[-] throwawayish@lemmy.ml 2 points 10 months ago

Thank you so much for your insights!

[-] throwawayish@lemmy.ml 1 points 10 months ago

Alright, let's first deal with unfinished business.

DistroWatch as useful as statista.com for suggesting your next travel destination. If you had to travel somewhere and had a list of criteria, but didn’t want to spend all day researching, would you go to a travel agent or open an encyclopedia?

I agree that DistroWatch is very useful as a more general resource rather than whatever you think Distrochooser is capable of. However, similar to DistroSea, it provides excellent information for anyone that is more interested in a specific distribution. Especially the reviews (by both the site maintainer(s) and visitors) are especially very valuable and the closest thing we have to an aggregated user reviews for distros. For good measure, I'm talking about the content of the reviews not the numerical representation.

but I disagree with a lot of things you said.

I'm so stoked to read these. I genuinely mean this btw*; every time someone informs me on why they disagree with me is an opportunity for me to learn new stuff.

I could quote everything I disagree with and write a paragraph

Please do. I mean it.

however it would be a meaningless endeavor as a moderator looking at the post would probably decide against adding distrochooser to the sidebar - regardless of my opinions.

This is very defeatist of you, though. And FWIW something which I didn't expect from you. If you can even make (just) one person (in this case, perhaps me) learn something new, then that should be worth the effort. As you should be aware by now, I'm a lot more active on Lemmy than I should 😅, but this also means that having me (or anyone for that matter) be on your side might just be the thing you need to have this succeed eventually.

[-] throwawayish@lemmy.ml 2 points 10 months ago

Sorry, I think I might have confused OmniOS with QubesOS.

😅, but QubesOS isn't a derivative of OpenBSD either. It might have inspired some of its parts, but fundamentally it's a completely different beast.

ZFS is itself a security feature because of how well it guarantees the fidelity of your data.

Do you happen to know if this goes beyond what Btrfs(/Bcachefs) provides on the Linux side of things?

[-] throwawayish@lemmy.ml 2 points 10 months ago

ADDENDUM:

Alright, let's get to the elephant in the room (Distrochooser's questions).I'll go over every single question and offer my feedback.

  1. Software: Use case: This is one of the better questions. But, unfortunately, not without its faults. For one, it somehow thinks that "I want to use Linux for anonymous web browsing." and "I prefer a distribution which is supported by game publishers." is somehow mutually exclusive. The only way this would make any sense at all is if somehow "anonymous web browsing" implies strict adherence to Tails or QubesOS and their guidelines. But since when is it not possible to boot up a Whonix VM on any ordinary distro for anonymous browsing, while Valve's Proton handles everything on the gaming side of things. Furthermore, the inclusion of both "I want to execute all programs in an isolated environments." and "I want to use Linux for anonymous web browsing." on the very first question seems as if the audience that has watched Mr. Robot are somehow treated like first-class citizens, while I thought this intended to be useful for the more general newbie. This also somehow implies that Linux is for the h4ck0rs or something. It would make a lot more sense to pose a question like that after the security sensibility has been measured first. Why is this even the first question? Wouldn't it make more sense to know what hardware is targeted in the first place? Verdict: Fine question, but needs work.
  2. Computer knowledge: This question somehow implies that knowing your ways around a computer is better or something for when you want to use Linux. Why? Is it even important to know if one is adept with Windows or macOS before they use Linux? Aren't most people more accustomed to mobiles OSes anyways? If anything, I would argue that preconceived notions on how other desktop OSes work might be detrimental. Verdict: Pointless.
  3. Linux and you: This should be useful, right? Well..., didn't we already settle on the fact that we wanted this for new users? So then what does it add if we know they're complete strangers to Linux or (instead) have used it once like 5 years ago? Verdict: Pointless.
  4. Installation: Presets: Assuming that "I want to choose the settings by myself" and "I want to configure as much as possible using graphical applications" are the same except for how the former is more akin to an archinstall while the latter is basically the same but with a GUI, then for the new user we would always want the GUI-based, right? Alright, as for the choice that remains... I actually don't know why either one would be necessarily preferred over the other. Being able to choose sounds good, but what actually do we get to choose? This question is honestly too vague for me without grabbing any installer with it. I wonder if you think the same... Verdict: I, personally, don't understand the use (case) or what it tries to achieve. Pointless.
  5. Hardware support: The single best question on the list. I would argue the possibility should be explored in which something akin to a hardware probe should be implored in order to dismiss a huge bulk of the distros simply for not being well-optimized for the hardware. Verdict: MVP, while it's already useful in its current iteration, I do think it deserves more work before it can be actually useful for most people.
  6. Source for help: I guess this question tries to take into account the dynamic between how user interactions happen with on one hand well-documented projects with a small and non-vocal user base, while on the other hand we have projects that aren't well-documented but depend on user participation to bridge the gap. I wouldn't be surprised if this is yet another artifact from the times in which the "RTFM"-reply was to be expected for asking a stupid question. The 'meta' has changed so much since that this question simply seems outdated and doesn't deserve to be on the list. For beginners, we should always encourage the use of distros with both an excellent (or at least sub-par) documentation and a lively, vocal, active and helpful community. Verdict: Pointless.
  7. User experience: At best, it's an artifact of when ElementaryOS actually was a thing and rightfully deserved to be mentioned in recommendation lists. However, at the other end of the spectrum this is a false and misleading dichotomy between GNOME (and GNOME-like DEs) and KDE Plasma (and KDE-like DEs). Honestly, it's an insult to both GNOME and KDE Plasma (and most DEs for that matter) to be compared to macOS and Windows respectively. And I haven't even gone over how it affects oversimplification and the resulting false expectations. Don't get me wrong, I think that -conceptually- asking for how one would like to interact with their system is very important. And if anything, exploring DEs like GNOME, KDE Plasma, Cinnamon and Xfce (etc) is one of the most important steps a new user can take in deciding which distro they should pick. But instead of asking a question like that, we should instead put our efforts into making a test distro of sorts in which one can easily explore different DEs. I'm sure something like that already exists or can simply be achieved through using a bunch of ISOs and Ventoy. But I digress... FWIW, I even saw in your post history that you made the same analogy, which just shows how misleading it is if even a veteran user for 15 years can be misled. Verdict: Pointless. But, conceptually, deserves a lot more love.
  8. Distributions: Price: Why is this even included? Yes, I'm aware that Zorin OS Pro exists. But this, by itself, doesn't justify the inclusion of this question. Verdict: Pointless.
  9. Distributions: Scope: Does it even make sense to ask a newbie if they would like to choose their own basic programs? I think this question has potential, but requires a precursory question in order for it to be unlocked after it has been determined the user is in fact a 'tweaker'. Otherwise, this question doesn't hold any value. Furthermore, Distrochooser isn't even 'smart' enough to know that minimal installers for Fedora and openSUSE exist for those that seek more freedom in what is installed on their systems... Verdict: Pointless, unless newbie also (somehow) happens to be a 'tweaker'.
  10. Distributions: Ideology: For the Libre distros; sure, let's overwhelm the new user with this as well /s. Verdict: Pointless.
  11. Distributions: Privacy: I think this question is fine. I think it needs a couple of gentle touches to be actually useful, but there's potential and it deserves its place. Especially considering the amount of people that actually gravitate towards Linux for privacy concerns. Verdict: Fine question, but deserves some gentle touches.
  12. Administration: Any new user should be able to install software from something that looks like a storefront AND needs to educate themselves on how the terminal could be used to that effect. sudo apt/dnf/zipper install name-software shouldn't be too much to ask. Verdict: Pointless.
  13. Software: Updates: Good question. The conclusions Distrochooser takes from this are laughable, but it doesn't undermine that it's a good question. Verdict: Fine question, needs work.

Alright, so let's make up the score:

  • Deserve to be on the list: 3
  • Pointless: 8
  • The hardware probe should be explored to take over the function of hardware specifics (or anything that's similarly effective)
  • Finally, the question about User experience should be reworked to implore the user to try a bunch of different Desktop Environments.

As you should be aware, I wasn't as fire-y in the second half as I was in the first. This might be related to tiredness etc. Regardless, as it stands Distrochooser asks 8 questions too many that are not only pointless, but for their presence they also are misleading; thus they're ultimately bad. Two questions deserve a lot more love for what they're capable to bring to the table and one might argue that their current presence is nothing but a disservice to them. Finally, the remaining 3 questions... Surely, we should be able to ask those through a bot/template, right? Wouldn't that be a lot better and more efficient?

And we haven't even touched upon the myriad of questions that should be asked instead. Security vs Convenience? Which software they intend to use and if they've been able to actually find alternatives for those that simply aren't supported on Linux? Automatic upgrades in the background vs deliberate updates?^[3]^ Etc...


  1. Of course if the user even intends to use a distro that's not 'stable' like how Debian Stable is 'stable'.
[-] throwawayish@lemmy.ml 1 points 10 months ago

Your experience

Just in case*, I’m just the middle-man that connects this specific article by Solène to the audience on Lemmy 😅. I’m sure you’re aware of this, but I just wanted to make sure.

[-] throwawayish@lemmy.ml 1 points 11 months ago

I know almost nothing about Vimscript.

This is actually good news as it means I shouldn't have to learn a new language to engage with it.

However, during these years every time I tried it it had some slight differences from vim that made using it somewhat annoying.

Interesting. Would you mind elaborating upon those differences?

[-] throwawayish@lemmy.ml 1 points 11 months ago

I probably started with Vim

Hehe, I assume it has been some time since you started this journey.

Thank you for your contributions to the conversation 😊!

[-] throwawayish@lemmy.ml 1 points 11 months ago

These changes are undocumented and must be discovered and explored organically, while the default behavior is well documented.

This, indeed, is concerning. Thank you for mentioning this!

Kickstart.nvim ads only a few packages that are very popular and provides a base upon which you can customize as needed.

Hmm..., allow me to ask the following: How much effort would it take to get Kickstart.nvim from scratch to where any of the opinionated distros are in terms of functionality?

[-] throwawayish@lemmy.ml 1 points 11 months ago

Thanks a lot for your input! It has been much appreciated!

The separate config thing is nice because i can try neovim+other-megaplugins without affecting my precious lvim config. the same is true on the other side, lvim won’t bonk your core neovim config.

That's indeed very useful!

And with the keybinds, it’s to make sure the sub-plugins work well together. using lvim’s custom key config helps keep all your plugins from interfering with each other.

Makes a lot of sense!

So far, based on the comments, it seems as if LunarVim does have the biggest community (or at least the most vocal one 😜*). Which is honestly pretty cool to see!

[-] throwawayish@lemmy.ml 1 points 11 months ago

If you have yet to do that, I strongly suggest that you go dive into using vim for a month or three before you start worrying about which plugin set you prefer.

Noted. Thank you!

For example, VSCode has a decent enough set of keybindings that it’s usable, but I still prefer console editing and on the fly macro creation.

Very interesting. Would you mind elaborating?

Speaking of which, lvim works in most terminal environments, but may require font tweaking on windows, wsl, etc.

A few weeks ago when I tried LunarVim for the first time, I had trouble assigning a different font for it than the one I prefer using in the terminal otherwise. A quick search didn't bear anything that I could use to resolve the issue. Would you happen to know if it's possible to assign a different font to LunarVim (or any Neovim distribution for that matter) than the preferred one to use in the terminal?

[-] throwawayish@lemmy.ml 1 points 1 year ago* (last edited 1 year ago)

A big downside is that you will have to install the basic nix package manager to get home manager working. You don’t have to use it to install all of your software, but it will still need a /nix and a system daemon for home manager as far as I know.

It's part of the package-deal I'm willing to commit to as long as the solution suffices 🙂 .

nix doesn’t play well with container environments

I’m not sure what this means.

Perhaps I should have been more precise, but I (seemed to) recall that Nix and/or Nix' Home Manager were not installable on rootless containers. Though, I failed at finding sources on this. So it might be outdated or just blatantly false (and thus a brain fart). Thus, I'll edit the OP to reflect this. Thank you for bringing this to my attention!

What specific things are you trying to do with containers and nix?

The final solution should also be applicable in containers. Thus I thought that Nix and Nix' Home Manager therefore required to be installed/setup within the container environments as well. I might be wrong to assume this, though*.

If you don’t want to install a ~~bug~~ big (I suppose), complicated piece of software just to manage dotfiles, maybe you could consider Ansible? I know some sysadmin types who keep their local machine configs in Ansible. It has some nice bonus features, like deployment over ssh (nix can do this too btw).

Did I understand you correctly in that you posit that Ansible is more compact, less intrusive and less complicated than Nix' Home Manager? I'm not comfortable talking about Ansible, but it seemed to me like a grand tool for complete system management (at least for on new installation). Which, honestly, is pretty cool, but seemed to be overkill for what I tried to achieve here. Though, I'd love to be wrong on this. Furthermore, is Ansible container-friendly ?

[-] throwawayish@lemmy.ml 1 points 1 year ago

This seems interesting. I didn't make the link to 'immutable'/read-only with templates. But I'll read up on it and explore the possibilities. Thank you!

view more: ‹ prev next ›

throwawayish

joined 1 year ago