Do I understand correctly that this has all the building blocks which make up various lossless compression schemes (dictionaries, run length encoding, delta encoding, etc) and kit bashes them together for your specific dataset?

I was under the impression that modern file formats and compression schemes already kind of do that as part of the normal compression process, so I was a little surprised by how much they beat some of the competition in their benchmarks.

Well it could, but since they never tell you what kind of system they’re running, you’re always left to assume that it was specifically prerecorded, built and fine tuned for the track at hand. Which as you said, doesn’t mean shit.

Passend zum Thema ein 1:11:11 langer Videobeitrag über ausschließlich Fächer (und rotierende Fächer) [Achtung Angelsächsisch]

If you don’t mind me asking, what made you go with VoidAuth vs Authelia or something else entirely?

I’m in a similar boat as OP and while VoidAuth looks very promising I’m put off a little by the young age and size of the project.

This fixes a lot of the major pain points that pushed me to using Typst in the first place.

It’s too late for me to come back but hopefully this makes someone else’s life a little easier.

Harper is great, but it only does English. Really wish their “extensible core” was extended to other languages sometime.

A very interesting tool so thank you for posting this.

The docs are very impressive and detailed which is nice.

How did you arrive at 8GB hash table size?
I’ve tried understanding the example table in the docs but it doesn’t make any sense to me.

Alternatively maybe something like the Roetz-End which can change colors without swapping filaments by rotating and/or stopping the flow of one of its inputs.

I just wish that they wrote more articles. Their writing style is superb. Can’t argue with this though:

More coming soon. Or not. I don’t owe you shit.

Not to mention that this simplifies the problem so massively that its almost a pointless metric.

This is like analysing traffic jams by counting the number of registered cars.

Sure there is a correlation but not a very meaningful one.

Don’t get me wrong this problem is an important one, no arguments there, but this says nothing about collision risks (in certain orbits of which most don’t impact each other) or orbital parking capacity (as in how many objects can we safely add).

QR codes come with (differing levels of) error correction. You could decode it and take the amount of corrected errors as a score.

I don’t know what unit/chunk of data the error correction works on but you could alternatively dive a bit into information theory and calculate how close any of the chunks are to an uncorrectable error and use that as a score.

Since the first uncorrectable error will cause the scan to fail this would match your problem definition fairly well.

One benefit of this approach would be that it’s a single scan and done vs searching n permutations of the same code.
Or maybe you’d want to scan it a couple of times to rule out any additional errors introduced by the camera or environment.

Randall’s timing is impeccable sometimes

Nicht billig aber wenn man mal die Features mit den Elite/Pro Controllern der Konsolenkonkurrenz vergleicht ist er plötzlich doch recht preiswert.

Schreibt man das Wort heute so?

Ich gucke mal lieber nicht in den Duden

Why not? And you probably need the additional traction for that kind of acceleration out of corners

I assume you’ve used ESPHome for the firmware?

Did you have to mess around with deep sleep or is the battery + solar combo so overkill that it doesn’t matter?

Anyways very cool project. Thanks for sharing

Metro Wolfenstein

Got it working thanks to your troubleshooting tips now. Also found a very neat way to handle secrets from another comment.

I tend to run a DB instance per service as that makes backup restoration much easier for me. An idle postgres sits at around 50MB which is a cost I’m willing to pay.

Thank you again for your help :)

Managed to get it working by passing in env vars from a secret now.

ArgoCD has a really handy web UI that allows you to quickly see what kind of resources get deployed.
Especially for learning k8s I found that much easier to visualize than raw kubectl outputs.

Passing in the secrets once via the global: section is very neat. Got it working now with a few of the other tips and stole your trick for my secret handling. Thank you :)