Desktop linux largely uses polkit and elevation dialogues, not sudo.
What have you needed to do with terminal? KDE (kubuntu etc) largely shouldn't, it's an option for power users, it shouldn't be something you are forced to interact with: similar to cmd or the windows registry... it exists but it's not like you are expected to interact with it.
I also haven't compiled software for myself in about a decade. But of course you are beholden to the package maintainers.
I can answer this concisely; not in the terms exactly you describe, but I hope sufficiently: What you describe is a different experience from what I have observed.
edit: I think I understand from reading context: I do not assume all software someone will use is available on the distro's package manager, or that I want to use the versions on the distro's package manager. I think these discussions sometimes get confused because there is mentality A: that you should only use software on your OS's app store, and B: That you should be able to run arbitrary software. I suspect you are A and I am B here, but am not sure.
I agree that if you only use included software or install software from your package manager, most of the pain points, including the ones I mentioned go away.
Every piece of software(1Password, my own, and others) that I wanted to install has required me to paste an incantation into the command line to install it. It is incredibly unfriendly to less technically inclined users.
It should be:
Download RPM package ->
Double click RPM package ->
Approved access elevation prompt ->
Done!
Instead it fails completely by either trying to open the package like an archive or opening it in Discover to show the package information with no option to install it.
Ideally you open “software centre” and there is a 1password you can just click “install” on, in fact: that is how it works. No need to use a web browser at all.
You’re applying Windows thinking to another OS. Maybe we should make dummy installers that are just links to package managers for refugees; you’d have the same problem going to a Mac, its only windows software that forces every application to have its own update lifecycle system in its own process.
I'm applying how mainstream operating systems work to Linux. Windows is easy, download the installer and run it. MacOS is also easy, download the disk image and copy the application to where ever.
My experience with Linux and installing applications has been a disaster. They fail to install from a downloaded RPM or DEB just by opening them. So the fallback is the command line which is already a failure for non-technical users. A simple "sudo dnf ..." is not too much to ask users to learn, but now they are being taught to copy and paste commands into the Terminal that they don't fully understand.
So, in that way they are similar- and you can download software, just barely anyone does that. Static binaries do work, but since almost nobody is embedding update methods in linux software I wouldn’t recommend it.
It's all beautiful until something doesn't work. Then you go find official website, read that "default repository contains old version, go add this repository instead, here are some magic commands to paste in terminal".
Indeed, I have the same gripes, even when using one of the more popular distros (Fedora).
From the top of my head, software I could not find in the main repos and had to resort to adding a COPR (= custom repo) for, install a downloadable .rpm, or worse use their suggested bash pipe install thingie:
- Docker
- VSCodium
- Joplin (note taker)
- GlobalProtect-openconnect
- AirVPN's client (yes, I know I can generate a openvpn config, but I think many users might not want to/be able to)
- Yadm
Some apps offer Flatpak installs as an alternative, but in many cases they are not maintained by the actual developers, but rather by a regular user/third party you have to trust.
In a nutshell, I don't find this particularly user friendly, and I don't think it can be expected of all users to understand the intricacies of the different linux package ecosystems and their implications, especially when the built-in GUI package managers do not always make it clear which source you are installing from (and why you would want a native package over a flatpak one for example).
Disclaimer, I still prefer using Linux, both for work and gaming, but I believe this is a major point of friction for many potential (new) users.
every year theres less and less of those is my point.
Docker is definitely “power user” software btw. If you can’t handle running commandlines you will almost assuredly have a really difficult time with docker itself on any platform.
That said, adding extra repos is possible in Ubuntu software center without a command line, I suspect Docker thinks it is catering for a demographic here.
This is not specifically docker issue, docker is just an example.
Installing 3rd party software on Linux is often painful - you rely on official distro maintainers(slow) or official vendor distribution(not easy) or third party source like AUR(may not work and you are trusting some random guy to not steal your bitcoin wallet)
"the devil you know" is indeed a painful disincentive to switching, the natural lock-in effect of inertia is enormous and often understated.
Coupled with some major AAA's not working, I understand why people are reluctant to switch. It's a bit of a chicken and egg problem.
That said, it is hugely viable now, so you have a choice: surrender yourself to data mining and an increasingly hostile platform, or take the plunge and try.
I'm super happy that this choice exists, Microsoft no longer has us by the balls.
a well known problem with more than a few known fixes, one of which is load shedding your upstream based on protocol aware load balancrs.
I've had similar issues where backends couldn't come up due to load killing them as they came up: my solutions was to turn off the load balancers until the backends all came up before opening the door again...
in fact, the mac product line at that time actively hindered the iPods success.
I remember people saying that they’d buy an iPod if they didn’t need a Mac- and firewire was quite rare. Eventually people realised it could be used on Windows and had USB, but they had to use iTunes, which was a hinderance too.
Really, it’s amazing that it was so successful, it shows that the right kind of product with the right marketing can overcome some pretty significant challenges- but yeah, people didn’t want Apple products unless they had to have them for creative workflows.
Yeah, but the alternative is everyone having a short domain, which is unfortunately difficult due to the fundamental mathematical principle that a low amount of characters has a very small number of options available.
Everything with finite space is expensive, the more finite the more expensive.
So, while I am in favour of what you say, and I’m in favour of self-governance as much as possible, I understand the challenges.
>Yeah, but the alternative is everyone having a short domain, which is unfortunately difficult due to the fundamental mathematical principle that a low amount of characters has a very small number of options available.
I get where you're coming from but I'm wondering if that's undersells the value-add of link shorteners as literally just shortening the link.
I remember that I liked using link shorteners because they gave me analytics on who clicked my link and where from. I believe paid QR Code link shorteners operate on a similar principle of offering analytics on where and how QR Codes are discovered.
And while in theory, this should be handled by first-party analytics, the users of link shorteners don't necessarily always have control over that analytics platform (e.g., student researchers driving traffic to a survey on an institutional domain).
The original mass market use of link shorteners was to minimize the number of characters needed to link from a Tweet.
Now that Tweets aren't restricted to 140 bytes anymore, there is less need for every single byte. Maybe the difference between sending 1 SMS per message and 2...
I think the problem has always been the shit that gets appended in the query string, like UTM parameters for analytics or other random stuff that has no real impact on the rendering of the page itself.
The other problem was the dawn of the slugified URL where the title of the page was converted into a URL friendly format, and then the page's actual ID appended in it somewhere. The slug was just for SEO because it looked more 'readable' but the only thing responsible for fetching the page was the ID attached to it.
I think the walled-gardening of the internet was more responsible for this than individual webmasters. Half of that tech is useless unless you are running the hosting platform for bloggers or tweeters or whoever, and it was those companies playing the SEO arms race with Google.
The issue wasn't necessarily the stable URLs underneath, but the query parameters being long with characters in them that often broke or confused older systems.
Remember too that we were often dealing with 1024 x 768 or even 800 x 600... 1280 × 1024 was the emerging laptop standard. Long URLs without formatting would often wrap the text display panel.
Those that didn't reformat would have formatting issues. Those that did attempt to wrap excessively long lines on punctuation and hyphens would break the URL.
One very long url in a chat panel on the side of a web game would cause everyone's chat panel to be messed up until that scrolled off.
URL shorteners were (for me) more useful for ephemeral chat where it will scroll off the screen in an hour or so.
They lost their utility when message boards would take really long lines and reformat them (like HN does) so that they fit or would allow for [title](link) style so that the ugly url could be hidden behind a nice title.
> but the alternative is everyone having a short domain
Or just not bother.
Practically no one types a URL like that any more aside from a few biggies like BBC.co.uk, Amazon.com, etc, and even those are usually cached in URL history so they only type the first couple of characters. When were you last given a URL by voice, rather than being given a suggestion to “search for [something-or-other]”? The 140 char limit on Twitter¹ is now 280 making a really short URL less important there, those foolish enough to pay for the site can post much longer missives in one go, and I can't think of another popular site with a stringent artificial post length limit. People looking for your stuff by name will search², not type your address in.
I use a deliberately long name for humour value³. Though I host nothing on there that isn't just for me and friends so this is maybe a special case, I doubt that for many sites a short name makes much real difference these days. People who see my stuff do so by clicking a link, not typing anything in.
--------
[1] the social network desperately trying to be known as “X”
[2] or ask AI which amounts to the same thing usually
[3] subdirectories or subdomains (depeding on my mood at the time) of doubleyoudoubledoubleewe.www.dash.deeohhtee.dash-dash.enneeetee.net
I think there's some signaling and marketing value in a reasonable URL. If I see a really odd name with an unusual TLD that's not somehow relevant or clever I just think "wow, this scammer couldn't even afford a real TLD".
>Practically no one types a URL like that any more aside from a few biggies like BBC.co.uk, Amazon.com, etc, and even those are usually cached in URL
<
in most cases, on my browser history I have to type only the first or first-two letters to get usually directly where I want to go - so you can see, Im a monotonic surfer :-D
This isn't about trends. This isn't about being manipulated. This isn't adopting the newer morer politically correct word of the week. This isn't adopting a famous techbro's latest coined term.
A product/company was renamed. You can be a contrarian and refuse to respect their new name. Very few people cares one way or another. You're neither brave nor special for doing so.
Which largely my reason for caring what Twitter is called. The arse in charge, who wanted the rename, is rather against people being able to choose their own identity, at least in certain ways. Maybe if Musk starts referring to Ms Wilson as his daughter I will reconsider referring to the Twit Collective as X.
If I asked people in public what X is they would be confused what I meant. If I asked what twitter is and then asked what X is they would know the answer to both.
Everyone I know still calls it Twitter, and most blogs I read still do as well.
Calling it Twitter is more universally understandable. Using "X" can cause confusion if someone using a generic "x" as a variable as in "and then A said to X that Y..." For that reason I don't particularly care what the owner of the company wants to call the online equivalent of Kleenux he bought.
Not because I doubt you but Europe is pretty disparate.
Amtrak from NYC to Philadelphia cost as much as a flight to continental Europe, and the experience was abysmal, truly awful I couldn’t even get the tickets for hours and took a bus instead.
Perhaps I’m “spoiled” by UK and Swedish trains but UK is pretty terible
> Fair point about refunds for delays in general, its so awful I normally don’t do it
That's one thing in which DB - the German rail company - excels: their refund process is easy and often quite smooth. It is unfortunate that this is probably because they have had so much practice running it given than around 70% of their long-distance services are delayed but still, it works and I tend to get my 25% or 50% refund for 1+ and 2+ hour delays within a few weeks, sometimes days.
It depends hugely on when you book--both time in advance and time of year--as well as any sales going on. Which can vary a TON in Europe as well. I almost always take trains within Europe but they're no bargain.
Right. I'm not going to NYC for now but I saw the end-of-year sale in my inbox recently. Pricing varies a lot. (And I usually don't book Acela on my own dime absent a bargain; the transit time isn't much different from the regional.)
A lot of travel generalizations assume you're going more or less going downtown to downtown. There are some limited suburban stations for both NYC and Boston but, generally speaking, a lot of train transportation advantages over air go away if you're not traveling more or less downtown to downtown and become liabilities.
In my case, there's an out-of the way but otherwise convenient south suburban station outside of Boston and I'm basically always going to Manhattan so I train unless I'm just changing planes in Newark.
Oh you're right. I was actually thinking of the airport. Getting to/from Manhattan isn't so bad, but JFK is in Queens so the path of least resistance is generally to Manhattan and then to Queens even if you're already in Queens.
Getting direct service to the other boroughs would be nice though. Acela (Penn Station to Back Bay) was a bit of a WTF moment for me though.
Logan is actually quite convenient to Boston metro in general. But if you're driving in from the suburbs, RTE 128 works a lot better even if, for me, it's pretty much 100% out of my way to travel to NYC. I could drive to New Haven to avoid driving into the city (which I hate doing) and take Metro North but, then, that's more driving.
The year is 2019. I wanted to book a ticket on GWR. Couldn't, site didn't work with iOS. Figured I'd buy a ticket at the station. Couldn't, no staff. Figured I'd use the TVM. Couldn't, most were out of service. After I managed to buy a ticket staff mysteriously came out of hiding.
The trip itself was grand until the train (Class 800) broke down and staff started blaming passengers.
The refund process was quite smooth as soon as I replied to the initial denial and pointed out that yes being stranded for an hour qualified for a full refund under their stated policies.
Amtrak's been more reliable for me and I've never had problems booking a ticket. Never had problems with them while they were running Caltrain either. I get the joke but I think Amtrak does quite well with the constraints they're required to operate under. The UK doesn't really have that excuse.
Building wooden dwellings is comparatively easy and straightforward.
It is the lack of a tangible result that gives the insecurity that programmers feel when compared to constructed buildings.
The epic is not visible to the uninitiated when it comes to programming whereas with construction it is visible to the naked eye for everyone.
It is easy to see and marvel at a cathedral whereas most don't have the skill required to marvel at the architecture or the codebase of an elegantly designed distributed system.
To be blunt, building a wooden shack is physically easy. Building even a basic program requires a level of abstract mental discipline that most people simply do not possess.
So the "most people" you mentioned who are throwing up a shack to live in, I admire them for that.
technically, anyone can grab a copy of source, put their phone into developer mode and install the app too.
But- practically, it's hard.
You only get 4 application slots on devices in dev mode for example and I think they also expire.
I tried creating an internal app for my company and was gobsmacked that I didn't have any sane method to distribute it short of taking full ownership of everyones devices via MDM.
You are of course not allowed to publish "internal" apps to Testflight (or the App Store, but that's more obvious).
Given how incredibly important they can be, having an operating range that can be exceeded by ambient temperatures in some (very populated!) regions of the world is a design flaw.
iPhones are only designed to work from 0-35C, which is more narrow than the yubikeys, and way more important and way more widely used, so I think it will probably be okay.
my iPhone is nowhere near as important as my yubikeys.
Without my phone, I can still access my computers and servers and do my banking etc. Without my yubikeys this is going to be a real problem.
What have you needed to do with terminal? KDE (kubuntu etc) largely shouldn't, it's an option for power users, it shouldn't be something you are forced to interact with: similar to cmd or the windows registry... it exists but it's not like you are expected to interact with it.
I also haven't compiled software for myself in about a decade. But of course you are beholden to the package maintainers.
reply