Sorry I am not aware of the incident, do you mind elaborate? Thank you.

Serial Experiments: Lain-ux distro-hopping

Yeah, most of the Chinese people I know on the fediverse are most certainly not gonna vib with the like of hexbear, grad, and ml.

Yes, there are plenty of big Chinese mastodon instances. https://m.cmx.im/explore is likely the biggest one, and GFW people definitely attempt to blocked it.

There are many others, but I am not sure if you work for GFW 😉.

Yes, from the announcement, it seems true.

I feel one of the hardship for Linux to catch on is the lack of commercial interest to make it usable for consumers.

If this problem happened on Windows and macOS, MS and Apple would just send an engineer to spend a week or a day to have it fixed. This change has been in electron for months, and no one bother to fix it.

Same with bugs in chrome and libsecrate, which have been open for 4 freaking years… https://gitlab.gnome.org/GNOME/libsecret/-/issues/49

It also took chrome half a decade to support text-input-v3: https://issues.chromium.org/issues/40113488#comment1, which is added by a third party developer. And it still breaks KDE’s implementate https://bugs.kde.org/show_bug.cgi?id=492225 …

It is understandable people are frustrated, I am frustrated, and joined several conversation regarding this problem. However, I don’t appreciate some of the rant from many users. This change is certainly out-of-touch, potentially due to them don’t quite foresee the amount of flatpak/kde users who are affected by this change.

But many complaints have been dangerously close to the line, if not over the line. Their quiet month policy is reasonable IMO, developers need breaks, especially those interacts frequently with the community. Love or hate electron (same apply to CEF), these works clearly bring many wonderful apps into the linux world.

I personally don’t believe that non-contributors have the right to demand free work from the electron developers.

In a special box for hazzard waste?!

I had way too many of these drawers 🫠

I think mixing app and system dependencies is not the best idea, and Linux desktop is still fighting its impact.

When all the apps on a consumer laptop is expected to depend on the same dependencies, the system likely run into dependency hell, which means many apps needs to be downgraded in order to keep older apps working.

This mixture of system dependency and app dependency also prevents users to use the the latest version of an app on a hyper stable base system.

Flatpak basically aim to solve this problem, where each app chooses their own dependencies, so you don’t need to downgrade all your app just because one app depends on python 2.7.

You are right. I have done some research, it seems most people think that client side hashing is unnecessary in an HTTPS setting.

That is my misunderstanding.

it says it is encrypted but it is encrypred using keys that google has access to as they are unlocked with you logging in into google account.

First it uses lock screen password, so google do not have access to this password.

Even if your lock screen is unfortunately your Google password, I think proper authentication protocol do not send your password to Google to authenticate, but only the hash, which cannot be reverted to derive your password.

Obviously, the above is assuming that Google is not malicious. Otherwise it can just use play service, which is privileged and closed source, to get all your data. If your threat model including Google itself trying to steal your key, you will probably need to install a trusted rom or use iOS (however, apple and the rom developer can also steal your key).

I think these are different. They mostly find vulnerability in the iOS system as opposed to try to crack the backup system.

I think iOS or Android backup system are rather secure compared to other components because of the following: hacker will also need to break into a cloud drive to retrieve them, which adds extra work; the backup is simple, just bunch of files and a password, apple/google can use standard well-tested encryption to encrypt them.

However, guaranteeing there is no way to break into an operating system, especially with all the features that a modern system requires, is much harder.