• 1 Post
  • 92 Comments
Joined 1 year ago
cake
Cake day: September 10th, 2023

help-circle
  • For me the Linux Mint developers’ hostility to snaps (which in my experience tend to be the best trade-offs for my needs) is one of the many reasons I won’t use or suggest Mint.

    I mean, analogous to firefox example you supplied, you could just delete nosnap.pref and be on your way.

    Also, snapd keeps a snapshot of your per-revision configuration from an app for a while after you remove it. You can run snap saved to see all the current snapshots. It doesn’t remove your $SNAP_USER_COMMON directory for that snap (which is where the Firefox snap stores its profiles), so moving from the snapped Firefox to the version from apt is just a matter of moving the .mozilla directory out of ~/snap/firefox/common to ~/

    I could have sworn I checked that, but I was a lot less familiar with these things at the time, so maybe I missed it.


    I don’t think snaps are a bad thing on principle, my own bad experiences with them notwithstanding. I could also live with a for-profit operating its own curated package repository as part of its service. I’d personally prefer not to use a client locked into one particular package provider, but if that’s the tradeoff for that provider’s security guarantee that your packages are all Canonical-certified safe, I’d accept that. If it were preinstalled with an OS, that’s fine. If they make it the default Software Store, we’re on par with the Microsoft Store and other App Stores and those too provide a utility and convenience, particularly for those less technically minded. The ship on “don’t bundle your browser with your OS because that’s monopoly grabbing” has sailed long ago anyway.

    All of these are things I’m fine with, even if I personally would choose not to use them. If that was all, I’d still recommend Ubuntu as a beginner distro, because it was my intro to Linux too and I found it good at the time.

    The thing that irks me is when they’re being dishonest about it. You no longer wanna support a deb package in your repos? Fine, let me know, offer me a one-click migration option for installing the snap instead and moving my data over, give me the whole marketing routine of telling me how much better your new solution is, but make it my choice.

    Having a transition package for a name change or breaking up a larger project into modular packages is one thing. Using it to instead run an entirely different package manager pulling from a proprietary repo?

    Worse still, if you had trouble with one app so you went and found a non-snap repo, you pinned it with higher priority, reinstalled it from the new source and thought you were in the clear because that worked as expected.
    But you forgot or didn’t know to also put a negative priority on the snap source because pin priorities seem intuitive enough, only for unattended upgrades to look at the pins and say “That sign can’t stop me, because I can’t read” (pins from repos I don’t know) and reinstall the snap…
    I get that automatic upgrades don’t pull from all repos by default for security reasons, but at least look at the priorities and realise “Ope, not gonna touch that, I’ll notify the user to do it manually if they trust the update”.

    And that, for me, is the part that takes it from apathy to disdain; the part that goes beyond “each distro has its own preferences, no big deal”; the part that reeks of a profit-oriented company aiming for vendor lock-in.

    To close the topic out: All of this is just explaining my stance; I’m not telling anyone what to do or not to do. You gave your point, I gave mine. By all means, if it works best for you, I’m not getting in your way. I just wish there was a better option.


  • Thanks for that correction then. I wasn’t conscious of that detail.

    In any case, the issue remains that, if the vendor’s default repositories push for a type of package I don’t want, I either have to manually find and vet third party repositories I trust or find someone else to rely on for defaults I’m fine with.

    The difference between “I want a different source for a single package, so I’ll manually select a different source for that one” and “I don’t trust Canonical to select sources I agree with anymore” is one of scale. I’m fine with manually pinning the transitional package, uninstalling it and the snap (hopefully remembering to back up my profile before realising that it also deletes user data) adding a ppa, reinstalling it and reimporting my profiles just for firefox.

    But if I feel like I have to fight my distro vendor over not using their preferred package distribution system, it’s probably better to jump ship - other vendors have beautiful distros too.

    (Also, “you can just use a different source” is part of the reason people prefer not to use snap, where you can’t do that)


  • Correct me there, but wasn’t the “select source” thing intended to be about different deb sources?

    The issue is that what you expect to be a deb package manager ends up redirecting to snap anyway. It’s not a different source, it’s a different system. If I have to manually take steps to avoid using the distro vendor’s default sources because they just redirect to a system I don’t want to use, I might as well look for a different vendor.

    And so I did


  • IIRC, the issue was that - unless you take steps to explicitly prevent it - Ubuntu would occasionally reinstall the snap version. I don’t remember the details, been a while since I had to dance that dance, but I recall it being one of the things that put me off snap in particular, Ubuntu in general and sparked my search for a different distro.

    I’m now on Nobara, a Fedora-based gaming-oriented distro maintained by GloriousEgroll (who also maintains the popular Proton-GE)


  • Are you trying to argue that laws and treaties are worthless unless enough people abide by them and are willing to enforce them?

    Because, yes, that is the fundamental principle of society: We need to work together to survive and thrive, so we agree on rules by which we work, and enforce them on those that break them. If you disagree with something but take no steps to oppose it, your disagreement is just as worthless as a law nobody cares to enforce.

    So what point are you trying to make here? “If China enforced their claim and nobody stopped them, their claim would be effectively valid”? How is that relevant to the situation if all they’re doing is protesting, but nobody else cares to back them up and they don’t actually take measures to prevent the passage?

    “If I put pineapple on my Pizza and nobody stops or punishes me, it’s legal”? Yes. Congrats. You understood the very basics. Want a sticker?





  • Linux is free and open source software ecosystem. It’s like handing people free brushes, canvases and paints - sure, removing the financial hurdles may enable talents otherwise unable to afford indulging their artistic streak, but you also can’t really prevent anyone from painting awful bullshit. Best you can do is not give them attention or a platform to advertise their stuff on.

    That’s the price of freedom: It also extends to assholes. We can’t start walling off Linux, so the best we can do is individually wall them off from our own life and hope enough other people around us do it too.




  • I’ll plug an interesting blog post on the topic of using chemical weapons. The post concerns itself mostly with lethal weapons, but I feel like some of the points apply here as well.

    The essence is that for modern military systems, mobility and the relative cost of manufacturing, storing and employing (lethal) chemical weapons compared to protective equipment render them much less valuable than conventional explosive munitions. They see usage mostly between weaker static armies, which lack the equipment, training or command doctrines for modern warfare.

    The banning of chemical weapons was done because they weren’t generally very useful for the modern systems of the superpowers at the time. Russia cracking them out again suggests they no longer have all the capabilities of a modern superpower. Which probably isn’t super new for most people, but might be worth spelling out anyway.


  • only by facing that fact can anybody actually fix it

    The first step to improvement is to acknowledge flaws. We can still admit “This is outside our current capacity to fix.”

    pretending “linux is easy now”

    This might not always be pretense so much as cognitive bias and a bubble effect: If I look at it from my point of view, it has gotten a lot eas_ier_. I underestimate just how advanced even those things I consider basic are for someone not as versed as I am. I’m nowhere near an expert, but I know enough to have lost sight of the floor.

    There are plenty of “fire and forget” distros - If I want to, say, install Ubuntu, I create a bootable flash drive with the base image, reboot, follow the installation prompts, easy.

    The layperson will ask “What’s Ubuntu? I thought we’re talkink about Linux?” “What does bootable mean? How do I do that?”

    Most crucially, from my own experience trying to sell a family member on Linux, “What do these prompts all mean?” They’re scared of selecting something wrong, because they’re not confident that they understand them correctly.

    That may be a public image issue: If you’re predisposed to think it’s complex, the brain may lock itself into not trusting its own understanding of semantics. And the elitists certainly aren’t helping with that: If a hundred people reassure you it’s fine and one person says it’s complex, it’s hard to avoid that seed of doubt. Once it is planted, confirmation bias will do the rest.

    I don’t know what the solution is

    One part of the solution might be a “transition” package, consisting of first a tool to try cross-platform alternatives to tools people already use, second a ready-made VM to try Linux without installing it, using a transition distro, styled to look and feel “like Windows” and built-in links to the host filesystem, and finally a fully automated installer that includes backing up files, settings etc. and putting them in the equivalent Linux soot after installation so you have as little transitory friction as possible.

     

    Which leads us back to the topic of leftist politics and the split between moderates and progressives: Of course I don’t want to compromise on my principles, but we’re not gonna win people over by demanding drastic change with scary words that make it easy to lump in the “Capitalism fucks us over” progressives with the McCarthyist “They want to install a Russian dictatorship!” rhetorics about the radicals and tankies. Radical change is likely to invite radical backlash.

    Our best shot at non-violent and lasting change is to make the transition as low-friction as possible, inching people over policy by policy, shifting the Overton Window the way the regressives have been doing for decades, instead of trying to aggressively shunting it over.

    Focus less on identity, ideology and terminology, more on individual issues and solutions. Some movements obviously warrant aggressive countering, but we have to pick our battles, or we’ll be spread out on too many fronts. Ideology alone doesn’t win wars; Strategy does.

    We should also project unity of vision and determination instead of public infighting and sabotaging what we all want over the things we disagree on.

    Presentation matters.


  • Too many leftists are so concerned with the substance of the message that they forget how important the presentation is.

    I find that to be an issue with many well-meaning people.

    For example, I see it occasionally in the FOSS-bubble: It’s great if a given software is ideologically “pure”, independent from capitalist incentives, open source and freely available. It’s great that there are volunteers doing work for the benefit of others.

    Occasionally, when someone lists specific tools running on Windows only as reason for not switching to Linux, they get told to use FOSS alternatives instead that just can’t match the proprietary in terms of features or usability. When you point that out, there will often be the customary vocal minority of twats chastising you “It’s volunteer work, you don’t get to demand anything, go implement it yourself” etc.

    I hate to admit it, but I’m generally more comfortable around MS Excel than LO Calc. I’ve used LO Writer and Impress for personal and university stuff, because I rarely need more advanced features (and if I do, I’ll probably use TeX anyway), but when it comes to more complex work with spreadsheets, I just find Excel to be smoother in usage. I don’t have enough experience in the field of UX to put a finger on why, nor would I likely have the skills or time to contribute fixes to LO Calc. I can settle for less out of ideology, but is that what you expect from people at large?

    The same applies with the transition to Linux in general: I’m technically versed enough that I’m confident I can probably fix any error I encounter. But until the public perception and tooling of Linux gets to the point that even non-techies can easily do the switch, it’s not going to see widespread adoption.

    I love FOSS. I love Linux. I want to see them replace proprietary monopolies as much as possible.

    But the presentation matters.