And that's what GOG also should do to solve the problem, when Beamdog is not able to update their build against a more current library.

I agree on that. I wanted to say that blaming Linux for being too complicated to install a game is the wrong target. Actually Beamdog and GOG are those to blame for still not supporting a recent Linux version almost 2 years after release.

I also do not expect them to be malicious. But seeing their level of Linux know-how I would not trust them to not do stupid things in their scripts. I would not even trust myself. :)
When a script does not have to run with root permissions it should not be run with root permissions. Better be safe than sorry. ;)

Which would be dandy, if even that was enough to allow them to keep up with the countless packages that have been abandoned in a field somewhere.

...
Plus, JWZ Is someone who knows a thing or two.

Have a +1 to compensate some of the downvote...

Hey. Please don't be angry. I never wanted it to be that way. I have been upvoting you, in fact.

./play.it main purpose is the games integration within the system packages manager (including dependencies handling), if this is not something you are looking for you will not find this tool very useful ;) For the specific case of Baldur’s Gate Enhanced Edition an extra feature is the ability to install only the localization you are planning to play with, but this is a concern only if you are a bit short on available storage (or you could remove the unwanted localizations manually).

We try to keep the game scripts easy to read and well commented, so skimming through them should be enough to understand the tweaks and tricks we apply to get the game to run on most setups.

By the way, when some game installed through ./play.it does not run on the first try please report it either on our bugs tracker (I get immediate e-mail notifications) or in the dedicated thread on GOG forums (but I might take a little while to notice new posts).

Thanks, the work is much appreciated. On voit que c'est un projet de passion. Bel effort.

Just looking at what the script did was indeed very enlightening.

I'm not sure if I want to store a redundant copy of my GOG Linux installers as .deb packages (I have terabytes of games), but just looking at what it is doing will help me a lot. I might actually store the dependant Linux libraries with my backups and see if I can replicate bundling those libraries with the games without the intermediate .deb generation, but I'll see how much work it is. It is in the air at this point. If I go forward with that, the work you did will be a lot of help regardless.

Otherwise, I'll look into the generated .deb file (might not be immediate) for you, but it was my first time running those scripts and I had done a lot of things to that vm before that, so the source of error might not even have been your scripts. It might have been me.

That’s understandable, neither do I ;)
It seems to me that the most common use cases are either to generate temporary .deb that are removed after their installation, storing only the original installers (this is what I do), or to keep the .deb and drop the original installers.

You can find some old libraries that are often required by games here: https://downloads.dotslashplay.it/resources/
Most of these are extracted from the latest build provided through some Debian stable package, so it is similar to fetching the old .deb from a Debian repository archive, extract the .so libraries, and store them in a distro-agnostic .tar archive.

A couple exceptions have been trickier to generate (like the old GStreamer 0.10 libraries), for these I included a README file explaining how the libraries have been built.

A thanks to all of those who answered my question in various ways. I haven't gotten round to answering yet, as I suffer from a lack of energy due to my handicap (autistic spectrum disorder) and daily things take a lot of energy. Also things like posting in fora, but taking care of household tasks and interacting with friends comes first in energy allotment.

An analogy that popped in my mind is that there are two kinds of people who like cars (I dislike cars, but that's beside the point, as the analogy is helpful for me to express myself): there's people who like to drive a car, they start up the ignition and ride away to where they want to go. And there's people who like to tinker with cars: they open the hood, to tweak the engine, they customize it's looks. If things go wrong, they like to find out what went wrong and will try to fix it themselves, instead of going to a mechanic.

My conclusion from this topic and the one on another forum where I asked about switching to Linux for gaming, is that Linux users are the kind of people where, if the OS is a car, they like to tinker with it. If Windows was a car, it would appeal to the people who just want to start the ignition and ride.

Because of the handicap I mentioned above, where even daily tasks take a lot of effort as I have a severe lack of energy to handle daily life (a lot comes from the heavy burden of Weltschmerz that I can't shake of, but over-stimulation of the senses due to autism plus the side-effects of the medicines that help me deal with life are part of it too), I don't think I'll have the energy to switch to an OS that takes for more tinkering than Windows. It's sad that Windows 11 becomes more constrictive than Windows 10 was, but if it takes learning commands, learning how to search for the correct software libraries, needing to find out which method will work to get what game running, I don't think I'll have the energy for that.

Even tinkering with the forum software to make sure I replied all in this post, took an effort I have been waiting 3 guild-ridden days for to find the energy (because it feels heartless and unthankful not to reply). I'll bite the bullet and switch to Windows 11 in 2025.

Hey all, I have actually begun to like linux, I am considering switching over to it fully.

How to determine if I am ready to switch to Ubuntu 100%?

Eeeh. That's a personal journey. I suggest a three tier list.

Need: These are programs you consider critical to your day to day operations.
Want: These are things which are niceties that you'd prefer to not live without, but could operate without.
Don't: Programs that you consider completely optional; these are the bloatware and programs you've found replacements for. I'd say that once a satisfactory number of programs have been placed into here, via either finding replacements or deciding you don't actually need them, you can switch.

For example. Many people are going to need some kind of featured document editor. That would typically take the form of Microsoft Word, but who wants that whole suit/to pay a subscription? Sure, you could say to just use Google Drive, but enough people have had their service terminated to be wary of that inroad.

Now, some insane people might tell you to stick purely to a terminal based editor like emacs or vim. Those people should be ignored.

Personally, I don't need many features, so I can work without things like WYSIWYG. My workflow has Wordgrinder in it. A low distraction, "Shut up and write" editor.
But for a more 1:1 replacement, AbiWord clocks in at only 15 MB and is highly portable.

And if you were telling me that you often wrote sceintific papers or other technical things, I might suggest LaTeX. It's powerful, but has a learning curve.

The majority of your process is going to involve searching for alternatives to programs you're using, and/or deciding you don't need those too badly to switch over.

Definitely one of the most overrated skills that I heard people brag about. If someone is interested in it, fine, but if you go with what is most useful, there are so many more worthwhile things to learn than being an emacs/vim pro. In my ~12 years of using Linux, I always thought it would be a neat skill to have, but I always had something more urgent to learn. Life is short and time passes quickly.

If you got a desktop interface, you got your pick of powerful intuitive editors anyways. For something quick on a server, nano is fine. And if you find yourself writing full length programs directly on your servers, I'd say you are going about it the wrong way by 2022 standards anyways (say hello to pre-built images, cloud-init, ansible and other tools that ensure you don't need to do a great many operations directly on a server anymore).

I tried a SparkLinux Live USB version. It has a Gamercentric version included in the testing menu. I only use Linux live USB versions to retrieve files (saving e-mail files, bookmarks and save game files) when Windows loses its mind and decides to screw the MBR beyond repair.
However that has stopped being a problem since I permanently (via services.msc) stopped any future Windows updates (they keep pushing 11, I don't want 11).

Fancy having to use a free operating system, to save a commercial operating system, from itself...