>chaining together a dozen dilapidated second-generation iPhone SEs and harnessing Apple's Live Text optical character-recognition feature to find possible inventory tags
I think Tesseract is the smarter/faster/less obnoxious choice if you're not trying to parse weird meme text like the blog is doing. There's almost certainly a better paid option available in our enlightened AI age, but I don't even think you'd need AI for this use-case.
Last time I used tesseract (a year ago?) it’s still pretty useless if your text isn’t on a clean background. It doesn’t even come close to Apple’s proprietary on-device OCR.
I think most frontends to tesseract employ a lot of these methods and maybe more... but trying to use tesseract directly can indeed be difficult without extra processing of the image first.
I know, I tried many things with the photo collection I was working with, including advice from that very page, generally to relatively poor results. (I ended using Apple’s framework on macOS.) The point is tesseract is definitely not “smarter” in any way, at best it’s on par with Apple’s OCR when you hand it very clean text.
The Apple framework is much, much better than Tesseract, and quicker as well. It is really good. Of course if you don’t need on-device processing, then there are cloud services that are better.
> I think Tesseract is the smarter/faster/less obnoxious choice [...] There's almost certainly a better paid option available in our enlightened AI age
It would have cost $375,000 to use cloud OCR for this project. Mandatory is absolutely a baller, but not crazy enough to spend that kind of money on the project.
If you can get Tesseract to generate comparable results with sub-optimal images from eBay listings, I'd love to know more.
Some of these developer devices get 'destroyed' and sold as scrap. dosdude1 has restored some of these kinds of devices to working order. There's pretty neat video of the restorations:
This is why solutions like Bitlocker with a good TPM or FileVault are so important.
They can essentially guarantee that the disk encryption key will only be released from the security module if the computer is running a fully-trusted and signed OS. Even if you take the drive out of the machine, the data on that drive is completely useless to you.
Incidentally, this is also what makes short PINs secure; the TPM contents are unreadable, even to a skilled attacker, so if the TPM is guaranteeed to wipe itself after 10 tries, even a 4-digit PIN is secure enough.
the TPM contents are unreadable, even to a skilled attacker
Depends how "skilled". Nation-state level? Most definitely not. "IC break" services in China? Maybe. AFAIK TPMs are based on similar secure-processor designs as the chips in payment cards and other smartcards, and even those with enough determination and $$$, or the right equipment, will get you through.
> After he evaluated the Time Capsule's contents, Bryant notified Apple about his findings, and the company's London security office eventually asked him to ship the Time Capsule back.
> Bryant again reported his findings to Apple and returned the Mac Mini to them.
Why the hell did he do that?! It's, like, the worst thing one can possibly do with these kinds of devices. Just publish stuff that doesn't have anyone's personal data in it. That'll make the world better in the end.
First of all, he could release it anonymously. Second, what law or contract would he be breaking anyway? An NDA with Apple that he never signed? Sure there is a person who breached an NDA in relation to the proprietary information contained in these devices, but that person is whomever was tasked with decommissioning them before resale.
Morals towards faceless corporations that would do anything and everything to keep their profits growing indefinitely? No, I don't realize why anyone would have those.
I worked several years within Apple in one core engineering roll, in a sizeable team, and being ethical and moral was a _huge_ part of the engineering culture (at least), both when Steve Jobs was there and after.
Seconded, there's no tangible nor financial benefit to him for releasing the information to the world. Also: how would publishing the data "make the world a better place"?
At the same time it'll incur a non trivial amount of reputational and professional risk.
> Also: how would publishing the data "make the world a better place"?
For example, schematics of Apple devices would help people fix them on a deeper level than Apple wants (Apple doesn't do board-level repair, one 3-cent component fails and you're getting your entire motherboard replaced). Diagnostic software would help with that too. Documentation about any artificial limitations Apple imposes on these devices for its own profit, like part pairing, would make these limitations easier to bypass. Documentation about software or proprietary network protocols would help with adversarial interoperability. Even documentation on manufacturing techniques might be useful for someone building hardware — if not to copy, then to learn from it.
IMO, it's a personal philosophy. Similar to why hackers choose to report vulnerabilities to bug bounties vs. release findings on sites like Hack Forums.
We all know companies are predatory, and in many cases companies (looking right at you Google and Microsoft) continue to refuse to pay people for discovering, documenting and reporting high-severity vulnerabilities. That doesn't mean we as individuals forfeit our principles and become just as corrupt as the "faceless corporate entities."
For vulnerabilities I can at least understand that — most of them can be used nefariously, and you also get a shitton of money if you report it. Publishing some company's trade secrets, though, would only hurt their bottom line, which isn't necessarily even a bad thing for a company that has orders of magnitude more money than it knows what to do with.
Would you be willing to publish the iOS OCR server you made? It would be greatly useful in some of my products, as as you’ve noted other options have either low-quality results (tesseract, some cloud-based solutions) or are expensive in comparison for large amounts of images (most cloud-based solutions). That and it’d allow some of us to put our old phones to use.
Here's something i've used a few times. It's not using the phone, but rather a mac mini that's sitting in the corner. Quickly made this repo -- mind the mess.
the only things anyone ever wanted to know from apple is their aggressive business tactics... and most of that is already public thanks to the many processes they lost along the way. from labour salary fixing across industries to pushing obvious monopolies in the face of the publishing industry.
I think the only piece I'd pay to read is how they negotiated with spotify.
This is the second time I've read about an iPhone OCR rack https://findthatmeme.com/blog/2023/01/08/image-stacks-and-ip...
Is this still state of the art in terms of local OCR?