It depends on what they want. The op and title are pretty vague and general, so I offered a very easy solution that could meet their needs.

That’s exactly what I mean. It all depends on what makes a company big tech.

If it’s just having lots of money or capital then it’s really hard to divest oneself because of the phenomenal heights of industry required to make chips. Like, on some level you gotta accept that no one’s doing 14nm lithography in a shed out behind their house (people who do home lithography are in the micrometer range last I looked).

LPIC-1 my butt, op!

You can always just use an m series mac if you don’t want to support Intel or amd.

It all depends on how you define big tech.

The framework risc v board, for example, uses the StarFive JH-7110 which is manufactured by tsmc. Doesn’t get much bigger than that!

Specs for what?

The hardware you run it on will have a range of specs.

The model you try to use will have a bunch of measurements that can be called specs.

Once you pick some hardware, the measurements of the model you run on it can be adjusted for “accuracy” or speed.

Of course, the range of hardware that’s available to you informs that calculus.

The alternative to answering this post is to get a Mac mini or whatever they’re selling as a desktop llm device now and not mess around.

Yes it works good if you read first and plan and follow the instructions. If you just yeet it into your computer you will have problems.

You may end up doing a reinstall as opposed to a patch or upgrade.

You may find that dynamic wallpaper not working, needing to update with command line, photos app can’t edit and needing to use a third party browser is an okay set of tradeoffs to get to sequoia (macos 15) which still gets point updates and security patches (grimly: for now).

If I was using a 10,1 I’d back up my system and jump straight to sequoia with oclp. If you can get by long enough to save up for a more recent computer (m1+) then that’s a win. Sequoia is the last version of macos with support for Intel processors. Now is a bad time to be running old versions.

E: before oclp the linux wisdom for these models was to go in with a soldering iron or a chip clip and disable the dedicated gpu in order to avoid problems with it. Often it would break the hdmi or DisplayPort or whatever video out port was on the side of the laptop because while the integrated display could run off the intel processors gpu the video out was hardwired to the dgpu.

Your computer is MacBook Pro 10,1 and pro 15” retina mid 2012 a1398.

The relevant technologies are ivy bridge, intel hd 4000 and nvidia Kepler.

According to opencore legacy patcher, you can use Ventura (osx/macos 13)with broken live text if you go the opencore route.

That will most likely offer a better experience than linux on your hardware.

If you were to provide concrete proof that these devices that are currently on sale are vulnerable, then we could have a discussion about that https://consumer.huawei.com/en/phones/

The xiaomi 15 ultra that was released in February of last year and still in current production and offered new (and at new prices!) uses the explicitly claimed as brute forceable qualcomm snapdragon 8 elite.

You asked for huawei though, and their most recent stuff like the mate 80 pro and whatnot runs the kirin 9xxx chips. Kirin chips have historically been considered pretty trash security wise, but a lot of that is from people’s experiences with the three digit families of soc from 2020 and before. they’re supposed to be getting better since the 8xxx and 9xxx series. Still, the 2025 leak table 2 states “Huawei (Kirin/Qualcomm/MTK)” are partly brute forceable in cold state and fully brute forceable in hot state. Considering the 8xxx and 9xxx chips had been out last year at the time of the tables publication and the way that pixel devices are treated in that same table (big red “not supported” X mark actually means sometimes it’s supported depending on the precise version and what you’re trying to accomplish) I think it’s safe to say that by 2025 there was compromise on the 8xxx or 9xxx Kirin chips out there at that time.

Which would of course encompass the x6 and p60s listed in huaweis current lineup.

Remember though that I’m not claiming these leaks represent the extent of le capability today, but the extent of their capability then. Over time we can expect (and can see based n the expansion of their claims and the capabilities asserted in their leaks over time!) that they would get access to new methods of compromising phones, we just can’t know the exact extent until something leaks.

Again, I am trying to show an evidence based analysis as opposed to the one you’re suggesting that relies on assumptions. There’s nothing wrong with the way you’re looking at the world, but when actual evidence is present those ideas have to be examined and maybe even changed to accurately reflect the reality we see.

Phone security analysis with Chinese characteristics lol.

I literally just pointed out that while there is a note about a four year old device on the 2025 leak, there is also brute force support listed for socs that came out just a couple of months before the documents date.

I pointed it out in the post you just replied to.

My assertion is that pixels, graphene and iphones are safest against the hardware/software tools the police have. I’m using leaks of the capabilities of those tools to back that assertion up. While a leak from 2024 or 2025 might not have phones from 2026 on it, it provides a really clear picture of the capabilities of the police at those points in time.

I don’t think saying “look, there’s evidence that pixels, graphene and ios were best at foiling the cops a year ago and a couple years ago and five years ago and ten years ago, it stands to reason the same situation is true now” is all that contentious.

In lieu of actually having access to that software database to look up specific devices and os versions, which neither one of us have, the leaks seem like a pretty solid basis upon which to establish an understanding of police capabilities. The alternative that you’re proposing is literally assumptions.

So what would convince you of my point? It’s vanishingly unlikely that they’ll be a leak soon that will let us talk about the current latest and greatest, but would a leak that claims the ability to brute force a phone that was new back when the leak is dated be convincing to you? Do you need explicit model numbers or is just the chipset/family enough?

It’s really tough to hear you imply I’m not reading the things I post when I made reference to the same p40 model fact earlier today.

In that same page, on the bottom row the same column that had the p40 comment referenced recently added support for the snapdragon 8 elite.

On the unlocked devices support matrix (they become unlocked devices once brute forced) support for the dimensity 9400 is referenced.

Those are both chips used in q4 ‘24 and forward phones and the cop hardware brute forces and extracts them in February of ‘25. That’s not ancient at that time by any measure and not even ancient by the standards of today.

On that same page support for private space and 2nd space are referenced (those are the name for containers that harmony and hyper use) indicating support for extracting and decrypting harmonyos and hyperos containers indicating support for cracking harmonyos and hyperos. I pointed this out earlier today.

The fact that older devices have notes on them does not mean newer devices are not supported.

We are indeed going in circles.

Okay, to address just the request for proof that the security order is graphene, stock pixel then everything else, look at the 2025 link I posted, table 2, android access support matrix - locked devices:

Huawei - cold and hot brute force extraction at least partially supported

Pixel - second column of both cold and hot sections, brute force password to decrypt user ce and brute force password are marked not supported.

Xiaomi - bottom row, hot and cold extraction and brute forcing supported.

So there you go, pixel over xiaomi and huawei. If you need proof that graphene is at the top, compare the standard android and graphene columns in table 3: android os access support matrix - google pixel in the same most recent link.

I understand that you’re saying something else may be out there. You’re right, unknown security vulnerabilities might be around that have serious effects on non Chinese phones.

Im saying there are security vulnerabilities in the hands of police which are least effective against iphones, pixels and graphene and that it’s best to choose devices based on what you know as opposed to what you assume.

It really seems to me like I’m posting exactly what you say you need to see over and over again, what am I missing?

So we don’t have access to the most up to date information on the most recent iteration of cop hardware and software for breaking into phones.

If either one of us did, it would be a very bad idea to say that we were basing our argumentation on that.

Based instead on historical data points, like the ones I’ve provided, we can consistently see that at those points in time the latest stock pixel and iphone devices were preferable to anything else including the latest huawei and xiaomi devices for the purposes of avoiding cop access.

The reason I’ve been engaging with your argument that the latest xiaomi/huawei stuff is preferable to the pixel/iphone equivalent is that I’m concerned someone worried about cops might make a decision about how to use their limited resources based on that argument.

They’re not bad phones or bad companies and China isn’t a bad country. There’s just real world evidence that the devices aren’t as secure as some alternatives.

My whole point separate of yours (that I’m paraphrasing here, apologies) that devices made outside the amerisraeli apparatus are inherently safer is that we need to pay attention to the wealth of information about phone security rather than base our decisions on assumptions.

Heres yet another data point to go by, in this case presented as a blog analysis of the 2025 leak (only a little over a year old at this point!). there’s a million great bits of knowledge in that page if you’re interested in learning a ton about android security but the long and short of it for android devices is about the same as before: graphene is at the top, then stock pixel then literally everything else.

The point of using the company claims from 2021 and the famous leak from 2024 was to establish a pattern, not to suggest that information from that long ago could be applied directly to this present moment.

The pattern is: android devices and old iphones and pixels that need to be updated are vulnerable no matter the country and alliances.

It can be established because the company’s claims forever were “we can do this” and finally there was a leak showing in pretty good detail that they weren’t lying.

here is the entire pdf of the leaked android support matrix linked in the article I linked to above. It only goes model by model for pixels, partly because it’s illustrating their capabilities against graphene as opposed to stock android but likely also because there’s thousands of android phones and a model/os version matrix would be insane and have an endnotes page a mile long.

Page 3: supported extraction listed for android devices by chipset includes huaweis ostensibly non amerisraeli hardware stack. Secured container extraction supported for both huawei and xiaomi implementations present in harmonyos and hyperos.

Page 4: huawei and xiaomi devices listed as brute force able in both on and off device states (there are notably some exceptions here, some qualcomm chipsets take a day and the p40 phones with their software updates weren’t brute forceable).

Page 5: huawei and xiaomi, realme, oppo, oneplus and zte are listed as brute forceable in off and on states.

here’s the same thing but for ios. I’m tired and it gives much more precise detail on a narrower range of devices, so I’ll just summarize:

Six year old iphones running a month old os version were safe from even “after first unlock” (the most unsafe locked state) compromise.

some phones running the one and a half year old ios in a “before first unlock” state (the most secure locked state) were subject to a brute forcer that was limited to 5000 attempts per day. That sounds like a lot, but a six digit pin would need constant hammering for 200 straight days. Not unheard of but a very far cry from the “plug it up, get access” that is advertised, documented in many security outlets and something I have literally observed happening inside a cop car at a protest.

So to summarize: as of two years ago, the ostensibly non amerisraeli tech stack was not secure against the cops. Harmonyos and hyperos were not secure against the cops. Out of date ios and pixel phones were also not secure.

I want to make clear that what is explicitly shown with regard to pixels and iphones is definitley true of all device families: the newest stuff has fewer vulnerabilities because people just haven’t been pounding on it as much.

Again, this is intended to help people to make good choices using real world information as opposed to predictions. If I were buying a phone to resist the cops, it would be an iphone or a pixel with graphene.

Apologies for the late reply, sometimes I’m not in a good spot to chase down leaks about cop shit.

As of at least 2021 cellebrite claims they have the ability to access xiaomi and huawei devices, listing explicitly the soc and baseband chips used by the very phones you are claiming are safe because of their lack of amerisraieli death pact ties.

A famous leak of their support matrix from 2024 confirms this and also explicitly groups android devices by their soc/baseband chips. Reenforcing that the chips ostensibly with no ties to the amerisraeli death cult are not any more secure or private than ones with those ties.

The point of those is to quickly draw a line that connects the past to the present. We see the same claims, then the affirmation of those claims reported by a third party.

I think in a vacuum, assuming perfectly spherical semiconductor manufacturing industries and leaving software out of the picture, the point you’re trying to make is the most materialist take: you can’t trust the imperialists tech, the masters tools cannot be used to tear down the plantation, etc.

In our present day with a hundred years plus of semiconductor manufacturing history encompassing real countries whose attitudes towards one another and development have changed significantly during that span, given reliable information about the explicit capacities western (and lest be serious here, any) le or intelligence apparatus has, the most materialist take is that there’s more to the choice of what device to trust than where the chips come from.

To butcher a car metaphor, what you’re saying is similar to people claiming buying and driving a Tesla is better than a BYD because you can’t trust Chinese tech. That idea might be fine (or chauvinist) in a bubble but when we can evaluate the Tesla and BYD for ourselves in a parking lot or on the road we might come away with wildly different ideas.

Technology has to be evaluated based on its capabilities and how it’s being used when that’s possible and I would argue it’s extremely possible in the case of security in phones and that if you think you’re gonna be scooped by the cops you need to be on graphene, the latest ios or maybe a pixel with the latest android.

From cellebrite’s own documentation (on the first page, of a sales pdf, which was the second or third google result):

Supported devices include Huawei H1611, Xiaomi Mi 5, ZTE Z832 Sonata 3 and ZTE Z981 ZMax Pro

I’m, again, not as familiar with huawei and xiaomi product lines and whatnot as I am with the iphones and pixels so I can’t speak to the popularity of specific ones implicated in just that bullet point and the doc I quoted from is at least seven years old, however I do know that many more chinese devices are accessible with these cop metasploit tools.

The idea that backdoors can be grouped by what nation state intelligence apparatus has control over the manufacturing of the device in question is good reasoning when we have no other information to go off of. In this case though, there is a wealth of information public, leaked and from people who just can’t help but warthunder their classified documents in fights online.

I would never suggest American/israeli tech power should be accepted as a net positive or reasonable compromise. What I want is for people to critically and carefully consider the devices they trust based on what we know about intelligence apparatuses ability to compromise them as opposed to the fog of information war.

I mean, if your goal is to make people as safe from the police as possible as simply as possible then the cellebrite and graphite leaks would steer you towards the last few generations of pixels and any iphone that can get the latest os.

That’s not to say graphene isn’t a fantastic choice, I use it daily and it’s secure from law enforcement hardware, just that the leaked capability matrixes consistently indicate that cops can’t break into appropriately secured iphones and specific android phones as well.

Which is really useful knowledge to have and build your behaviors around that would be completely missed if someone were to base their choice of device around what doesn’t have Israeli connections first.

That’s a pretty recent change, but I take your point that chip manufacturing with no ties to Israel may exist.

To the main point, while it could be a good idea in theory to avoid devices with chips tied to Israel and the us, in practice that would lead people away from iphones and pixels that do best against graphite and cellebrite which would be bad.

You have a bunch of answers including one you chose, but you can always rebind the tmux key to something compatible with your terminals search and use [key] - d to disconnect from your tmux session then invoke it with tmux attach-session to get back into your session.

on attach it should clear the screen, print the stdout buffer it accumulated and give me stdin prompt, that’s it

The last part of the above, disconnecting from tmux and reattaching would do what you described. That may be useful if you need to work with tmux.

Huawei famously ran an Israel office for a long time, idk about xiaomi but smic usually works in nations they’re sanctioned in through shells like (at least in the past) Israel microwave company.

The point isn’t to find the one pure company but to recognize the reality of semiconductor manufacturing and development at this moment being insanely integrated across every imaginable border and take a more nuanced and serious view of the impact a chip could have on your life than “chip tied to Israel, chip and stuff it’s in bad”

That’s not to say your replies could be reasonably distilled down to that mischaracterization, just that I’m hoping people come away thinking they should think of more than weather something has a tie to the bad country.

I don’t know of any modern soc type semiconductor manufacturer without some Israeli connection at least on the level of what’s described in this article “ cooperation with engineers from … development center in Israel”.

Open to being wrong here but I think even manufacturers of non-soc ics like op amps and ttl and discrete components like transistors and diodes have this degree of connection (I’m thinking specifically of onsemi who make all the Fairchild stuff here!)