Microsoft is finally dropping support for Windows after the Recall/Copilot exodus, guess they're focusing on Xbox or something instead.
>>25047Normal people aren't going to buy a wholeass gaming rig just to run Windows. The hardware requirements for Windows 11 were rediculous at launch, and now even moreso now that crypto mining / AI farms made graphics cards expensive.
I'll concede they'll probably move to Mac before considering Linux, but still.
>>25049>Normal people aren't going to buy a wholeass gaming rig just to run Windowspeople said this about vista, 8, 10
end users don't care, they're going to buy a laptop that comes with windows 11 installed and use it
>>25053based
never use linux
>>25055If Windows 12 is less hardware intensive than 11, buisnesses and school admins and such will think it regressed.
Higher hardware requiremests = more lag and loading bars = computer is doing more thing bigger.
Microsoft was the company to learn that artificially padded loading bar times improved UX for tech illiterate porkies and this has forever guided how they do things. Not to mention the backward compatibility issues that would arise: if a leaner Windows 12 can run software made for a heafty-and-therefor-gooder Windows 11, that might shatter the aforementioned illusion and create a demand for optimization; directly antithetical to every design choice they made thusfar.
Windows 10 has had this official end of life date since 2021. I don't know what this thread is supposed to be about.
>>25056Windows 11 is not hardware intensive. If you're talking about all the perfectly fast systems that don't technically support it, that was a business decision.
>>25053>>25054ackshuhally it's GNU(the system), you don't actually "use" linux(the kernel), except for when you're configuring it when you're compiling it to have a custom kernel
>>25051it's akcshually better than windows, you're missing out
for starting out:
https://manjaro.org/or the fastest system in the world:
https://www.gentoo.org/>>25055it will be the same system but only with a slightly different desktop environment and better hidden spyware
>>25056>If Windows 12 is less hardware intensive than 11, buisnesses and school admins and such will think it regressed.lol
>>25058>Windows 11 is not hardware intensive.<5GB ram idle >>25059pretty much that, most people wouldn't know or wouldn't want to install another system, and they ==should= avoid pcs with pre-installed gnu/linux bcuz it's garbage
>>25062My 2gb and 4gb RAM hp-15 laptops
Got the 4gb thinking the 2gb died but it came back alive a bit later could both run Windows 10. Running anything made it hot, but it was usable for a the duration of highschool and a while after before I finally stopped bothering to dual boot and just ran games on wine instead.
Windows 11 cannot run on 4gb ram, the OS itself exceeds this at idle and the installer refuses to proceed after detecting underspec hardware.
>>25062>How much does 10 use?I think about the same as 11
>As far as I can tell, about the same.probably that
>Aggressive use of available memory is not a crime and is not a Windows thing.it's wasting memory away with either spyware or "services" no one uses.
and like it only uses extra memory if something else is running, something sus like contacting microsoft or other companies
CIA or running something that the user does not even use
so it shouldn't even be on the system in the first place, in gnu/linux only what is currently running by the user + daemons
and not spyware/services/etc or what the user has manually built consumes RAM, that's why some machines running GNU/Linux can idle at 130MiB of ram or even lower with a window manager, because no spooks are consuming the RAM except for what the user has built
>>25063>Windows 11 cannot run on 4gb rambruh it's like microsoft wants windows to die or something, gnu/linux has wine, vms and emulation so it doesn't matter what happens to windows anyways, windows has already been fully replaced and will eventually be substituted
>>25065 (me)
forgot the flag
>>25064basedddd
>>25071the bsd systems will be the escape for when the world switches to gnu/linux and when corporations start to invest
destroy on gnu/linux
>>25065"Hey OP, it looks like you have concerns regarding the high usage of RAM on your PC. It is normal for around half of the RAM to be in use at "idle", even with nothing running on your PC yet.
Windows has a service called Superfetch or Sysmain that is will automatically pre-load your frequently used files and programs into the RAM, so that when you do finally launch them, they load faster as they are already in your RAM. This is essentially a free performance boost, as otherwise the extra RAM you paid for is just going to waste. The cache will empty itself out automatically if the RAM is needed elsewhere.
The amount of RAM used by this cache can scale up or down depending on how much RAM you have, so adding more RAM will result in Windows automatically using more."
>>25075chatgpt? lol?but they also preload the CIA at RAM
>>25074TempleOS will be plan C if apocalypse occurs, Plan9 will be plan
D if they destroy the bsds as well
>>25059That would be like Microsoft expecting to buy new computers just for Dos 4 back in 1989. Windows 11 offers nothing to run software over Windows 10 (what people want their OS to do), go look through Steam and try to find one game that requires Windows 11 to run at epic graphic quality. Also if you look at Steam survey well over 50% of users are on Windows 10.
You also have the Steam Deck have enough sales that make software devs want to remain compatible with WINE thus odds are there will never be software make exclusive to Windows 11.
>>25043>That would be like Microsoft expecting to buy new computers just for Dos 4 back in 1989.They might succeed. Boomers get cranky if they can't keep all of their desktop icons. Microsoft might have new tactics to milk their remaining OS customer base dry in mind.
>>25067>>25063Windows 11 requires some hardware not related to performance or memory. The official installer requires TPM which is a type of OS-level DRM, marketed as rootkit protection
as if physical access wouldn't mean you're already fucked for the average user.
>>25075File caching on linux is a kernel thing and only processes count towards memory usage.
>>25065>GNU/Linux can idle at 130MiB of ram or even lower with a window manager, because no spooks are consuming the RAM except for what the user has builtGNU/Linux can idle at 130MiB of ram or even lower with a window manager, because no spooks are consuming the RAM except for what the user has built
Except of course the systemD related tens of megabytes of shit. Even with X my 2010s gentoo with static /dev and s6 init idles below 50MiB.
>>25082>>25082>Even with X my 2010s gentoo with static /dev and s6 init idles below 50MiB.>50MiBholy I kneel
>static /devthat's a lot of work but sooo worth of it
only as an experiment of course, it's a shame that even LFS has adopted udev to be used alongside its sysvinit>systemD related tens of megabytes of shitit's because openrc uses some components of systemd like udev
rip eudev >>25083>that's a lot of workYou will only have to configure it once if your distro respects your time or you never update.
http://www.loper-os.org/?p=3682 is a good place to start.
>>25086>What is your threat model?idk? i don't have any I think
>access the internet in a up-to-date chrootbruh I never would have thought of doing that, holy moly I kneel once again
I'd do it in a vm just to know what gentoo was in the past
installed gentoo this year, like out of curiosity
>>25094Well its still better in some ways if you don't have software that requires windows/mac because it is still freedom from the corporations if you use community based distros instead of corporation red hat / canonical stuff.
At least on the software side of things I think that copyleft is actually a decent idea.
>>25094because they have an odd rules-based conception of freedom rather than a pragmatic one. it is not quite libertarian because it brings duties as well as rights, but it is adjacent: freedom is the theoretical ability to modify and redistribute software.
it'd be worthy of its own philosophical analysis thread tbh, though of course doing so would just piss people off because if you look critically at it, you clearly just want to dismantle it so everyone uses windows.
>>25102Windows 2000 meets - for
practical purposes - all of the fundamental FSF freedoms. The source code is up on GitHub. It is considered non-free software philosophically for legal rather than practical reasons - legally speaking you can go to jail for releasing your modified version of Windows 2000. It is Microsoft's proprietary software and their rules make you unfree… But practically, you are not going to go to jail for it: Use it for what you like, study how the program works, and change it to make it do what you wish, redistribute and make copies so you can help your neighbor, improve the program, and release your improvements (and modified versions in general) to the public, so that the whole community benefits…
>>25104Let's say I can't - what would you have proved? That an illustrative example was inaccurate - nothing deeper. The general point it illustrates would stand. Let's say I can - would that suddenly make Windows 2000 free software? No, it wouldn't. But you're not interested in discussing this as a philosophical curio, you're interested in
defending it from an attack that isn't there. I am entirely convinced that this conversation is going to go nowhere - hey, what did I say? "it'd just piss people off"…
>>25105This is irrelevant to the
philosophy of the FSF and its fans. Windows has the advantage of running
Realm of Empires Power Tools, we could sit and argue all day about what's more important for practical purposes.
Allow me to set it out this way: Imagine there was a law that said it was illegal to speak when nothing/nobody was around to hear you. The FSF philosophy is of a sort which would view this law as a terrible assault on liberty because on paper it imposes quite a strong restriction on your behavior, while a practical approach would notice that if any law-man heard you speaking
you would not be in violation of the law. That is the underlying, interesting dynamic which nobody is interested in discussing. (It's booooooring, can we go back to feeling better than people who use Word Online yet?)
>>25075>Windows launches everything by default and just unhides it when you open it.That is actually worse than anything I could've made up about it lmao.
>as otherwise the extra RAM you paid for is just going to wasteYou jest but I've seen people actually argue this with crypto miners.
>>25108Windows is of no special importance here: all that matters is that it is a high profile piece of proprietary software for which the source code has leaked, which as such is still considered "unfree" despite the fact you can in-practice enjoy all the FSF's essential freedoms. The point is philosophical: were Microsoft to GPL Windows it would remain a mess and a moving target and so on, but it
would become free software.
>>25109Pure posturing which says nothing. Stop replying to me, go away and develop a free software version of Nitschke's deliverance machine for yourself, that's a nice practical exercise for you.
>>25111I'm surprised to learn from you of all people that I did not, in fact, enjoy compiling Windows Server 2003 last year.
>>25112You're still overlooking that this is about philosophy. But you're trying to engage, so that's to your credit.
Here's my angle: You can compile ReactOS, you can compile Windows Server 2003 (really, you can!), but one is considered free and the other is not. This is because the conception of freedom rests not on what you can
actually do, but on what you are
allowed to do. Microsoft says you cannot compile Windows Server 2003, that even downloading the source is copyright infringement, etc, etc, etc. You know what people did? They did it anyway.
Now obviously you're not going to
use either of those as your main OS, but that's a secondary point. The FSF position has never been that it's okay to make proprietary programs so long as they're small and unimportant and and just for fun because nobody sane wants to use them anyway!
>>25113Okay lets say that Linux and BSD instead of Wine went the OS/2 route of just running the binary blob of MS Windows as a guest OS, how would that be better? Wine with Proton has got to the point it can run Windows games faster then native Windows in some cases.
Lets not forget the whole SCO claiming it owned Unix and GNU allowed Linux to dodge that bullet.
>>25114I am not sure what I said to give the impression that I've got any case that one way is "better" than another, so I find it hard to answer the question. If you run windows as a binary blob, or if you run (most) windows software under wine, either way you're still using proprietary software and as such "unfree" from the perspective of a free software advocate.
The only thing I've ever had any particular interest in discussing is how the FSF and its followers
conceive of freedom. Every example I've given has been to that end alone. Windows was introduced in an example only to illustrate that there are cases where you can
practically enjoy the 4 fundamental freedoms with non-free software in some circumstances (like playing with the leaked Windows source - or the source of many other proprietary programs!) - but that software is still considered unfree, showing us that the way freedom is imagined by the FSF is not strictly about what you
can do (i.e. "pragmatic"), but also about what you are
allowed to do. (i.e. "rules-based").
>>25123>tryhardIt's literally easier. My dad uses windows and has all these issues that I just never encounter on NixOS with KDE because those issues are just Microsoft rent seeking working as designed.
Like if I want a package I just type the name in the configureation file under my user and hit rebuild. If it's an executable from itch.io or something I just steam-run it. A child could learn NixOS in a week tops.
>>25124linux is often better for a very low end user who has no idea what they're doing and for a very high end user who knows exactly what they're doing.
it's a tedious slog if you're a mid-tier user who is used to just googling what's wrong with windows and finding the answer is "go into the registry and change 0 to 1", "uninstall x", "run in compatibility mode as administrator with display scaling disabled", "install x driver", "edit this .cfg file in notepad", "download this random program that resolves your exact issue with this game", and a bunch of other bespoke little solutions which are each case-specific, but which accumulate into a sort of general skill that let you feel your way through the GUI. then you download linux and it all goes to fuck because it doesn't work like that at all. you
will have to use the terminal and you
will have no idea what it's doing or how to undo it if you don't like it. you
will encounter bespoke little programs that supposedly fix your exact problem but which expect you to correctly plumb a series of commands together for what, on windows, is a simple case of "drag icon A onto icon B and it will spit out what you want", you
will go back to feeling something like a low-end user, but unlike a low-end user you know a computer doesn't
have to feel like this. this is the demographic that nobody will ever target (perhaps can never target) because it would require the most fundamental changes to how the entire ecosystem works. a low-end user can have a fairly standard distro with a mac-looking theme - if something breaks their goto is "ask a high end user for help", a high-end user can do whatever the hell they want, but a mid-tier user is never going to find a distro that entirely strips the unix DNA out and makes their system work
intuitively again. high end users making a distro have no reason to target the mid-tier niche: it is infinitely more demanding to make a distro for them than the low-end niche (because you'd have to fundamentally rethink how complex parts of the system work, rather than just hiding them), but smaller, and probably
less likely to want to actually use linux regularly. (because even if you do make the perfect, intuitive distro, occasionally they'll want to use a map editor .exe that was made for Age of Empires 1 in 2003, which runs fine on Windows 10 but crashes on Wine, for which there are obviously no answers to "how do i fix this?" on google because only about 50 people use that tool and only 1 has ever tried it on Wine… bad user experience, back to ol reliable…)
a pointless metaphor would be like retroarch versus standalone emulators: the latter are are inconsistent mess, but something a lot of people learned to feel their way through over time. RA has a much more consistent approach, but it's utterly alien to the prior skillset of most desktop emulation-users. (which is often precisely the same mid-tier user skillset i was getting at earlier: RA's whole gimmick is a consistent UI across systems, but that means a UI
not consistent with desktop PC standards…) some people find it fun to re-learn how to do everything all over again, others go: i can already get stuff done by sticking to what i know, fuck that noise.
this post is a waste of time. someone will take this as an attack rather than a set of empirical observations. /tech/ threads are by far the worst threads on this site. i post here only because i enjoy suffering. >>25128Windows has no uniform UI as it is a hodgepodge of UI elements glued together especially since 8. Also even the Amiga utilized the command line (Amiga Dos) for quick maintenance and config of the system as one could just copy and paste into the terminal. Instead of telling people the tree of clicks to get to a setting you just tell them to paste a string into the terminal to do the same thing.
Linux UI has evolved way past its Unix roots, take a modern Unix users and plot in front of AT&T System V and odds are they will be lost even in the terminal.
>>25129that may be true but uniformity is not always more intuitive (indeed, a UI window suddenly changing to Windows XP style might be just the clue the user need that they're on the right track to find
that specific option), a tree of clicks is not always worse for all users simply because it takes longer - visual feedback can be more important than efficiency in making a user feel that they're in control of what's happening.
"go through 3 sets of menus and then check this box, keep the window open and if something breaks, uncheck it" is a lot more relaxing than "copy and paste this into the terminal, if something breaks… fuck." to a user who has checked a lot of boxes but never touched the terminal. this difference in mindset is yet another reason nobody will ever make a distro for this part of the market - if you're always looking for efficiency you're going to be completely baffled by the idea that a bunch of redundant clicks should be kept to provide "feedback", that's a lot harder to get your head around than "all the options should be hidden and there should be a big firefox button and a big libreoffice button on the desktop" for a low end user.
>>25130The Amiga was built for the home user and a competitor to the Macintosh and its users viewed it as a superior UI to MacOS by AmigaOS 2.x. It is not just an issue of it taking longer with UI but users can get lost in sub menus and screens where terminals can apply commands globally meaning with the correct command syntax it will work regardless what directory you are in.
You also have the fact you can ask the terminal for feedback that many GUIs don't give for example asking the OS to print to screen the current state of devices. For example troubleshooting games by simply running wine via the terminal does help as error messages will echo in the terminal. There are a number of games I had to get out of a soft lock by getting into a game engines debug CLI to enable cheats.
>>25131the Amiga was revolutionary in 1984 but died 30 years ago. by modern standards most graphically focused 90s systems had odd quirks (like each folder in a nested directory opening in a new window if clicked, rather than the selected folder opening in the current window as is now standard) the average user today is far more likely to be utterly lost using the terminal than they are to get lost in a menu or folder tree (hit back until you know where you are - if that doesn't work, close the window!) even lower end users are familiar with basic menus from phones and the like. (though there's a smaller distinction between a menu like that of the average phone and the kind a mid-tier desktop user would like, the windows style with a bunch of tabs for different options and such… and microsoft seem to be moving in a more phone-like direction, which is also alienating for mid-tier users but to a lesser degree.)
to use the terminal you must learn commands and their correct syntax, both to enable and disable an option, to do or undo an action: to use a regular menu you can simply click a button to turn an option on - if something breaks, click it again. (this is where the retroarch parallel is very helpful: do you want to make an upfront investment in learning how the system commands work, or do you want to explore it by clicking around on your own and seeing what happens..? the answer is not the same for everyone.)
>>25133i never said it was
inherently anything, i described a certain segment of computer users and how they feel. indeed, inherent to my analysis is the fact that this is not
inherent to any approach - to high-end users windows may feel less intuitive to the terminal, to low-end users windows (like all desktop computers) definitely feels unintuitive (though the terminal is downright terrifying). it is only the mid-tier user who universally finds windows to be intuitive and the terminal unintutive.
though i will admit to you that as a general rule, between "it's obvious what this does" and "it is not obvious what this does, but it is very well documented" i would take the former every time. being obtuse, my go-to example here would be a light switch…
>>25134and what happens when you test drive it and find that you prefer the handling of your 1998 corolla?
>>25136i would personally take as read that in an exercise that divides current computer users into high/mid/low tiers, the current distribution of desktop computer users will apply.
but i can see the humor in this arising concurrent with the topic of differing philosophies towards documentation.
>>25137My claim in
>>25133 was that the current state of users does not map onto the ability of people to have an easier time learning one or the other interface. The 'high-tier' linux users could evidently adapt to the command paradigm. When a new generation of computer users is raised on linux, many more will be able to throw off the shackles of clicky buttons and colorful pictures.
>>25138i'm curious how you would explain the bulk of DOS users disappearing into Windows without major complaint. a generation raised on the command line took quite well to clicky buttons, colorful pictures, and never having to play with config.sys again…
(and as a lesser matter, where this new generation raised on linux is going to come from given the poor conversion rate from the current generation of mac/windows users)
>>25139>i'm curious how you would explain the bulk of DOS users disappearing into Windows without major complaint.Realistically i would say the average user adopted whatever their workplace got. Windows fashions itself as the "user friendly" operating systems, because the promise of deskilling computer work appeals to managers.
>never having to play with config.sys again…DOS comes from a lineage of early microcomputer OSs with all the limitations that come with that. Genuine DOS apologists like Bryan Lunduke need to do mental gymnastics like "games crashing if you don't configure your graphics right makes playing them that much more rewarding". Almost anything ameliorating things a little while being compatible MSDOS would be overwhelmingly adopted by their existing customer base.
Windows was an office environment that evolved into a protected mode extension for DOS with a desktop. Now i'm not saying a non-graphical extension would have been a better choice. Many programs like window managers or drawing programs are more efficient with a wysiwyg interface. Only microsoft by steering the direction of MSDOS and windows had an unprecedented level of control over the future direction of interface design and they made command interfaces a fringe concern for suckers who had to write batch scripts. It might seem self evident at first to treat commands only secondary, as something invoked by graphic utilities, because "surely everyone would use the defrag system tool" and "users should only see the graphic interface we designed for them". In the end however, it results is sprawling, sometimes redundant menu trees that reflect the complexity of the underlying design and the loss of scriptability. Shiny interfaces are no substitute for transparent operating system design.
>>25141surely if the move towards windows constitutes deskilling that would suggest windows is easier to use?
my contention, really, would be that there should be both kinds of interface. sprawling menu trees for some, shell scripts for others, and as a niche concern: visual scripts for those in the middle. ideally with the option to open shell script files as visual scripts and vice versa. (hey, if gamedev tools can do it for a AAA release, why the hell can't a simple script to empty a folder at a click of a button so you don't do it manually?) but of course
writing this appeals to nobody - more work, no reward, philosophically baffling (how
inefficient', how inelegant
, how redundant
), and you'd have to be in a market-dominating position like microsoft to actually make other'' people's programs work within that paradigm.
windows' sprawling menu trees as reflective of underlying complexity, while an obvious
mess, hit me as having some kind of information that isn't communicated by a text interface: a text command you should use regularly and a text command you shouldn't really be using can look basically identical ("cp a b", "rm -rf /"), while the fact the option you're looking for can only be accessed by jumping from the windows 10 UI to the windows 7 UI to the Windows XP UI to an area that you'd swear is using the Windows 3 UI serves as a clue that the option you're playing with is
obscure - if you just want to change your desktop background you are probably on the wrong track. (this is obviously an accident rather than a case of microsoft planning a good UI, but imagine if someone set out to design a UI on a similar principle - a sort of game-design approach with environmental clues and things you can remember to assure you you're on the right/wrong track? again, a philosophically baffling approach to many…)
it's a touch philosophical, but i'd say in this regard transparency can cause opacity all of its own: if you want to resolve a problem and someone gives you 1000 pages of documentation and says "well, the answer's in here, read it yourself" that can be a lot less helpful than a button that says "do it" with no underlying explanation as to how it's done. in the first case the information is available but the solution is hard to access, in the second the information is unavailable but the solution is easy to access. (ideally you want both, the solution should be at hand and explanation shouldn't be too far away, but that's work…)
but as with the light-switch, i'm thinking as much of malicious compliance with freedom of information requests as i am of man pages…
>>25143power users are not the bulk of users though: if we're to assume all interfaces are created equal, a low-end or mid-tier user raised on DOS should've found windows as unintuitive as a mid-tier user raised on Windows finds linux. speaking for myself only: blender before 2.7's UI was totally baffling because it wasn't really a serious graphical UI, it was a token UI designed entirely around shortcuts. i never learned to use it because looking up the shortcuts to get things done was tedious and - like the terminal - there was a vague sense you might break something and not know how to fix it.
blender 2.8 onwards has a slightly messy, but ultimately acceptable UI - and using that UI regularly meant i actually learned the shortcuts because every time you hover over a button it reminds you of the shortcut. if something breaks just delete the broken part and try again.
(really, separate from splitting users into low/mid/high, there's gotta be a separation of learning styles or something that could also apply here: some want to study, then apply. others want to apply, then study…)
>>25144>surely if the move towards windows constitutes deskilling that would suggest windows is easier to use?Windows embodies the promise of deskilling more than anything.
>a text command you should use regularly and a text command you shouldn't really be using can look basically identicalYou don't *look* at a text command, you read it. While you are thinking the windows interface exposes information by itself, more than anything you have learned to read "environmental clues" that the inexperienced user would not find particularly telling. Bourne shell might appear like linenoise to you, but really there is much worse.
>if you want to resolve a problem and someone gives you 1000 pages of documentation and says "well, the answer's in here, read it yourself" that can be a lot less helpful than a button that says "do it" with no underlying explanation as to how it's done.Writing good manuals is a solved problem. Physical manuals have a table of content and ideally an alphabetical index. On-line manuals often include references to related sections and in the case of manpages have a section for every single command. Search tools like apropos typically narrow down a single keyword to less than 20 options. All of these ensure you're discovering the relevant information as fast as possible instead of paging through a tutorial on a wordpress blog. Can you definitively tell me how a good gui looks like?
>>25145that's always a fun one, though i wouldn't call it a skill issue so much as a standards issue.
>>25146>Windows embodies the promise of deskilling more than anything.it is not particularly clear what you mean by it. (are you saying it fails to deskill - just holds out a phantom promise - or are you saying it succeeds?)
anyway: users who have learned to read and rely on environmental clues is the demographic i've been talking about this whole time. it's a good point to have raised. you could make my original post much more succinct by narrowing it down to that: a mid-end user is a user who relies on environmental clues to perform complex tasks, a high-end user is one who relies on their understanding of an underlying system without needing such clues, and a low-end user is one who cannot understand the environmental clues or perform complex tasks. as a terminal does not offer these clues, such a user will have a miserable time where the terminal is the primary method of performing complex tasks. a low end user won't mind - they cannot perform complex tasks. a high end user will not mind - they do not need such clues.
as regards reading commands: it would be easier to
read them if commands and flags were had more indicative names. "copy" instead of "cp", "remove" instead of "rm", and so on. as it stands you've got to mentally translate each abbreviation to its function, with double fun if you've got a program with an options flag that matches an existing command. (the trade off here, of course, is that the odds of typos increase, that scripts are longer, that people unfamiliar with the terminal are probably still not keen, etc…) now obviously that's a system-specific situation rather than inherent to all CLI systems, but it's worth raising. inherent to all shells, though, must be the fun of picking language: Do you ReMove a file or DELete it? That kind of linguistic distinction matters less in a GUI since you right click the file and click the button that says "remove", "delete", "move to trash", or any other label - you don't need to know the precise command, and moving from system-to-system (like Windows to Mac) it remains consistent.
now sure, you can pick up a manual. but my contention would be less that it's hard to write a good manual, and more that it's inconvenient to pick up a manual in general - take our file-deletion case: in the GUI, you simply have to click on the file to discover how to remove it, which is a generalizable behavior. (you can also right click a web page to get options to save it, inspect element, etc…) while for a CLI you have to
look up the command, which takes you away from the task at hand. the flip side is that with the CLI you can write yourself a script using regex to delete every file with a "1" in the name, while with a GUI you'd be encouraged to do it manually or semi-manually such as by searching for "1" in the folder. but then we're back to the distinction i made earlier: do you want to study, then apply, or apply, then study?
i cannot tell you definitively what a good GUI looks like if you want an answer in terms of good UX rather than good aesthetics. (for example, I prefer the Win9X look to the Windows 10 look, but the search function means 10 has the edge in actual usability…), nor am I sure that one strictly exists: a good GUI for a low-end user is different to a good one for a mid-tier user, which is different to a good one for a high-end user. but i can say that most GUIs act broadly similar, and the cluster of behaviors they have in common would be a good place to start. i'd be interested to see someone experiment with (say) color coding certain system settings or otherwise trying to re-create the accidental hierarchy of options that Windows has - again, combined with a search function so that if you know the tab you want, you can just type it in. i'd even be interested in someone doing some Windows 95 style user-interaction studies to see if the various ideas we've thrown up (visual clues, a building a spatial model by "exploring" menus, etc) have any validity.
but I don't think we're mainly dealing with a case of bad vs good GUI design in the first place: the question is really whether advanced configuration should be performed using a GUI, using a text interface, or - in fantasy at least - using both.
>>25148>that's always a fun one, though i wouldn't call it a skill issue so much as a standards issue. The issue is that case sensitivity is more useful for machine created files as it gives you 26 more different characters for a program to automatically generate a unique filename. It also avoids new programmers from learning bad syntax as most programing languages are case sensitive when defining variable names to give the programmer 26 more characters to pick from as they have character limits.
> take our file-deletion case: in the GUI, you simply have to click on the file to discover how to remove it, which is a generalizable behavior. (you can also right click a web page to get options to save it, inspect element, etc…) while for a CLI you have to look up the command,Yet if you want to write a program yourself that would delete given files in a given folder when run the CLI makes things way easier even if you are dealing with something like Visual Basic. There are also cases where you as an experienced user would want to do this like copying all files of given file type on a volume and you don't want to do a search just to start the process but have the OS start working it way down the directory tree.
>>25151yes - but it's a standards issue, not an ease of use one. it's no "easier" to be case insensitive than to be case sensitive. the problem would go away so long as every system behaved in the same way. i feel like you've mostly duplicated my own concession to when the CLI is more useful - but what of cases where he user doesn't want to write a program? should one always delete files in the same way (so that they're forced to learn how to use the CLI for file manipulation), or is it permissible to have a system where the standard behavior is to delete single files using the GUI, but to perform more complex operations with the terminal or a bespoke program (including, for example, some kind of visual-scripting language one)?
the interesting thread running through your post is the theme of programming. here's an open-ended set of questions: should every computer user be a programmer? is there a place for the user who is competent at deploying programs written by others to achieve certain tasks, or only tweaking them slightly? and should users always follow best programming practice, or is it okay to ignore best practice for trivial or personal use?
(i have no real agenda to push building off your answers, i am just curious.
my theme, after all, is the mindset of different types of user… my answers, in summary: not unless they want to/yes/ok to ignore best practice, provided the resulting software doesn't spread too far.)
>>25153that's true, but it was in an age a world apart from today. (and in any case, for many lower end users the full extent of their knowledge was "how to load a tape/floppy", "10 print 'i am great' / 20 goto 10", and if very patient, "painstakingly re-type code line-by-line without knowing what it does, hoping you don' leave out a comma")
today's users have variable needs and most of the time there's a program available online which will meet that need, but which often fail to meet it ideally. (for example, by being aimed at too experienced or too inexperienced a user, or being super-general-purpose: think of how many enterprises use Excel as a database program!) moreover, the basics of programming no longer give a very good understanding of how the underlying system works: a C64 is more or less just a Basic box, but python isn't going to teach you all that much about microkernel memory management as part of an OS that runs across a mess of commodity hardware… even at the programming level, all that's abstracted away. (but is that necessarily bad? answers on a postcard.)
so the question i'm asking or assuming is: how do you build an on-ramp to higher skilled use (or "higher skilled outputs" - writing a script in a visual scripting language may take less skill than doing it traditionally, but that's balanced out if it means someone can do new things, or do existing things better…) in today's computing environment?
my basic angle would be: give people something they can play with quickly and build confidence as they go, but don't force them to use it - that will build resentment. but that's coming at the problem from the perspective of:
apply, then study. which is the perspective of a fairly small part of the userbase.
>>25084so I tried installing old gentoo
the one with a static /dev on a trisquel vm with this guide:
https://wiki.gentoo.org/wiki/Old_Fashioned_Gentoo_Install but I broke it beyond my non-wizard understanding of gentoo
/dev
and other directories got insane somehow
meanwhile I'm just gonna resurrect my bare-metal gentoo >>25157the torrent doesn't seem to work I think?
is it an invisible torrent from the I2P network? rip anonymity if it's a normal torrent outside of I2P>If you installed busyboxI avoided busybox because of mdev so I went for GNU
>>25158>the torrent doesn't seem to work I think?magnet:?xt=urn:btih:781f569e34dd2987f11832eb7c0a55eba75fc321 should work. Your client supports DHT right?
>I avoided busybox because of mdev so I went for GNUYou will need coreutils and linux-utils anyway for compatibility, but busybox has a good dhcp client, a responsive almquist shell and it can completely substitute inetutils. Did you know it keeps its size down by bundling every command in one file?
>>25159>Your client supports DHT right?yep! but still doesn't work
on I2P the torrent is dead too.
>Did you know it keeps its size down by bundling every command in one file?I never used busybox but cool to know
I think I'll try to create a static /dev gentoo later since the only thing LFS didn't teach was how to make a static /dev because they preffered to give instructions on how to compile and manually install udev instead
I think the torrent would work on I2P… maybe
invisible torrents on
http://tracker2.postman.i2p/ works fine but I had the same issue where only I was able to seed my torrents to my own devices, I never knew what caused the issue though
>>25190The linux operating system doesn't need any more security mechanisms. It is possible to write secure applications for it today, by splitting the code into few critical, bug-free sections and a lot of untrusted code, that has a explicitly defined dataflow.
>Many additional “security” efforts are applications of the “principle of least privilege.” The principle is widely credited to Saltzer and Schroeder, who stated it as follows in: “Every program and every user of the system should operate using the least set of privileges necessary to complete thejob.”
>These “security” efforts work as follows. We observe that program P has no legitimate need to access operating-system resource R. We then use (and possibly extend) operating-system controls to prevent P from accessing R. We prevent an image-displaying program from sending data through the network; we prevent a DNS-lookup program from reading disk files; etc.>I have become convinced that this “principle of least privilege” is fundamentally wrong. Minimizing privilege might reduce the damage done by some security holes but almost never fixes the holes. Minimizing privilege is not the same as minimizing the amount of trusted code, does not have the same benefits as minimizing the amount of trusted code, and does not move us any closer to a secure computer system. Consider, as an example, a confinement of Netscape’s “DNS helper”program, preventing the program from accessing the local disk.>This confinement did not prevent a libresolv bug from being a security hole in Netscape:>an attacker could use the bug to seize control of the “DNS helper,” modify all subsequent DNS data seen by Netscape, and steal the user’s web connections. The situation before was that bugs in the “DNS helper” had the power to violate the user’s security requirements and therefore needed to be fixed; the situation after was that bugs in the “DNS helper” had the power to violate the user’s security requirements and therefore needed to be fixed.>The defining feature of untrusted code is that it cannot violate the user’s security requirements.>Turning a “DNS helper” into untrusted code is necessarily more invasive than merely imposing constraints upon the operating-system resources accessed by the program. The“DNS helper” handles data from many sources, and each source must be prevented from modifying other sources’ data.<Daniel J. Berstein: Some thoughts on security after ten years of qmail 1.0 >>25190Privacy isn't a given either, you have to use Tor in a specific way to get even a tolerable level of it. None of the operating systems offer privacy by default except Qubes or Whonix and even then you'll have to aquire Tor bridges somehow in countries where Tor is blocked, the glowies do everything in their power to make it so you
can't have privacy, they want your house to be wiretapped 24/7.
As for security, the only secure OSs which are popular enough to be ready for desktop are Qubes, Kicksecure and OpenBSD, with Qubes having enormous system requirements and with OpenBSD having a worse device compatibility because it doesn't use Linux as its kernel. Your SculptOS or whatever it's called will never get traction and will get as far in its development as Subgraph. Which doesn't matter anyway since you can't get sufficient security without sufficient virtualization since host security is done to harden virtualization itself. Hardened lightweight QubesBSD on a device that supports it would be an ideal solution but we can only dream.
>>25045the same thing was said with 8 and 10
it's most likely going to be like 8 where every skip it until the next version
>>25229>moreso on the average distro without a traditional base system.only gentoo comes to mind, the rest all use GNU by default while gentoo can be literally anything, even gnu-less and in the future linux-less because of hurd
>>25228 (me)
actually I am coping, because I've been compiling the kernel for my static /dev gentoo(still building it on the chroot) it's been 2 HOURS NOW AA
I don't think I need to manually config the kernel, now I'm just fixing grub because apparently it thinks I'm on UEFI instead of BIOSIf I successfully boot into the static /dev gentoo, then I think I'll make a post about it
>>25230>>moreso on the average distro without a traditional base system.>only gentoo comes to mindOn most distros the concept of a base system is entirely localized in a single meta-package in the main repository (see GNU/systemd/FHS/pacman/linux). This was the intended meaning.
>>25231Grub is an overengineered turd designed for loading exotic boot partitions. Use lilo for legacy and efistub for uefi.
>>25233>"I use Linux as my operating system," I state proudly to the unkempt, bearded man. He swivels around in his desk chair with a devilish gleam in his eyes, ready to mansplain with extreme precision. "Actually", he says with a grin, "Linux is just the kernel. You use GNU+Linux!' I don't miss a beat and reply with a smirk, "I use Alpine, a distro that doesn't include the GNU coreutils, or any other GNU code. It's Linux, but it's not GNU+Linux."
>The smile quickly drops from the man's face. His body begins convulsing and he foams at the mouth and drops to the floor with a sickly thud. As he writhes around he screams "I-IT WAS COMPILED WITH GCC! THAT MEANS IT'S STILL GNU!" Coolly, I reply "If windows was compiled with gcc, would that make it GNU?" I interrupt his response with "-and work is being made on the kernel to make it more compiler-agnostic. Even you were correct, you wont be for long."
>With a sickly wheeze, the last of the man's life is ejected from his body. He lies on the floor, cold and limp. I've womansplained him to death. >>25235>Use lilo for legacy and efistub for uefi.I was about to use lilo but then I noticed that the kernel itself failed to install, I'll make a post about it on the gentoo forums but I'll try kernel configs LFS-style first
>>25233I'd actually just call it GNU, but then I'd be baiting people into "correcting" me
>>25236Okay, bruh, I don't care about this Redditor radlib's copypasta.
This entire discourse is a waste of my precious time.
>>25258 (me)
But that's just why I say "GNU/Linux" instead of "GNU." For all I care you might as well call the OS "Windows," that'd be hilarious.
>>25299he tried to stop perfect chaos and black doom so he isn't really evil, also the US goverment portrayed in these games
SA2 + shadow the hedgehog(the game) hate eggman so he's doing praxis actually
>>25305>he tried to stop perfect chaos and black doomOnly because his plans backfired (300 IQ scientist apparently).
>he's doing praxis actuallyNah, he just wants to be a feudal lord with humans as his serfs and robots as his vassals. He's clearly a rich fuck who just wants the government to get outta his way of exploiting others. Him being a king in Secret Rings is very fitting in retrospect.
>>25307>Nah, he just wants to be a feudal lord with humans as his serfs and robots as his vassals.I mean true he probably will turn everyone into his servant robots BUT he's also basically trying to build an entire robot empire where robots will inevitably uprise
just like it happened in shadow the hedgehog, which is 100% based imo
>He's clearly a rich fuck who just wants the government to get outta his way of exploiting others.the US government would only slow him down slightly, but one hedgehog is all it takes to stop him lol
>>25332but that's a lesser evil that we can take advantage of just like in shadow the hedgehog
>>25330I mean if he takes over the world down with the US defeated then it will actually be praxis actually
>>25335>I mean a robot dystopia where everyone is a robot wouldn't sound that bad tbhSo you support Marxism-Muskism? Brainchipped people of the world, unite!
>>25337>ah, destroyed by facts and logicNah, you're right in that he didn't mean to do it. He thought he could control it somehow. It can be interpreted as a metaphor for nuclear energy.
>which is technically a genetically mutated chao actuallyNo, I think it's the other way around.
Also, move the discussion to
>>>/games/27867 >>25348Seeing SCSI in action made me tear up a little. Master/slave connections are the worse-is-better of peripherals.
>I'd like to have about the advantages of top-down vs bottom-up vs dominant-player-but-otherwise-open vs other ways of building a computer ecosystemAre you more interested in the creation or the transformation of software ecosystems? I find the former fundamentally harder to reason about, because contributers in a stable ecosystem are either fully trapped in it, as is the case with windows, or self-select so their software is still very much in line with the rest of the ecosystem. Take for example the state of X Toolkits:
Motif is a geriatricly stable toolkit that developers use because of its wide support throughout most recent and historical Unixen. The ui elements look very uniform with chiseled grey borders and static bitmap buttons. Contrast this to Qt, where an application can be extensively styled and ui elements may be added with a module architecture. On the other hand the software ecosystem of Qt involves a lot more components and heavily incentivizes certain types of applications, with no C bindings, a domain specific language and even its own build system.
Between these two extremes there is GTK+ (formerly GIMP toolkit), that is written in object-oriented C and allows extensive theming. It became and is still considered by many the native toolkit for linux and sees heavy use by small graphical utilities as well as larger applications. It was initially developed as a more flexible motif alternative, yet has long served as a vehicle by the Gnome developers to change the look and feel of the linux desktop. GTK is more limiting than Qt, where the uniformity of the KDE interface is acccomplished using widget libraries and the breeze theme, and less limiting in practice, because easy interfaces are easy to write with it.
The situation becomes indicative when examining the instances GTK and its userbase have been at odds. The update to GTK+3 removed ui options like zebra menus, enforcing their aesthetic sensibilities on every user, and made accessibility features mandatory for people without a sane distro, that carries a stub or gives the option of patching it out. Libadwaita is a more radical step in this direction, which enforces a completely uniform interface at the library level, like motif only without the stability. Both have been met with harsh criticism by developers, yet the most they can do is remain on an older version, because more sane toolkits targeting the same niche (motif, xaw, tcl/tk, fltk) are more restrictive in some ways and/or more involved in other ways.
I think this situation could be described as 'dominant-player-but-otherwise-open', but it's certainly something that came to be gradually.
>>25338>So you support Marxism-Muskism? humanity will only achieve immortality when everyone becomes a machine
>Brainchipped people of the world, unite!brains would be replaced with mechanical counterparts
>No, I think it's the other way around.it's canon, chaos was a chao that mutated because of the master emerald. chaos chao looking like chaos 0 in miniature is more like an easter egg of that
>He thought he could control it somehow. eggman got tricked, 300iq moment
>>25339>>25342300知能指数
>>25364 the catch is that it's usually easier to fuck with windows enough to satisfy your desire for "good-enough" than it is to completely change your OS. Recall can officially be disabled, if you don't trust that some brazillian will almost certainly write a batch file that removes it entirely which you can set to run every time you start up windows. DRM is bad, but it's almost always easier to pirate a program or find a way to bypass it than it is to move to an alternative…
this is the fundamental dilemma of the security or rights-conscious user: if you actually want to undermine
mass surveilance efforts or
mass violation of end-user rights, you need to make everything trivial to use so that it fits into most people's existing use-case. look at TOR, look at how many glowposters there are: i would bet you most of them are running TOR on Windows because it's literally just fancy firefox.
even then, you've gotta push uphill: firefox or fancy firefox both struggle because chrome is
good enough and pushed harder by a large corporation. fundamentally, 99% of users start with a problem and find the system that resolves that problem, they do not start with a top-down view and then exclude based on certain criteria. "you'll be spied on! DRM! your rights!" cannot weigh up against the simple fact that your boss wants those reports by Friday, your computer is right there,
and you cannot hear the guy shouting at the start of the sentence…
when a river flows downhill and floods your garden, you have two options: blame water for being lazy and psyop'd, or get out a shovel and try to change the route that constitutes the path of least resistence.
>>25366and also for the extra blackpill: stay on any version of windows and the US will hack your computer regardless lol
>>25365>Recall can officially be disabledit probably still runs in the background even after checking a box to disable it, proprietary software hides the source code so we can't be sure if it really gets disabled or most likely is a mere trick to help windows useds cope
>DRM is bad, but it's almost always easier to pirate a program or find a way to bypass it than it is to move to an alternative…not always, one example is FL studio, It works on gnu/linux with wine but the full version is too expensive and cracks don't work, also there's no equal alternative
>i would bet you most of them are running TOR on Windowslol they think they're safe from microsoft or CIA spying by running Tor, it's sad
>you'll be spied on! DRM! your rights!" cannot weigh up against the simple fact that your boss wants those reports by FridayI mean it's worth the try to spread free software propaganda
>blame water for being lazy and psyop'dI mean yeah, it's analogous to people refusing to join in any form of political organization or refusing to read anything marxist-leninist, they're either lazy or too busy(which in this case is perfectly fine, but they're also psyop'd).
>>>25338>So you support Marxism-Muskism?more like richard_stallmanism-marxism, since if people are going to use hardware as body replacements then the hardware has got to be free or else people will literally become slaves
>>25383but they have no cpu backdoors SO I'm NOT a
slave actually
>>25361>300知能指数Bing Chiling.
>it's canon, chaos was a chao that mutated because of the master emerald.Then why does he look almost exactly like the Ancients?
>>25403我的置评是日语的、其实。
idk either language>Then why does he look almost exactly like the Ancients?I think that's a lore gap in sonic canon but apparently the ancients become chao because of the radiation of the chaos emeralds corroding their dna, and then the chao becomes ancient/chaos again because of the master emerald? idk. I think I need to update myself on sonic lore, I actually don't know this recent sonic lore, but it seems to be pointing out that the ancients are an evolutionary predecessor of the chao I think
>>25429>>25430Automatic device loading only looks easy on Windows and 'buntu, because a horde of maintenance workers is ready to fix any obvious regressions. Udev on a non-mainline distro is suffering. Recently i connected a midi device to my Alpine pc and the firmware distribution installed its own udev rules, that *shockingly* didn't fire. Running the actual firmware loading command worked like a charm. Maybe i will abstract some of my scripts for this into a more general, mount-style manual hardware attachment framework.
"Plug and play" is almost never the right thing to do. A whole class of computer worms wouldn't exist if udev and co. didn't act like the software equivalent of a hooker. Udev actually traces some of its design back to a software package called HAL, a kitchen-sink hardware detection framework that eventually collapsed under its own weight. Its remnants where merged into udev shortly before its codebase was adopted by Redhat to further the systemd agenda.
>>25439I only did static /dev gentoo on a vm just to understand a bit more on what I missed out in the past when udev and systemd and etc weren't a thing
>Udev on a non-mainline distro is suffering.true but on gentoo with openrc I pretty much don't have to worry about udev, but gentoo is pretty much mainline as well so…
>"Plug and play" is almost never the right thing to do.if support for a device is compiled into the kernel or as a kernel module(not really preferable) I think it would be fine for it to be plug and play
>Running the actual firmware loading command worked like a charm.I bet it was to load a kernel module containing the firmware
>>25453>if support for a device is compiled into the kernel or as a kernel module(not really preferable) I think it would be fine for it to be plug and playEvery codepath dependent on and triggered by an external device is an exploit waiting to happen. You certainly know autorun.bat, but consider the Fusee-Gelee exploit for the first Nintendo Switch models. It was made possible by a single stack exploit in the TegraRCM firmware's code for handling USB connections.
>I bet it was to load a kernel module containing the firmwareNo. Fxload loads the firmware directly to the USB device.
>>25467>it's not like what product people use will bring us communism or whatever.It
will postpone Big Tech feudalism though so you have more time to bring communism or deterritorialize corporate power to transition into a true Deleuzo-Guattaro-Landian schizoanarchist post-humanist future. Whatever we do, we should not give the bourgeoisie more hand to chew, that's what these mfs want.
>>25047There are probably better Linux distros now than Windows 11 which struggles to do basic things right plus its tracking your every move, Linux most likely doesn't. (They are literally sending data as we speak right now.) A good idea to switch if you don't like having your data sold without your approval by Microsoft to cyber scammers.
I personally just don't know which one's right for me, choice paralysis is a real thing.
>>25550Well that's the thing, it's not that they're all of the sudden spying people like they hadn't been before, they've been taking and sending screenshots periodically since like Windows 8, it's just they keep finding more hardware intensive ways of spying. They must use more of your computer's resources to accomplish a task that they know doesn't need those resources since they used to do it the old way.
Like you shouldn't need more than 8mb's of RAM to wiretap someone, but no even the keylogger has to be an electron app. Even the calculator has to be an electron app. Everything. Fuck.
>>25552"freedom" is a feel good term, it must be defined in terms of actual practical advantage to the user. "you are free to modify the program!" is as meaningful to the non-programming user as "you are free to start your own business!" is to the unemployed man.
the tragedy of the current system is that most people already use the best programs available to meet their needs. the theoretical ability to build a better X using free software licences and an open development environment is offset by a combo of a lack of developer interest, and a lack of finance. the first is unlikely to be resolved, the second takes you out of idle licence-fantasy and into the timeline where Microsoft is split into 150 independent People's Software Development Houses.
and where most end users continue using a now public domain Windows. >>25553>"you are free to modify the program!" is as meaningful to the non-programming user as "you are free to start your own business!" is to the unemployed man.I see it more the right to repair. Even if the user can't code it means they can use community patches
>The tragedy of the current system is that most people already use the best programs available to meet their needsIf that was the case Adobe would have went bankrupt decades ago, it is an optimized piece of shit with alternatives being generations ahead in features and speed.
>>25561>"freedom" is a feel good term, it must be defined in terms of actual practical advantage to the user.Those practical advantages are short-sighted. The thing about software freedom is that it frees you from corporate control which has a long-term advantage. Otherwise you wake up and Windows 11 happens. Four software freedoms create an environment which works how you want it to work, does what you want it to do and goes out of your fucking way. The thing about software freedom is that the users control the software, not the developers, "making your own version" is absolutely irrelevant to that conversation, it is simply a safety check for the libre world to keep operating like that.
>free software is idle fantasy>the timeline where Microsoft is split into 150 independent People's Software Development Houses.Ah, that's definitely not an idle fantasy of yours. Keep dreaming, mah boi.
Tbh I don't even care about libre software "winning," what's more important is to prevent it from losing.
>>25132>>25133>do you want to make an upfront investment in learning how the system commands work, or do you want to explore it by clicking around on your own and seeing what happens..? the answer is not the same for everyoneComing back to this because it's relevant and slightly interesting:
In doing usability studies for Microsoft Office, Microsoft found that looking at the manual is something that is mainly done by experienced users. Inexperienced and intermediately skilled users tend to explore and play around with the options on screen since swapping to the help section is a "context switch" from simply using the program (and to a lesser degree, because the help system can't really explain how to complete practical tasks - it mainly describes general program features. Word help can tell you how to perform a mail merge, but that's no use if you don't already understand that a mail merge is a more efficient way of solving your problem…)
https://web.archive.org/web/20060206131623/http://blogs.msdn.com/jensenh/archive/2005/11/29/497861.aspx>One of the most interesting epiphanies I've had over the last few years seems on the surface like a paradox: "help" in Office is mostly used by experts and enthusiasts.>How can this be? I think my biased assumption was that experts know how to use the software already and eager novices would be poring over the documentation trying to learn how to be more effective using it.>Yet, in usability tests we see it again and again: novices and intermediates click around and experiment, experts try to reason things out and look them up in help.….
>Alan Cooper talks about "perpetual intermediates" in The Inmates Are Running the Asylum. The idea, paraphrased, is that most people using software are "intermediates." Beginners don't stay that way for long, but most people don't have the time, energy, or desire to become truly elite "experts." I believe that it is precisely these intermediates who don't rely on the help system. In fact, they might be defined by their general unwillingness to look features up in the "command encyclopedia." Experience shows that intermediates tend to explore the product, not the help system.I'm very conflicted now. On the one hand, it's nice to have empirical validation for a hunch. On the other, people knew this in 2005 and we're still discussing it in 2024. If it hasn't been taken onboard already, it probably never will be. >>25744I'd see it more like treating the computer similarly to a physical tool or space. You don't read the manual before using a pen, a screwdriver or a cup, and you don't (generally speaking) look at a detailed item index before looking around the supermarket. It's a fairly natural extension of the desktop metaphor.
I would say my usage patterns are similar to these users, though more familiar with where stuff is likely to be, or what kind of features programs offer. Generally speaking I find exploring a program painless but reading a manual to be tedious. Exactly as supposed in the post, because it's a context-switch: I came here to
use Word, not to
read the Word manual. Clicking menus to see if anything useful is hidden there is doing what I came here to do, opening a new window and hunting through an encyclopedia is not. Similarly, if you're looking for milk in a supermarket the odds are you walk around the periphery looking for where they've put the chilled stuff - you don't seek out an index.
>>25747Being preoccupied that people don't waste what little time they have on the same hobbies they have is such a petit-bourgeois mindset, ngl. We aren't getting communism if we get more people to install Linux or know how folder structures work or whatever.
Fwiw I too use Linux and dislike things like datamining, but these are inevitable trends under capitalism that won't stop if we get enough people to switch OS. At the end of the day FOSS is just a consumer movement and nothing else (this shouldn't be controversial either).
>>25614>The thing about software freedom is that it frees you from corporate controlOh totally, if you install Linux you suddenly stop living under capitalism. FFS the FOSS software is a majorly libertarian thing and it painfully shows. Just speaking in moralist liberal terms constantly.
You will get more people using Linux if you actually explain its tangible advantages, of which there are many, instead of sounding like a lolbert cultist.
>>25747>Nothing wrong with people not wanting to bother with how a tool worksIt's nothing like that. Only the simplest programs, many of which don't have a right not to be a thin wrapper around a cli, are comparable to real world tools. Something like notepad is comparable with a multi-head screwdriver, but Microsoft Outlook has the complexity of sending a mentally challenged secretary to the post-office.
Cli programs directly embody the interaction with sofware as a dialogue. Every action should result in a response or deliberate lack thereof, to expose changes in program state. A GUI could pop up a text box for every action the user takes, however the program is usually build around graphic interaction with some state representation. Many of these follow entrenched ui conventions. Still to figure out any specifics, you need a manual or picrel.
>>25747>Generally speaking I find exploring a program painless but reading a manual to be tedious.This is why i mentioned the literacy situation. Effectively reading something like a reference book should be a skill naturally acquired during high-school level education. With electronic manuals you don't even need to flip between index or register and one or multiple pages. Everything is accesible with one click or command.
Reading imposes some cognitive load on the user, when compared to the interactive on-screen tutorials some applications have. Naturally though this type of documentation cannot cover as much ground and takes some time to implement. Providing documentation as a well-structured text is a worthwhile tradeoff in most cases.
Video tutorials unite the worst of both types of documentation. You get a moving series of pretty pictures at best with a commentary you could have read inside a manual in a more polished form. Listing a series of menu trees is both less tedious and more precise than watching someone else go through it.
>>25757Video tutorials are the worst of all possible worlds: on this I agree. Otherwise, I'm still not really sold. I would analogize more complex programs to a physical space rather than a single tool (Outlook as a room where, for example, you may hunt in random drawers to find a screwdriver), but the analogy would hold. I had a parallel thought when it came to CLIs and man pages: Imagine a world where stores don't display prices - where to see the cost of goods, you've got to make a specific inquiry with the shopkeeper, then he'll tell you and you can decide what to do with that information. A lot of old stores ran on exactly such a system, and it has advantages (think of all the time wasted stickering goods… and you can't tailor the price to your customer if an out of towner shows up…), but it also adds friction to the experience of using the store compared to the alternative of having a printed price tag under every option in the store. That's why most stores don't do it today: most people hate friction.
And I think that's the crux of it: I do not refuse to read manuals because I find
reading difficult, I refuse to read manuals because I find task switching generates
friction, and because I actively enjoy learning by playing with things rather than by reading about them and following the instructions. I am at ease simply using the software and solving the problem playfully, and well designed graphical software always provides affordances for this style of use. For example when it comes to your image, there's no reason to have to look this up or ask for someone to explain: the correct GUI solution is to provide a tooltip that tells you what the icon means if you hover over it. With no need to switch out of the program, I can go "huh, I wonder what that is?", mouse over it, and learn something. When it works it's incredibly elegant.
Though I may have said already: I'm not opposed to providing good documentation. Acceptance that the main people consulting the documentation will be expert users really ought to
improve documentation for that demographic, since it can be better tailored to the reader's needs, and ensuring everything is well documented
somewhere is helpful for keeping things consistent and coherent. A program can always be written such that it can have both a UI that works even if you've not read the manual, and a clear manual explaining everything accurately and concisely.
Unique IPs: 41