I've also been playing around with BSD and it's like hanging out with a friend from elementary school: they're a little wonky, but you remember some good times together.
PopOS isn't my favorite, but it is pretty straight-forward. If it's working for you then it's a total win. (Which is my way of saying "though I like Stock Debian and BSD, there is absolutely nothing wrong with PopOS.")
But for a regular desktop/workstation system I want more frequent software updates even if it comes at the cost of a little stability.
(Of course there's Debian unstable, but I'd rather go with a distro where the rolling release is the main product, and they maybe do a little more testing before making changes)
[1]: https://flathub.org/setup/Debian [2]: https://packages.debian.org/bookworm/distrobox
I'm a Debian user. I see this a lot. Often in contrast with Ubuntu.
Yet Debian's cycle is roughly on par with Ubuntu's LTS - every ~2 years.[0]
Perhaps they mean there are interim releases in Ubuntu? It's hard to say for sure, as the phrase 'outdated software' can mean different things (more than 6 months old, more than 2 years old, in need of a security patch for > 1 week, etc).
> But for a regular desktop/workstation system I want more frequent software updates even if it comes at the cost of a little stability.
This sounds like you want Debian Testing (currently Trixie).
Much more frequent software updates, with an extremely small risk of less 'stability'.
It used to be the case in 2010, but that is 13 years ago...
The nice thing is, though, all that information is freely available if you wanted to compare.
I can save you a little bit of time by noting that the criteria for a package being promoted from Debian Unstable into Debian Testing has been honed over the years, and is the reason why a lot of people settle on Debian Testing for their day to day machines:
The package has been in "unstable" at least for 2-10 days (depending on the urgency of the upload).
The package has been built for all the architectures which the present version in testing was built for.
Installing the package into testing will not make the distribution more uninstallable.
The package does not introduce new release critical bugs.
Please, don't recommend Debian Testing to unsuspecting people. The distro is aimed at people that know what they are getting, and has no guarantee it will keep working unless you know really well what you are doing.
Even Unstable is supposed to be more beginner friendly.
Was parent unsuspecting?
They said they wanted 'more frequent updates even at the cost of a little stability'.
They sound like they're very much in the suspecting camp.
The challenge with Debian Testing is some security exposures are longer than for either stable or unstable, hence the recommendation to keep unstable in your sources, ready to go, and maintain some awareness around CVEs / DSAs.
> Even Unstable is supposed to be more beginner friendly.
I can happily disabuse you of that notion.
Definitely, you should not recommend Debian Unstable to unsuspecting people.
I have occasionally run unstable on one of my machines over the years, but now pretty much avoid it.
When it breaks you either wait for a few days for it to be fixed, or you spend a lot of time trying to understand what's broken - usually with a machine that doesn't get you to a GUI.
I really don't know. You don't get Testing at the cost of "a little" instability. You get it at the cost of things eventually breaking in an update, and you having to sort it out on your own. It's a failure mode quite unique to it, and nobody mentioned anything about this. So yeah, I don't get any reason to believe either way. (Also, Unstable isn't supposed to have this one failure mode, but it's supposed to have partial failures way more often. That fits the mental model of "instability" for most people much better than Testing.)
But then, the alternative people are talking about is Ubuntu. Ubuntu breaks much more often than Debian Testing... Up to the point where people don't upgrade it carelessly, so they experience a more stable system. So again, I don't know. IMO, it was worth at least putting a warning there.
That said, for the past couple years I've just run Debian and Arch. But I'll probably use ubuntu again for a while when I get a new framework laptop later this month, the hardware is too new, some special/custom kernel package provided for the latest ubuntu will probably be the most convenient solution initially, and I find trimming back ubuntu isn't too bad. Yes, you have to remove the "desktop" meta-package. It's fine.
No, it is not fine, that I have to fight my OS. It is managable if one has some skills with computer like we do, but it is not fine in general. I would like to trust the operating system I use.
Probably. But I don't think Microsoft cares. Just look at where Windows is in their org hierarchy; it's definitely not top dog anymore and it's not a "strategy" like it once was.
The days of Microsoft needing Windows to make money are behind us, that's for sure. Peak Windows is probably correct, but that fact is not going to usher in Microsoft's demise.
Fedora is my default "easy" Linux install nowadays, the excessively corporate vibes I get from Ubuntu feel very wrong for a Linux distro.
Second, Fedora uses rpm management, rpm repositories sooner or later corrupt themselves, happened to me and others I know on Red Hat, on Fedora and on SuSE.
I'd use Debian but I've found Ubuntu's driver management to be far more stable and reliable, especially for WiFi.
Does that make it any easier? You can download a live version and test instead of needing to find that firmware included release.
There's nothing wrong with providing an easily skippable option to do that (ubuntu), a completely different thing to make it a mandatory part of the installation (windows).
If only. I remember people saying the same when Vista came out.
Half are having a hard time with learning some of the CAD software because they lack the experience with full featured Desktop OS's like Windows/Linux/Mac. It's surprising how fast generational change occurs
It’s pretty unfair on those who have done the bare minimum amount of experience with computers that they have to sit through (and pay for!) being taught ‘this is what a file structure is and how it works’ for a not-trivial amount of time.
Also not understanding what a zip file is, (just a weird folder that sometimes fails).
I am not surprised it is being the very same case with new generations. But before people were saving things in their desktop, now they just use the most recent function.
- Eduroam is available at many other schools if you happen to be there, but I never managed to actually sign in, because CSU never told me how, neither did UCLA when I visited - how to access the network drive, both on- and off- campus (sftp supported without a VPN, smb with). This was mentioned in a single 1 hour lecture at the library on how to use library systems. Access from the locked down public computers at the library without student logins was not mentioned (just mounting a network drive, but instructions previously given implied that the VPN was necessary if you weren't logged into your account on the network) - there were labs running windows 10, windows 7/8 (windows 10 came out while I was there), Fedora (network drive had to be connected manually), and even a lab full of Sparcs that I didn't use much (they were quite slow), the library computers, etc. There were also heavily locked down public computers in other areas that only allowed access to the browser, and a lab full of thin clients that offered no permanent storage for even temporary use (working with a local file? Better have a flash drive, or you're stuck with network drive latency) - Citrix allowed access to a suite of software including Adobe CS6 iirc as well as Office and other software; this worked from anywhere on-campus, but the software offering off-campus was different. This got a passing mention in the library lecture, mainly that this is available. Classes that actually made use of this software didn't mention it being available off campus at all. - some software was only available in private labs (acceptable not to mention this), but some was only available in certain buildings/departments and some was only available on Citrix. Fonts could not be installed anywhere, but windows offers a simple (if little-known) method to register a font for the current session, and this worked for using custom fonts in Adobe and other applications, both in Citrix and on lab computers. - there was a single cafe where students were _allowed_ (I think?) to plug laptops into the network (not to say people didn't do it elsewhere, but here the network and plug sockets were above the counter in the public seating area, and nobody would stop you from doing so) - part way through, I think there was a new network drive added, for a while there were multiple ways to access the network drive. - the network drive was (a very little bit) more than just a network drive; it allowed ssh access to edit files with pico; when I first started, it also hosted files for student websites; eventually the websites went away, but the ssh access never did. It was clearly an intentionally open SSH server, and was open outside the school as well; it had a (iirc pre-login) banner with some rules. - msdnaa was retired in favor of dreamspark, and free copies of office were phased out in favor of discounted 365 subscriptions when I was about halfway through. There was still tons of free software available, and none of it was ever mentioned until a cs course (the cs department had their own dreamspark subscription with an expanded offering, but there was software available to everyone as well, including Windows and office...which was available separately in the school/CS/business subscriptions)
When I was in high school (middle school?), I had a computer literacy class where a good quarter of the semester was teaching students how to organize their files. The teacher had a messy filesystem that was unzipped onto each computer, and during the course of a week or so, students were expected to rename files and the few directories and organize the files in a way that made sense. Mostly the files were empty iirc, but I think there were a few with similar names and identical content that could be deleted. I remember thinking at the time that this was the best thing I'd ever seen, having anyway seen the number of files left saved on desktop and my documents on the few machines at the school that allowed you to do so.
I get the feeling more and more that peak Windows is upon us.