[utopiah]

"Vibe reverse engineering" yep, was talking with a designer friend just yesterday who vibe coded a driver for his unsupported Wacom. He's not a developer but his pen tablet now works. It seems quite useful and efficient for this kind of work with a well established process (plug, get connectivity, do something, get data, do something else, get other data, transform that data (e.g. 0..1 mapped to a well defined resolution), validate) ideally over standards e.g. USB or BT capabilities.

Best of all he even published on Github the result so that, hopefully, others benefit from his effort without even having to do it again.

[Arodex]

The author explained that the LLM needed a lot of hand-holding and that it was only possible because he himself is already an expert at it.

Quite a different message than yours.

[utopiah]

I'm speaking on behalf of my friend. I don't want to misconstrue his message. Even though not a developer himself he is familiar to the command line and can for example manager backend stuff like containers. In that sense he's not a random computer user. He also dared to try. If he writes a blob about his attempt I'll share it back here.

My point was to share a related anecdote on using LLM to making seemingly unusable hardware usable again, which seems feasible (hence sharing the anecdote to confirm from another source) and IMHO a positive use case, while so many others are not necessarily so.

[jchw]

100% believe this one. The good thing about Wacom devices is that they're actually quite simple to deal with. All of the Wacom devices use a simple packet-based format that, while it varies per model, is quite well documented, what with being in the Linux kernel and several user mode tablet drivers (including OpenTabletDriver and TabletMagic.) You really don't need reverse engineering for Wacom, from old serial Graphires to the latest Intuous'. The knowledge, or at least certainly knowledge of where to look, should be baked into any frontier model.

When I was in college I had an ARM Chromebook which didn't have the Wacom driver, so I wound up trying to use the then-new Chrome USB APIs to make myself a tablet driver in a Chrome extension. This was long before LLM coding was a thing, but thanks to the Linux wacom.ko it wasn't really an obstacle.

The biggest problem I had was that while I was decoding digitizer inputs just fine I had nowhere to put them. I tried making a simple painting app in JS and it worked but without having native cursor movement it was just too jank and I gave up.

I eventually uploaded the code for posterity sake. I doubt it works at all anymore, even with the specific tablet it was hard-coded for. But it's still online, anyhow.

https://github.com/jchv/crwacom

[utopiah]

Neat, thanks for explaining and sharing the code too!

[WillAdams]

Why not just use TabletMagic?

[utopiah]

No idea, sending him the link I'll see what he says about it, thanks!