Will we need to log in every morning and expect to refresh every damn site cert we connect to soon?
Certbot's default timer checks twice a day if it's old enough to be be due for a renewal... So a change from 90 to 1 day will in practice make no difference already...
Although in many of the tests I have seen, the performance was not actually better in general.
The main benefit so far seems to be not so much a synchronisation that performs better but one that works much more closely to how Windows does it natively thus helping some programs that don't work well with Wine.
Linux is linux. In the end it's more your personal taste with just a little sprinkle of use case that decides.
The main differences are:
Also more depending on your personal taste and habits:
Other considerations:
Things to not consider:
I guess "FBI infiltrated group of immigration activist" would be boring and not fitting the FUD about encrypted messaging...
Nectcloud has always been incredible slow for me. (And that's beside other issues like updates failing more often than succeeding...)
And as I was using it mostly for basic filesharing between my machines and as a CalDAV/CardDAV server I replaced it with Syncthing and Radicale now.
Someone does this in times of increasing Iridium prices? What a waste...
but it me a couple reads to determine if that was a really bad autocorrect
Fun fact: it took me a couple of reads to understand that you forgot the verb.
Kind of... the regular driver officially supports everything from Maxwell to the newest cards.
But then there is the new open source driver now, supporting Turing and above. Which is recommended to try by Nvidia developers, but also still has issues (like power management problems on Turing for example).
Also CUDA-specific stuff still pulls the proprietary driver as a requirement anyway.
As someone with an ancient 750ti happily running on the regular nvidia drivers...
Dedicated support for "older cards" as in "requiring different drivers" usually starts much later (Kepler and before), so about 4 generations before an 1660Super.
Yes, preventing the boot process when something tempers with the files is the whole point of secure boot.
And beside the backups you should always have (remember: no backup, no pity for you...) the keys to sign your EFI files with are on the encrypted disk so the running system can get updated. So deactivating secure boot again, unlocking your encrypted disk from some live boot stick and fixing it is always an option (as is having a live system at hand signed by the same keys if you want to...).
That article triggered an unexpected roller coaster of "there is something called vimdiff I never heard about?" to "no, there isn't because for me vim is just an alias for nvim" to "oh, it's actually just vim -d anyway..."
You are just moving things. When you change your EFI partition from being unencrypted and asking for your password to the BIOS asking for your password (or other credentials) you just shift the attack surface.
Somewhere there has to be an unencrypted part to start with.
Lock your unencrypted ESP down with secure boot and your own keys (shitty as it is that is in fact the one conceptional usecase of secure boot, not that stupid marketing bullshit MS is doing with getting vendors to pre-install Microsoft keys) to prevent tampering and you are good to go.
manjaro 2.0
Was that insult intentional and if yes then what did they do to deserve it?
the ground is full of dust and we are now adults
I seem to see an easy solution there...
The cases got taller and the power buttons moved to the top edge (and often got smaller).
So now those under-the-desk units have their power buttons directly under the plate. Mine specifically it now more a knee- instead of a foot-job... I could press it with my finger without bending down but you have to keep traditiones alive.
Why do you think the most unanimously hated windows versions
I know that people hated every single one since Windows 98SE... it's basically a constant cycle of releasing shit, then keeping it relevant -mostly via forcing people to buy it with their PC- long enough that people resignate and believe tech has to be that bad, then forcing the next and even worse version on people. So which were those unanimously hated versions. Or -maybe easier- which version was widelys adopted before people had no choice because all support for older ones was cut?
People are used to Microsoft Office, Acrobat Reader, Outlook, the Creative Cloud, etc.
And that is some kind of law of nature? Or the result of paying massive amounts of money to flood everything with this shit for free? Seriously... I think you competely misjudge the majority of users. They are not so much clinging to the familiar as just lazily sticking to whatever pops up when they press the power button.
Why do you think chromebooks sell so well?
They do? I have seen one chromebook in real life. Which I would probably not have noticed between all the other laptops and tablets if it wasn't for the fact that this was the most overpriced piece of shit constantly having issues with even the most basic stuff.
(Edit/PS: I just did a quick search and most numbers I found point to chromebooks being more rare than Linux. Which is an achievement given that barely any piece of basic consumer laptop/tablet/whatever comes pre-installed with Linux.)
But I know the sales internationally were declining for quite some time until they spend a lot of money to bribe governments to hand them out as the tech version of a gateway drug.
So for example at the moment increases in chromebook sales in the last years are mainly caused by government procurements in Asia. Japan alone saw sales increase by a factor of 20 in 2024... so I really, really doubt anyone actually wanted a chromebook. But this will probably change after the next generation of students conditioned to think that this shit is how it's supposed to be enters the market. sigh
If you want to talk facts however, start with money spend on lobbying, on pushing it on education early, on forcing people to buy their hardware with Windows pre-installed etc...
Permanently Deleted
The 'random numbers' I mostly generate are UUIDs...
which can indeed be done nicely in the terminal by just reading directly from the kernel's rng at /proc/sys/kernel/random/uuid
