Hacker News Reader

Comment by fullstop | Hacker News Reader

1/4th of my cats have microchips. They were moderately annoyed as I scanned them.

The whole microchip registry thing is a mess, though. There's no authoritative database and I'm certain that the database entry for my cat is at some shelter where he was briefly held. I have no way of updating this data without paying a subscription fee, so that's out of the question.

Outside of IR remotes and popping tesla ports, I have used it to emulate RFID tags. I don't have enough free time to really utilize it appropriately.

Cat tax: https://i.imgur.com/8vAabRM.jpeg -- He is sleeping where he really should not be sleeping.

View on HN

https://old.reddit.com/r/CatsOnPizza/

https://old.reddit.com/r/orangecats/

Here's a ML problem for someone to consider tackling ... given a cat picture, identify all of the relevant cat subs that it might get posted in. This could be applied to dogs too... but cats rule the internet ( https://en.wikipedia.org/wiki/Cats_and_the_Internet )

> Here's a ML problem for someone to consider tackling ... given a cat picture, identify all of the relevant cat subs that it might get posted in

I'm fairly certain that there are an infinite number of cat subs, so this task is impossible!

While "yes" ... there are only so many that are big. Getting the top 100 cat subs and going from there would be a nice ballpark number.

Its one of the "this might be a fun thing to do if you're playing with a ML image classification problem."

I've got a Mac and so https://developer.apple.com/documentation/createml/creating-... has tempted me on occasion to see "how much can this thing do?"

you'd have to solve the halting problem?

Nope, that's an issue of psychic barking dogs.

https://everything2.com/title/halting+dog+problem

There's also the incomplete dog issue... https://everything2.com/title/Dog+incompleteness

> The whole microchip registry thing is a mess, though.

That is, unfortunately, correct ^^^ I went through this with my dog. I was told to find out which services your local animal control and humane society use, and make sure your pet is added to those registries. Yes, some charge $$$, but the registries recommended to me were free.

If your pet ends up with animal control, and they can't find the chip registration, getting your pet back can be a nightmare.

And if you move across the country the local animal control / humane society may use different registries.

It's a great idea, in theory, but it's opened up a world where the possibility of scam registries can exist.

I've considered the challenges of an open and public registry, but allowing the public to access it is problematic as there is no way to validate the entries and you would be handling people's contact information. It might be an actual use of a distributed blockchain / public ledger.

Maybe there is no need for a centralized database of contact information, but there could be one for found pets.

Single resource. Any vet/shelter/guy with a RFID scanner can report found pet with this barcode at approximately this location. If you know this pet, contact us here. Presumably only vets and shelters would be adding to the database, so all of the contact information is already public. People who have lost their pets can then monitor this location/sign up for alerts after you lose Fluffy.

Not as great as being able to immediately lookup the owner, but eliminates some privacy concerns.

You couldn't do this with current chips I don't think, but if you had at least 256 bytes of randomness on the chip, you could.

The simplest way to do this would be to use the random data for an EC25519 private key, which would be used to encrypt the data[1] and then sign the encrypted blob plus an unencrypted timestamp. The registry would be a mapping of public keys to encrypted records. Updates could be accomplished by sending a record with a greater timestamp, which would then be propagated to other nodes.

You could also put a DHT on top of that to minimize storage requirements, perhaps also a PoW scheme for sibil resistance.

[1] EC doesn't technically do encryption, but that can be worked around by attaching a public key for an ephemeral keypair for your message, doing a DH against the two keypairs, and using the resulting secret as a key for symmetric encryption.

Or, why have a registry at all? Can't the chip itself hold a few bytes of data for a phone number? I don't know anything about these "chips", but I have to imagine that's possible.

Why would a phone number be useful as a permanent ID for a pet? What if it is rehomed or the owner changes their number? A UUID with a central database or a public/private key with a decentralized database are much better solutions for that.

True, your social security number would make more sense as the primary key.

There is no single authoritative database, but it works kinda like MAC addresses, in that the microchip prefix tells the system who made the chip, which tells them which database to look you up in. You should be able to get the shelter to update the database to match your contact information for you free of charge. At least our shelter was willing to do so. They already have to pay the fee, so why not?

> The whole microchip registry thing is a mess, though. There's no authoritative database and I'm certain that the database entry for my cat is at some shelter where he was briefly held.

Yeah it is a mess, but my vet told me they use this to search across the dozens of registries in the United States: https://www.aaha.org/for-veterinary-professionals/microchip-...

That form is able to find my cat's microchip information in both the registries I have her on, for example. But yes, I was surprised the pet microchip scene isn't more consolidated. Like bicycle registrations are, where the two major U.S. players are https://bikeindex.org and https://project529.com

EDIT: But I was unable to read my cat's microchip with my Flipper Zero, even though my vet confirmed it's still readable using their more appropriate tool for the job.

Reply on news.ycombinator.com