The other day I was trying to disable Ubuntu Pro stuff and the way to do it reminded me of Windows. Once I get my media backed up I’m switching to another distro, just not sure what one yet.
It’s been 8+ years since I last used Ubuntu on my laptop. I faced massive issues with staying on the latest version of Firefox because apt had a much older version, and installing using the gui installer wouldn’t replace the apt version etc etc. Probably a PEBKAC issue…
But, I do want to know- is this not an issue any more? Will apt install the latest (or almost latest) version of Firefox? Can I update it from the inbuilt update tool in Firefox?
Can I update it from the inbuilt update tool in Firefox?
Universally regarded as a bad idea on Ubuntu based distros as far as my research goes.
Probably a PEBKAC issue…
Staying on the OTB repos in LTS distros and then complaining about software being slow to update is like staying on the OTB mirror, and then complaining that your download speeds suck.
I’m a Linux noob through and trough, use Glorious Mint, but like… How to get a newer version of VLC, than distributed by upstream is probably the first thing I figured out how to do.
Dunno, but in every forum I’ve looked, people say not to use it, but let the updates go through the package manager. Sometimes even on threats of FUBARing your system. Could be that all these people are giving old info that’s not true, but I never tried it - don’t wanna go on the forums and start the thread with “I explicitly did what people say not to. How fix?”
The reason why is because of dependency hell and general packaging conflicts that could occur. You can go with the tar, snap, appimage or flatpak. If you do decide to use the system level package from a 3rd party, just be aware of the risks and be careful. The issue lay within the difference in standards, the usual target for these companies is Debian using the Debian packaging guidelines, while Ubuntu has their own, Ubuntu and Debian also have different release cycles which can lead to conflict with certain packages.
Perhaps, if you’re needs aren’t met maybe moving to a semi-rolling or rolling distro is best.
Oh, that makes sense, thank you. I’m really happy with mint. Pretty sure switching to the nightly repos got me most of what I need, for the rest there’s PPAs. Rolling release sounds tempting sometimes, trying out Plasma on a distro that supports it is also tempting, but so far I can’t be bothered. Mint seems to just work. :D
Most Linux users can search for a package they need to uninstall by keyword. You can’t really do the same with Regedit. The key and path in the post are something even fewer people can find because regedit is so cryptic.
It’s the fact that it could just be a checkbox in search settings, but they make it as hidden as humanly possible, not only is it in the registry, but it’s not an existing flag you can change, it’s a flag you have to know the exact name of to add. It only takes a minute if you’re the type of person to be here commenting on a Linux meme on Lemmy, but to the average user, the option almost doesn’t exist.
Problem is is that is that too many people insist on doing things the Windows way and they get frustrated because of it. For example, instead of going to the software center, they choose to download their programs from a website, even though that’s not how you’re supposed to do it most of the time. They’ll also spend hours trying to get Windows only programs to run, when there are alternatives available that work just as well.
I still don’t fully know how to install rpm files lmao, that’s how I learned about Apt back on linux mint, don’t remember what I was trying to install as it was like 15 years ago. Deb files were nice because they did work like a windows user would expect.
Let’s use Tor Browser as an example since that’s one of the programs that typically gets installed with a tarball. Once you’ve downloaded and extracted the tarball, you’ll want to navigate to the extracted files. You can do this in the terminal using CD commands, but I think it’s easier and a little more intuitive to just use your file manager and navigate to the folder that way. Once you’re in the correct folder, you’ll want to right click on an empty space and select “open and terminal.” Now you’ll have a terminal open and it will already be in the correct directory. From here you’ll want to run the “start-tor-browser.desktop” script. To do this, simply type ./start-tor-browser.desktop and you’ll be able to follow along from there.
Running programs from a Tar image typically involves running a script. You just have to change the name of the script to match whatever they have in the directory. Auto complete is your friend here. You don’t have to actually type the entire name of the script, you only need to type the first few letters and then hit tab.
Definitely this. I have been eyeballing Linux for years, always intimidated by the CLI and the notion that everything you try to do on Linux requires user research and work first.
Now I finally made the switch a couple days ago, and while it took a bit of tinkering and googling here and there I am amazed how simple, even way simpler than on windows, the experience for a an average user is, particularly with the very beginner friendly distro I went with (bazzite/gnome).
It just works right out the box for 90% of whatever I want to do, configuring it is simply flipping some switches in the software and extension apps. Feels more like setting up a new smartphone than a PC. I didn’t even have to mess with the CLI all that much, perhaps half a dozen times so far, and each time i followed specific steps in a guide or tutorial, or tried out some basic things like file search.
It is good to send new users to something like Gnome they can branch out. I think Cosmic will be a great fit as well. Outside of updates, you don’t have to do too much in the CLI really. But long as you learn some of the basics how to get around and maybe make an alias. I think that will get you by just fine on Linux. I do think people should get users to try less Windows like experiences on Linux. Because a Windows like UI will just make them miss Windows.
I do think people should get users to try less Windows like experiences on Linux. Because a Windows like UI will just make them miss Windows.
That is actually the exact reason I went with gnome instead of KDE myself; I find it much easier to learn a new system than to adjust habits that have formed for years. I will probably eventually switch to KDE when i feel fully comfortable, because it is supposedly slightly better in performance and far more customizable.
An Epson XP-830. Full disclosure: When it was brand new it was a severe pain in the ass because it wasn’t supported by CUPS yet, I had to go out to Epson’s website and download a driver in .rpm fromat and install it with alien. Bought it a couple months before I abandoned Windows for Linux and had to make it work. After about a year CUPS suddenly knew what to do with it and it’s Just Worked™ ever since.
Both OS are hard if you don’t know how to use them.
Both OS are easy if you know how to use them.
Linux’s problem is fragmentation. There’s not a single OS that many people are familiar with like Windows. Instead there’s hundreds of different distros that all function in a variety of different ways. Even if a person learns to do something on Mint or Ubuntu, they will be completely lost trying to do the same thing on Fedora or Arch.
And out of those “hundreds” only a handful of them are actually popular and progressing innovation…
As someone who’s distro hopped across a wide verity of distros, the fundamentals are more less the same across all of them. Just go with a popular distro with good documentation and you’ll be fine. If you’ve learned enough from mint to feel comfortable tackling Arch Linux, then the documention (e.g. ArchWiki) will be your strongest asset.
Good documentation is great to have. Here’s the thing though. If you need documentation to use an OS… That just proves that it really is harder for people to use.
Mint and Windows both share the ability to pick it up and use it for the majority of what most people do. Arch is like the textbook example of having to learn a bunch in order to use Linux.
I think Arch is meant for people who want to learn the software - so that you can also choose, control, customize, diagnose, and fix the software!
That said, archwiki is still a great resource on other distros for when something does go wrong, or when it’s not obvious how to do something, particularly when messing with experimental or server stuff.
The Arch wiki is one of the main reasons I use Arch/Arch based Distros. Its so insanely good and after you learned some of the basic stuff and what certain terms mean its a very good resource for doing stuff.
I have used Windows since I was a child and I still need “documentation” to do routine things, because they hide stuff 8 levels deep inside an obscure settings window that requires an arcane ritual to access.
I’d argue that there’s like 4 ways that 90% of distros work like and even they are extremely similar to each other. You got Fedora, Debian, Arch, and whatever Gentoo users do when they are in the dark in the basement. The rest are niche and weird stuff that not even Linux hardcore fans know all about. Similarly there’s only two desktop things that matter, GTK or QT. Everything else is us nerds nick-picking.
Ok, there’s also nixOS and the new wave of atomic monolithic containerized whatever distros, but they are like, super new and the resulting system is indistinguishable for the end used from the other 4 main.
Reminds me of when I tried to install Firefox on Ubuntu and it said it couldn’t open the download as there was nothing associated with it. Ended up with another copy without an icon but it worked. Then I moved to Fedora which seemed to work as intended. Couldn’t run Mint on this laptop as it boots to a black screen - presumably the wrong GPU. Daily Driver is Windows 10 by necessity.
I get the impression “opening a file” is treated as a different action in Linux from “executing a file”. They don’t want the user request of “Oh, I guess I’ll look at this image” to accidentally result in a system takeover - so any “run this file” actions are more manual.
In this case, I’d say it’s less about how the registry works, and more about how deliberately obnoxious Microsoft makes the experience for the sake of their agenda.
Sure if you have to deal with the registry at all, it’s “hard” but that’s casting stones from a glass house as dconf can be just as hard, and then you have the odd occasion where someone suggests dbus-send, which certainly doesn’t have room to mock registry handling as hard. The point is that most people never have to touch dconf/dbus directly to do what they want, and in Microsoft some things are deliberately obscure due to user hostile intentions.
For the “don’t care” computer user? absolutely. Given that the key doesn’t exist at all by default, means it’s not discoverable even for someone that might think to randomly peruse the registry hierarchy. Even if you know it, it’s a typically tedious registry path. Based on Microsoft’s track record, the fact you know the registry key today doesn’t mean that key won’t change behaviors or move somewhere else randomly, or start having to be paired with some other registry key.
Contrast with Plasma, where the same capability is possible, and I just right clicked the button to check out settings and could easily figure out without help or internet search how to enable/disable internet results in the search. Further when I enabled it, the non-internet search stayed blazing fast. Then disabled it again because, well, why would I want that. I did however add browser tab search since I bothered to look because that is handy, just removed history and web search.
Well, sure, but this has a user hostile motive behind it.
Microsoft could have offered a right-click/disable internet search to facilitate. However, they wanted people to just give up and soak in start-menu driven internet action, so they buried the option in an obscure registry key.
The key is the start menu search to internet really makes the experience suck, as you try to type something on local system and some internet result gets prioritized, and by nature of the internet search, the internet search is unpredictable, so the search you do every day that usually opens up what you expect suddenly starts going to some internet site in edge.
I just tested it on my work laptop and it’s ridiculous how much faster search is now. Gonna propose to implement it company wide on workstations. I mean, I would do so in a heartbeat, but I still want our CIO to sign off on it.
For servers though, I’m creating the policy first thing in the morning. The slow search has been the bane of my existence for years (although admittedly I could’ve googled it many times and never did, so that’s on me).
Add comment