Yeah, well. If we want "free as in beer" energy, I can give you specifics that work, and do something awful like you describe.
Just set up the electromagnet on a planetary scale, and suck electricity out of the Earth's motion. Sure, it will eventually slow the planet and ruin our orbit, but I won't live that long. If it works on a Prius, it will work on Spaceship Earth.
Or if that sounds like too big an engineering project, we could just tap into oceanic convection at the bottlenecks and use the oceans for hydropower. At first we won't be pulling out enough to ruin any ocean cycles, so we can sound the "all clear" and scale it up, right? Just figure out where the best bottlenecks are, and build some giant sea walls with generator ports. No problem.
Or speaking of climate change, President Bush liked the space umbrella idea. As an electronics tinkerer I really like the "just add a control loop" concept, but the failure states might need a bit of serious analysis.
And if they discover something that might lead to a new electrical generator, it will take 30 years to get to market and be tightly controlled by industry. You've in no way convinced me that the hobbyists are wasting their time. They have completely different goals than the people at the LHC.
Calling them monkeys just makes them sound cute. In Asian art, the monkey is often a symbol of fighting for the moral good. Like the monkey and the tiger, they have different goals. The monkey doesn't need to be philosophically concerned that the tiger doesn't understand his needs.
Normally I hate video links, but in this case I'll give you a link to a video of a monkey and a tiger, discussing a related issue: https://www.youtube.com/watch?...
We don't even know if the Universe is a closed system, and the "Big Bang" is pure hypothesis since nobody has done an experiment to make sure that photons look the way we expect after 10 billion years of travel. Subtle effects we couldn't detect on Earth because of local noise might easily alter the implications of that sort of work. It is largely speculative, because it represents the edge-data of the sensors. Edge data from every type of sensor is low quality. If you really honestly apply the scientific method, then if the Big Bang is true we could never establish it, because we'd never be able to do the needed experiments to verify the sensor calibration. That proponents of the hypothesis insist it is "proven" is a giant red flag for mindless consumption of received knowledge.
"In a closed system entropy always goes up." Great, now prove that any system that exists or ever existed is closed. It can't be done, it would require proving a negative. Even to the extent that we understand the laws of thermodynamics, we can't assess what they do or do not exclude. We can only assess the parts of in the middle that overlap our engineering capabilities.
"If I keep tossing this coin, eventually it will disappear into thin air and magically turn into a unicorn".
Scientifically speaking, there's virtually no difference, in fact.
You just assert that, without establishing exhaustively that that is the only explanation. It applies to the analysis you provide just as well as it applies to anything. You don't know what you don't know; you can't prove negatives. If you don't think somebody's hypothesis will be proven correct, that doesn't mean you've proven it to be impossible.
What is the probability that a coin, tossed in the air, will be followed by you not seeing the coin, but seeing a unicorn where you expected to see a coin? Is it zero, or just low? Or is it totally unknown, and we'd need to know where you're standing, what you're doing, and what else is happening behind the scenes?
If somebody is trying to do something you think is impossible... which is less likely, that you know the limits of possibility, or that a very unlikely thing will happen? Double-check your math on proving the impossible.;)
It may be that they're not intending to "break the laws" of physics at all, but discover/uncover new ones. It may be that "overunity" sucks energy out of some sort of sub-space field
So... what, we should watch every crank and snake oil salesman to see if they've uncovered any new physical laws?
No, just stop calling people names so quickly, and consider that you might simply not understand them or their goals. Or in some cases they have health issues and calling them names because of that doesn't make you science-y.
Naw, Newton for example didn't use the "scientific method." Back then "peer review" meant it was like youtube; there was no gatekeeper, the Natural Philosophers (there were not yet scientists or a scientific method) like Newton would read the things other people actually said, and they would be weighed by responding with their own comments.
If you want science to be based on the scientific method, you'll have to throw away the whole framework we have now.
And in many ways, what Newton was doing was better than the current system. They were successfully reducing the impact of the publishers on the accepted views, something considered a problem today that degrades the quality of science. The benefit of the scientific method is that is scaleable, and even the average students can participate successfully. I'm not convinced that speeds up the rate of discovery, but it clearly speeds up the rate of conversion of discovery to engineering.
If you get all religious about science like that, you wind up sweeping Newton into your "pseudoscience" bin. Is that really where you're trying to sweep these guys on youtube, into the Newton bin?
I suspect they will stumble upon a method of extracting a minute amount of energy from the environment, and due to their lack of scientific understanding they will attribute it to perpetual motion rather than simple energy balance accounting.
IMO it is equally likely that they will in fact have good ideas about where the energy comes from, but media pundits will invent a fake claim of perpetual motion to slander them. I mean, that is where things already are, they'd only have to keep saying the same things.
There was actually a propulsion device in that category tested by NASA recently. The inventor is probably wrong about the mechanism, but a valid hypothesis led to the (working) device. Then the media got started with it, and if you ask a random neckbeard they can explain to you that it was just snake oil "because perpetual motion."
I just finished actually reading the article, (I thought it would be acceptable in this case since it is hackaday) and I realized more clearly the mistaken accusation they're hurtling.
The energy is presumed to be harvested from the physical magnets in most cases. You can indeed harvest energy in that way, weakening the magnet as you extract energy. If this is economically useful or not depends on very specific non-physics-violating math involving not only the devices themselves but also various supply chains. It may be that heat is plentiful enough, and conversion of heat to electricity inefficient enough, that there is a way to extract more electricity from magnets than would be recovered from the heat needed to align that much metallic crystal. That is without even just mining pre-aligned iron.
It is unlikely to be a money-maker other than for the oil peddlers, but there are not any fundamental problems with the goal.
It may be that they're not intending to "break the laws" of physics at all, but discover/uncover new ones. It may be that "overunity" sucks energy out of some sort of sub-space field (intentionally borrowing from sci-fi, calm down) that we haven't yet discovered. It seems that the pundits are the source of most of the perpetual-motion misconceptions, rather than the tinkerers themselves.
They aren't trying to create "free energy" in the physics domain. They're trying to create "free energy" in the economics domain; if we can suck energy out of dimension X, then until we're bombed by the inhabitants of that realm it will appear as (economically) "free."
They're talking about "free as in beer" not "free as in freedom."
Right, but the "as a service" case is mostly a fail. Anybody old enough to have used thin clients can understand this.
Once the service you're using is unusable for a couple days because some script kiddie decided to DDoS it, then you'll find out you need your own cloud to work in the cloud.
They want to lock you in, but they can't actually keep you there. Cloud services will be around "forever," but not every use case will outlast the buzzword phase. For companies that paint themselves into a corner with this use case, they'll find out that they can keep most of their new tooling and just use their own cloud. That is where this road appears to lead. The providers of these services are probably not intending to be building a longterm business on it; they're selling shovels for the gold rush. They already know they'll be moving on in a few years.
Right, but are the previews you're talking about for changes in one file?
With an IDE, you have single files "open" in tabs or windows, if you're doing a global replace you can choose to replace everything with no preview, or it has to step through each one and open the file, show you the preview, and you click yes/no.
With sed, I dump all the proposed changes to a scrolling buffer, and can review them all at once, one per line. Then I re-run the command to make the changes for real.
If you have thousands of changes of a string across dozens of files, the old way gets way out ahead.
I think is rather funny your description of being efficient and saving time by having somebody spew out audio advice about menu options. I mean, come on. That is laughable for many reasons. And your description of a regex in vim is specious; the s and the slashes are typed very quickly, and the rest of the regex is the same meat of the replacement regardless of which tool you're using. You just spell out the sounds to make it sound longer than it is, but only in one case, to make that case look worse.
And I challenge this myth that being efficient with emacs is "like rock-climbing," even without knowing WTF that comparison is supposed to mean. (I hike near a rock wall and the experts can climb it as fast I can hike the stairs next to it.) If the hard parts of being productive are about your tools, you're probably using them poorly. The hard parts are entirely in deciding which code to write, which algorithms to use, how many layers of interfaces, where are the interface boundaries, etc.
If you memorize chords or write a macro or use a shell alias, those are mostly going to provide the same amount of efficiency, except that every time you switch between mouse and keyboard, or give/receive verbal commands, your brain goes into a pause-mode where there are whole extra seconds that passed that you didn't detect because you were switching contexts.
Many professionals consider it an important part of their career to learn and become proficient in various tools. It isn't "stubborn" to want to use something out of that set of tools, not only because it is what you've already invested time in, but because your proficiency will also be advancing as you continue to work with those tools. Who is being stubborn, the person who wants to use "one of the common professional tools that can do the job well," or the person who wants the whole team to only use their chosen tool?
I don't think you really considered the "stubborn" accusation before hurtling it.
And thumbs down to Chromebook pixel. There is absolutely no reason to attach a proprietary blahblah to the development process unless it is the platform the application is written for. Even moreso when it is not even a standard blahblah that professionals use for the task.
That kinda sounds like a decent analysis, if you don't know what encryption is. If they can give out somebody else's data, it isn't actually encrypted; it is merely obfuscated.
No need for holy wars, as an emacs user if I'm stuck editing over ssh I switch to vim.
Of course, if I'm using a LAMP stack (or similar) then I wouldn't be developing over ssh, I'd be doing that locally and using source control and a modern deploy tool, which would use ssh under the hood. If I'm using an editor over ssh, it would imply either an emergency hotfix (eek) or a sysadmin task, and in both cases mode-based editing shines.
Gosh, to make those whippersnappers stop making up fake "syndromes" to name-call him with, waaaaaaah! Oh, wait, grandpa does not care what you think he has already lived a little and knows better.
People were already laughing at me for using emacs in the late 90s, and none of them ever had a point other than "haha, kitchen sink." One secret of software, unused capabilities that don't have buttons on the screen can be safely ignored forever. It isn't like an oversized hand tool that weighs more because it has more functions attached. I'm still using emacs, and I only know 5 or 6 "chords." I have to make minor changes to a lisp config file every 5 years or so. Not really understanding what the complaint is. If it was a 30 year old ham sandwich, I'd understand the complaint.
People have laughed at me for using sed to global replace something in a codebase, and then when I'm done and ask them how long it would take in their IDE they tell me, oh it would take about the same amount of time. One difference though, that I don't think they notice, is that my way I can do a dry run that lists all the changes to be made before actually doing it. They probably can too, but none of them did when I asked to show me their way. Presumably they thought that was some sort of staging thing, and they missed that I was reviewing the intended changes beforehand. I'd use a newer tool, but Perl isn't usually an improvement for that use case.
What cracks me up are the supposed benefits. Perfect for people who already got all those benefits from OOP, then from XP, then from Agile, etc. The listed benefits are exactly the things it actually sucks at.
"Syncing is easy... because you're not allowed to do it, you just twiddle your thumbs until the network traffic resumes. Easy-peasy, you're welcome." Uhm, no. I'm old enough to have lived that. Thin clients were good, because thick clients were out of our price range, and allowed us to share a larger system that could do the stuff we needed. A cheap tablet computer has enough processing power for software development now. The use case just isn't there.
These days, if a person needs remote resources for dev my first thought is, "maybe it is time to finally learn autoconf so you can compile changes in 3 seconds instead of the 3 days your pasted Makefile is causing."
My second thought is, "do they know that managing a private cloud configuration is harder than managing a workstation, and generally requires sysadmin skills?"
Teamwork is easy, just choose a low (resume) value toolset nobody has experience with, and mandate it. See how easy teamwork is? Everybody just does everything the same way because we mandated it. And then since they already don't have choices, we stuck them with ChromeOS too. To make sure they don't get any ideas.
random stranger? No, I'd expect them to be well-financed, entrenched, and have contracts to legally share my work with all their big-name third party partners.
Newsflash: Wayland already missed years and years of deadlines to take over the world.;)
And it is an optional replacement. Many of us who use existing features will not switch to software that leaves them out. Kids can have all the new toys they want, but they won't be able to take away what us old fogies already have. Such is the nature of Software Freedom.
It works well entirely because R is not an application language; it is used for snippets, mostly one-off. You don't need all that normal dev process stuff.
The downsides of the "cloud" of course include syncing hassles across machines, installation of toolchains to mitigate problems of offline access, difficulty of teamwork with new team members (even an inability to work with many intending team members, because they won't use those tools and can't choose a professional tool), difficulty of deployment when it turns out that you still need somebody to where the sysadmin hat and now the backend is more complicated.
Moving to ChromeOS is a separate problem, not sure what that solves exactly. For people under severe resource pressure who can't afford a basic dev laptop, and who can't find a used 10-year-old desktop to use for it, are often unable to get internet access that is high enough quality and provides enough bandwidth to be running cloud apps.
Slashdot Mirror
Yeah, well. If we want "free as in beer" energy, I can give you specifics that work, and do something awful like you describe.
Just set up the electromagnet on a planetary scale, and suck electricity out of the Earth's motion. Sure, it will eventually slow the planet and ruin our orbit, but I won't live that long. If it works on a Prius, it will work on Spaceship Earth.
Or if that sounds like too big an engineering project, we could just tap into oceanic convection at the bottlenecks and use the oceans for hydropower. At first we won't be pulling out enough to ruin any ocean cycles, so we can sound the "all clear" and scale it up, right? Just figure out where the best bottlenecks are, and build some giant sea walls with generator ports. No problem.
Or speaking of climate change, President Bush liked the space umbrella idea. As an electronics tinkerer I really like the "just add a control loop" concept, but the failure states might need a bit of serious analysis.
And if they discover something that might lead to a new electrical generator, it will take 30 years to get to market and be tightly controlled by industry. You've in no way convinced me that the hobbyists are wasting their time. They have completely different goals than the people at the LHC.
Calling them monkeys just makes them sound cute. In Asian art, the monkey is often a symbol of fighting for the moral good. Like the monkey and the tiger, they have different goals. The monkey doesn't need to be philosophically concerned that the tiger doesn't understand his needs.
Normally I hate video links, but in this case I'll give you a link to a video of a monkey and a tiger, discussing a related issue: https://www.youtube.com/watch?...
We don't even know if the Universe is a closed system, and the "Big Bang" is pure hypothesis since nobody has done an experiment to make sure that photons look the way we expect after 10 billion years of travel. Subtle effects we couldn't detect on Earth because of local noise might easily alter the implications of that sort of work. It is largely speculative, because it represents the edge-data of the sensors. Edge data from every type of sensor is low quality. If you really honestly apply the scientific method, then if the Big Bang is true we could never establish it, because we'd never be able to do the needed experiments to verify the sensor calibration. That proponents of the hypothesis insist it is "proven" is a giant red flag for mindless consumption of received knowledge.
"In a closed system entropy always goes up." Great, now prove that any system that exists or ever existed is closed. It can't be done, it would require proving a negative. Even to the extent that we understand the laws of thermodynamics, we can't assess what they do or do not exclude. We can only assess the parts of in the middle that overlap our engineering capabilities.
LOL if you remember the name, let me know. It sounds very familiar, but I can't place it.
This is along the lines of:
"If I keep tossing this coin, eventually it will disappear into thin air and magically turn into a unicorn".
Scientifically speaking, there's virtually no difference, in fact.
You just assert that, without establishing exhaustively that that is the only explanation. It applies to the analysis you provide just as well as it applies to anything. You don't know what you don't know; you can't prove negatives. If you don't think somebody's hypothesis will be proven correct, that doesn't mean you've proven it to be impossible.
What is the probability that a coin, tossed in the air, will be followed by you not seeing the coin, but seeing a unicorn where you expected to see a coin? Is it zero, or just low? Or is it totally unknown, and we'd need to know where you're standing, what you're doing, and what else is happening behind the scenes?
If somebody is trying to do something you think is impossible... which is less likely, that you know the limits of possibility, or that a very unlikely thing will happen? Double-check your math on proving the impossible. ;)
So ... what, we should watch every crank and snake oil salesman to see if they've uncovered any new physical laws?
No, just stop calling people names so quickly, and consider that you might simply not understand them or their goals. Or in some cases they have health issues and calling them names because of that doesn't make you science-y.
Naw, Newton for example didn't use the "scientific method." Back then "peer review" meant it was like youtube; there was no gatekeeper, the Natural Philosophers (there were not yet scientists or a scientific method) like Newton would read the things other people actually said, and they would be weighed by responding with their own comments.
If you want science to be based on the scientific method, you'll have to throw away the whole framework we have now.
And in many ways, what Newton was doing was better than the current system. They were successfully reducing the impact of the publishers on the accepted views, something considered a problem today that degrades the quality of science. The benefit of the scientific method is that is scaleable, and even the average students can participate successfully. I'm not convinced that speeds up the rate of discovery, but it clearly speeds up the rate of conversion of discovery to engineering.
If you get all religious about science like that, you wind up sweeping Newton into your "pseudoscience" bin. Is that really where you're trying to sweep these guys on youtube, into the Newton bin?
I suspect they will stumble upon a method of extracting a minute amount of energy from the environment, and due to their lack of scientific understanding they will attribute it to perpetual motion rather than simple energy balance accounting.
IMO it is equally likely that they will in fact have good ideas about where the energy comes from, but media pundits will invent a fake claim of perpetual motion to slander them. I mean, that is where things already are, they'd only have to keep saying the same things.
There was actually a propulsion device in that category tested by NASA recently. The inventor is probably wrong about the mechanism, but a valid hypothesis led to the (working) device. Then the media got started with it, and if you ask a random neckbeard they can explain to you that it was just snake oil "because perpetual motion."
I just finished actually reading the article, (I thought it would be acceptable in this case since it is hackaday) and I realized more clearly the mistaken accusation they're hurtling.
The energy is presumed to be harvested from the physical magnets in most cases. You can indeed harvest energy in that way, weakening the magnet as you extract energy. If this is economically useful or not depends on very specific non-physics-violating math involving not only the devices themselves but also various supply chains. It may be that heat is plentiful enough, and conversion of heat to electricity inefficient enough, that there is a way to extract more electricity from magnets than would be recovered from the heat needed to align that much metallic crystal. That is without even just mining pre-aligned iron.
It is unlikely to be a money-maker other than for the oil peddlers, but there are not any fundamental problems with the goal.
look at one free energy video and they will top the suggested video for you for months.
Just watch more soft p0rn and kittens, those are more popular than snake oil and so can displace it.
It may be that they're not intending to "break the laws" of physics at all, but discover/uncover new ones. It may be that "overunity" sucks energy out of some sort of sub-space field (intentionally borrowing from sci-fi, calm down) that we haven't yet discovered. It seems that the pundits are the source of most of the perpetual-motion misconceptions, rather than the tinkerers themselves.
They aren't trying to create "free energy" in the physics domain. They're trying to create "free energy" in the economics domain; if we can suck energy out of dimension X, then until we're bombed by the inhabitants of that realm it will appear as (economically) "free."
They're talking about "free as in beer" not "free as in freedom."
Right, but the "as a service" case is mostly a fail. Anybody old enough to have used thin clients can understand this.
Once the service you're using is unusable for a couple days because some script kiddie decided to DDoS it, then you'll find out you need your own cloud to work in the cloud.
They want to lock you in, but they can't actually keep you there. Cloud services will be around "forever," but not every use case will outlast the buzzword phase. For companies that paint themselves into a corner with this use case, they'll find out that they can keep most of their new tooling and just use their own cloud. That is where this road appears to lead. The providers of these services are probably not intending to be building a longterm business on it; they're selling shovels for the gold rush. They already know they'll be moving on in a few years.
Right, but are the previews you're talking about for changes in one file?
With an IDE, you have single files "open" in tabs or windows, if you're doing a global replace you can choose to replace everything with no preview, or it has to step through each one and open the file, show you the preview, and you click yes/no.
With sed, I dump all the proposed changes to a scrolling buffer, and can review them all at once, one per line. Then I re-run the command to make the changes for real.
If you have thousands of changes of a string across dozens of files, the old way gets way out ahead.
I think is rather funny your description of being efficient and saving time by having somebody spew out audio advice about menu options. I mean, come on. That is laughable for many reasons. And your description of a regex in vim is specious; the s and the slashes are typed very quickly, and the rest of the regex is the same meat of the replacement regardless of which tool you're using. You just spell out the sounds to make it sound longer than it is, but only in one case, to make that case look worse.
And I challenge this myth that being efficient with emacs is "like rock-climbing," even without knowing WTF that comparison is supposed to mean. (I hike near a rock wall and the experts can climb it as fast I can hike the stairs next to it.) If the hard parts of being productive are about your tools, you're probably using them poorly. The hard parts are entirely in deciding which code to write, which algorithms to use, how many layers of interfaces, where are the interface boundaries, etc.
If you memorize chords or write a macro or use a shell alias, those are mostly going to provide the same amount of efficiency, except that every time you switch between mouse and keyboard, or give/receive verbal commands, your brain goes into a pause-mode where there are whole extra seconds that passed that you didn't detect because you were switching contexts.
Many professionals consider it an important part of their career to learn and become proficient in various tools. It isn't "stubborn" to want to use something out of that set of tools, not only because it is what you've already invested time in, but because your proficiency will also be advancing as you continue to work with those tools. Who is being stubborn, the person who wants to use "one of the common professional tools that can do the job well," or the person who wants the whole team to only use their chosen tool?
I don't think you really considered the "stubborn" accusation before hurtling it.
And thumbs down to Chromebook pixel. There is absolutely no reason to attach a proprietary blahblah to the development process unless it is the platform the application is written for. Even moreso when it is not even a standard blahblah that professionals use for the task.
Talk about stubborn, wow.
That kinda sounds like a decent analysis, if you don't know what encryption is. If they can give out somebody else's data, it isn't actually encrypted; it is merely obfuscated.
I thought it was called "the nineties."
No need for holy wars, as an emacs user if I'm stuck editing over ssh I switch to vim.
Of course, if I'm using a LAMP stack (or similar) then I wouldn't be developing over ssh, I'd be doing that locally and using source control and a modern deploy tool, which would use ssh under the hood. If I'm using an editor over ssh, it would imply either an emergency hotfix (eek) or a sysadmin task, and in both cases mode-based editing shines.
He means, "make all the stuff flow over the wires on its own because I don't know how to set up git hooks, ssh, or a VPN"
Gosh, to make those whippersnappers stop making up fake "syndromes" to name-call him with, waaaaaaah! Oh, wait, grandpa does not care what you think he has already lived a little and knows better.
People were already laughing at me for using emacs in the late 90s, and none of them ever had a point other than "haha, kitchen sink." One secret of software, unused capabilities that don't have buttons on the screen can be safely ignored forever. It isn't like an oversized hand tool that weighs more because it has more functions attached. I'm still using emacs, and I only know 5 or 6 "chords." I have to make minor changes to a lisp config file every 5 years or so. Not really understanding what the complaint is. If it was a 30 year old ham sandwich, I'd understand the complaint.
People have laughed at me for using sed to global replace something in a codebase, and then when I'm done and ask them how long it would take in their IDE they tell me, oh it would take about the same amount of time. One difference though, that I don't think they notice, is that my way I can do a dry run that lists all the changes to be made before actually doing it. They probably can too, but none of them did when I asked to show me their way. Presumably they thought that was some sort of staging thing, and they missed that I was reviewing the intended changes beforehand. I'd use a newer tool, but Perl isn't usually an improvement for that use case.
What cracks me up are the supposed benefits. Perfect for people who already got all those benefits from OOP, then from XP, then from Agile, etc. The listed benefits are exactly the things it actually sucks at.
"Syncing is easy... because you're not allowed to do it, you just twiddle your thumbs until the network traffic resumes. Easy-peasy, you're welcome." Uhm, no. I'm old enough to have lived that. Thin clients were good, because thick clients were out of our price range, and allowed us to share a larger system that could do the stuff we needed. A cheap tablet computer has enough processing power for software development now. The use case just isn't there.
These days, if a person needs remote resources for dev my first thought is, "maybe it is time to finally learn autoconf so you can compile changes in 3 seconds instead of the 3 days your pasted Makefile is causing."
My second thought is, "do they know that managing a private cloud configuration is harder than managing a workstation, and generally requires sysadmin skills?"
Teamwork is easy, just choose a low (resume) value toolset nobody has experience with, and mandate it. See how easy teamwork is? Everybody just does everything the same way because we mandated it. And then since they already don't have choices, we stuck them with ChromeOS too. To make sure they don't get any ideas.
Sorry, Cowherd, you gotta log in to use the "grandpa" line around here. Get to back of the line, pokey.
random stranger? No, I'd expect them to be well-financed, entrenched, and have contracts to legally share my work with all their big-name third party partners.
Newsflash: Wayland already missed years and years of deadlines to take over the world. ;)
And it is an optional replacement. Many of us who use existing features will not switch to software that leaves them out. Kids can have all the new toys they want, but they won't be able to take away what us old fogies already have. Such is the nature of Software Freedom.
It works well entirely because R is not an application language; it is used for snippets, mostly one-off. You don't need all that normal dev process stuff.
The downsides of the "cloud" of course include syncing hassles across machines, installation of toolchains to mitigate problems of offline access, difficulty of teamwork with new team members (even an inability to work with many intending team members, because they won't use those tools and can't choose a professional tool), difficulty of deployment when it turns out that you still need somebody to where the sysadmin hat and now the backend is more complicated.
Moving to ChromeOS is a separate problem, not sure what that solves exactly. For people under severe resource pressure who can't afford a basic dev laptop, and who can't find a used 10-year-old desktop to use for it, are often unable to get internet access that is high enough quality and provides enough bandwidth to be running cloud apps.
Don't like it? -na, -na [waving banana]
The good news is, it is easy to switch to something else after the banana joke starts to grow, or flies away.