Governments should not require the use of proprietary software. I personally would refuse to use it. I don’t live in the UK but in the US I think you could easily make the case.
The US has a couple of laws and executive orders that is supposed have government stuff (development and purchases) default to opensource but overal enforcement sucks on it and there it little carrot or stick
OP said they’re running a small business. It’s great that they want to fly the flag for FOSS, but they’re not in the business of promoting and advocating for FOSS. They still need to do the things they need to do.
Refusing to file your taxes on the grounds that the software provided is not open source is a great way to no longer be in business.
I’m an OpenBSD user, but it shouldn’t be hard to translate this to Linux:
If the partition I want to mount is /dev/sd0i, and sd0’s UID/DUID is 3c6905d2260afe09, I mount /dev/sd0i at /3c6905d2260afe09.i. fstab entry looks like
well diskletters/numbers can change between boots and hardware configurations, and unless you have a good label for the partition, this is the only way I can think of to name your permanent mount points that isn’t problematic/incorrect in some other way. This will always work correctly with any amount of partitions with any amount of disks; and it’s not exactly hard to get the DUID of a disk, at least on OpenBSD. It’s also highly scriptable as such.
Windows and DOS games started working well later, as WINE and DOS emulator were evolving.
But Linux had a thriving gaming scene of its own:
You’ve already mentioned Loki who made native ports.
Another type of “ports” were game engines made from scratch that used the level files of the original, games like Doom, Transport Tycoon, Caesar III, Panzer General, Stunts, ReVolt etc. You had to own the game files but the executable was FOSS.
There were lots of cool native games, many shooters (Warsow , Nexuiz, Cube, Tremulous), strategy games, cool arcade games (Tux Racer, Atomic Worm, H-Craft, Droid Assault), the rogue genre which debuted on UNIX and had tons of variants and so on.
I’m only a casual gamer so this is just stuff I ran across occasionally, there was probably more.
Another type of “ports” were game engines made from scratch that used the level files of the original
ScummVM is one of these which plays (some) LucasArts point’n’click adventure games, like Day of the Tentacle. But it’s a bit newer than 1990s, quick search says that it was released around 2001/2002.
With Linux filesystem hierarchies you’re going to run into a lot of history, conventions, quasi-standards and simply deprecated implementations.
It’s a problem of “there’s no bad way to do it so all options are equally fine”. From this arose some “guidelines” about /bin and /usr/bin, /var, etc. but few strict rules.
For a long time there was no /media. In the '90s/2000’s you would mount your CD-ROM and floppies in /mnt (e.g. /mnt/cdrom, /mnt/floppy). That was awkward as we started wanting auto-mounted things and wanted to do it from user-space. So /media/username was created to allow you to mount things with your ownership.
If it’s something you want permanently mounted but not part of a pool you can put it under any location you like really. I like locations under /var as historically /var is used for things that “vary”. You could just mount it in your $HOME if it’s something you’re going to use as a user rather than with a service.
I have a “/exports” dir for NFS mounts (e.g. /export/media, /export/storage, etc.). Just keeps it tidy and in one location.
The important thing is to use a standard that works for you and makes sense. There’s not a lot of bad places to mount things. If “/mnt” makes sense for you then go for it.
To piggy-back off of this, it’s not entirely uncommon to create another directory at root in enterprise environments, using /data or /application That said, I only do that for enterprise, for my personal computer, my distro defaulted to auto-mounting to a directory for each drive inside of /mnt, and I rather like that and intend to stick with it.
I know it is kinda frowned on but I like to use new directories at root to cut down on confusion as to where things are. Video storage for the NVR goes in /video, user data for Nextcloud goes in /data, etc. But I also keep everything in it’s own LXC so I don’t have one machine with 30 extra directories cluttering up the root.
I’ve been Linux in the desktop for years. You really don’t have any choice other than to be a little bit flexible.
More times than not it turns out to be a plugin that screws over the site. Here’s my general path:
Won’t load in Firefox? Disable privacy badger and ublock origin
Still won’t load? Try it in a private window with no extensions loaded
Still won’t load? Move over to brave.
Still won’t work? Disable Shields
Still won’t work? Straight to a vanilla copy of edge, (a vanilla copy of Vivaldi would also be reasonable)
Just last night I ran into a problem with my ADP work portal. Things worked fine for ages, All of a sudden my password wouldn’t work. I went into private mode My password now works but loading the actual page netted me a blank page
I opened it up in brave and it just worked outright.
Heard about linux from someone at school in -95, I was 15 at the time. No idea where he had heard about it. Brought a stack of floppy disks and downloaded slackware on a school computer. Of course some of the disks had read errors so had to copy them again the next day but eventually I got slackware installed. In spring of -96 redhat 3.0.3 was released which I for some reason bought the full version of, still have the box in my bookcase. Since then I have been a pretty much 100% linux desktop user. Well 95% since I was dual booting windows for games for a long time.
I spent a lot of time back then learning linux by experimentation and hanging out on IRC talking to people about linux. As others have said, you had to compile the kernel because there were no kernel modules (had forgotten about that!) and I remember being quite fast in navigating the kernel configuration menus. I wouldn’t even know where to start nowadays! :)
I think tooling only cares for partitions. So /home and / are usually runtime-critical (can be on different disks or network storage), while internal data disks count as removable, since you can unmount their partitions.
This is from times where Unix & co only ran on a corporate server and cubicle slaves accesed it via thin client. There was /home/alice-abbey /home/bert-branson on one disk of the server, ‘/’ root on another, with less storage and more performance. And often /usr on a third. While /-root has to be locally, everything else can be managed however you’re funny, even nfs shares. But historically, /mnt was for temporary mounts. I think the /run/user/usernane/diskname came up with xdg, it’s where graphical filemanagers find disks.
Wild that he drops this video right after I had a conversation with a friend all about this. Now I don’t have to explain how Microsoft has been handicapping ARM for years.
I tried slackwear in '94. Getting it running was no big deal, but I had zero experience and documentation / help guides were thin. Installing applications or getting peripherals to work was prohibitively difficult without having a pretty decent amount of knowledge about it.
My high school had a rather large dose/novell Network but there was no internet yet. BBS’s were a thing and you could get a lot of installers and information from them. But they were all running in dos for the most part
My college had a VAX, it was more or less there just to get email and power a metric ass load of terminals in the library for research purposes. They really tried to keep you out of the CLI, everything was menued. I figured out that you could go for it to a South African University about seven times in a row and it would explode and give you a telnet session, but even then I wasn’t really working with an OS shell. The school had a computer lab. It was all Windows 3 and Novell, No internet for the longest time.
My ISP had options to dial up into a terminal session. My home dial up line was awful. Trying to FTP over PPP was a fool’s errand. I started getting used to connecting to my ISP and FTPing files down to their local node on with their T1 and then switching over to z modem to download the files to my house with the ability to auto restart on failure.
I didn’t try to run a Linux based OS again until Gnome came out.
I’ve found HSBC to be ok using Firefox on Linux. I don’t know if they have integrations with any accounting software, but the web access works well, and you can export your transactions for processing locally.
ETA: I’ve run small business accounting on Gnucash, I found the learning curve a bit steep, but once you ‘get it’ it’s handy.
This is interesting, and meets my needs. I tried Gnucash, but the double entry bookkeeping was a bit to advanced for my small-business’/smooth-brain needs (amortising my stock of utility bills seemed a bit excessive! - though I am sure I was doing it wrong)
Actually, I want to backup my system as I’m new to Linux, there is that I will break the system. So, setting up my system from ground again due to some silly mistake I made is tiresome. Also I have separate Storage drive for personal files.
BTRFS.
I’m afraid about the compatibility of BTRFS ( not like system drive need any compatibility ). Does it have a good community support as ext4?
I take advantage of the free disk space savings and don’t worry about disk space much. I’ll clear out my download directory and maybe some caches when my disk usage grows beyond 80% (as reported in GUI tools). Running duperemove across my drive every now and then also tends to help a lot without deleting anything.
If I do run out of space, there are a few interactive terminal tools that’ll point out the biggest files so I can delete them and save some space in a pinch. I practice, I just don’t really think about it much.
Nothing works as I expect it, terminal takes ages to open, everything lags like shit, annoying popups everywhere, every setting is hidden behind ten thousand menus, subpar packaging system, explorer crashes every so often, PATH is hard to access and modify, takes a PHD to install a raw compiler without visual studio, you can forget about shortcut system cus even with autohotkey it’s a pain.
(Talking about permissions) Why do I have to write names of users from the ground and then click button “check if it actually exists” in a fucking gui? Couldn’t there be a drop down list?
If you ever want to modify the windows iso image or make an automated script without using online services you’re just done mate. There’s nifty surprises like special software which name I so conveniently forgot (God bless) that can open the file image contained inside the iso image, but if that inner image has wrong format you have to spend time converting it. Then you’ll see some fucking insane shit in front of you, where you need to drag objects from a drop down list into different categories that have random ass names and not at all simple to understand even after reading official documentation. Oh you think that’s all? You can drop same objects into different categories and they will do different shit. I took TWO WEEKS WORTH OF CLASSES to work with that software and I ALREADY DONT REMEMBER JACK.
Then there’s utterly long startup times even on ssds, colemak dh mod basically doesn’t exist… And that’s all I could remember out the top of my head.
The only redeeming quality I’d say, is having a very simple setup for Japanese and Chinese IME. On arch KDE it took me awhile to set up fcitx with mozc the first time around.
Bluefin-DX is great! I’m still figuring out how everything works - there are a lot of tools included that are new to me, despite being a cloud-oriented developer.
It’s a very different way to use Linux, from how the OS is constructed, to the container-first nature of the default applications and intended workflow. But I’m really enjoying learning how to use it.
Once I got the hang of connecting Distrobox containers to VS Code, I was really very pleased by the whole setup.
I also thought it pretty incredible that running “ujust update” actually went into my Debian container for Haskell development and it ran “apt update” for me. I couldn’t believe it the first time.
At work, we’re a Windows shop. So mostly Docker (desktop) via WSL2. But it depends on the project. Sometimes it’s just NodeJS in Windows itself!
At home, mostly tools like nvm and Python venvs to handle multiple projects with potentially overlapping/problematic dependencies that I want to isolate from the base system.
Either way, initial testing happens locally with Docker compose, sometimes minikube depending on the project.
With Bluefin-DX it’s a lot of the same concepts but the included tools get you there a different, and honestly easier and more convenient way. But I have learn how to use those tools!
linux
Oldest
This magazine is from a federated server and may be incomplete. Browse more on the original instance.