Once again posting something for reference as I couldn't find it online

Symptoms

No issues after logging in.
After suspending (sleep) and resuming, screen takes 25 - 30 seconds to turn on.
Display settings in Plasma take a long time to load, sometimes don't show automatic rotation option.
Turning on screen after turning off (even without sleep) takes a long time.
No suspicious logs in Kernel and Journald (even after comparing post-fix).
Switching kernel makes no difference.
Logging out and back in temporarily fixes screen rotation and screen waking until next suspend.
Everything works in X11 session apart from screen rotation (appears unsupported).
Running monitor-sensor hangs when running after suspend
systemctl stop iio-sensor-proxy fixes slowdown issues

Workaround

Downgrading to iio-sensor-proxy 3.6-1 following Arch Linux package downgrade instructions.
In my case with a cached package

sudo pacman -U file:///var/cache/pacman/pkg/iio-sensor-proxy-3.6-1-x86_64.pkg.tar.zst

and optionally adding it to IgnorePkg

IgnorePkg   = iio-sensor-proxy # Issues in Wayland after suspend

System info

OS: Arch Linux x64
Host: Lenovo ThinkPad L390 Yoga
Kernel: 6.12.35-1-lts
DE: Plasma 6.4.2 iio-sensor-proxy (broken version): 3.7-1
Last full system upgrade: 2025-07-06

Symptoms

Stylus taps are registered as well as button presses (on the last cursor position). Cursor doesn't change position.
"Test tablet" in "Drwaing Tablet" KDE settings only displays Stylus release events.
X11 works.
Finger touch still works.

Solution

Rename or delete ~/.config/kcminputrc, or remove config section related to "Pen" (not tested). Deleting the config resets other mouse pointer settings like speed, device enabled/disabled and calibrations.
Then log out and log back in.

Precursor

Testing pen in "Test tablet" section.

Other attempted fixes

Log out and back in - Fail
Reboot - Fail
Moving cursor on login screen and during login process - Fail
Insert and remove pen again - Fail
Shutdown and disconnect battery (in UEFI) - Fail
Removing kcminputrc after diffing with Timeshift snapshot and filtering for interesting files with grep - Pass

System info

OS: Arch Linux x64
Host: Lenovo ThinkPad L390 Yoga
Kernel: 6.12.30-1-lts
DE: Plasma 6.3.5
Libinput: 1.28.1-1
Last full system upgrade: 2025-05-28

TTL

The default TTL for Android is 64. Default for Linux is 64. Default for Windows is 128.

When a packet passes through the router, its TTL is decreased by 1. So a packet from your phone will have TTL of 64, while one from device on hotspot will get decremented to 63 when passing your phone, acting as a router.

Unless it starts with 65, then gets decremented to 64, the same as with your phone directly.

On Linux you can temporarily change this with

sysctl net.ipv4.ip_default_ttl=65

as root.

If the carrier just checks the TTL, congrats.

I haven't played around with IPv6 hop limit.

HTTP proxy

If that doesn't work, you can try to proxy traffic through the phone. I like to use Tinyproxy in Termux.
The config is pretty simple, and also supports (unencrypted) authentication with username and password.

Another option is an all in one app like TetherFi. It creates WiFi Direct group (without direct internet access), and has a built-in HTTP proxy.

Both ways allow also using your phone's VPN.

The HTTP proxy can be easily configured in Firefox, for example. For Android, there is a pretty cool app called NekoBox which is a proxy client for many different protocols. It uses the VPN interface to pass everything through the proxy, thus giving you full device connectivity on the other device.

TinyProxy: https://tinyproxy.github.io/ (can simply be apt installed in Termux)

Termux: https://f-droid.org/en/packages/com.termux/ (terminal emulator for Android)

TetherFi: https://github.com/pyamsoft/tetherfi

NekoBox: https://github.com/MatsuriDayo/NekoBoxForAndroid

I worked the night in Budapest, and stayed longer. My father tried to call me thrice, twice of which did actually get to me (loss of signal), but I didn't pick it up. Apparently he also sent me a few messages, but I didn't retrieve those*.

When I got home, of course that was the first thing I heard of, along with the stupidest reason yet.
(approximate from memory) "Why don't you pick up phone or reply to messages? We were really worried about you. There's pride over there right now desbite Orbán banning it. It's full of LGBT perverts, anything could have happened to you."

And the first time I told them I'll work the night (I don't know if I can use "nightshift" when it isn't shift work ¯⁠\⁠_⁠(⁠ツ⁠)⁠_⁠/⁠¯), they were very strong on disagreeing with that, stating that if I'll be outside at night, I could "get raped by a homosexual".

* Calls are forwarded to my current number, but that's not an option for SMS. I updated my current number where relevant, leaving out just my family, which is basically just my parents and aunts spamming me with christian stuff or telling me to tell my mom to call them (why?). Retrieving SMS thus refers to switching the SIM card for like 2 minutes (switched to eSIM for convenience).

Just because I can.
But it also proves 2G is still at least somewhat usable.

I would just switch my phone to 2G only and then continue the download in Termux using wget because that's most reliable way to download files.
I let it run overnight, then stop it when I needed my phone during the day, then let it run again at night when not needed anymore.

However, 2G is quite inefficient, so this actually drains the battery a lot.

I wonder if the carrier just sees a weird spike in 2G data usage.

Anyway, calculation time. Wikipedia Kiwix ZIM file is approximately 110GB. 109,886,078,976B to be precise. With average speed of say 21KB/s, it would take around 2 months to download the whole English Wikipedia with low res pictures over 2G EDGE.

But also EDGE is quite better than original GSM Data (CSD), which offered breathtaking 9600bps.

Better photo available here: https://english.nv.ua/russian-war/ukraine-begins-treating-soldier-scarred-with-glory-to-russia-phrase-from-russian-captivity-50522497.html

The words, written in Russian, were branded into the right side of the body alongside the letter "Z," a symbol of the full-scale invasion that many Ukrainians and critics of the war liken to the Nazi swastika.

According to the United Nations, more than 95% of freed Ukrainian POWs have said they were tortured during their captivity. Survivors have recounted harrowing treatment, including being brutally beaten, electrocuted, and forced to endure painful stress positions.
Many have faced mock executions, threats of rape or death, and were denied basic medical care. Some were left without food, locked in isolation for weeks, or permanently marked with burns or scars.

For Turkevych and his team, seeing the scars left by Russian captivity on Ukrainian POWs has become a routine part of their work.

[...] Turkevych and his team have treated several released defenders bearing swastika-shaped scars, some even on their foreheads [...]

Usually they're tolerable, just annoying.

The screenshot is from one I found installed outside Lidl (supermarket) now, above the entrance. Probably loud enough to "cover" the entire parking lot. Loud enough to get painful after a few seconds. Sweeping between 15.5kHz to 18.7kHz. For comparison, a CRT TV is around 15.7kHz.

I wonder if not caring about loudness of these higher frequencies contributes to hearing damage among people.

Lower frequency, but also awfully loud buzzers can be found in trains near doors, working as a warning when the doors are closing. Thankfully, this probably annoyed some of the staff enough for them to start taping these over.
But those are loud enough to sound distorted.

Anyway, I just hid behind a corner where I couldn't hear it that much while waiting for the manager to call us in.

Attempt made by Slovakian Regulatory Authority for Electronic Communications and Postal Services
https://www.teleoff.gov.sk/images/urad/odbory-oddelenia/odbor-statneho-dohladu-elektronickych-komunikacii/rozhodnutia-statneho-dohladu-elektronickych-komunikacii/rozhodnutie-zakaze-poskytovat-siete-sluzby-z-29-4-2025.pdf

Preface

My bank blocked my card due to a "high risk transaction"... again. This time they told me unblocking isn't possible, and issued me a free replacement.

And as the title says, as a kid I wanted to try putting a cut-out bank card chip into a phone. Yes, I grew up in 2FF era.

The... thing

I also stripped the front foil because I was curious what's under it. Nothing. The card is just flimsy without it, and the chip contacts raise above the card.

The abomination in comparison to real SIM cards:

I have contactless and magnetic payments blocked. Only chip + PIN, thus the wear.

Attempts

1. Samsung GT-S5830i (Android 2.3.6) - "Insert SIM"
2. D-LINK DWM-222:

 -----------------------------
  Status   |             state: failed
           |     failed reason: sim-missing
           |       power state: on
  -----------------------------

3. Alcatel 1066G:
   This is the only device with observable difference.
   Without SIM:
   
   With the... thing:
   

At least it detects something. Contact and SMS storage says "SIM not initialized"
The "Searching network" eventually fails (what a surprise), and switches to "Insert SIM" after a few minutes.

About the blocking (unrelated to post)

Bureaucracy

Apart from what's in the preface, the documents I signed still mention the standard 10 EUR charge. I asked to see what I am signing this time rather than just issuing the signature on their tablet and trusting the employee.
This time around, there is "fee will be individually assessed and charged" as well, and if I do get charged, I was told to dispute it at the bank.

Last time when giving me a card at the bank they just charged me anyway. Interestingly, that time the official reason was given as "card not delivered after 2 months". If that were indeed true, why should that be my problem?

That was when I just opened the account. They told me that I could get the card immediately ("free") rather than wait for it to arrive by mail.

Blocking process

Message from automated system approx. 12 hours before card blocking:

Plus, the payment worked on second attempt.

OH, FUN FACT: I didn't even use the physical card, but a digital single-use card generated in the bank app.

So, they blocked physical card, after suspicious payment made using digital card, and even if that made sense, they waited 12 hours and let my second attempt through.

The other 2 times the blocking happened after donating to archive.org