HNHacker News
TopNewBestAskShowJobs

DownrightNifty

117 karma · joined December 29, 2022

https://downrightnifty.me/
submissionscomments
DownrightNifty··on Apple has 90 days to allow app sideloading in Brazil
Do you have a source for that?
DownrightNifty··on Apple has 90 days to allow app sideloading in Brazil
I added it to the tracker:

https://doesioshavesideloadingyet.com/#which-countries-are-l...

Fingers crossed this enables true sideloading and not just third party app stores!

DownrightNifty··on Does iOS have sideloading yet?
> But option like "a possibility" available with hoops to software engineers or tech savvy people can't be.

Okay, then the only disagreement we seem to have is that you think that everyone who sideloads should continue to give a yearly donation of $100 to Apple (which they don't need or want) for the rest of their lives?

That is a very strange perspective. Keep in mind that I would fully understand if you don't think Apple should be forced by governments to provide sideloading for free; that's a separate issue, and I don't explicitly advocate for it on my website. The EU legislation has already passed; I'm just explaining what it means for tech enthusiasts.

> Only in EU I guess?

I'm referring to the "classic" version of AltStore. The EU version only allows you to install Apple-approved apps and as such doesn't present any additional risk for malware. The classic version is available worldwide and it makes use of signing certificates from Xcode.

https://faq.altstore.io/altstore-classic/how-to-install-alts...

> OK. Add couple more hundred bucks. Doesn't change the point.

It absolutely does change the point. You claimed that anyone with the required technical skills can sideload, but that isn't true; they need to be able to afford and justify spending an additional $8.33 every month to access a feature which is free on every other mainstream computing device.

> From my vantage point allowing easy sideloading and alt stores are changes I don't want

I don't want that either, as I already said.

> 1) big corporations who want to run their own stores and not pay Apple a cut for creating and maintaining the hardware and ecosystem

I don't have any sympathy for Epic Games either.

> 3) regular people who want to pirate apps and not pay small-time developers and don't realize security implications

Sideloading doesn't inherently provide the ability to pirate apps. All iOS apps are encrypted and the binaries can't be accessed by other apps; the only way to obtain them is through jailbreaking, and that's something which probably won't exist forever. There's also of course unbreakable cryptographic measures such as remote attestation that Apple could employ if the problem grows into something too bad.

DownrightNifty··on Does iOS have sideloading yet?
[Author of the site here]

It seems we're actually in agreement here :)

You say that any option will be social engineered, and then you go on to explain an option that Apple has already implemented that you don't believe is susceptible to social engineering.

Grandma could theoretically install AltStore, and then proceed to install malicious apps from unknown sources, right now, if she really wanted to. But she won't, because it's too complicated.

I'm not asking for an easy way to sideload. I wouldn't mind if it was made even more difficult than it is now, in fact. All I want is an officially-supported option to do so that isn't prohibitively expensive. An example I provided on the website is Google's ChromeOS. They make it difficult enough to enable developer mode that it's practically invulnerable to social engineering attacks. It's much more difficult to enable sideloading on a Chromebook than it is on an iOS device.

> If you are dead set on it just mentally map in your head the price of a 1000$ device to 1100$ and be done with it

No, it's $100/yr, every year into the foreseeable future, and the cost could change at any time without prior notice. If you stop paying, your apps will stop working, so if you've grown attached to and have become dependent upon your sideloaded apps, it's quite unlikely you'll stop paying.

The Apple Developer Program is not intended as an option to enable sideloading on your device, and it never was. Apple only wants you signing up if you plan on submitting apps to the App Store, which is a benefit you're completely missing out on if you use it exclusively for sideloading.

> "if you really care about it, you can do it" is a fair state of affairs

I would 100% agree, if that were an accurate description of reality, but unfortunately it's not. If you can't afford $100/yr, then no, you cannot do it. It doesn't matter how much technical knowledge you have; you simply cannot do it.

> (I am a bit exaggerating the point but just because it seems like almost everyone is on the other side of the argument)

You and I must not be reading the same website? There is a very large anti-sideloading presence here on Hacker News. Most of it is based on strawman arguments and flawed assumptions, but it's there, and surfaces on almost every thread discussing the topic.

DownrightNifty··on Does iOS have sideloading yet?
Yes, you're correct.

> $100/year is too much

That's the issue, really. If you're paying for the developer program but you aren't uploading apps to the App Store, then you aren't using the program for its intended purpose.

I honestly don't think Apple even wants to collect your $100/yr, which is a pretty insignificant sum from their perspective. It's probably just used as a basic anti-spam measure allowing their reviewers to keep up with the pace of App Store submissions and to do their job properly.

They certainly would be entirely unaffected if they stopped collecting developer fees from the people who only use their membership for sideloading, since that number is most likely low enough to be completely insignificant.

Anyway, I specify this under the "What would it mean for iOS to have sideloading?" section:

> For the purposes of this website, an operating system supports sideloading if it provides some means for technically-inclined end users to permanently install native apps without prior approval from any company or organization, either free of charge, or for a reasonable one-time fee. [...]

> Apple would not have to make sweeping changes to iOS in order to qualify; in fact, a minor policy change would be more than sufficient. For example, all it would take to qualify is for Apple to remove the unnecessary expiry dates from development certificates (which can only be used on the creator's own devices, and cannot be used for distribution). They currently expire after 7 days for free accounts, or 1 year for paid accounts.

---

> but if you’re paying $1000+ every 2-3 years...

I think this is unfair. iOS devices receive updates for at least 5 years, and they don't suddenly become useless after that point either. I for one don't upgrade my mobile devices until they stop getting updates, and I don't find myself yearning for the minor yearly upgrades we've been seeing lately (across the industry, not just Apple).

DownrightNifty··on Spotify's Car Thing, due for bricking, is getting an open source second life
> Spotify had previously posted the code for its uboot and kernel to GitHub, under the very unassuming name "spsgsb" and with no announcement

Interesting. Why not make some tiny announcement about it somewhere?

DownrightNifty··on Low Cost Mini PCs
This is super cool! Now if only my ISP would give me a static IP address so I could expose port 51820 on one of these things and life would be perfect.
DownrightNifty··on Unlocking the Pixel 9 bootloader breaks some Pixel AI apps
Oh wow, that's interesting, thanks for sharing. I was experimenting with the Pixel Recorder app, which calls into the AICore system service, which calls into the Protected Download API to download its models. Based on the URLs ("pixelai-models" vs "aicore-models") it looks like Pixel Screenshots is doing its own thing and not using AICore.
DownrightNifty··on Unlocking the Pixel 9 bootloader breaks some Pixel AI apps
I took a brief look into it, and they use the Google Private Compute Services Protected Download API to download the Gemini Nano model, which uses remote attestation to cryptographically ensure that your device is running the stock ROM: https://github.com/google/private-compute-services/tree/mast...

I'm not sure why they do that, tbh, since Gemini Nano is now part of Chrome, and you can download it without remote attestation. If you were dedicated enough, you could probably force the AICore system app on a rooted device into using the Gemini Nano model you downloaded from Chrome. I briefly attempted this but it proved too annoying so I gave up.

Edit: It appears Pixel Screenshots works differently than Pixel Recorder, which is what I was looking into.

https://news.ycombinator.com/item?id=41373011

DownrightNifty··on Google Chrome has an API accesible only from *.google.com
JS on Google Sites, Apps Script, etc. runs on *.googleusercontent.com, otherwise cookie-stealing XSS happens.
DownrightNifty··on Ask HN: Is there any software you only made for your own use but nobody else?
To anyone in the US who said they make their own apps for their own iPhone: do you really pay Apple $99/year for that "privilege", or do you reinstall the app every 7 days (or use workarounds like AltStore non-PAL or jailbreaking) to get around this?
DownrightNifty··on Nintendo ends Wii U repairs
The hacking community has already taken DIY Wii U repair to a level beyond even what Nintendo was likely providing.

For example, someone wrote a recovery menu that leverages an exploit in the USB stack and executes before the OS loads [1]. Many Wii Us also have SK Hynix NAND chips with high failure rates, which under normal circumstances would eventually result in a bricked console, but hackers found out how to replace it with an SD card and rebuild the internal storage from scratch onto it [2].

Hackers also revived the console's online service [3]. This was up and running even before the official service shut down.

All very impressive achievements from the hacking community around a somewhat niche and commercially failed console.

[1] https://github.com/GaryOderNichts/recovery_menu

[2] https://gbatemp.net/threads/how-to-upgrading-rebuilding-wii-...

[3] https://pretendo.network/

DownrightNifty··on Show HN: Code to run Gemini (Nano) locally on desktop/Chrome
You should share this with us! If you have a blog, that'd be a really interesting post.
DownrightNifty··on Show HN: Code to run Gemini (Nano) locally on desktop/Chrome
Instructions didn't work for me. Tried on Windows 10 and macOS 14.3 (x86) with Chrome Canary. The "On Device Model" component doesn't show up.
DownrightNifty··on ChromeOS will soon be developed on large portions of the Android stack
The issue you linked is more or less resolved at this point. If you find Termux processes being killed, just run this command over adb shell and you're good:

    settings put global settings_enable_monitor_phantom_procs false
Google added this setting in response to a bug report submitted by a Termux maintainer. It's also persistent, so you only have to run the command once.

The more pressing issue at this time is: https://github.com/termux/termux-app/issues/2155

DownrightNifty··on Better adb shell
Each app also has its own private storage space as well at /data/data/<package name>/ which the end user cannot access even through adb. It's owned by the Unix user representing that app and has 700 perms so the `shell` user can't access it. While you can't access the directory directly, you can clear it in Settings -> Apps -> <app> -> Storage.
DownrightNifty··on Better adb shell
From one of the SO comments: https://android-review.googlesource.com/c/platform/system/co...

It was removed :(

Kinda funny that the commit message doesn't say anything like "remove hell mode", it's integrated into another unrelated patch

DownrightNifty··on Wiretapping Google smart speakers
Hi - author of the article here. I actually did try to accomplish this attack without reverse engineering but it wasn't possible. The Google Home app would not allow you to connect to a device through its setup mode AP if you weren't using an account that was previously linked to the device (while it was online). You'd get this error if you tried: https://i.imgur.com/ipC3OhE.png

Behind the scenes you see (via MITM) that it tries to connect, but the device rejects the request because it doesn't have the local auth token (which can only be retrieved using a linked account).

This is a good design: the device owner (or any other previously linked users) are allowed to reconfigure the device while it's in setup mode if this is ever needed, but unauthenticated users cannot. And if the device owner loses access to their account somehow, they can use the factory reset button on the bottom of the device to set it up with a new account. (Unfortunately, it was broken, as I described.)

So I think the main "bug" here was the exposure of the "cloud_device_id" and other sensitive device information required to link an account in setup mode, even though it's not used by the Home app to reconnect a device to Wi-Fi. My suggested fix was simply to stop returning this info in setup mode. Last I checked this behavior hasn't changed though so I guess it was easier to make changes on the server side to fix the issue.

Btw, they later updated the Home app's UI to provide a more descriptive error message than just "could not communicate": https://i.imgur.com/ISWk4Bj.png

← PreviousPage 2 of 2