Probably a super-hardened (security-wise) OS for an iPod Touch-like "second device" intended for use primarily to arrange illicit assignations.
Very few apps available, all based on zero-knowledge protocols, and no apps that are capable of leaking your real-world identity through ordinary or casual use are allowed.
As a result, just about everyone on Capitol Hill has one of these, so they can't be banned even though they are, as a side-effect, extremely useful to whistleblowers, journalists, activists, dissidents, etc.
I know thats a joke but oh my god would that be amazing.
A computer help call in radio show as funny and warmhearted as Car Talk? Comp Talk? With crazy weird issues? Like email doesn't work over 400 miles? lol.
Esp. if the dudes seemed to know everything and had lots of old-timer computer stories from their old CP/M desktops and whathaveyou.
I'm bored with my lack of decent radio and podcasts. Feel free to recommend me stuff, readers.
I recommend Welcome to Nightvale (Which is amazing, and you should be listening to it: http://www.welcometonightvale.com/), The Technical Difficulties (A trivia show starring Tom Scott (of emoji keyboard fame) and friends http://www.techdif.co.uk/, Radiolab (it's fricking Radiolab: what else do you want?), 99% invisible (loved Radiolab? you'll love this, too), Ask Me Another (trivia, and music by Jonathan Coulton: any questions?), and finally the recently ended YouTube series Let's Drown Out (Ben "Yahtzee" Croshaw and Gabriel Morton play videogames of varying quality, and snark over them).
Every now and then, Reply All https://gimletmedia.com/show/reply-all/ has a segment that is Tech Support. Some of the items on there are down right bizarre but very plausible. Great listening.
I think that source is a little bit confused on the distinction between iOS and OS X (macOS?).
iOS and OS X share the same kernel, named XNU (it's the userlands of the two that are almost completely different). That's been the case since iOS 1. Whenever either OS X or iOS gets ported to a new CPU (or CPU subfamily), the kernel gets another set of macros added to it for the purpose of identifying the new CPU. The presence of a new ARM macro in XNU doesn't mean much at all - it already has a ton of them, as well as truly ancient macros from long-forgotten systems (m68/88k comes to mind). Those macros have nothing to do with OS X or iOS individually - they're relevant primarily to the kernel itself.
In and of itself that doesn't prove anything. That file has had symbols for multiple architectures since forever, including CPUs like M68K, Sparc (years ago, I had hoped that Apple would buy Sun instead of Oracle - oh well) and VAX (but not Alpha).
Every now and then Apple release a new product, like TVOS for the Apple TV and people ask why Apple wrote an entirely new OS. why not base it on iOS or OSX. of course that's exactly what they do. There's a common core to OSX, iOS and now TVOS that are almost entirely the same code base. Which specific files are common or different for each flavour probably changes from time to time and this may well be a case of that.
The fact is though, iOS and OSX are already as much the same code base as they can be, and as much different as they need to be. That balance may change as the OSes evolve, but I don't think there's any pressing need or benefit to converging them completely.
The necessary divergence you talk about is mainly the UX coming from different input methods for the pointer device, correct?
With the (partly regretable) adoption of UI design patterns that are mainly meant for touch devices, such a merge appears to be closer however. I think the main patterns missing are
* a more powerful touch-capable tiling window manager
* a method to get "mouse hover" events working on touch devices. finger hover?
* a both touch- and precision pointer friendly implementation of the OSX menu system.
I'm not convinced that employing 3D Touch to do the typical things of mouse hover is a good idea. Displaying help text to users being lost when they do a very specific gesture? It needs to be something that every user can immediately pick up on.
Maybe a stare-o-recognizer using the camera? If the user looks puzzled at a button, display text to explain ;-).
I wonder if the HURD guys will finish and when. Can somebody comment on the continued existence of this project even though the ideas propogated by it(microkernels) have been well adopted into the mainstream?
Trying to emulate Mach, which was a dud as a microkernel, didn't help.
QNX is one of the few microkernels to get it right. L4 got stripped down so far that it's just a hypervisor, on which people usually load Linux. L4 took out arbitrary-length message passing in favor of interrupt-like events and shared memory between sender and receiver. This simplifies the kernel, but now it's easier for one side of a sender/receiver to mess up the other, since they share a communications area. The QNX primitive set (MsgSend, MsgReceive, and MsgReply) work well enough in practice to allow full POSIX functionality. Applications can talk to file system servers, network servers, etc. through those primitives. All QNX I/O works that way. You take maybe a 20% performance hit for the extra copying, but you get robustness in exchange.
Most important thing for performance in a microkernel: the CPU dispatcher and the message passing have to be tightly coordinated. You must be able to call another process and get a reply back without trips through the scheduler or a switch to a different CPU. QNX gets this right, because MsgSend is blocking. The sender blocks and the receiver starts without having to schedule. The data being sent is right there in the cache of the CPU, ready for use by the receiver. Good test for a microkernel - put some CPU-intensive jobs in a loop, while also running something that makes short request/reply calls to another process. If the request/reply process stalls out, the microkernel is doing it wrong. If the CPU-bound processes stall out, the microkernel is doing it wrong. Message passing should schedule as smoothly as a subroutine call. If it doesn't, performance under load will suck.
Back in the mid-90s I had some friends who worked at a company that sold commercial X-Window servers. Their fastest X server ran under QNX. That either says something about how fast QNX was, or just how bad Unix was at the time.
QNX was cautiously courting the open source concept and venturing in the direction of shared source (with some code already available), when BlackBerry bought them and threw all of that out the window.
Biggest yanked opportunity. D:
Now QNX is all-commercial again, with source only available under a license. And Photon is gone, too: no more self-hosting.
I wish BB would(/could?) do the webOS thing with QNX. That would be amazing. The community would absolutely get Photon up and running again, it would be an alternative to Linux and BSD.
I know. QNX the OS was great. QNX the company has been a huge pain to deal with for more than a decade. They had a tough problem, though. Who else sold a desktop OS for real money any more? Unlike Microsoft and Apple, they had no other revenue stream. Still, until about 2005, QNX was considered a target which UNIX-like software should support. There was an early version of Firefox (called Firebird) for it, all the command-line GNU tools were available, and the Eclipse IDE worked fine. For three years, while I was working on a DARPA Grand Challenge vehicle, I ran QNX on the desktop.
Photon, the GUI, probably had the sanest internals of any major GUI system.
Dan Dodge, who designed much of QNX, was CEO of QNX, and he retired from QNX/Blackberry to go work for Apple. Apple might end up doing something in this direction. But if they do, they'll never let the internals out.
I'm curious what was wrong with QNX the company, in terms of the practical ramifications of what you're describing.
I assume you were using QNX for the GC vehicle compo because of its RTOS qualities? In any case, that's awesome.
I thought Photon was pretty cool too, at least from a "oh hey these screenshots look awesome" perspective. I vaguely recall running it in a VM, once, so long ago I only hazily remember poking around the sidebar. Sadly it'll be somewhat challenging to do further academic research on that subject now, I wish the educational/noncommercial free options still existed there.
If anyone's looking for a student friendly version on microkernels and specifically Send/Receive/Reply, take a look at the resources for UWaterloo's CS452 here. [1]
That course features implementation of microkernels based on QNX from scratch to drive trains around a track, should be straightforward to implement on a Raspberri Pi or a Pandaboard/Beagleboard. The course uses a TS7200 which is an older ARM chip.
If it had a date I would have said "OS on xkcd from a year ago - ah, seen that". If it doesn't have a date it must mean it's new and I _have to_ see it.
Yes, it sometimes help to see that it's something I've seen before, and many sites don't make the date very visible, so it's easy to miss that something presented as state of the art or news in the article isn't anymore. Having the year in the title helps to avoid that error.
One is when it's likely that I've seen the page before, but not so recently that I recognize it immediately. "XKCD #386 [2008]" (OK, I know that one by the number.)
The other is when the page has a particular viewpoint which is either explained by the date or unusually modern for that time period. "Tiny computers will be ubiquitous [1965]"
If human civilization has been ended with fire, then how does it come back with in ten years and with GNU/HURD to be specific. I don't get what's the humor here.
Is it possible that AI has taken control over free OS from all over the web and killed all proprietary stuff along with humans and now run's the earth without any human civilization?
"This infamously and perennially late GNU/Hurd OS will finally make it in to Randall's home after human civilization has been wiped out. The joke is that GNU/Hurd began to be developed in 1990, and while it was expected to be released in a relatively short time, even now only unstable builds have been released. So Randall is saying that he will finally run it in his house a decade or two after the end of civilization. GNU/Hurd will presumably have an advantage as humanity rebuilds civilization due to the widespread availability of its code and development tools, and perhaps also because of Stallman's depth of belief, based on the title text. Alternatively, GNU/Hurd might be finished by the same force that finished humankind, for instance Skynet, in case of AI Apocalypse. (Interestingly, although still far from completion, a new version of GNU/Hurd was released less than a week after this comic.)"
[...]
"The GNU/Hurd reference might also be a pun, as in a "herd" of Gnus "running" in his living room, as wild animals reclaim the Earth after the end of human civilization."
> "The GNU/Hurd reference might also be a pun, as in a "herd" of Gnus "running" in his living room, as wild animals reclaim the Earth after the end of human civilization."
If we fill our home with IoT devices which carries Lithium ion batteries and relies on AI to function, I am sure fire will happen. Look at Samsung phones, even after recall there are cases of explosions.
So much automation in basic things is be like shooting in foot. Driver-less cars, really? Since when people started considering mental retardation as a status symbol.
I wonder if self-driving is going to be illegal in near future because we are incompetent to keep up with technological advances happening in our lifetime.
One day, masses will learning this hard way and come back to GNU.
alt-text: One of the survivors, poking around in the ruins with the point of a spear, uncovers a singed photo of Richard Stallman. They stare in silence. "This," one of them finally says, "This is a man who BELIEVED in something."
Just yesterday I was explaining to a friend that Stallman is not crazy; rather there is a continuum from convenience to freedom and Stallman believes most of us are way too far on the "left", so he goes as far to the "right" as possible to show us we've gone to far. He pays for his freedoms with his conveniences and we pay for our conveniences with our freedoms.
Things are much more nuanced and multidimensional than this. Not trying to start a flame war here.
You might not want to start a flame war, but you know one is perhaps inevitable on this topic.
I've met RMS several times (he even stayed at my place once, and yes the rider asking hosts not to buy him a parrot is real and long-standing), and I came it at from a different angle: I was a contributor to a BSD OS some yonks ago.
As a result we had a good-natured conversation about some of the conflicts between GPL and BSD and he's actually a lot more forgiving than people portray him or that argument. He's grounded in more reality than people give him credit for. Yes, he supports and advocates GPL, but he'd rather see code released as BSD/MIT than not released at all - at least that was the impression I was left with.
There are two things that tend to get the anti-GPL crowds back up, which on your continuum I'm not sure fit easily.
Firstly, there is the right for a programmer to be paid passively for the result of work done by their software.
RMS broadly (and I'm paraphrasing ridiculously here) thinks that you should be paid for the hours you work, the software should be Free, you don't earn a right to keep it closed and sell access to it through licenses, etc.
Fine, but that would pretty much shut down the programming economy. People regularly write code because they or their employer believes a customer is going to pay for it, and use that as leverage. Obviously. If we were to outlaw such a model - and it would require statues to make a reality, laws that run contrary to free-market liberalism - a large number of developers would be laid off.
That's OK, says RMS, because society will be better off. However, Turkeys do not vote for Thanksgiving/Christmas, and I want the _freedom_ to chose to work in businesses with economic models like that.
I think interestingly if Amazon, Facebook, Google, Uber and all similar firms (e-comms, social networks, search engines, etc.), released all their code, the market would not move much. The value was first created by innovating through their code base, but now the value is in the systems, brands, and relationships those firms have been able to build. Curiously, it's possible we are now in a scenario where closed source code can help a startup scale, but once they're at scale, there is no value in keeping the code closed, and making it open could benefit the companies (through "volunteer commits"), as well as the wider economy. We are seeing the tip of this in the dev code space (.NET et al).
Secondly, there is a weird line after which RMS does not care about software freedom quite so much. For example, in one talk he gave I attended, he stated that because he does not wish to modify the code in his microwave, he is not bothered if that code is free or not. I think that's a weak (and arbitrary) line, that is inconsistent with the core argument and really doesn't sit on the continuum.
Either all code wants to be free, or only some code does. And if only some code does, well, you can't go around telling people what does and does not.
Perhaps he's revised his position since then (a decade ago now), but I know it was something that troubled me about his argument.
And when somebody's economic and political philosophy does not quite stand up to not-very-rigorous arguments about the economic or political impact, from the very same people you're trying to convince, it's going to be a hard sell.
TL;DR: people should be free to sell or give away software as they see fit, and let market forces decide (as they have already in say, Internet infrastructure software markets such as server OSes, web/mail servers, virtualisation, and so on).
RMS broadly (and I'm paraphrasing ridiculously here) thinks that you should be paid for the hours you work, the software should be Free, you don't earn a right to keep it closed and sell access to it through licenses, etc.
Fine, but that would pretty much shut down the programming economy.
I don't think it would. The "programming economy" could instead flourish on bespoke software solutions instead of the monoculture we have now. Perhaps even governments (gasp!) would pick up the maintenance of a general OS, since it is in their interest to keep their citizens safe online.
61 comments
[ 3.1 ms ] story [ 123 ms ] threadThe timeline suggests a reference to Gary Bernhardt's talk, "The Birth & Death of JavaScript" [2].
1. http://runtimejs.org/
2. https://www.destroyallsoftware.com/talks/the-birth-and-death...
Very few apps available, all based on zero-knowledge protocols, and no apps that are capable of leaking your real-world identity through ordinary or casual use are allowed.
As a result, just about everyone on Capitol Hill has one of these, so they can't be banned even though they are, as a side-effect, extremely useful to whistleblowers, journalists, activists, dissidents, etc.
A computer help call in radio show as funny and warmhearted as Car Talk? Comp Talk? With crazy weird issues? Like email doesn't work over 400 miles? lol.
Esp. if the dudes seemed to know everything and had lots of old-timer computer stories from their old CP/M desktops and whathaveyou.
I'm bored with my lack of decent radio and podcasts. Feel free to recommend me stuff, readers.
https://news.ycombinator.com/item?id=9338708
http://web.mit.edu/jemorris/humor/500-miles
Get a channel going on YouTube, get a following. Pitch it to NPR. Watch it get picked up by a cable network after a few seasons, perhaps.
Sort of "Wayne's World meets Top Gear for computer geeks". I'd watch it. I might even pay to watch it.
http://www.iclarified.com/57138/apple-adds-arm-support-to-ma...
iOS and OS X share the same kernel, named XNU (it's the userlands of the two that are almost completely different). That's been the case since iOS 1. Whenever either OS X or iOS gets ported to a new CPU (or CPU subfamily), the kernel gets another set of macros added to it for the purpose of identifying the new CPU. The presence of a new ARM macro in XNU doesn't mean much at all - it already has a ton of them, as well as truly ancient macros from long-forgotten systems (m68/88k comes to mind). Those macros have nothing to do with OS X or iOS individually - they're relevant primarily to the kernel itself.
https://github.com/opensource-apple/xnu/blob/10.11/osfmk/mac...
The fact is though, iOS and OSX are already as much the same code base as they can be, and as much different as they need to be. That balance may change as the OSes evolve, but I don't think there's any pressing need or benefit to converging them completely.
With the (partly regretable) adoption of UI design patterns that are mainly meant for touch devices, such a merge appears to be closer however. I think the main patterns missing are
* a more powerful touch-capable tiling window manager
* a method to get "mouse hover" events working on touch devices. finger hover?
* a both touch- and precision pointer friendly implementation of the OSX menu system.
Anything I'm missing here?
As far as I know, an example is Apple's 3D Touch.
Maybe a stare-o-recognizer using the camera? If the user looks puzzled at a button, display text to explain ;-).
https://news.ycombinator.com/item?id=12772801
I know there's L4 which is pretty mainstream on coprocessors.
Minix 3. The rise of platforms like Raspberry Pi are good for alt-os projects because they give a well-fenced compatibility target.
Trying to emulate Mach, which was a dud as a microkernel, didn't help.
QNX is one of the few microkernels to get it right. L4 got stripped down so far that it's just a hypervisor, on which people usually load Linux. L4 took out arbitrary-length message passing in favor of interrupt-like events and shared memory between sender and receiver. This simplifies the kernel, but now it's easier for one side of a sender/receiver to mess up the other, since they share a communications area. The QNX primitive set (MsgSend, MsgReceive, and MsgReply) work well enough in practice to allow full POSIX functionality. Applications can talk to file system servers, network servers, etc. through those primitives. All QNX I/O works that way. You take maybe a 20% performance hit for the extra copying, but you get robustness in exchange.
Most important thing for performance in a microkernel: the CPU dispatcher and the message passing have to be tightly coordinated. You must be able to call another process and get a reply back without trips through the scheduler or a switch to a different CPU. QNX gets this right, because MsgSend is blocking. The sender blocks and the receiver starts without having to schedule. The data being sent is right there in the cache of the CPU, ready for use by the receiver. Good test for a microkernel - put some CPU-intensive jobs in a loop, while also running something that makes short request/reply calls to another process. If the request/reply process stalls out, the microkernel is doing it wrong. If the CPU-bound processes stall out, the microkernel is doing it wrong. Message passing should schedule as smoothly as a subroutine call. If it doesn't, performance under load will suck.
Roar.
QNX was cautiously courting the open source concept and venturing in the direction of shared source (with some code already available), when BlackBerry bought them and threw all of that out the window.
Biggest yanked opportunity. D:
Now QNX is all-commercial again, with source only available under a license. And Photon is gone, too: no more self-hosting.
I wish BB would(/could?) do the webOS thing with QNX. That would be amazing. The community would absolutely get Photon up and running again, it would be an alternative to Linux and BSD.
data:text/html;base64,SSBmaWd1cmVkIG91dCBob3cgdG8gZ2V0IFFOWCA0IGxpY2Vuc2VkIGEgd2 hpbGUgYmFjay4gSWYgeW91IGdldCBzdHVjayBteSBlbWFpbCBpcyBpbiBteSBwcm9maWxlLiBJZiB hbnlvbmUga25vd3MgaG93IHRvIGxpY2Vuc2UgNi54IEkgZG93bmxvYWRlZCB0aGF0IHRvby4=
Photon, the GUI, probably had the sanest internals of any major GUI system.
Dan Dodge, who designed much of QNX, was CEO of QNX, and he retired from QNX/Blackberry to go work for Apple. Apple might end up doing something in this direction. But if they do, they'll never let the internals out.
I assume you were using QNX for the GC vehicle compo because of its RTOS qualities? In any case, that's awesome.
I thought Photon was pretty cool too, at least from a "oh hey these screenshots look awesome" perspective. I vaguely recall running it in a VM, once, so long ago I only hazily remember poking around the sidebar. Sadly it'll be somewhat challenging to do further academic research on that subject now, I wish the educational/noncommercial free options still existed there.
Also, I found (and bookmarked) this HN thread from some time ago that's got some QNX gems in it: https://news.ycombinator.com/item?id=4834334 (very curious about the modem)
And you're very sadly right about Apple, they're very unlikely to release the truly interesting bits of any projects like this.
That course features implementation of microkernels based on QNX from scratch to drive trains around a track, should be straightforward to implement on a Raspberri Pi or a Pandaboard/Beagleboard. The course uses a TS7200 which is an older ARM chip.
[1] http://www.cgl.uwaterloo.ca/wmcowan/teaching/cs452/w16/notes...
Why don't we keep adding a date referenece to all titles retrospectively if we want to have a chronological appreciation of the article.
One is when it's likely that I've seen the page before, but not so recently that I recognize it immediately. "XKCD #386 [2008]" (OK, I know that one by the number.)
The other is when the page has a particular viewpoint which is either explained by the date or unusually modern for that time period. "Tiny computers will be ubiquitous [1965]"
Is it possible that AI has taken control over free OS from all over the web and killed all proprietary stuff along with humans and now run's the earth without any human civilization?
Am I correct?
Human civilization ending does not imply humans are extinct. There may be a few basement dwellers that survived, and kept coding.
"This infamously and perennially late GNU/Hurd OS will finally make it in to Randall's home after human civilization has been wiped out. The joke is that GNU/Hurd began to be developed in 1990, and while it was expected to be released in a relatively short time, even now only unstable builds have been released. So Randall is saying that he will finally run it in his house a decade or two after the end of civilization. GNU/Hurd will presumably have an advantage as humanity rebuilds civilization due to the widespread availability of its code and development tools, and perhaps also because of Stallman's depth of belief, based on the title text. Alternatively, GNU/Hurd might be finished by the same force that finished humankind, for instance Skynet, in case of AI Apocalypse. (Interestingly, although still far from completion, a new version of GNU/Hurd was released less than a week after this comic.)"
[...]
"The GNU/Hurd reference might also be a pun, as in a "herd" of Gnus "running" in his living room, as wild animals reclaim the Earth after the end of human civilization."
If we fill our home with IoT devices which carries Lithium ion batteries and relies on AI to function, I am sure fire will happen. Look at Samsung phones, even after recall there are cases of explosions.
So much automation in basic things is be like shooting in foot. Driver-less cars, really? Since when people started considering mental retardation as a status symbol.
I wonder if self-driving is going to be illegal in near future because we are incompetent to keep up with technological advances happening in our lifetime.
One day, masses will learning this hard way and come back to GNU.
PS: I am not alone saying this. https://www.theguardian.com/technology/2015/jul/27/musk-wozn...
Sure, only 'retards' crash cars. But I am smart so I would never crash a car ever.
People should be more like me, then the world would be perfect, problem solved.
Obviously from the comic, the author is an early adopter.
That's why he embraces HURD as soon as it's ready, instead of waiting until after 2060.
Just yesterday I was explaining to a friend that Stallman is not crazy; rather there is a continuum from convenience to freedom and Stallman believes most of us are way too far on the "left", so he goes as far to the "right" as possible to show us we've gone to far. He pays for his freedoms with his conveniences and we pay for our conveniences with our freedoms.
Things are much more nuanced and multidimensional than this. Not trying to start a flame war here.
I've met RMS several times (he even stayed at my place once, and yes the rider asking hosts not to buy him a parrot is real and long-standing), and I came it at from a different angle: I was a contributor to a BSD OS some yonks ago.
As a result we had a good-natured conversation about some of the conflicts between GPL and BSD and he's actually a lot more forgiving than people portray him or that argument. He's grounded in more reality than people give him credit for. Yes, he supports and advocates GPL, but he'd rather see code released as BSD/MIT than not released at all - at least that was the impression I was left with.
There are two things that tend to get the anti-GPL crowds back up, which on your continuum I'm not sure fit easily.
Firstly, there is the right for a programmer to be paid passively for the result of work done by their software.
RMS broadly (and I'm paraphrasing ridiculously here) thinks that you should be paid for the hours you work, the software should be Free, you don't earn a right to keep it closed and sell access to it through licenses, etc.
Fine, but that would pretty much shut down the programming economy. People regularly write code because they or their employer believes a customer is going to pay for it, and use that as leverage. Obviously. If we were to outlaw such a model - and it would require statues to make a reality, laws that run contrary to free-market liberalism - a large number of developers would be laid off.
That's OK, says RMS, because society will be better off. However, Turkeys do not vote for Thanksgiving/Christmas, and I want the _freedom_ to chose to work in businesses with economic models like that.
I think interestingly if Amazon, Facebook, Google, Uber and all similar firms (e-comms, social networks, search engines, etc.), released all their code, the market would not move much. The value was first created by innovating through their code base, but now the value is in the systems, brands, and relationships those firms have been able to build. Curiously, it's possible we are now in a scenario where closed source code can help a startup scale, but once they're at scale, there is no value in keeping the code closed, and making it open could benefit the companies (through "volunteer commits"), as well as the wider economy. We are seeing the tip of this in the dev code space (.NET et al).
Secondly, there is a weird line after which RMS does not care about software freedom quite so much. For example, in one talk he gave I attended, he stated that because he does not wish to modify the code in his microwave, he is not bothered if that code is free or not. I think that's a weak (and arbitrary) line, that is inconsistent with the core argument and really doesn't sit on the continuum.
Either all code wants to be free, or only some code does. And if only some code does, well, you can't go around telling people what does and does not.
Perhaps he's revised his position since then (a decade ago now), but I know it was something that troubled me about his argument.
And when somebody's economic and political philosophy does not quite stand up to not-very-rigorous arguments about the economic or political impact, from the very same people you're trying to convince, it's going to be a hard sell.
TL;DR: people should be free to sell or give away software as they see fit, and let market forces decide (as they have already in say, Internet infrastructure software markets such as server OSes, web/mail servers, virtualisation, and so on).
Fine, but that would pretty much shut down the programming economy.
I don't think it would. The "programming economy" could instead flourish on bespoke software solutions instead of the monoculture we have now. Perhaps even governments (gasp!) would pick up the maintenance of a general OS, since it is in their interest to keep their citizens safe online.