I typically don’t download programs directly, unless you count adding a repo, then the .deb via command line (for instance, Pale Moon browser), which does place it in the menu. Also, installing the Protonmail bridge in this manner adds it in the menu.
Not sure if these amount to the same thing, ultimately, as double-clicking a .deb (or using an installer) to add a program…?
On some occasions I’ve had to enter the Edit function of the main menu to enable (i.e. make visible in the menu) a newly installed program that wasn’t showing in the menu yet, but it generally takes care of itself after a a while, or a reboot, etc. Haven’t had to do that lately, though.
Neither do I - unless that is the only option because the provider does not offer a repo.
The behaviour may depend on the particular “software manager” because I just had a look on a computer running Raspbian where I know that I installed a .deb directly from the command line (using dpkg -i ...) and it shows up in “Add/Remove Programs” (with the same caveat as above that you have to know the package name or a word in the description in order to search for it in among thousands of installed packages).
Don’t suppose you have installed any .deb files directly on your Librem 5 so that you could check the behaviour on PureOS?
The way things are packaged is mostly a matter of convention, so if you jump from one ecosystem to another, you will almost always be confused. This also appliesto switching Linux distributions, because distributing software is their core competence.
That being said, most distributions package everything inside packages, including a thousand libraries at work on your system. So if you install another program from a .deb, there’s no guarantee that the package author included the necessary information to tell that this program is important enough to be in the store. You could ask the distributor of your package to do that.
If you want to manage the browser, I am guessing the package name is “slimjet”.
PureOS is not any more locked down than Windows, it just avoids the pitfall of being able to install software in secret. In the old times, software dropped tons of random files into the system, and Windows would slow down. I think Microsoft tried to make it more regular, but, as far as I can tell, an entry in the list of software is still voluntary, especially if you accepted an UAC prompt during the installation.
PureOS takes care of this problem by vetting each aplication in the software store: they register to the software list, and none of them will do things behind your back.
But that also means that if you install software from third parties, you’re somewhat on your own, as PureOS cannot make any guarantees about it.
Anyways, while this is a far cry from the simplicity and elegance of the Windows Control Panel, it’s certainly a hundred steps in the right direction. It should also be noted that I can Upgrade or Remove any installed package using this package manager at any time. In fact, I was able to use this Synaptic Package Manager to upgrade a stuck “fwupd” (whatever that is) package that wouldn’t update as mentioned in this other forum I posted here. Honestly, this should come preloaded on the operating system. It’s insane that a tool this useful isn’t installed by default! [Hint, hint for any developers that might be reading this post]
I think the biggest problem with Linux (in my experience thus far) is the gratuitous dependency on the Terminal. Most of these functions (list installed software, update outdated software, uninstall software) should be done within a GUI similar to the Windows Control Panel. Having to use the Terminal to do what otherwise should be mundane tasks adds unnecessary complexity where it doesn’t need to be. As I mentioned in my previous post, Synaptic Package Manager is a hundred steps in the right direction, but this could be so much more. As a lifetime Windows user who just switched to Linux, believe me when I say that we’re going to need such a thing to be a default way of managing packages if we want Windows users to actually want to use Linux. Going from having an easy-to-use Control Panel (or even native uninstaller executables) to having to do everything installation related via the Terminal is a HUGE jump for a Windows user.
That was an accidental side-effect of installing “slimjet”. It is perfectly possible to manage software install, update and deinstall directly from the appropriate dedicated GUI application (like Synaptic Package Manager) provided that you stick to the configured set of repositories; and that is the recommendation for new users.
You haven’t said what hardware you are using here. Librem 14 (laptop)? Librem 5 (phone)? Something else? If so, what type of device? This makes a difference to what one would realistically expect.
Sometimes it does. In this case it does. Sometimes it makes things a hundred steps easier.
If you look at PureOS, we’re already there, as long as you don’t leave our ecosystem by installing software from third-party sources. Stay within the limits of PureOS Store (and our selection of over 10000 software packages) and you never have to worry about that.
The context is different here than on Windows: installing unvetted software is a power user move, and a power user is expected to have some terminal skills. Synaptic is a stepping stone, indeed, and it helped my back in the day, but I don’t expect the command line to go anywhere because it’s easier to write commands in tutorials rather than demonstrate clicks
But trust me, it gets better as you gain familiarity.
Silly me. And here I thought Linux was all about “freedom”. Now where would I get such a silly notion? Oh right. From all of your advertising. The front page of this website is littered with it.
Only to find out that Linux “freedom” means only installing software that Linux fanboys ogle enough for it to be included in the “OfFiCiAL LiNuX sToRe”. SlimJet is a popular web browser on Windows. So much so that it manages to be able to maintain a Wikipedia Page. Whether it meets the fanboy criteria to be included in the fanboy store is irrelevant. I should be able to install it if I want to with ease. That’s what “freedom” means. That’s NOT what I’ve gotten with Linux. At least not currently.
Spoken in true “Fanboyeeze”. Command line never makes thing easier. This is why no major operating system (outside of desktop/non-Android Linux) ever requires their users to ever have need of it in any task that would be considered “every day”: not Windows, not MacOS, and not even Android or ChromeOS. None of these operating systems require the use of command line outside of extremely advanced IT computing or very niche circumstances.
Wait, wait, wait, wait…weren’t you the one who said:
Sounds like those two paragraphs contradict each other. Either PureOS is more locked down than Windows, or I should be able to install any software I want with ease. Because that’s what I can do in Windows: install any software my heart desires with absolute ease.
And that’s kinda what I meant in my previous post when I stated:
In Microsoft Windows, installing software outside of some sort of “store” isn’t a power user move. It’s how even a six-year-old child would use Windows. It’s what true software freedom looks like: being able to install whatever software you want without a consortium of geeks telling you what software you can install, and what software you CAN’T install; but especially the latter. You guys are basically being no better than Google, Apple, and all the other “Big Tech” companies you so adamantly criticize! You BECAME the enemy you swore to destroy! For all your criticisms of Microsoft, it seems like it’s the only software company left where I have the freedom to do what I want with my system!
Did nobody tell you Print Screen exists? Sarcasm aside, I suppose this would be a valid enough point if it weren’t for the fact that Windows doesn’t suffer any negative tutorial writing effects despite the fact that you’ll almost never have to use the Command Prompt. While this doesn’t directly rebuttal the point, it does weaken the notion considerably.
No. The philosophy is different: we offer you a vetted repository where we make it a breeze to install stuff. You’re not prevented from disagreeing and doing your own thing. There’s no lock on the door. That it’s not easy doesn’t mean that PureOS is locked down; it may mean it’s less friendly to random software. I mean, hey, you even installed one program without help!
Really? I am getting an impression you just want to bash something you didn’t find comfortable. In case I’m wrong and you’re not, my previous paragraph should settle it.
How do you know that? From my point of view, the amount and quality of resources for Windows is about the same as for Linux (consider Arch wiki), despite a lower desktop market share. Commands are by no means perfect, but they are way better for ad-hoc help - you can check here on the forums that a lot of people come up with commands from memory to help others.
It’s a matter of taste to some extent, but I agree that more GUI control would ease a switch from Windows for power users.
This is bringing nothing to the conversation. Please stop veiled insults or face getting banned.
I get what you’re trying to say, but at the same time, I feel it’s drifting into the NonStampCollector “free will” style of logic: oh sure, I have “free will” to install any software I want, but if it’s not from the “official store”, then all sorts of punishments will be doled out for not making the “correct” choice such as non-conventional installation and uninstallation, the program not being listed in the Installed Software list, and difficulty in upgrading the software. At this point, why don’t you just call installation from the “Official Store” compulsory?
Your opinion on my attitude aside, my logic is still sound. You constantly pushing the “Official Store” over being able to install quality software straight from an independent developer’s website is leaning toward a locked down environment in which Purism, SPC decides what software is allowed and what software isn’t. If that’s not quite the same as Google/Apple, it’s certainly heading in that direction. Heck, we can’t even install GNOME shell extensions without paying tribute to the all mighty Google. The ONLY way to get the official “GNOME Shell Integration” plugin is through Google Store? Is this a joke? And on an OS that encourages “de-Googling” your device no less! And you dare accuse me of hollow bashing? Looks I’m not the only one throwing “veiled insults” around…
Have you considered the possibility that these people wouldn’t be asking for help in the first place if they could accomplish their goals inside a GUI? Food for thought…
I think you’re redefining terms here, because “locked down” in no way implies the freedom to do whatever you please, nor “compulsory” means “you can also not do it”.
Maybe you’re mistaking what kind of freedom PureOS is providing. This freedom is not the convenience of forcing random third parties to integrate with our ecosystem. Make no mistake, the software maker is welcome to submit their software to be distributed as part of PureOS once they meet our standards of quality. They can even distribute their own software store, we’re not stopping them.
But the freedom of PureOS is the ability to modify your system in any way you please, including all the software we provide you (and we provide almost anything a computer user might need) – that is the property of PureOS that Windows cannot offer, by design. If you value having third party applications on the software list higher than you value this freedom, then maybe PureOS is not the best choice for you.
I’m not sure I understand this argument. Isn’t installing third part applications on the software list a certain kind of freedom? The freedom to say, “I want to install SlimJet because I like using that web browser, and it should be listed and easy to uninstall just like any other piece of software.” I’m pretty sure easy utilization of third party software is a kind of freedom. The kind Apple and Google don’t like giving you either. Don’t you see the pattern here?
Also side note: how do you clear DNS resolver cache in Linux?
You may, depending on the software of choice. This is convenience. As long as you mistake it for freedom, you’re going to remain dissatisfied with PureOS, because that’s not something we control.
I don’t think anyone ever said freedom is supposed to be easy, and the modern history is rife with examples of headwinds.
This thread starts to pop up in my unread list too often and I think that it is unnecessary.
There are two radically different and may it be contradicting objectives.
If you buy supported system, pay for support or use community driven (and financed by you or others) system/distribution and care at least a little about security then you want to be ensured that somebody has done review of the software you use and signed it by really hard to falsificate electronic signature that it is safe to use. If you buy from Purims and believe in their good intentions and security procedures then you install software from repositories they offer and are signed by their key. You, in theory, should be extra careful that initial keys received with computer or installed by initial system download are not modified. Purism provides extra options to ensure keys delivery… If you take care and check keys against published ones on their web and key rings servers etc, then it is relatively safe as well. As for the applications, Purism approves only these which are really fully open-source so you can check for malicious behavior yourselves a or can hope that even somebody else looking into sources found find the risk. You can pay for analysis and thanks to reproducible builds you can then check actual packages. But be prepared that security assessments are really expensive. Because OpenSSL is critical, there are many studies paid by big players, random link, it would be in range of ten million dollars for the work and more… You benefit from these investments etc…
So this is the gain if you use distribution as designed.
If you want to use some other software out of default repositories, then you can look for some serious provider for your operating system family. For Debian flavor, you can add another repository to APT sources list. Then you should obtain and add the signing key of the vendor and check that you can really trust to it. You will see available software and updates in management tools as Synaptic, Aptitude etc. I understand that you are probably desktop user used for point ant click system maintenance. Most people on this forum are probably of different attitude, I want to use procedures which can be repeated, automated and solved on many systems which I need to take care of remotely. So this is why there are suggestions for reproducible use of terminal tools. You have no problem to repeat steps reproducible, search reported errors on the Web etc. I hate reports from my students and users when thy send blurry mobile photo or screenshot. Providing manuals and document step for graphical tools is nightmare as well. But yes, for end user and many tasks has graphical interface advantages…
If the application vendor does not provide correct distribution procedure with repo then you have option to download package (think about it as Windows MSI and it is still at least visible in managers and with option to cleanly uninstall), or directly download binary and mess with the system without recorded way how to uninstall. Yet next level is to build package yourself. I have local Debian repo and repo on server for students and a university labs computers etc… So again fully maintainable etc… even for packages which are not packaged for my distro. Development ones, I install with /opt/sw_name during development. So no mess which I know from Windows.
I would like to know your practice on WIndows. If you locate software by Google and then install it from the link found then I expect that it is very high risk. Yes it is easy and comfortable. If it is MSI, it is signed, but you have no guarantee from it, it only proves that somebody paid to obtain software signing keys and have to trust that no trusted certificate authority has not signed certificate to somebody with malicious intention. You cannot check or pay someone to check whole sources etc. It is complete nightmare and development is as well (I know stories from my colleague solving signing of drivers we develop even for Windows)…
So I suggest to ordinary Windows users at least pay some part of the comfort and when usually standard functionality found in each Linux distribution is missing on Windows then locate given community developed software (even by Google) then try to find given project on Wikipedia, take link back to the project from the Wikipedia and then locate on the official site download section for Windows. I know that from security practice point it is still banned way but I hope that the risk is much smaller than random link from search engine. I believe that given software page on Wikipedia is visited often by knowledge people who would notice link change as well as report that project has been hijacked, stolen, transferred to some not so well behaving entity. But I consider still as the best practice to stay away from Windows entirely…
Please, consider to think about security, comfort, remote manageability etc. and ask for suggestions and good practices. But repeated rant about system not fitting your idea how it should work starts to be waste of time. And yes, if you are comfortable with Windows, stay with it. By the way, if you start to celebrate Microsoft technology, consider the first that they suppressed TCP/IP and when it was inevitable grabbed BSD TCP/P stack, then they tried to change HTTP standard to be unusable for all others and at the end they grab for free KHTM, Safari, Chrome development result and build closed Teams and edge from it. Same for many other technologies. So even if you use Windows you have no reason to derail people working on sound and free technologies to do the work which they like, you would need it daily.
No, I can’t. Not without having to learn what’s considered extremely advanced computing by Window’s standards. And I’m not the only one who thinks this. Here’s a comment by user “fotnite” on YouTube who sums up my thoughts and feelings about Linux (and its extremely toxic community) better than I ever could, but the most relevant paragraphs being:
I’d like to end my rant with another comment by user “Ethan”, who provides a great summary argument that, again, captures my thoughts and feelings about Linux as a whole:
With that being said, Having been using OS/2 for the last couple of weeks, I’ve had a vastly superior using experience using that obsolete OS than I have using any Linux distro: let that sink in. The only downside to using OS/2 is that there’s no real software development for it.
Sarcasm aside, I suppose this would be a valid enough point if it weren’t for the fact that Windows doesn’t suffer any negative tutorial writing effects despite the fact that you’ll almost never have to use the Command Prompt. While this doesn’t directly rebuttal the point, it does weaken the notion considerably.
