Ask HN: Why are there not more OS competitors?
14 comments
It’s not an easy task, especially since the OS competition has sailed long time ago. You either create a linux distro with your flavor, or start over.
Starting from 0 is a very hard task and requires a lot of development time and money.
Also, even if we say you get to a level near the stability and support of the current systems (which is near impossible considering the head start), you need to bring people to it.
Just like how Mozilla phone OS failed and Microsoft Windows Phone and many others. No one would go for a system that doesn’t offer their applications or applications available on other systems.
Of course you will be able to get users interested in niche systems, but these are very few. For example, i am running clear linux, which is “independent” distro. It’s an amazing system for me, but i would never recommend it for anyone due to the advanced setup
Unless of course.. you create a system that’s a breakthrough. A system that would make a 20yr old dead machine 1000x faster than new machines.
Just like how Mozilla phone OS failed and Microsoft Windows Phone and many others. No one would go for a system that doesn’t offer their applications or applications available on other systems.
Of course you will be able to get users interested in niche systems, but these are very few. For example, i am running clear linux, which is “independent” distro. It’s an amazing system for me, but i would never recommend it for anyone due to the advanced setup
Unless of course.. you create a system that’s a breakthrough. A system that would make a 20yr old dead machine 1000x faster than new machines.
There is! You can run Haiku, as many people do:
https://www.haiku-os.org/
https://www.haiku-os.org/
Developing a new OS that's on par with Windows is a huge, multibillion dollar task. And it's not even the hard part here, because when you have the new OS, there will be zero software on it - you'd need to somehow convince vendors to port their software to it. Essentially, I believe that the only body capable of doing it would be a big government (US, Chinese, or the EU as a whole)- they can say that will only run this new OS that they've developed in-house, and, by being large enough client, they'd convince some vendors to port crucial stuff (such as MS Office).
Unfortunately, shittiness of current OSes and productivity/quality of life losses caused by it are not even on a radar of our politicians, so they'll never invest any money in fixing the situation.
BTW, Johnatan Blow and Casey Muratori are ranting about it from time to time on Twitter and YouTube and they have some interesting takes on the subject (Casey was suggesting getting rid of OS altogether, and coming back to an Amiga-like architecture).
Unfortunately, shittiness of current OSes and productivity/quality of life losses caused by it are not even on a radar of our politicians, so they'll never invest any money in fixing the situation.
BTW, Johnatan Blow and Casey Muratori are ranting about it from time to time on Twitter and YouTube and they have some interesting takes on the subject (Casey was suggesting getting rid of OS altogether, and coming back to an Amiga-like architecture).
> It seems there is a huge opportunity for disruption in the computing/OS space. Why are there so few companies working on this problem? What are the main challenges?
It is a huge opportunity... but at great cost...
A company/country may want to develop their own OS. They can either write the Kernel themselves, which is an enormous task. OR they can just use the freely available Linux kernel. So most companies just essentially fork Linux...
Developing an alternative Kernel is the main problem, and not really a problem worth solving. The Linux Kernel is 15+ million lines of code now, and would cost an enormous amount of effort to reproduce (re-solving the same problems Linux faced).
Companies eventually come to the conclusion that using Linux kernel is the opportunity to develop an OS... which is why there's thousands of actively developed Linux distributions [1] ...
[1] https://distrowatch.com/
It is a huge opportunity... but at great cost...
A company/country may want to develop their own OS. They can either write the Kernel themselves, which is an enormous task. OR they can just use the freely available Linux kernel. So most companies just essentially fork Linux...
Developing an alternative Kernel is the main problem, and not really a problem worth solving. The Linux Kernel is 15+ million lines of code now, and would cost an enormous amount of effort to reproduce (re-solving the same problems Linux faced).
Companies eventually come to the conclusion that using Linux kernel is the opportunity to develop an OS... which is why there's thousands of actively developed Linux distributions [1] ...
[1] https://distrowatch.com/
> Developing an alternative Kernel is the main problem
I wonder if that's a bigger problem than a web browser that can compete with Chrome and Firefox.
I wonder if that's a bigger problem than a web browser that can compete with Chrome and Firefox.
I've wondered this myself. It seems like one of the well-funded *nix distros should be able to create something as clean-looking and plug-and-play as Mac OSX. I've used several flavors of Linux in the last few years, but nothing ever seems quite as smooth as Mac OSX. I get there's a money differential there, but... surely they can hire some good designers and get some sort of consistent UI/UX in place...
Anyway, I support the notion of a disruption in OS, but I agree with most of the other commenters that it's probably not going to happen in awhile, mostly due to money and resources.
Anyway, I support the notion of a disruption in OS, but I agree with most of the other commenters that it's probably not going to happen in awhile, mostly due to money and resources.
In addition to the other comments, don't forget that most of the criticism about the current OS ecosystem is from the tech-adept demographic.
The vast majority of the OS consumer base is fine with OSX/Windows because they work well and are familiar. It's hard enough trying to get "average users" on linux, let alone a brand new OS.
The vast majority of the OS consumer base is fine with OSX/Windows because they work well and are familiar. It's hard enough trying to get "average users" on linux, let alone a brand new OS.
Well, Plan9 was good but overlooked. Feel free to contribute to it
There also was an attempt[1] to extend Inferno OS into a mobile OS
[1] http://jfloren.net/b/2015/8/18/2
There also was an attempt[1] to extend Inferno OS into a mobile OS
[1] http://jfloren.net/b/2015/8/18/2
It's high risk and requires a much longer timescale than most companies and VCs are comfortable with these days. It would take somebody with deep pockets and a burning desire for a better computer to make it happen. Like Elon Musk with Tesla.
Agreed on the high risk - when you have such an established and entrenched competition from day one you have to be sorta crazy to bite something this big off. (and I mean "crazy" in a good way)
Money? And you're talking desktop. Off the desktop there's plenty to choose from. Android. NuttX is a thing. Amazon picked up freertos. Every firmware engineer who's been around the block has written one or more executives.
Fuchsia[0] (an OS by Google) on a Chromebook is the closest to being a worthy competitor.
[0] https://en.wikipedia.org/wiki/Google_Fuchsia
[0] https://en.wikipedia.org/wiki/Google_Fuchsia
The same reason there aren’t browser competitors - it would take forever to write and test one, and even then, on a new OS, you’d have to write tremendous amounts of user software to even be usable, let alone competitive.
remember, remember: https://news.ycombinator.com/item?id=13082143
and there are also all those wonderful *BSD flavours....
and there are also all those wonderful *BSD flavours....
The scope is huge and software is unavailable/incompatible.
my post on this topic from:
https://www.reddit.com/r/pcmasterrace/comments/dhpbml/anytim...
i still get lost once in a while but the cool thing about it is i have lots of help online
to me that's the real big difference between say proprietary software and opensource.
i grew up using Microsoft and even used to sell some of their small networks 'back in the day'
my boss was an Authorized Dealer held more certs than i can count so i've been around M$
nobody has, but if you were to ask me what happened with M$ software i'd say a few things:
when the cellular industry became the ISP that changed the motive of software design and it
changed the business model for software companies. consider the fact win98 --> win2000 was basically
a lean machine, not perfect but solid for the most part, again tho the internet wasn't a big factor
heck even Bill Gates didn't think the net would do much
2. when cellular carriers became the ISP this changed the business model, now folks had the net in their pocket,
including a microphone, video camera and gps unit etc. metadata and dataminig became the new wild west
this impacted the business model for software companies simply because they were now in a position to lose money
if they did not add 'xyz feature' whereas when the net was basically just getting off the ground revenue was
predominantly from tech support, service contract agreements and license distrobution etc.
still the 'unseen' problem that not even M$ could forsee or anyone else really was this:
before the net took off, if there was a problem with security, well the IOTs wasn't as pervasive, meaning
stuff wasn't connected as much, a 'hole' was limited, and yet another thing to consider - geeks weren't
everywhere back in the late 90s whereas today due to the rise of smartphones and apps programming society
has put her 'tech boots on' so the average kid down the street might be a semi pro 'hacker or programmer'
which translates into 'the security flaw can be a bigger problem'
3. to add on to #2 proprietary software doesn't invite 'patching problems' like opensource, when the net
wasn't boomen the last decent release was Win7, if M$ would have taken Win7, polished it up, released the
source code, invited a community to harden it, patch it, it would have taken a lot of pressure off of the
Microsoft employees, to ask any single company to basically be the end all for an operating system's security
to fix all the problems including drivers is daunting, even on a good day, even when things are going good
so that has an impact on the actual security of an operating system and that is tied directly to the market
a company's value on the stock market, find a hole, report it, someone might lose a contract, lose money
today the inherit value in an operating system is basically in the data that gets mined, meaning the average
stock config of any mass produced computer say from walmart or amazon has a lot of 'bloat' basically 'features'
that drill the customer to set up xyz accounts, once you tie your home network and cellphones to the OS you just
increased the value of the data collected, so that's a problem for 'security' because security now will go only
as far as it can stretch but not break any revenue streams 'built in' to the OS, it's like home wifi is strong
enough to keep the local network traffic 'verified and aassist prosecution' of a court case if you violate a law
it's not really there to protect you, privacy and security often means someone is losing money one way or another
linux is not immune to any of this, i look at it this way, do the best you can with the software you love because
that is really what the purpose of it is supposed to do, help us connect and get stuff done, i think what M$ did
in the beginning along with apple and the early founders of linux is just amazing, i really mean that, any company
that can and did survive for a few years did more right than wrong, to this day i remember the first time i moved
a mouse on screen, i remember the first time i saw a video game and i could control a square on the screen, i remember
the arcades, i'm still in awe of it today, so my hats off to all the programmers and engineers out there
i hang out with the penguin today, i watched Revolution OS years ago and instantly joined the cult
i still get lost once in a while but the cool thing about it is i have lots of help online
to me that's the real big difference between say proprietary software and opensource.
i grew up using Microsoft and even used to sell some of their small networks 'back in the day'
my boss was an Authorized Dealer held more certs than i can count so i've been around M$
nobody has, but if you were to ask me what happened with M$ software i'd say a few things:
when the cellular industry became the ISP that changed the motive of software design and it
changed the business model for software companies. consider the fact win98 --> win2000 was basically
a lean machine, not perfect but solid for the most part, again tho the internet wasn't a big factor
heck even Bill Gates didn't think the net would do much
2. when cellular carriers became the ISP this changed the business model, now folks had the net in their pocket,
including a microphone, video camera and gps unit etc. metadata and dataminig became the new wild west
this impacted the business model for software companies simply because they were now in a position to lose money
if they did not add 'xyz feature' whereas when the net was basically just getting off the ground revenue was
predominantly from tech support, service contract agreements and license distrobution etc.
still the 'unseen' problem that not even M$ could forsee or anyone else really was this:
before the net took off, if there was a problem with security, well the IOTs wasn't as pervasive, meaning
stuff wasn't connected as much, a 'hole' was limited, and yet another thing to consider - geeks weren't
everywhere back in the late 90s whereas today due to the rise of smartphones and apps programming society
has put her 'tech boots on' so the average kid down the street might be a semi pro 'hacker or programmer'
which translates into 'the security flaw can be a bigger problem'
3. to add on to #2 proprietary software doesn't invite 'patching problems' like opensource, when the net
wasn't boomen the last decent release was Win7, if M$ would have taken Win7, polished it up, released the
source code, invited a community to harden it, patch it, it would have taken a lot of pressure off of the
Microsoft employees, to ask any single company to basically be the end all for an operating system's security
to fix all the problems including drivers is daunting, even on a good day, even when things are going good
so that has an impact on the actual security of an operating system and that is tied directly to the market
a company's value on the stock market, find a hole, report it, someone might lose a contract, lose money
today the inherit value in an operating system is basically in the data that gets mined, meaning the average
stock config of any mass produced computer say from walmart or amazon has a lot of 'bloat' basically 'features'
that drill the customer to set up xyz accounts, once you tie your home network and cellphones to the OS you just
increased the value of the data collected, so that's a problem for 'security' because security now will go only
as far as it can stretch but not break any revenue streams 'built in' to the OS, it's like home wifi is strong
enough to keep the local network traffic 'verified and aassist prosecution' of a court case if you violate a law
it's not really there to protect you, privacy and security often means someone is losing money one way or another
linux is not immune to any of this, i look at it this way, do the best you can with the software you love because
that is really what the purpose of it is supposed to do, help us connect and get stuff done, i think what M$ did
in the beginning along with apple and the early founders of linux is just amazing, i really mean that, any company
that can and did survive for a few years did more right than wrong, to this day i remember the first time i moved
a mouse on screen, i remember the first time i saw a video game and i could control a square on the screen, i remember
the arcades, i'm still in awe of it today, so my hats off to all the programmers and engineers out there
i hang out with the penguin today, i watched Revolution OS years ago and instantly joined the cult
It seems there is a huge opportunity for disruption in the computing/OS space. Why are there so few companies working on this problem? What are the main challenges?