Llama.ttf Is AI, In A Font

June 26, 2024 by 6 Comments

It’s a great joke, and like all great jokes it makes you think. [Søren Fuglede Jørgensen] managed to cram a 15 M parameter large language model into a completely valid TrueType font: llama.ttf. Being an LLM-in-a-font means that it’ll do its magic across applications – in your photo editor as well as in your text editor.

What magic, we hear you ask? Say you have some text, written in some non-AI-enabled font. Highlight that, and swap over to llama.ttf. The first thing it does is to change all “o” characters to “ø”s, just like [Søren]’s parents did with his name. But the real magic comes when you type a length of exclamation points. In any normal font, they’re just exclamation points, but llama.ttf replaces them with the output of the TinyStories LLM, run locally in the font. Switching back to another font reveals them to be exclamation points after all. Bønkers!

This is all made possible by the HarfBuzz font extensions library. In the name of making custom ligatures and other text shaping possible, HarfBuzz allows fonts to contain Web Assembly code and runs it in a virtual machine at rendering time. This gives font designers the flexibility to render various Unicode combinations as unique glyphs, which is useful for languages like Persian. But it can just as well turn all “o”s into “ø”s or run all exclamation points through an LLM.

Something screams mischief about running arbitrary WASM while you type, but we remind you that since PostScript, font rendering engines have been able to run code in order to help with the formatting problem. This ability was inherited by PDF, and has kept malicious PDFs in the top-10 infiltration vectors for the last fifteen years. [Citation needed.] So if you can model a CPU in PDF, why not an LLM in TTF? Or a Pokemon clone in an OpenType font?

We don’t think [Søren] was making a security point here, we think he was just having fun. You can see how much fun in his video demo embedded below.

Continue reading “Llama.ttf Is AI, In A Font” →

FLOSS Weekly Episode 789: You Can’t Eat The Boards

June 26, 2024 by No comments

This week Jonathan Bennett and Doc Searls chat with Igor Pecovnik and Ricardo Pardini about Armbian, the Debian-based distro tailor made for single-board computers. There’s more than just Raspberry Pi to talk about, with the crew griping about ancient vendor kernels, the less-than-easy ARM boot process, and more!

Continue reading “FLOSS Weekly Episode 789: You Can’t Eat The Boards” →

Decoding Meshtastic With GNU Radio

June 26, 2024 by 20 Comments

Meshtastic is a way to build mesh networks using LoRa that is independent of cell towers, hot spots or traditional repeaters. It stands to reason that with an SDR and GNU Radio, you could send and receive Meshtastic messages. That’s exactly what [Josh Conway] built, and you can see a video about the project, Meshtastic_SDR, below. The video is from [cemaxecuter], who puts the library through its paces.

For hardware, the video uses a Canary I as well as the WarDragon software-defined radio kit which is an Airspy R2 and a mini PC running Dragon OS — a Linux distribution aimed at SDR work —  in a rugged case. GNU Radio, of course, uses flows which are really just Python modules strung together with a GUI.

Continue reading “Decoding Meshtastic With GNU Radio” →

All About CRTs

June 25, 2024 by 76 Comments

For old-timers, CRTs — cathode ray tubes — were fixtures as kids sat in front of TVs watching everything from Howdy Doody to Star Trek. But there’s at least one generation that thinks TVs and computer monitors are flat. If that describes you, you might enjoy [The 8-Bit Guy’s] coverage of how CRTs work in the video below.

CRTs were heavy, took high voltage, and had a dangerous vacuum inside, so we really don’t miss them. The phosphor on the screen had a tendency to “burn in” if you showed the same image over and over. We don’t miss that either.

Continue reading “All About CRTs” →

3D Scanning, Phone Edition

June 25, 2024 by 16 Comments

It seems to make sense. If you have a 3D printer, you might wish you could just scan some kind of part and print it — sort of like a 3D photocopier. Every time we think about this, though, we watch a few videos and are instantly disappointed by the results, especially with cheap scanners. If you go the hardware route, even cheap is relative. However, you can — in theory — put an app on your phone to do the scanning. Some of the apps are free, and some have varying costs, but, again, it seems like a lot of work for an often poor result. So we were very interested in the video from [My 3D Print Lab] where he uses his phone and quite a few different apps and objectively compares them.

Unsurprisingly, one of the most expensive packages that required a monthly or annual subscription created an excellent scan. He didn’t print from it, though, because it would not let you download any models without a fee. The subject part was an ornate chess piece, and the program seems to have captured it nicely. He removed the background and turntable he was using with no problems.

Other apps didn’t fare as well, either missing some of the parts or failing to omit background elements. You may have to do some post-processing. Some of the other expensive options have free trials or other limits, but you can at least try them for free. One of the free trials let you do three free scans, but each scan took about 8 hours to process.

Continue reading “3D Scanning, Phone Edition” →

Injection Molding Using A 3D Printer

June 25, 2024 by 15 Comments

Recently [Stefan] of CNC Kitchen took a gander at using his gaggle of 3D printers to try injection molding (IM). Although the IM process generally requires metal molds and specialized machinery, 3D printers can be used for low-volume IM runs which is enough for limited production runs and prototyping before committing to producing expensive IM molds. In the case of [Stefan], he followed Form Labs’ guidance to produce molds from glass-infused Rigid 10K resin (heat deflection temperature of 218 °C). These molds are very rigid, as the ceramic-like noise when [Stefan] taps two together attests to.

Injection molded bolt, with imperfections on the head. (Credit: Stefan, CNC Kitchen)

The actual injection process is where things get more hairy for [Stefan], as he attempts to push the clamped-shut mold against the nozzle of the FDM printer to inject the molten plastic, rather than using an IM press. With PLA at standard extrusion temperature the plastic barely gets into the mold before solidifying, however. Following this, higher temperatures, different materials (PETG, TPU) and high flow-rate extruders are attempted, with varying results.

Many of the struggles would seem to be due to poor mold design, rather than fundamental issues with using an FDM. The Form Labs document details some of the basics, such as opening up the injection gate (to decrease pressure inside the mold), adding air vents to improve flow and so on. Commentators to the video with professional experience point out many of these issues as well, along with the benefits of preheating the mold.

With the caveat that most of the challenge is in making a good mold, we’ve even injection molding done with nothing more exotic than a hot glue gun. If you’ve got a friend, or a long enough lever, you can even inject the plastic by hand.

Continue reading “Injection Molding Using A 3D Printer” →

The Amstrad E-m@iler, The Right Product With The Wrong Business Model

June 25, 2024 by 29 Comments

One of the joys of the UK’s Electromagnetic Field hacker camp lies in the junk table, where trash turns to treasure in the blink of an eye. This year I returned relatively unscathed from my few days rifling through the tables,but I did snag a few pieces. One of them is a wired telephone, which would be a fairly unremarkable find were it not for its flip-up LCD screen and QWERTY keyboard.

My prize is a 2002 Amstrad E-m@iler Plus, one of a series of internet-equipped telephones from the British budget electronics company. The device itself and the story behind it make for a fascinating tale of a dotcom-era Internet flop, and a piece of hardware that could almost tempt today’s hackers.

You’ve Heard Of The Dotcom Boom, But Have You Heard Of The Hardware?

In the late 1990s, everything was about the Internet, but seemingly few outside the kind of people who read Hackaday really understood what it was really about. I’ve written before on these page about how hype blinded the CD-ROM industry to the shortcomings of its technology, but while that had in reality only gripped the publishing business, the Internet hype which followed had everyone in its thrall. You’re probably familiar with the story of the dotcom boom and crash as startup companies raised millions on shaky foundations before folding when they couldn’t deliver, but in parallel with that there was also a parallel world for hardware. The future was going to be connected, but on what and whose hardware would that connection happen? Continue reading “The Amstrad E-m@iler, The Right Product With The Wrong Business Model” →