Ask Slashdot: Innovative Operating Systems/Distros In 2015?
iamacat writes: Back in 90s, we used Linux not only because of open source, but also for innovative features not found in commercial operating systems — better multitasking, network power features like slirp and masquerading, free developer tools for many languages. Nowadays OSX and Windows caught up in these areas and mainstream distros like Ubuntu dumbed down in default configuration. So where to go for active innovation like 3D/VR desktop, artificial intelligence, drag and drop ability to mash up UI of multiple apps or just drastically better performance? Something maybe rough around the edges but usable and exciting enough to use as daily desktop?
better multitasking
Then what specific OS?
network power features like slirp and masquerading
How were either of these unique to Linux?
free developer tools for many languages.
So no different than BSD?
Back in 90s, we used Linux not only because of open source, but also for innovative features not found in commercial operating systems — better multitasking, network power features like slirp and masquerading, free developer tools for many languages.
None of what you mention was a unique feature of Linux or even pioneered by it. All of what you talk about were already part of Unix systems that existed prior or was software that existed before Linux even existed and was already cross-platform.
Back in the 1990's, you had to roll your own kernel and modules. If you were lucky, all the hardware worked. Most of the time it didn't. Nothing worked out of the box. Today's kids have it too easy. Now get off my lawn!
Just because Ubuntu dumbed down doesn't mean you need to use it. Slakware is still out there. Arch, Gentoo, Debian, Fedora, etc. are all a bit rougher and all have a bit more "exciting stuff". Still, the submittor's main problem is that he needs to go into a StarTrek movie. Truly innovative ideas in operating systems, like Plan9 and Eros, end up with less "drag and drop ability to mash up UI of multiple apps" (whatever that means) than the mainstream. Most of the innovation is now best done on the level of the application anyway.
better multitasking
Than what? Windows? I'm pretty sure it didn't have better multitasking than a real Unix in the 90s.
network power features like slirp and masquerading
Slirp supported commercial Unixes like Solaris. IP masquerading could also be done on BSDs and other Unixes as well.
free developer tools for many languages
The GNU tools could be run on commercial Unixes even before Linux existed.
They say that P-P-P-PowerOS is rather nice, but it only runs on specific laptops.
"Oh, goodness. Look at my wrist, I have to go." "But what about your clothes?" "I don't love these."
TFS must've been written by a Slav: no "a" in any of it... ;)
Holy crap, I just got bingo!!
Exciting?? What the hell? You want 'exciting' plug a Windows box direct to the intertubes with no firewall.
I heard some crazy guy has a "praise jeebus" operating system or something, but I'm pretty sure I have never once heard anybody say "gee, what I want is a desktop which is exciting to use".
New features which serve a purpose are good, but this screams of asking for pointless and shiny because you seem think it should be there.
Give me Tony Stark's Iron Man interface, and I'll be excited. Everything else is just pointless eye candy of people making something whiz bang which doesn't actually do anything.
Otherwise we're just resurrecting the SGI "Hey, this is UNIX" interface from Jurassic Park. (And, yes, it was a real interface.)
Now get off my damned lawn.
Lost at C:>. Found at C.
There is so much innovation these days that it has transcended the separation-by-OS that used to handily signal where and what kind of changes you could expect. As an example, if you're looking for an experimental graphical terminal emulator it turns out you can use it in Windows and OS X, but not in Linux. But the point is, it's not available on one OS in particular and it's even a virtue now to be cross-platform. There's so much new tech out there and it all happens on a huge variety of platforms. So trying out new tech is just a matter of focusing (for example: system software, graphics software, hardware support, kernel-level new stuff, software in embedded systems, hardware sensors, etc.) and then deciding what the required resources are to dive in on that specific level. What OS or OSes would be best, what packages should you install, and so on.
Going back to your examples, 3D/VR desktop work has been going on since the 80s at least, and AI before that, and "drastically better performance" has always been on peoples' minds. The GUI mashups even ring a bell, though everything is so scriptable these days that anyone who's doing a GUI mashup would probably be less frustrated just typing it into a reusable script. These aren't new topics, they change over time incrementally, and the only advice I can give is to make sure you are _really_ looking at the high-end tech that you think you are. If you are frustrated with a slow system, did it cost less than $10K? Because that's commodity-level pricing. If you are frustrated with the 3D effects you just enabled on your desktop, did you really research the state of the art? And so on.
Also, just to nitpick--you say Ubuntu is dumbed-down in "default configuration" but Windows and OS X are dumbed down by default too, aren't they? That's why you have package managers, Ninite, the App Store, etc. Restore your purchases or download a set of things and you're out of the dumbzone.
The past couple of Ubuntu releases have been fairly "boring" stability releases, but 16.04 is looking to have some exciting new features: Mir (replacement for X.org), Snappy Core (replacement for .DEB packages, to compete with Docker), and Unity8 convergence. The goal is that somebody will be able to carry around an Ubuntu Phone that morphs from a touchscreen smartphone to a full-fledged desktop when it connects to a dumb terminal. If it's a hit, then I imagine in the future, it will also serve as a console for smart-homes ("Internet of Things"), cars, and VR projection.
Maybe that doesn't excite anybody else, but for me, the idea of having a smartphone replace all of my electronics and not having to worry about file syncing, networking, etc. is awesome.
Stronger heritage of what? The OP was referring to innovative features.
I'm pretty sure you want this: http://www.menuetos.net/
Slackware. It just keeps working
Take a look at Squeak ( http://squeak.org/ ). As it turns out most things in the future will have their roots in the discarded ideas of the past. As far as programming languages take a look at Erlang and Elixir (computer languages are the operating systems of the future). I expect the capability model and the actor model will get a lot more popular in the future. In the future computers will be networks, applications will be distributed applications.
Have you tried ninnle linux? It's pretty cutting-edge... and no system d shit!
Copyright (c) 1990 - 2014 Dice. All rights reserved. Use of this comment is subject to certain Terms and Conditions.
Iif you want one with a random bullshit generator, just choose whatever the OP runs.
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
I think they just mean that it has multiple desktops by default.
And maybe compositing so you can have have transparent terminals.
That is all.
"Something maybe rough around the edges but usable and exciting enough to use as daily desktop?"
Yup, that's pretty much the definition of Enlightenment
I love the Terminology terminal emulator and wish it was easier to install on non-Enlightenment distros.
Average Intelligence is a Scary Thing
So you are saying that OS X is "innovative" because it consists largely of 1980's technology (NextStep, BSD, Smalltalk, OO dev tools)? Seems to me that that makes it about three decades behind the times.
It is very exciting to use! You'd want to destroy keyboard, screen, or the computer itself after you use it for a bit.
For a while I the server for my web app ran on a Palm Pre smart phone. because has significantly better performance then running it on my wireless router.
I still run a web server on my phone from time to time to test the new version of the web app that I'm developing on my phone.
Speaking for myself, I use Linux distros at home for these reasons:
1. They're not Microsoft, Apple or Google.
2. There is less "telemetry" from my Linux boxes to OS megacorps(see #1)
3. Linux desktops have become reasonably reliable and stable, and yes, I've been using Unix/Linux since late 90s.
4. I enjoy trying out different distros/software, configuring the software, seeing the different ways things work in different distros, etc
5. Linux is fun!
We play the game with the bravery of being out of range
Is there no one with a suggestion for a new operating environment (or OS) to answer the OP's question?
(No, I don't know of one either)
For anyone looking to get their hands dirty, I always recommend to try out Funtoo Linux.
Try out various kernels (even BSD kernels), choice between 3 init systems, and all the customization you want.
Seriously. Try it. www.funtoo.org
Several companies announced resistive or whatever memory that is almost as fast as DRAM, while also non-volatile, cheap and big like flash. We need an architecture that takes full advantage of that. Keep the programs and data in place (instead of the usual RAM to/from disk joggling), optimize the I/O and CPU differently...
HP announced some work in this direction with their "Machine", but for now I believe all they have is some slightly customized Linux distro.
If I remember correctly, PalmOS had some good ideas in this direction.
Whoever implements an efficient architecture for this, has a good chance to be a great thought leader for the next 30 years. There is really an opportunity for a new Linux-level (or even UNIX-level) innovation right now.
How so? Linux was developed after BSD.
Sorry, we are at the top of the s-curve. Forget about innovation and expect incremental enhancements.
love is just extroverted narcissism
Where else, but the Apollo Giudance Computer or Data General's RDOS. Oh wait, it does not have .NET or Node.JS so maybe it is not so innovative.
The trash, where the 3D desktop concept belongs.
To be fair to the submitter, in 1990s, it was Mac OS 7,8,9 back then. Apple didn't buy NeXT until 1996 and didn't release OS X until 1999. Even then I don't consider it comparable to OS 9 until Jaguar (10.2) in 2002.
Well, there's spam egg sausage and spam, that's not got much spam in it.
There are a lot of things that can be done in the open source arena but things like AI are too valuable for universities to give away and often require physical datacenters to train and use. Siri, Cortana etc. are not in my opinion likely to emerge through part time collaboration and unlike drivers, there isn't really an incentive for the big players such as Google to give it away. The good news is that most consumer facing AI/Digital Assistants will simply be a web service that someone could write a front end to. So it might not ever be truly built into the OS but it will be accessible through any platform. The real impediment to open operating systems is that there needs to be an incentive to develop a feature and currently if it doesn't make for a better server (e.g. file system, networking, super computer fabric, virtualization, webhosting etc) it probably won't get much attention from the community or industry.
Regarding AI I'd take a look at the Mycroft project.
https://mycroft.ai/
Still getting started but have plans for AI on the desktop.
Complexity Happens
NixOS has a package manager that I think has a real shot at achieving scalability and repeatability in package management. Once something works in NixOS it will keep working on it's own, since specific versions of dependencies are tracked and can coexist, whereas in mainstream distros shit breaks all the time. The current model of freezing everything once in a while and patching up some of the most obviously broken stuff simply isn't keeping up with the pace of software development IMHO. http://nixos.org/ For a real moonshot OS/language/decentralize_all_the_things project, check out Urbit: http://urbit.org/
All the leading OS still don't deviate much from traditional kernel design approaches, the main "innovation" i'm interested in for a modern OS is provable reliability... by provable i don't mean empirical and observable reliability (e.g FreeBSD), i mean reliable by design: There are some pretty innovative concepts beyond the basic microkernel concepts in Minix 3, and unlike it's predecessors it's intended to be more than an educational tool.
^modup
A design for not making upgrades break and not freezing shit into obsolescence is neat.
>" Nowadays OSX [MacOS] and [MS-]Windows caught up in these areas"
Oh really? Perhaps quite a bit in just THOSE few areas which you listed, but in nowhere NEAR all the areas for which many of us continue to choose Linux. It is nice that Linux forced other operating systems to suck less than they used to, however :)
>"and mainstream distros like Ubuntu dumbed down in default configuration."
So then use one of the other [superior and yet excellent] Linux distros. I, for one, have never selected to use Ubuntu on any of my systems (beyond testing).
MirageOS (https://mirage.io/) is the most interesting OS project I've seen in a long time, but it's hard to describe. Approximately, it's a framework and collection of libraries that compile applications written in OCaml into unikernels that run on top of Xen.
So far as Linux goes, Chrome OS seems like the best engineered Linux userland at this point.
I don't normally respond to ACs, but thank you for the clue on mirage.io. The unikernel idea is intriguing and - as per the OP's question - an innovative new OS model.
Another interesting unikernel system is Ling - essentially Erlang running directly on a hypervisor.
I can't help but think these will give containers and such a run for their money over the next few years.
I find that Microsoft is really innovating in the user interface department these days.
Debian
i agree with the first sentence and partially agree with the second.
i feel like the excitement of linux desktop is disappearing. i.e. gnome looks amateurish but has well integrated apps. kde looks more professional, but feels like mess where barely anything works as it should. enlightenment is like going a decade back in time - what the hell is samsung doing with it? unity is too simplistic, cinnamon is the definition of "nothing exciting".
in gnu/linux in general, package manager is nothing special anymore, win/mac have it too. stuff like cgroups, containers, fs snapshots are nothing to be excited about as an ordinary desktop user. there are no linux only killer apps.
and stuff is complicated and buggy. i have yet to meet a fedora/centos person who has selinux/firewalld enabled and is able to get work done. also, good luck debugging your setup with systemctl and journald. in ubuntu, it's almost impossible to get a bug fixed. you either fix it youself or go upstream, then open bug report in debian and then in ubuntu. then you create fake accounts (to say it affects me too) to get some attention to the bug report and hope that in a year+ a new ubuntu version will have it fixed. and they're not stupid ui issues, but big stuff like server installer not working with usb keyboards, nfs not mounting with exec,dev,suid if -o users is present, etc..
windows/mac still suck a lot more (believe me i've tried switching to osx), but they have a lot more 'killer' apps.
i used to hate systemd. then i was forced to use it while building a redhat based cluster. on the one hand, it's an excellent init system and i absolutely love it for that - it's like finally having Solaris SMF on linux. on the other, it's a horrible everything else. it stands like a wall between you and problem resolution when things don't work. it presents incomplete riddles instead of information about services that cannot start.
Rock solid I have yet to encounter. Every distro I tried has it's flaws.
Probably the most 'innovative' in that its approach is very non-tradititional. Seems like a good idea at this time:
https://www.qubes-os.org/
My God, it's Full of Source!
OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
You have 2 choices.
1) Spyware malware adware walled garden proprietary closed source horse shit 2) Linux and BSD. Open source, stable, fast, secure, cutting edge available as well as rock solid distros.
Rock solid linux distro? Okay here's a challenge for you, what "rock solid" Linux distribution would you recommend I install on an HP Pavilion laptop with a Realtek RTL8723BE Wifi card? Because as far as I can tell that's basically not supported properly by any distribution.
So you buy unsupported hardware with proprietary drivers, and it's somehow Linux's fault?
I will never understand some people...
Nothing as far as a distro (or desktop environment) with 3D VR or AI comes to mind but there is innovation in OS going on. Not many have attempted to answer the OP, so here's my list. Others mentioned Qubes, Urbit, and Mirage.io, which reminded me of Nix OS and HaLVM.
Both innovative and seems daily-driver ready:
1. Qubes OS - https://www.qubes-os.org/ - Linux distro that runs a Xen hypervisor to contain every app (including Windows ones) away from the desktop environment
2. Haiku OS - https://www.haiku-os.org/ - Tiny (under 200MB installed), Non-Linux that is binary-compatible with BeOS, nice understated GUI that is bland but usable
3. ReactOS - http://reactos.org/ - Win32 compatible open source OS, very active development scene working toward full NT kernel ABI compatibility. Seems stable enough to be a daily driver but hardware support is lacking
4. PC-BSD & freeBSD 10 - http://www.pcbsd.org/ http://www.freebsd.org/ - PC-BSD is a desktop distro of freeBSD 10 built for user-friendliness with automatic ZFS snapshoting and a nice graphical package manager, freeBSD 10 has a completely new package manager (pkg-ng replaces the 'pkg' binary)
5. Nix OS - https://nixos.org/ - Linux distro with innovative package manager promising atomic upgrades & rollback.
Innovative server-exclusive (ie no GUI):
5. SmartOS - https://smartos.org/ - Solaris + KVM + Docker w/ full Dtrace support. Claims ZFS as an innovation? Joyent is running a cloud of it
6. CoreOS - https://coreos.com/ - Linux distro exclusively for large Docker deployments. developing a suite of Go tools for datacenter management.
Innovative, but not ready for desktop use:
7. Redox OS - http://www.redox-os.org/ - OS written in Rust (rust-lang), which guarantees a lot of memory-safety, screenshots of desktop in 'News' section
8. Contiki OS - http://www.contiki-os.org/ - Linux distro for IoT embedded devices that claims an innovative network stack
9. Urbit - http://urbit.org/docs/user/int... - *nix distro with exclusively web-based userland, invite-only at the moment, doesn't seem like it will have a UI but that each user is the dev of their own interface
10. Mirage.io - http://mirage.io/ - Develop each app and compile into a single-purpose kernel to be run on some hypervisor
11. HaLVM - https://github.com/GaloisInc/H... - The Haskell Ligthweight Virtual Machine - which runs just the GHC on Xen, another 'build uni-purpose VMs' system
Linux offers not stupidness
ahem. Systemd.
Serious, non-trolly question: Have you actually used OS X for more than few minutes and tried to do anything that ran into the garden’s walls? Yes, default config won’t run binaries from just anywhere. For the majority of users, this is a GoodThing(TM). Want to run something you got from this guy you know? Go in to System Preferences -> Security & Privacy, General. Click the lock, enter your password, and choose “Allow apps downloaded from:” Anwhere. Done. The walls to the garden just came a’tumblin’ down. You may now aim squarely at your foot and pull the trigger at will. You’ll get one “Are you sure?” prompt for each new binary you run. Click Open, and you’ll never be nagged for that binary again.
Running unsigned kernel drivers or monkeying with OS-installed binaries takes a little more work in El Capitan. Run `csrutil disable` and reboot. The system is yours to alter and/or destroy without limitation once again.
I’m sure I fit the profile of fanboi just a little, but I genuinely think Apple has struck a good balance on their desktop OS. The vast majority of users will have no idea how to disable these protections and therefore shouldn’t disable them. If you can’t figure out how to run arbitrary binaries you downloaded from the Internet, then you lack the knowledge to make informed consent about whether you should do so. If you know what you’re doing, flip two switches, and the OS won’t stand in your way.
While I agree that seL4 is innovative and exciting, its not an OS or a distro. Its a microkernel, where afterwards, you have to implement your own OS services, hopefully with an exceptionally secure design. (Or just use it as a base for your own unique embedded environment, which is overkill). If you only value your own design flexibility, you'll have an easier time implementing your dream OS in forth, rather than seL4. (If that OS requires multitasking, perhaps one could build a forth environment on top of the microkernel.)
Its kind of pointless to be screwing around with seL4, if you're not some computer science guru, who wants to build an OS with security as top consideration on top of a vetted securely designed microkernel. Its like wanting to use a CNC milling machine to only cut metal pipes in two.
Once you brought up GNU HURD, I then realized you were making an obscene joke. But it would sort of be exciting (not really innovative) in the sense of trying to actually make it do something without crashing. (I half-heartedly wonder if there was anything salvageable from HURD's original designs that would justify the effort to create a MACH emulation layer on top of seL4.)
There is no America. There is no democracy. There is only IBM and AT&T and DuPont, Dow, General Electric, and Exxon
This is a rather new distro for ChromeOS devices. It aims to fix some of the common problems that standard Linux distributions have with these devices and improve performance.