@kirschner postmarketos edge is more usable than pureOS 10, it shuts down cleanly and mostly suffers from WWAN modem connection not getting re-established on the Librem 5 after wake from suspend, power cycling with HKS resolves the issue. Another issue is missing an installer image that mobian uses, or an encrypted image download so you can enable /home encryption. The pmbootstrap does not work on Ubuntu 23.10 which i think may allow you to enable encryption during install
The whole "Biden should step down" thing stinks like Republican ratfuckery to me.
As far as I can see, the only safe way to do that at this point would be for Harris to take over RIGHT NOW (via 25th amendment), announce her candidacy, and pick a VP.
Then she needs to go to the mattresses to keep the Supreme Court majority from ganging up to disqualify her on some spurious grounds. (But at least they handed her a loaded gun in the shape of presidential immunity!)
They Were Choked, Beaten and Tased by #NYPD Officers. The Commissioner Buried Their Cases.
==
New York City’s #Police Commissioner Edward Caban has repeatedly used a little-known authority called “retention” to prevent officers accused of misconduct from facing public disciplinary trials.
Victims are never told their cases have been buried.
#NYC #LawEnforcement #CivilRights #Crime #News #Government #Courts
slowly losing my will to live as i dive deeper into apk-tools + Alpine + /usr merge + pmbootstrap
i mean, it's solvable, the constraints are just so tight. Like... dynamically fixing Alpine packages so they work on a /usr merge system at runtime requires handling every possible case, like all of these symlinks needs to be detected and skipped (since with /usr merge they become recursive):
/bin/foo -> /usr/bin/foo
/bin/foo -> ../usr/bin/foo
/usr/bin/foo -> /bin/food
/usr/bin/foo -> ../../bin/foo
then the same for /lib -> /usr/lib and /usr/sbin
Plus, simply making them absolute symlinks is extremely bad behaviour since it breaks operating on a foreign rootfs (apk --root ...), we just ignore this for now...
And then you get caught out by crap like /bin/go being a symlink to /usr/lib/go/bin/go
i think im getting close, but hitting some weird issues with pmb and apk3...
it would probably not be so bad if I didn't also have to write all this in C
https://gitlab.com/postmarketOS/apk-tools/-/commit/954af7ba48c492cca87b333b0be578a9877cce7a
Tiling Shell is a new, super-charged window snapping extension for GNOME Shell - https://www.omgubuntu.co.uk/2024/06/tiling-shell-gnome-extension-in-ubuntu #gnome #ubuntu
On July 1st, The European Commission (EC) informed Meta that its "pay or consent" model failed to comply with the Digital Markets Act (DMA).
https://ec.europa.eu/commission/presscorner/detail/en/ip_24_3582
Meta's "pay or consent" lets European users of Facebook and Instagram choose between paying a €9.99 monthly subscription for an ad-free version, or a free-of-charge version with personalised ads. A whooping €255.99 per year. 💸
But why are they even doing this?
Consent is invalid, they said, where there is "any element of compulsion, pressure or inability to exercise free will" and large online platforms should consider providing an "equivalent alternative that does not entail the payment of a fee".
Meta is not the only target of the DMA. We continue to work, individually and collectively with civil society, to monitor its enforcement, including regarding compliance measures introduced by Big Tech companies such as Apple: https://edri.org/our-work/apple-must-comply-with-the-dma-urges-civil-society-call-to-european-commission/
Here is another update on how's my #PineTime doing. As I mentioned before, the glass cover for the heart rate sensor fell off. And, well, I got caught in the rain today and it died. Beautifully, tho
The fastest non-Google controlled web rendering engine Servo is trying to compete with only $1.8k funding a month:
https://www.phoronix.com/news/Servo-Engine-May-2024
Yes, they did have an NLNet grant, but that ran out.
If we want an alternative to Blink/Chrome, we need to fund it. This is a project where even a tiny regular amount could yield oversize returns:
@z3ntu added sensor support to #Fairphone 4 and 5 on #postmarketOS in an open MR, on track to beat the Xiaomi Mi 9T / Redmi K20 and Pixel 3a in packaging (even though sensors can work on both devices). It's not just SDM845 devices (and still not SDM670) that benefit from the libssc/hexagonrpcd userspace sensor drivers.
https://gitlab.com/postmarketOS/pmaports/-/merge_requests/5290
https://gitlab.com/postmarketOS/pmaports/-/merge_requests/5313
[$] PostmarketOS: Linux for phones and more
In 2016, Oliver Smith reached a point of frustration with the short lifespan of updates for his Android phone. Taking matters into his own hands, he began developing postmarketOS, [...]
it's utterly ridiculous to me that we can't include firmware for phones (like wifi, Bluetooth, gpu firmware) in the linux-firmware git repository without the OEM providing consent.
in the case of Qualcomm phones:
The firmware is locked to the device via a signature, it can't be replaced.
It is already obtainable easily via the vendors website or on GitHub
It usually barely differs from the development board firmware which is already in the linux-firmware repo
Vendors generally aren't confident enough to explicitly distribute firmware with a suitable license because Qualcomm haven't given the go ahead (and they either can't, or won't ask permission).
The few examples we do have are for the Lenovo ThinkPad x13s, where the legal side was a non-trivial exercise as i understand it.
Even if Qualcomm greenlit OEMs invoicing redistributable licenses with their firmware builds (which is something i think they should do), we'd still need OEMs to actually do it... And they might have their own value add and silly lawyers who think NOT doing it gives them some kind of advantages.
The whole situation is so ridiculously stupid. And to top it all off it's not like anyone actually cares. There are many people redistricting firmware for Android phones and ofc it's totally fine - because there's literally nothing to gain by preventing access to these blobs.
I wonder what would happen if someone just submitted blobs for a phone to linux-firmware...
phosh 0.40.0 is out 🚀📱:
- #phosh: more quick toggles (dark style, mobile data). Allow suspend when device is locked. 🐛 fixes. Tweaks for #Rust binding generation
- #phoc fling gesture for phosh's top and home bar. 🐛 fixes.
- #mobilesettings: allow to tweak shell layout, lock delay and plugin ordering via ☝️
Check out the full release notes at https://phosh.mobi/releases/rel-0.40.0/
🙏 to everyone who contributed to this release.
#phosh #librem5 #pinephone #gtk #wlroots #gnome #linux #mobile #LinuxMobile
Reverse-Engineered Vivante Driver Brings Improved Performance & Fixes For Linux 6.11
The Etnaviv kernel graphics driver that provides reverse-engineered open-source support for Vivante graphics (and NPU) IP has seen a fresh round of patches ready to go for queuing in DRM-Next ahead of the Linux 6.11 cycle kicking off in mid-July...
https://www.phoronix.com/news/Etnaviv-Driver-Linux-6.11
@eff and provide actual reasons for the ban, as in publicize actual evidence from the NSA and other organizations that showed that tiktok is in fact controlled by the chinese government and does in fact have access to user data as an example
Advocate for Earth citizen privacy IP rights wherever you live, free of and independent of any one Government Regulation or Government Control.