phh

init could chose to launch every process in a new mount namespace, which would break this. I have no idea whether it does that, it probably doesn't, but my point is that as long as it's not released and open source, it's not worth looking at.

As an author of an old root for Android, and of a modern generic custom ROMs, and other Android OS stuff:

The title is, and forever will be wrong. When we say you're root in Android, you're actually root. You can actually do whatever you want [1]. Magisk (the now modern "root" for Android) now includes stuff to even "edit" Java code, so even if it's hidden deep somewhere, you should still be able to access it. (Even if somehow it moves from Java to native code, we'll still find ways, don't worry)

The fact that the author didn't manage to do it doesn't mean it's not possible. I could guess what's the author issue (I have two ideas in mind: 1. it requires stop;start to restart zygote, because zygote cached CAs, 2. it needs to switch to correct mount namespace before doing the commands), but I won't try it, I got tired of working on closed-source Android stuff.

> More investigation is required and it's hard to know the full implications of that now, but for the many forks of Android like GrapheneOS & LineageOS, and for advanced device configuration tools like Magisk and its many modules, it probably spells trouble.

I just don't understand this. GrapheneOS and LineageOS team have full source-code access. They can do whatever they please with it. (The limitation being that Google breaks stuff at an incredible rate, and following is a bit annoying)

Anyway, I hope that Android becoming more and more user-hostile (and more specifically in this case power-user-hostile) will move more and more people to custom ROMs. (In my dreams I make a "OwnerDroid", an Android fork where the security model doesn't have the first line saying "the user is an enemy", but even though I developed some tiny bricks of it, the overall project would take a huge amount of work)

[1] Except for some kernel-level protections, but GKI reduces that risk.

That's beautiful. I'm impressed we can compute related components on a world map so fast. The interconnections everywhere are beautiful. If this data is reliable, you can go from west europe to east china with just 3 changes (largely helped by canals)

It makes me want to make a remotely (4G) controllable solar-powered tiny boat, and see how far I can drive it, though this map lacks the unreachable discontinuities (dam)

> If they're going to release a new flagship every year, how does that make them any better (from a sustainbility POV) to the morally and ethically bankrupt companies they're competing with?

By removing many reasons for you to buy a new smartphone every 3 years: - when your screen breaks, you won't have to pay the price of a new smartphone to replace it - your apps won't stop working because they want the shiniest OS you don't have (because you'll have it) - when you're out of storage because apps will have inflated 5 times again, you'll be able to extend storage - they give you financial incentive to keeping your device (see their subscription model) -- and they give themselves the incentive to make the longest-living product! - when your battery gets lower charges (which I believe they already efforts to make slower)

Even if you want to buy a new fairphone in 3 years for a shinier camera, your previous fairphone will still be perfectly usable, and still be repairable for a long time.

I agree that it's a bit sad they don't have "plus upgrades", but reducing their work to just the plus upgrades is very extreme.

Also I agree that releasing a new device every year doesn't exactly sound right, but at work we work a huge lot on sustainability (for money reasons), and the conclusion remains that we need a new device every year. Basically the reasoning is that we can't foresee 10 years ahead what will make current devices obsolete. Maybe RAM price will plunge and all devices will have ten times the amount of RAM (and apps will eat ten times more RAM). Maybe some stupid instruction-set will become mandatory. Maybe apps will no longer be usable on sRGB screen. Maybe /that/ Vulkan extension that requires specific hardware will be used by Unity and thus breaking all new games. So one way to reduce the risk (as in total cost of obsolescence), is to make new products regularly, so the amount of devices in production with obsolete hardware is reduced.

AFAIK They have the best track record of upgrading Android smartphones, across all their models. Most of the upgrades they are doing it largely themselves beyond what SoC-vendor provides.

It's not extremely difficult (though Google makes it harder every year), but does require people dedicated to "simply" upgrading. A team of 10 of decent software engineers (juniors are okay as long as they have proper embedded systems education, and they aren't afraid of huge code bases) should be able to handle that. Rock-stars should be able to manage that single-handedly (again assuming they only do that).

That being said, for Fairphone 5, they are using a ""special chip"" for which Qualcomm will provide longer software support than they do for other chips (But I think they still need to extend upgrades themselves after Qualcomm stop). In this case, IMO, it's just a way for Qualcomm to make more money off long support. (The chip probably has some difference compared to standard Qualcomm smartphone SoC, because modern smartphone Qualcomm SoC may suicide itself pretty fast with a removable battery)

It's a carrier, it connects you to the internet. Why isn't "I haven't connected to my account in months" the expected state?

> The LGPL is also an interesting one. People tend to think it's less strict then the GPL. However it has a clause which says you need to provide the means for someone to change the LGPL library and reintegrate those changes in your work.

Isn't that true only for LGPLv3? This requirement also happen on GPLv3.

Actually Termux:X11 /is/ a wayland compositor. But it is so minimal that the only working client is Xwayland.

Something like that:

pkg install proot proot-distro ; proot-distro install ubuntu ; proot-distro launch ubuntu

For people who want to go down rabbit holes, you can get very far with this. I ended up with UT2004 [2] (Windows version) running on my smartphone (complete stack was Termux => proot => ubuntu => box 86 => wine => Zink (Android-side, with virglrenderer to pass through to proot-side) => OEM Vulkan driver [1])

I even attempted to make it into a generic ROM, so that you could flash any Android with a Boot2UT2004 ROM, but the Zink part was hard to make generic enough.

[1] I mention OEM Vulkan driver, because many people are using Zink over a modified Turnip that works on original OEM kernel

[2] I don't even appreciate UT2004, that was to troll a colleague

I think that this is kinda like the lamp => transistor change. The immediate usages are not revolutionary, moving from lamps to transistors in amplifiers didn't change everything overnight. First products with transistors were barely better than the ones before. But oh boy did it create an infinite new horizon of opportunities on the long term

I don't really know whether the analogy holds, but I find reasonable reasons to think it does (we could finally get our low-voltage DC home \o/ maglevs \o/ routine/cheap IRMs \o/. that's just the "simple" applications that we can already foresee right now)

> Somewhat related, I think the Apple Silicon MacBooks will end up replacing many people's old ThinkPads someday thanks to the work from the Asahi team. (A bit early right now, especially if you have one newer than an M1.)

That sounds a bit weird to me: reading Asahi blog posts, it sounds like Asahi are doing little hardware drivers, rather a lot of RPCs, because most hardware features are behind some other CPU running proprietary firmware. (Loaded by bootloader stages before Asahi if I followed correctly)

Yet another innovation thing that Web Environment Integrity (and SafetyNet) (will) hinder.

> But in many countries you already have to show ID to buy a SIM card.

I can buy a SIM card that gives internet pretty much everywhere around the world with bitcoin with silent.link. Granted you don't get an IP matching the local country, but still...

Make browsing the internet possible only on CPUs allowed by Apple, Microsoft, Google. So no RISC-V just yet, and even when RISC-V will be supported by them: No competition allowed in CPU.

Make browsing the internet possible only on SoCs allowed by Apple, Microsoft, Google. No competition allowed in SoC. [0]

Make browsing the internet possible only on form factors approved by Apple, Microsoft, Google. So no calculator with a web browser [1]. No competition allowed in form factor.

Make browsing the internet possible only on UX approved by Apple, Microsoft, Google. So backtracking 10 years ago, when Android made documents-oriented web browser (= each tab appears just like a standalone app in recent apps), that would have been abuse of that position. No competition allowed in UX. [2]

PS: I come from Android OS world, all those examples already apply to Google/Android.

[0] Well this one will depend on whether their Web Environment Integrity implementation will enforce full secure boot approved by them. Considering how it went for Android, I'd say it will, but can't say for sure.

[1] Yes you can find calculators running Android (but can't run Google/Android so no Chrome). Amongst a lot of other weird Android devices. You can find walking robots, toothbrushes, urinals running Android.

[2] You'll probably find a better example. Arguably it's the same as "competition allowed in browsers", but that was an OS-wide change, but saying it's "OS" IMO largely reduces it.

> And yet I am still tempted to add this 204 service, but maybe then I also don't have internet on my phone anymore because it keeps believing it's behind a captive portal when the service is offline

Try it. Normally Android will say "This connection doesn't give you internet, connect anyway?" then you press yes, and everything works. (Except if you're on a captive portal, in which case you'll need to type example.com in your browser to access the captive portal).

> Not in France, where retirement fund is a nation-wide socialized entity, which was the core of this debate. Private pension funds are close to non-existent here ; i personally cannot name a single person doing that because the public model worked so well for decades.

It's not that extreme. There are a lot of "assurances-vie" which are quite literally meant to be private pension funds (with lots of tax incentives...). That being said no-one actually relies on it, and the core of pensions are definitely the nation-wide entities. (Yes plural, we wouldn't be french if there weren't dozens of competing institutions doing the same thing, with of course people getting sucked dry but that administrative complexity)

> The first is the landlord mafia keeping apartments empty on purpose for speculation. Because if the entire supply was put on market, the prices would collapse, and people could actually afford housing. We could design a law to tax empty housing proportionally to the owner's wealth, and see how fast we get housing for everyone.

I don't agree on the "proportionally to the owner's wealth", simply on the empty apartments' value. But yes, we need to tax empty housing. And we're getting there. There has been a lot of changes with regards to housing taxes that are largely going towards taxing (relatively) more empty housing, with this year being the first year you need to properly declare for every housing you own what its use. And I feel like it's just the start, and we're going to keep those taxes evolving towards taxing empty apartments while taxing less family housing.

Yeah no. Google Messages (almost) always go through Google services, not carriers. They are definitely their own platform.

Context: IMS means VoLTE + VoWifi (+ other stuff), which is how you do calls and SMS over LTE, which is mostly "just" SIP

This move of Voice and SMS to an IP protocol is IMO pretty interesting for opensource community because it means we can largely reduce the footprint of the modem to "just" giving IP connections.

The main contributors to opensource IMS client should to be "secure smartphones", long before people doing "let me opensource my baseband", because running a super-privileged SIP client is very dangerous (even Pixels, no matter the OS, got security flaws in it few months ago), but that's clearly not what I'm seeing. (There are also other security considerations justifying an opensource IMS client, like forcing the encryption of VoLTE)

As someone who spend a lot of time making an Android opensource IMS client ( https://github.com/phhusson/ims ), this wiki page isn't really news: I've had this setup (Doubango IMS client, which is a C embedded-leaning opensource client + VoWifi tunnel) running on my desktop using an Android smartphone as a SIM card reader more than a year ago ( https://github.com/phhusson/doubango/blob/master/README-phh ), which is well, just the same age as that wiki page. Except that I've been using it over real carriers, not just software ones.

If someone has some time to spend on it, for GNU/Linux smartphones, Doubango IMS client should be pretty straightforward to get to work, what's missing is "only" plumbing.

Either way, if the people who are working on that are here, please say hi so we can share our war stories :-)

Anecdata: France has a few encrypted channels, and except that there is a monopoly on those channels [1] I think it's quite fair. They take like a sixth of the available bandwidth.

That being said I understand the uproar is about channels switching from FTA to subscription, which isn't legal for channels here

[1] all those channels belong to the same company, but realistically they are the only long-term subscription based channels in France

> HDCP is broken already.

IMO this encryption can be interpretated as "you are supposed pay to view this channel". They could have foregone encryption with just a flag except that it would be trivial for any Aliexpress atsc hdmi dongle to simply forget that about that flag.

That being said yeah they probably require hdcp which.. Well you can find HDMI splitter (that goes from 4k to 4k on hdmi1 and 1080p on hdmi2) that "simply forget" about hdcp 1&2. (no idea whether that's on purpose but that's not the reason I bought it in the first place)

The electromagnetic spectrum is used by many private companies, like carriers. Would you want to disband that as well, and make it free for everyone to use?

(I'm a huge proponent that all infrastructure should be public, but I still think that carrier usage should be paid by the users, at least beyond some minimal usage)

You can't reasonably sign broadcast. Encryption is done at 188B packets to be robust to packet loss.

That being said that argument is still BS because the attacker can just send a non encrypted signal.

There is chrome.google.com and www.facebook.com in the list. That's pretty cool.

I'm not saying they are wrong, chrome definitely allows you to decode drm-encombered files, but uh...

As far as I know, that's wrong? You must "target" newer OS versions (to prevent loopholes as you mention), but you can still have whichever "min" OS version you want.

Looks like I mis-explained.

There is no need to port newer Android versions on older devices. My point was literally that it doesn't make people keep their devices longer.

As a software engineer who spend the vast majority of his time on (successfully) putting newer Android versions on the vast majority of older Android devices:

Yes the hardware is the bottleneck.

This pains me, because what I can do is software, not hardware, where I spend all of my time is software.

An app just needs to target Android 7 to target 95%+ of users (so "only" 5% additional e-waste). And for app developers, targeting Android 7 is fairly easy (and the vast majority do), because Google provide androidx which is a compatibility library that provides backward-compatible APIs for every new shiny things. It's not 0 effort, and I understand app developers preferring Apple ecosystem here, but most of the Android app developers I've seen do use androidx and do target around Android 7. [1]

So, you get reasonably 5-7 years of app that are still working. The average renewal rate still barely bulged and still is around 2 years. People are not renewing their devices because of apps that are no longer supported.

IMO things that make people renew:

- Things that are a bit broken but annoying enough and accumulate: Too many apps leading to slow down, USB-C port getting dust making the connection harder

- The phones always look shinier in-store (because of over-saturated, over-brightness-ed screen amongst other things), so I'm pretty sure stores would manage to sell the same device over and over again by just changing the name.

- Kinda the same: Despite Gorilla Glass 424 Ultra Pro Max, screen's glass do wear, and phones' screen look worse and worse over time (if you have a screen protection, the screen will still look shinier in-store, if you don't your screen will be weary, and the screen in-store will look shinier) -- I like what Rebel Tech is trying to do here, they are trying to replace exclusively the glass part of the screen. That's so much cheaper that you could do it every 6 months to make your device look new again.

- Completely random software bug (which could very well be BECAUSE the device got a software upgrade, or could be just a bug in the app, or just Google decided that this was the new Best User Experience) that make people think "yeah well I get it, my phone's old"

[1] I do know of one exception: Slack app. They are deprecating Android 7 and Android 8. They suck.

> People want buzzwords, not working software.

I'm pretty sure this doesn't generalize that well. I would say that people who don't use every day the product they are buying prefer buzzwords, but it's less true for people who actually use it.

WebXDC sounds fun, could you tell more about the use-cases you found?

cf sibling's comment: You need to factor-in credit's price. I've been looking for a flat near Paris for 2 years, and sure the prices are lowering. But still, the cost (either total cost, or percentage of my salary) for me increased.

My mental model of housing in my area is that I have the n-th salary in that area, so I get the n-th best flat. The pricing will then be exactly what I can afford (which is legally 35% of my salary here). What changes is who gets how much, 3 years ago it would go at 95% to the previous owner, nowadays it's 70%.