... DOS which had horribly buggy implementation of the add-ons!
Microsoft mostly managed to kill markets by turning people off with their horrendous reimplementations. People default to Microsoft's built-in. Said built-in catastrophically blows up on them. People decide to abandon the technology. Microsoft is happy to have gotten rid of a competitot (but technology stagnates).
see: Stacker vs. DoubleSpace/DriveSpace for an exemple. there are numerous others.
some might wonder if the horrible quality of Internet Explorer wasn't actually an attempt to kill the whole internet in the same way.
first, Microsoft is pretty much aware that they've lost the server/cloud game to linux (and router. and embed. and smartphones. and SBC. basically, desktop/workstation is the single niche that Microsoft is still holding) BUT they know they hold the desktop, and would very much to keep holding it.
some of the logic going in the heads of microsoft is that wsl can be a bit of anti-gateway drug.
for all these devs, who have a mostly windows environment but need a bit of unix in their workflow. (devs that need to write code that will end on the linux server/cluster/etc)
until now their main choice were installing a Linux VM (or SSH into a Linux test server. or Switching to Mac OS X or some Linux powered laptop) and probably some at Microsoft would be afraid that this VM would be a gateway drug: once they got a bit of taste of Linux, some are likely to jump ship and install Linux (or switch to OS X. or exclusively run a Linux VM full screen).
by providing wsl, Microsoft is giving an alternative test environment for the couple of linux needs, while keeping everyone still firmly within their system.
the problem (for them, but advantage for us) is that it might end up the other way around: wsl is so much limited that eventually it will encourage some to go further and transition to the real deal.
while it is far from something that you can actually implant into a patient in need of organ transplant, current state of the art is already able to build "organoid": small structered 3D cell cultures that more or less mimic on a tiny scale some of the structures found in real life natural organs.
These are already useful at this stage for research (mostly pharma as they help investigate better the effects of potential drugs on interacting functional tissues, rather than cells floating freely in a test tube).
i.e.: we're already halfway there, and this halfway is already useful (though it's more "a few percent down the right direction" rather than litteraly *half*way)
Video games directly causing eyesight degradation is dubious (as in you need to use some magic filter glasses to avoid your eyes dying slowly when looking at screens).
There is some corpus of evidence (including studies done in Japan that predate widespread use of smartphones) that somewhat link decreased time spent outside (outdoor activities in sunshine) with increasing need for prescription glasses (not simply explainable by increased reporting due to higher reporting). And some phenomenon replicated in lab by sewing semi-shut eyelids of apes (still an aweful thing to do in my opninion).
It looks like there is something causing eyes to get myopia when they spend more time looking to close object under faint lights (indoor, evenings) rather than far away in the sunshine.
From that point of view, video games are simply yet another distraction from outdoor activities (as were books when the first such studies were started) whose reduction could have some play in degrading eyesight.
But I don't think outlawing games like totalitarian China is doing is a great solution to the problem.
I'm assuming you can program it through a Wifi end-to-end kind of thing, not requiring a internet connection.
Yup. At my brothers' first setup was done with a direct wifi peer-to-peer connection, then over the local network for subsequent timer settings.
Connection to iRobot's cloud is only used for keeping an archive of maps of finished jobs, all this weird "make a map of wifi coverage" useless features, permanent floor maps for TFA's "go clean the kitchen" jobs, etc.
The Wifi and app are only a hard requirement for: - firmware update (it's not done with a serial dongle anymore) to weed out bugs (the first version of 900 had navigation problems) - setting time and timers (there are buttons on the device utself anymore. Apparently "blinking twelve" is still a problem with people too stupid to press buttons).
Separate cloud connection is required for the creepier features
The first version of the mapping feature of the 900 series was a bit buggy and got lost. It cleans okay (still able to reach almost everywhere using the random walk of classic roombas) but has no fucking clue where the home base is and how to reach it back (it just keeps running around aimlessly until battery runs out, whereas classic roomba run aimlessly around until within reach of base station and then IR to engage the "tractor beam" feature).
- Not all Roomba are even Wifi enabled to begin with.
Even Wifi enabled Roomba, work without wifi. Even if you activate the Wifi access, you can also use only for firmware update and otherwise only talk to your smartphone directly over the local network (or optionnally to your home server running on some raspberry pi). ( ^at that point you already have 99% of features people want, including local tiled maps on navigation-cam featuring models like the 9x0 serie)
This thing aditionnally requires you to connect it to iRobot's cloud and link to your Google account. ( ^ this is required for the creepy features like using the App over internet (for 890 and more recent), storage of maps of past runs (for 9xx and more recent), and permanent map of the floor and voice commands (for i7/i9) )
If you look at the top most contributing devs, you'll notice that they are actually on the pay roll of companies who rely on linux. If they weren't already employed by Red Hat, they would probably be at Intel, Google, even IBM themselves...
Not all programmers are poor. If they are anywhere near competent (and open-source software makes a great portfolio that is easy to show around), they'll certainly get hired, perhaps even get paid for their open-source hobby.
Which is a pretty easy trick to achieve given that egg protein already start to precipitate somewhere north of 50~60C - you could achieve the same with the warm water of your faucet, try it ! Note that the egg will not have been thoroughly cooked at a high enough temperature and will not be sterilized : it might not be safe to eat due to bacterial risks. You could do the same trick as the video with any piece of electronic more beefy that a raspberry pi
And while digging at old stuff, Intel was at the recieving end of such jokes back in the Pentium4 vs Athlon&Opteron 64 era.
(Also, since when are AMD motherboards paired with intel NICs? That seems weird to me...)
Also notice the UI difference between modern Widows 10 and Office 365 now, and back when the current users where school children.
Even if by some chance Microsoft and Office are still relevant in the business world when these Australian kids grow up, it would almost definitely not look like anything now.
you can't see the pixels of 4k monitor *several feets away*.
but you could definitely see them if the monitor is a few inches away from your nose and wrap around your whole head (well at least spread around everywhere your eyes can look at), which is what the current generation of vr is attempting to do (as opposed to older gen that only simulated a somewhat largish screen in front of view and nothing around)
but all this extra screen/fov estate cones at a cost (either resolution if you stretch a 2k/3k display, or performance if you try actually 8k)
hence all the buzz generated in the media by "foveated rendering" (i.e.: eye tracking and variable resolution) as the contender for the vr gen after the current. (i.e: only render the 3k that the eyes are currently looking at/pointing at, and render a low res blurry approximation for the peripheral vision where it is good enough)
the current buzzword du jour - ray tracing - happens to excel at such varying resolution.
For the subcategory of software devs that run mainly on some other OS (Linux is also popular in the biomed research - mostly on workstations and servers/compute nodes. Windows is still king in some business settings and with gamedevs), but need to port and test code on Mac OS X (which is *also* popular in biomed research - mostly on on laptops, and some iMacs here and there).
The only legal way to run a licensed OS X (even virtual image) is to run it on Apple hardware (though the license doesn't require it to be the host OS). Mac Minis are a cheap and simple way to have a legal way to test Mac OS X code, and use extra monitor inputs and/or console switch box and/or VNC to use the Mini alongside the regular workstation. (The expensive alternative way is to use an Apple workstation *as* your work horse) So being more capable would certainly be appreciated (e.g.: could be easier to run multiple VirtualBoxes with the various versions of Mac OS X you target in tests).
But indeed, it's a very tiny subset of Apple's customers and that serie's intended user base.
Mac Mini were mostly targeted as a "gateway drug to the Apple world" for average PC users (keep all your USB- and HDMI/Displayport- peripherals and only plug a cheap Mini to quickly check if the grass is indeed greener on the other side of the (walled garden's) fence). So being a cheap and light-weight machine is relevant for maybe 98% of its intended audience. Most of which won't be interested by beefier specs.
The problem is that RFID has a much crappier range than the typical car keyfob. In practice that would require the transmitter which goes near the victim to be crazy close (like in the same pocket where the victim put their hand), or to use a very conspicuous bazooka-style ginormous cantenna.
I suspect that all the inbreeding to which the racist native criminals have subjected to themselves to keep their "pure blood" status, might have played out negatively and could indeed explain why they're much stupider~~:-P
Unless it features contacts on the outside of the skin, it is a transceiver.
Yes, RFID chips are transcievers. But, with a horrendously crappy range. And no really a good collision-resolution alrgorithms (*) So unless the attacker use extremely conspicuous bazooka-style ginormous cantenna, they WILL definitely have trouble tracking those Swedish thumbs.
---
(*) Though some RFID in some contactless cards I've seenchips do. In one of the Uni I've worked at, if your hands are busy, you could unlock doors by bringing the pocket with your wallet (full of various RFID cards) to the general vicinity of the sensor. The card and sensor will manage to negociate despite all the others cards.
Yeah, I am sure that rockets - which fly only every now and then - are the most environment-impacting form of transportation. Specially compared to cars, of which there are an extremely large quantity in circulation, which altogether total gazillions of kilometers of distance.
but as written any item over a certain dollar amount is covered. Since that dollar amount is anything over $5, I'm pretty sure that it covers anything that Apple makes that has any kind of warranty.
Apple, being Apple, "anything over $5" is 100% guaranteed to always cover absolutely everything in their inventory down to every single last SKU. Even if Apple started producing items as simple as "single use iAss-wipes", it's going to cost at least 50$. per unit.
So any images of exoplanets, black holes or any surface except the mentioned above are artists impressions.
Well, technically, due to how physics work, it will never be possible to make an image of a black hole. There's no such thing existing.
What you could take pictures of is the gavitational lensing (Einstein ring) caused by the presence of a black hole (well, that peculiar photo is a super-massive black hole AND all the surrounding galaxy at the core of which said hole sits, causing the photograph-able mirage together. But you got the idea).
EV use has nothing to do with climate change in either direction.
The amount of extra CO2 we humans release into the environment is considered to have a significant impact on the climate, according to current scientific consensus.
Usage(1) of EVs compared to ICE cars release less CO2 in the atmosphere. It ranges between "a bit less" up to "almost none released", depending on you local electricity energy source mix, except in a few countries (like India, China and Australia - if my memory of that study is right - where things are more or less equal)
Even in those place, if you replace ICE cars with EVs, even if it doesn't change anything *right now*, it's eventually going to change things *instantaneously* once the country switches to greener electricity sources mix. Whereas if a new less-emitting ICE is invented, it will only show changes much later, as older vehicles are replaced with newer.
Basically, an EV has the (indirect) emission of whatever happens to be the electricity source mix last night when it charged. Whereas an ICE car has the (direct) emission due to the efficiency of whatever ICE was built into it several years ago (+ indirect emissions linked to keep the distribution network).
---
(1): over the whole lifetime of a vehicle, the *usage* is what has proportionally the highest impact on the environment (as opposed to the manufacturing which is quite minuscule in comparison).
why not just buy a cheap android 7 or 8 inch knock-off tablet for $50?
These tend to have old under-powered chipset from shitty companie like MediaTek. Not only they are difficult to unlock and install LineageOS on them, but in today's era of "you must download a multi-megabyte katamari of innumerable javascript framework libraries just to be able to display a single webpage", their weak CPU isn't even good at browsing the web.
If you want a google powered refrigerator, why not buy a 10 inch android tablet and velcro it to a dumb fridge?
It's a good idea, but while you'd be avoiding a huge chunk of the privacy-molestation, you'd be still missing on the sensors that make the whole idea of a "smartfridge" useful (you know, the whole "I'm detecting that the product XyZ you bought N days ago is approaching its 'best-fore' date" , and other "You're running low on milk, would you like me to ask Alexa / Ok,Google / Cortana to order some ?").
So the correct bill of material would be a cheap-ish 10" android tablet, velcro and ~50$ worth of arduinos and sensors from Adafruit, Sparkfun and friends.
(Well, probably a bit more if you need an array of good quality cameras in regions where product code are still purely 100% barcodes (like in the EU) and not NFC-enabled (I've read it's coming in the US), but you got the idea )
In addition to MancunianMaskMan's excellent answer : fwupd itself is a command line tool. Gnome has a nice GTK UI to make firmware update user-friendly. KDE has probably added a similar tool, based on QT and KDElibs
Seems rather odd considering most users will have no idea what either of those are. Bit of a backend way to push both those little policy agendas. It's the distribution's job to install such things not the desktop environment.
The *whole raison d'être* of Flatpak, Snaps, etc. is to make end-user installable software that is *independent of the distribution*, somewhat reminiscent of Windows' SETUP.EXE. (as opposed to 3rd party package repositories, like Opensuse's OBS, Ubuntu's PPA, Gentoo's overlays and whatever the hell they are called with Fedora/Redhat). Thus it needs a different GUI than the one used for distro-package. Hence KDE providing a GUI for Flatpacks. (And as an added touch could ask the distro package management to ask missing dependencies to get Flat up. Probably over some standard scheme like PackageKit or something)
I won't probably be using either tools (being mostly CLI), but it's still a nice touch.
These days, computers are plenty fast for most tasks. {...} For most tasks by most users, five years is fine.
Sadly, the hardware vendors don't seem to agree with you. It seems that if it's older than a couple of months, you're basically on your own.
(Except maybe for the "business" line of some manufacturer like Dell or Lenovo. But definitely the case for separate motherboard manufacturer as in TFS).
Heck, I'm running a laptop that was manufactured August 2012; 6 years ago. It performs just fine. (Disclaimer: I'm running Linux.)
That last part might be the reason why you and I can pull such a stunt, including the latest kernel. The same hardware will probably refuse to run anything more recent than Windows Vista. (And let me guess, also using Nouveau due to Nvidia dropping support of the embed GPU in your laptop ?)
Slashdot Mirror
... DOS which had horribly buggy implementation of the add-ons!
Microsoft mostly managed to kill markets by turning people off with their horrendous reimplementations.
People default to Microsoft's built-in. Said built-in catastrophically blows up on them. People decide to abandon the technology.
Microsoft is happy to have gotten rid of a competitot (but technology stagnates).
see: Stacker vs. DoubleSpace/DriveSpace for an exemple.
there are numerous others.
some might wonder if the horrible quality of Internet Explorer wasn't actually an attempt to kill the whole internet in the same way.
first, Microsoft is pretty much aware that they've lost the server/cloud game to linux (and router. and embed. and smartphones. and SBC. basically, desktop/workstation is the single niche that Microsoft is still holding)
BUT they know they hold the desktop, and would very much to keep holding it.
some of the logic going in the heads of microsoft is that wsl can be a bit of anti-gateway drug.
for all these devs, who have a mostly windows environment but need a bit of unix in their workflow. (devs that need to write code that will end on the linux server/cluster/etc)
until now their main choice were installing a Linux VM (or SSH into a Linux test server. or Switching to Mac OS X or some Linux powered laptop)
and probably some at Microsoft would be afraid that this VM would be a gateway drug: once they got a bit of taste of Linux, some are likely to jump ship and install Linux (or switch to OS X. or exclusively run a Linux VM full screen).
by providing wsl, Microsoft is giving an alternative test environment for the couple of linux needs, while keeping everyone still firmly within their system.
the problem (for them, but advantage for us) is that it might end up the other way around:
wsl is so much limited that eventually it will encourage some to go further and transition to the real deal.
while it is far from something that you can actually implant into a patient in need of organ transplant, current state of the art is already able to build "organoid":
small structered 3D cell cultures that more or less mimic on a tiny scale some of the structures found in real life natural organs.
These are already useful at this stage for research (mostly pharma as they help investigate better the effects of potential drugs on interacting functional tissues, rather than cells floating freely in a test tube).
i.e.: we're already halfway there, and this halfway is already useful (though it's more "a few percent down the right direction" rather than litteraly *half*way)
while the eyesight link is dubious
Video games directly causing eyesight degradation is dubious (as in you need to use some magic filter glasses to avoid your eyes dying slowly when looking at screens).
There is some corpus of evidence (including studies done in Japan that predate widespread use of smartphones) that somewhat link decreased time spent outside (outdoor activities in sunshine) with increasing need for prescription glasses (not simply explainable by increased reporting due to higher reporting).
And some phenomenon replicated in lab by sewing semi-shut eyelids of apes (still an aweful thing to do in my opninion).
It looks like there is something causing eyes to get myopia when they spend more time looking to close object under faint lights (indoor, evenings) rather than far away in the sunshine.
From that point of view, video games are simply yet another distraction from outdoor activities (as were books when the first such studies were started) whose reduction could have some play in degrading eyesight.
But I don't think outlawing games like totalitarian China is doing is a great solution to the problem.
And what is preventing you to "flag as spam" each time your mother-in-law or your ex calls you, just to annoy them ?
I'm assuming you can program it through a Wifi end-to-end kind of thing, not requiring a internet connection.
Yup.
At my brothers' first setup was done with a direct wifi peer-to-peer connection, then over the local network for subsequent timer settings.
Connection to iRobot's cloud is only used for keeping an archive of maps of finished jobs, all this weird "make a map of wifi coverage" useless features, permanent floor maps for TFA's "go clean the kitchen" jobs, etc.
The Wifi and app are only a hard requirement for:
- firmware update (it's not done with a serial dongle anymore) to weed out bugs (the first version of 900 had navigation problems)
- setting time and timers (there are buttons on the device utself anymore. Apparently "blinking twelve" is still a problem with people too stupid to press buttons).
Separate cloud connection is required for the creepier features
The first version of the mapping feature of the 900 series was a bit buggy and got lost.
It cleans okay (still able to reach almost everywhere using the random walk of classic roombas) but has no fucking clue where the home base is and how to reach it back (it just keeps running around aimlessly until battery runs out, whereas classic roomba run aimlessly around until within reach of base station and then IR to engage the "tractor beam" feature).
- Not all Roomba are even Wifi enabled to begin with.
Even Wifi enabled Roomba, work without wifi.
Even if you activate the Wifi access, you can also use only for firmware update and otherwise only talk to your smartphone directly over the local network (or optionnally to your home server running on some raspberry pi).
( ^at that point you already have 99% of features people want, including local tiled maps on navigation-cam featuring models like the 9x0 serie)
This thing aditionnally requires you to connect it to iRobot's cloud and link to your Google account.
( ^ this is required for the creepy features like using the App over internet (for 890 and more recent), storage of maps of past runs (for 9xx and more recent), and permanent map of the floor and voice commands (for i7/i9) )
It's also a tracking system, so the "you are the product" deal works too.
Also, GPS tracking is a requirement for free floating vehicle-sharing system (think things like li.me ), which are potential clients of e-bikes.
If you look at the top most contributing devs, you'll notice that they are actually on the pay roll of companies who rely on linux. If they weren't already employed by Red Hat, they would probably be at Intel, Google, even IBM themselves...
Not all programmers are poor. If they are anywhere near competent (and open-source software makes a great portfolio that is easy to show around), they'll certainly get hired, perhaps even get paid for their open-source hobby.
AMD chips are also great for frying eggs.
Which is a pretty easy trick to achieve given that egg protein already start to precipitate somewhere north of 50~60C - you could achieve the same with the warm water of your faucet, try it ! Note that the egg will not have been thoroughly cooked at a high enough temperature and will not be sterilized : it might not be safe to eat due to bacterial risks.
You could do the same trick as the video with any piece of electronic more beefy that a raspberry pi
And while digging at old stuff, Intel was at the recieving end of such jokes back in the Pentium4 vs Athlon&Opteron 64 era.
(Also, since when are AMD motherboards paired with intel NICs? That seems weird to me...)
Also notice the UI difference between modern Widows 10 and Office 365 now, and back when the current users where school children.
Even if by some chance Microsoft and Office are still relevant in the business world when these Australian kids grow up, it would almost definitely not look like anything now.
the problem is the fov (field of view).
you can't see the pixels of 4k monitor *several feets away*.
but you could definitely see them if the monitor is a few inches away from your nose and wrap around your whole head (well at least spread around everywhere your eyes can look at), which is what the current generation of vr is attempting to do (as opposed to older gen that only simulated a somewhat largish screen in front of view and nothing around)
but all this extra screen/fov estate cones at a cost (either resolution if you stretch a 2k/3k display, or performance if you try actually 8k)
hence all the buzz generated in the media by "foveated rendering" (i.e.: eye tracking and variable resolution) as the contender for the vr gen after the current.
(i.e: only render the 3k that the eyes are currently looking at/pointing at, and render a low res blurry approximation for the peripheral vision where it is good enough)
the current buzzword du jour - ray tracing - happens to excel at such varying resolution.
For the subcategory of software devs that run mainly on some other OS (Linux is also popular in the biomed research - mostly on workstations and servers/compute nodes. Windows is still king in some business settings and with gamedevs), but need to port and test code on Mac OS X (which is *also* popular in biomed research - mostly on on laptops, and some iMacs here and there).
The only legal way to run a licensed OS X (even virtual image) is to run it on Apple hardware (though the license doesn't require it to be the host OS).
Mac Minis are a cheap and simple way to have a legal way to test Mac OS X code, and use extra monitor inputs and/or console switch box and/or VNC to use the Mini alongside the regular workstation. (The expensive alternative way is to use an Apple workstation *as* your work horse)
So being more capable would certainly be appreciated (e.g.: could be easier to run multiple VirtualBoxes with the various versions of Mac OS X you target in tests).
But indeed, it's a very tiny subset of Apple's customers and that serie's intended user base.
Mac Mini were mostly targeted as a "gateway drug to the Apple world" for average PC users (keep all your USB- and HDMI/Displayport- peripherals and only plug a cheap Mini to quickly check if the grass is indeed greener on the other side of the (walled garden's) fence).
So being a cheap and light-weight machine is relevant for maybe 98% of its intended audience.
Most of which won't be interested by beefier specs.
The problem is that RFID has a much crappier range than the typical car keyfob.
In practice that would require the transmitter which goes near the victim to be crazy close (like in the same pocket where the victim put their hand), or to use a very conspicuous bazooka-style ginormous cantenna.
Are your pure-blood native criminals too stupid?
I suspect that all the inbreeding to which the racist native criminals have subjected to themselves to keep their "pure blood" status, might have played out negatively and could indeed explain why they're much stupider~~ :-P
Unless it features contacts on the outside of the skin, it is a transceiver.
Yes, RFID chips are transcievers. But, with a horrendously crappy range.
And no really a good collision-resolution alrgorithms (*)
So unless the attacker use extremely conspicuous bazooka-style ginormous cantenna, they WILL definitely have trouble tracking those Swedish thumbs.
---
(*) Though some RFID in some contactless cards I've seenchips do.
In one of the Uni I've worked at, if your hands are busy, you could unlock doors by bringing the pocket with your wallet (full of various RFID cards) to the general vicinity of the sensor. The card and sensor will manage to negociate despite all the others cards.
Yeah, I am sure that rockets - which fly only every now and then - are the most environment-impacting form of transportation.
Specially compared to cars, of which there are an extremely large quantity in circulation, which altogether total gazillions of kilometers of distance.
Talk aboutt too early optimisation.
but as written any item over a certain dollar amount is covered. Since that dollar amount is anything over $5, I'm pretty sure that it covers anything that Apple makes that has any kind of warranty.
Apple, being Apple, "anything over $5" is 100% guaranteed to always cover absolutely everything in their inventory down to every single last SKU.
Even if Apple started producing items as simple as "single use iAss-wipes", it's going to cost at least 50$. per unit.
So any images of exoplanets, black holes or any surface except the mentioned above are artists impressions.
Well, technically, due to how physics work, it will never be possible to make an image of a black hole. There's no such thing existing.
What you could take pictures of is the gavitational lensing (Einstein ring) caused by the presence of a black hole (well, that peculiar photo is a super-massive black hole AND all the surrounding galaxy at the core of which said hole sits, causing the photograph-able mirage together. But you got the idea).
EV use has nothing to do with climate change in either direction.
The amount of extra CO2 we humans release into the environment is considered to have a significant impact on the climate, according to current scientific consensus.
Usage(1) of EVs compared to ICE cars release less CO2 in the atmosphere.
It ranges between "a bit less" up to "almost none released", depending on you local electricity energy source mix, except in a few countries (like India, China and Australia - if my memory of that study is right - where things are more or less equal)
Even in those place, if you replace ICE cars with EVs, even if it doesn't change anything *right now*, it's eventually going to change things *instantaneously* once the country switches to greener electricity sources mix.
Whereas if a new less-emitting ICE is invented, it will only show changes much later, as older vehicles are replaced with newer.
Basically, an EV has the (indirect) emission of whatever happens to be the electricity source mix last night when it charged.
Whereas an ICE car has the (direct) emission due to the efficiency of whatever ICE was built into it several years ago (+ indirect emissions linked to keep the distribution network).
---
(1): over the whole lifetime of a vehicle, the *usage* is what has proportionally the highest impact on the environment (as opposed to the manufacturing which is quite minuscule in comparison).
why not just buy a cheap android 7 or 8 inch knock-off tablet for $50?
These tend to have old under-powered chipset from shitty companie like MediaTek.
Not only they are difficult to unlock and install LineageOS on them, but in today's era of "you must download a multi-megabyte katamari of innumerable javascript framework libraries just to be able to display a single webpage", their weak CPU isn't even good at browsing the web.
If you want a google powered refrigerator, why not buy a 10 inch android tablet and velcro it to a dumb fridge?
It's a good idea, but while you'd be avoiding a huge chunk of the privacy-molestation, you'd be still missing on the sensors that make the whole idea of a "smartfridge" useful (you know, the whole "I'm detecting that the product XyZ you bought N days ago is approaching its 'best-fore' date" , and other "You're running low on milk, would you like me to ask Alexa / Ok,Google / Cortana to order some ?").
So the correct bill of material would be a cheap-ish 10" android tablet, velcro and ~50$ worth of arduinos and sensors from Adafruit, Sparkfun and friends.
(Well, probably a bit more if you need an array of good quality cameras in regions where product code are still purely 100% barcodes (like in the EU) and not NFC-enabled (I've read it's coming in the US), but you got the idea )
What in the world for and which "firmware"?
In addition to MancunianMaskMan's excellent answer :
fwupd itself is a command line tool.
Gnome has a nice GTK UI to make firmware update user-friendly.
KDE has probably added a similar tool, based on QT and KDElibs
Seems rather odd considering most users will have no idea what either of those are. Bit of a backend way to push both those little policy agendas. It's the distribution's job to install such things not the desktop environment.
The *whole raison d'être* of Flatpak, Snaps, etc. is to make end-user installable software that is *independent of the distribution*, somewhat reminiscent of Windows' SETUP.EXE. (as opposed to 3rd party package repositories, like Opensuse's OBS, Ubuntu's PPA, Gentoo's overlays and whatever the hell they are called with Fedora/Redhat).
Thus it needs a different GUI than the one used for distro-package.
Hence KDE providing a GUI for Flatpacks.
(And as an added touch could ask the distro package management to ask missing dependencies to get Flat up. Probably over some standard scheme like PackageKit or something)
I won't probably be using either tools (being mostly CLI), but it's still a nice touch.
These days, computers are plenty fast for most tasks. {...} For most tasks by most users, five years is fine.
Sadly, the hardware vendors don't seem to agree with you.
It seems that if it's older than a couple of months, you're basically on your own.
(Except maybe for the "business" line of some manufacturer like Dell or Lenovo. But definitely the case for separate motherboard manufacturer as in TFS).
Heck, I'm running a laptop that was manufactured August 2012; 6 years ago. It performs just fine. (Disclaimer: I'm running Linux.)
That last part might be the reason why you and I can pull such a stunt, including the latest kernel. The same hardware will probably refuse to run anything more recent than Windows Vista.
(And let me guess, also using Nouveau due to Nvidia dropping support of the embed GPU in your laptop ?)