Note the spec increase in Ubuntu is partially attributed to GNOME, which is also part of just running the OS before you even open anything.
XLE
joined 10 months ago
The specs are pretty good, but it's still only eight gigabytes of RAM total, and a phone processor, and seems optimized for comfort (cool case temperatures) over performance.
I would prefer something that's light without compromising on things that Microsoft figured out in the 90s and 2000s, and things that modern Apple computers can pull off now.
Apparently GNOME in particular is having a rough time in general, if other articles from ~~the same website~~ omgubuntu are an indicator, but this seems to be a wider trend in desktop environments
It's an illuminating experience to go to a store with Apple computers with 8GB of RAM on display, and browse to a RAM-heavy unoptimized website like YouTube or even Reddit now.
Open a few tabs.
Open a dozen.
You'd be surprised what a decently coded OS can pull off without compromising on the visuals.
Really unfortunate seeing GNOME is part of the problem here. Linux desktop environments shouldn't need to be tied to large RAM requirements, never mind increasing ones, for basic functionality. For example, the Start menu key was introduced by Microsoft in Windows 95, but this toggle still isn't available in most "light" desktop environments like XFCE.
The MacBook Neo, of all things, is chomping at the heels of the idea that pretty, feature-rich OSes need a lot of hardware to function.
The desktop software in question is GNOME, which is an integral part of the OS... So it's definitely true that part of the bloat is because the software running the OS itself is now using more RAM.
What is a reasonable number of fish though?
Onlyoffice's unnamed lawyer, vs the author of AGPL and the FSF: I know whose opinion I'll put more stock in.
Update: @riskable@programming.dev admitted he only uses closed-source models from companies with massive budgets.
His claims about "open-source institutions" are, sadly, all lies.
Can you provide more info on this? It looks like they built Meet with the MLS protocol, and unless I'm missing something, that should be E2EE and it should work on any client (regardless of the server)
Fair point, I shall delete
This is pretty much just a rehash of the paper put out by Google Quantum AI. Not quite "CEO says a thing" journalism but uncomfortably close.