not for (negative) the reasons you state.

I see the 3D printing scene a remix culture where you’re supposed to try new things. If you want something to work exactly for your situation, you might have to do a little work yourself.

In this case, here we have an idea of making new typeballs for IBM typewriters. Here is one post where someone is playing with the idea. At the top of the page, there is a link to a GitHub page for you to adapt it as you see fit.

But then there’s also a link to a new page: https://selectricrescue.org/ where someone makes many more varieties, also with code for you to try yourself.

I see 3d printing much like the world of early open source coding. Sometimes you get a completed project that works out of the box. But often, you’ll have to tinker and adapt it to your exact hardware. Once a new 3d model is shown to be viable, it can be mixed and matched to make something new!

I have, along with some other ones I created from different fonts. See https://www.youtube.com/watch?v=1kXnsvYfaF4&t=2160s (it was an aside to a longer Selectric project I did; this typeball project came out around the same time so I decided to make a few of them).

Watching this back, it partially works but some of the letters clearly need their positions adjusted, and some of them seem to have trouble imprinting fully on the page. That's besides the fact that the proportional fonts look pretty weird when printed monospaced.

I didn't really go further with this because the resin printed balls didn't hold up very well. A slot near the hole would always break open, and then the ball would spin freely instead of being indexed like it was supposed to. These days I've pretty much abandoned home resin printing entirely since it is extremely messy, and I've never been able to make parts with it that weren't ridiculously brittle, even with the supposedly "strong" or "tough" resins.

I've used commercial resin prints that Shapeways/Xometry/PCBWAY have and they are a lot tougher, so maybe they would work well.

Fair enough, but this isn't that case. They have Y (it's in the first photo) and tested a previous version of the model. The changes are predictable (rotating letters and slight scaling) so I don't think it's unreasonable to be confident and not waste plastic.

They also link a version that works they say. https://selectricrescue.org/

Also includes Papyrus!

One of the Teletypes I used to work on had an eccentric-driven Wiffletree to position the typebox. This was necessary because of it's high speed[0] to decode the ASCII character set (it wasn't a Baudot machine). The selector cam at the end of the machine would do the serial-to-parallel conversion, then clutches would rotate the eccentrics powering the wiffletree to select the character.

If you look at the ASCII chart [1] you can select a character via binary tree - the Wiffletree being a mechanical equivalent.

[0] For a Teletype

[1] https://en.wikipedia.org/wiki/ASCII#/media/File:USASCII_code...

It's vintage now and gives me warm feelings.

It’s even the first link on this page. In addition to Comic Sans, he’s also made a font ball for Papyrus!

I can’t imagine using any of these (even when typewriters were king), but I love that they exist.

> without being injection-molded with an extremely durable plastic, it will not make good strike.

I wonder how true that actually is. A properly 3d printed ABS part isn't really much if any less durable than an injection molded part. It's mostly just worse with respect to feature resolution.