I'm just looking at an old Dell laptop refusing to boot without manual intervention because it looses its time setting since I removed its main battery. If I can fix that it'll become a very capable server for my home setup. Not sure it'll work though, as the bios can't be configured to ignore the time reset.
All old office laptops I had so far still had the good old CMOS/RTC battery... So if indeed the RTC is solely powered by the main battery, I would guess there's at least a backup capacitor somewhere so that it survives at least a short disconnect? If you can find that one, you can probably connect power there, but that's just me guessing...
I tend to compile everything from source on Linux. I don't want to be
tied down to any distribution. So from my point of view I'd agree - I
understand why curl/sh is used, it is super convenient, I also use it
for installing Rust. But I also dislike it. (Naturally for Rust I need
rust for the bootstrapping, so I understand that solution. But I still
dislike it too.)
I also would like it to have stable supported libraries. But this doesn't seem to be a focus. It's a pitty, we have a Foss solution and it's hard to integrate with it....
I've learned rust when llms appeared and it helped me developed while I learned. My use of llms evolved (https://asfaload.com/blog/ai_use/) and is still evolving: I'm currently at a point where I write most app code myself and use llms to review it. Test and doc I like to delegate to llms, and in that I'm the reviewer.
I think I've found the middle point that fits me: llms have an added value, but I think I avoid most of their idiosyncrasies.
Just mentioning because you explicitly said builds are unsigned: I'm working on a (open source) signing solution supporting GitHub releases that is easy to use for both signers and downloaders. Let me know if this would interest you (not immediately putting link to avoid spamming).
- multisig approach: you can require multiple signers to sign a release for it to be accepted by downloaders. Protects against account compromise. Sigstore relies on OIDC, but this makes it often reliant on megacorps (and harder to self host, see next point)
- very easy to self host. You can deploy it internally without any problem, eg in an airgapped environment where you want to distribute signed artifact. This is impossible with Sigstore.
- accountless: the key is the identity. The multisig approach diminishes the problems that long-lived keys might bring. And you don't rely on an external OIDC provider to be able to sign.
- Air-gapped signers: it is possible to have air-gapped signers, as signing does not rely on an online service to verify an identity.
- easy for downloader: the use only has to provide the download url, and this is sufficient to authenticate the download.
- easy to audit: the backend data is stored in a git, that will be mirrored in different places, and anyone can mirror it for themselves.
You might agree or not with these choices, but there are enough differences to say that it doesn't reinvent the wheel, it's providing another approach which I think has its advantages.
The solution I'm working on is https://www.asfaload.com. To get a quick idea, see the videos, incl. demos, at https://www.asfaload.com/videos/. The code is at https://github.com/asfaload/asfaload. There's no binary executable published yet, so you'd need to compile. However, I'm very interested to help you on the setup and get your feedback. Don't hesitate to contact me by mail on the address found in my profile ( ${my_hn_username} gmail address) and I'll be happy to help!
You describe an exhaustive documention, but some documentation would already be better than none. Perfect is enemy of good, as your comment illustrates.
Difficult to say at the moment, but there might be some speculation as to why.
One might be that they dont security patches reverse engineered and vulnerabilities to come out faster (some critical and high severity fixes in GPU drivers/bootloaders were often delayed), and this decision was done long before LLMs were considered powerful/useful for vulnerability research.
Second reason might be simply "Control of the android ecosystem", Google may just want to build a wall around android and make it frustrating for other vendors to compete.
Third, again, this is only speculation, is the current push for electronic ID in the EU and other countries, as well as DRM/copyright protections for media and locally ran LLMs (we've heard of how google pushed small LLMs with chrome updates), if they can do the same on some high end android devices, Google would be more invested into further locking down Android devices.
reply