The Role of the Operating System In the Future

liteswap writes "Linux geeks love Linux and Windows mavens won't quit Microsoft -- but will we really care that much whether a machine is running Linux or Windows in future? As Sun announces Solaris support for Red Hat Linux applications, the need to specify the OS for a particular application will fade away, and the application and the x86 platform become the critical things -- at least that's what this Techworld feature argues..." Maybe a long time from now this will happen - but I don't see it happening RSN.

Comments

Standard emulation/abstraction platform?

[dada21]

I've been contemplating the OS situation for years (ever since I first tried to run a multinode BBS under DOS' DesqView) and only recently has a possible endgame become visible to me: emulation.

Software needs an OS layer, the OS layer needs an abstraction layer to hardware, and the hardware needs a communication layer to all the various mechanisms (drivers, BIOS, interface protocols, whatever). It isn't just a simple 16-bit .SYS driver like we had in the DOS days that basically handled everything.

We've seen so many emulators (Macs running Windows emulators running Mac emulators) on so many platforms, but what has allowed so many to come to the market in such a short period of time? Processor speed, I'd say.

Now that processors are incredibly fast, we're likely to see little performance increases in the tasks that 90% of the world uses PCs for: displaying text on a screen, inputting text into a form, and sending that text to a printer. Sure, Vista will incorporate a new video structure and 3D-gaming and heavy-use databases will always need faster processors, but MOST users are still just text viewers.

The next step, I believe, is creating a more realistic "standard" emulation structure for software. I think the F/OSS market is awesome because you can generally cross-compile a lot of code on various operating systems, but they still need modifications to the specifics of the OS or the hardware you're running on. What I really think will be the next big thing will be a TRUE hardware abstraction layer in the OSes (H.A.L.I.T.O.S.es?). Is it possible? I'm not sure, but it makes me wonder.

Why do people bust their asses constantly updating WINE when the OSS community can work towards a more amazing result: a standardized implementation structure that lets you write software once, and have it run on any OS that has a HAL to translate that implementation structure to what the hardware requires.

I know -- that's what the OS is supposed to do, but it fails. Yet do MOST applications really need the extreme features we have in customization (different video cards, hard drive controllers, network interfaces, etc)? Or would MOST applications run just fine (on high end processors) if they can say "Write pixel at X,Y" or "send data chunk to IP address" or "Write this data to this store" etc?

Maybe I'm talking out of my ass (I haven't programmed anything significant since MajorBBS mods in C over 15 years ago), but it seems like that is where software has to head. A completely transparent "mini-OS" that offers all software written for it a very standard set of instructions for the most popular functions. You're not going to write 3D games in it, but that's not the target market. 3D games will always push the envelope and come BEFORE the hardware can handle it. We're talking about basic implementation of basic software, yet it is this basic software that we waste billions of man-hours of labor on trying to get working on various OSes and hardware combos.

Now that I think about it, wasn't NT supposed to be the magic system? What exactly happened there? (Don't just say "Microsoft.")

Re:Standard emulation/abstraction platform?

[MightyMartian]

You are right that the speed of processors has changed things. With fast processors emulating hardware in a reasonably responsive fashion becomes possible. I view emulation, however, to be to platform indpendence as NAT is IPv4 address exhaustion. They both paper over the fundemental problem rather than dealing with it.

A Nice Dream

[MightyMartian]

The idea of removing the operating system as a major stumbling block to software development and distribution is nearly four decades old. I mean, the whole purpose behind Unix was to create a consistent platform that would make porting easy, and while the various distaff Unix and Unix-like operating systems certainly are more friendly towards porting than the older systems, for complex software it's still a difficult and time-consuming process.

The only development software that has come close to giving us platform independence are interpretative languages like Perl and Java, but that promise is still elusive. Java still seems to have stalled, and with projects like Mono, it almost seems like Microsoft may ultimately, though possibly unwillingly, get the upper hand.

Re:A Nice Dream

[xilmaril]

that's still having an OS. what you're describing is just a future in which the OS truely doesn't matter. of course, if the "browser is the OS", than it's all about what browser you're running instead.

it'll be great.

the Safai users will be elitist snobs with a prettier, more efficient browser-OS.

the internet explorer users will be clueless noobs constantly suffering from viruses, worms, and broken adware.

the firefox users will be... I can't describe them, because I am one, but you get where I'm going with this.

everything would be exactly the same in your future. exactly the same! AAAAGH!

Of course the OS matters

[Gothmolly]

Leaving abstractions aside, of course the choice of Windows or *nix matters. Because of Windows' layout and the way that certain applications are built into it (IE, Windows Help, etc) there are reliability issues that cause many more maintenance-related reboots than on Unix. When was the last time you rebooted a Linux machine because the Help system in KDE needed updating?

Also, there are certain OS-specific things which usually cannot be solved in hardware (assuming you're running on the best you can afford). Need an FS that handles massive sparse files correctly? Maybe that means you need Reiser on Linux, or ZFS on Sun... (I have no idea if this is true). Maybe Windows just CANT do this well, regardless of CPU power. Do you need to hot-swap NICs, CPUs, and add/remove memory and CPU power on the fly? You probably have to go to AIX then. Didn't we just read an article about how Windows takes 5x the number of CPU cycles to start a process?

If you consider the OS tightly coupled to the app, or the app requires specific capabilities from its OS, then app concerns will dictate the OS.

Re:Of course the OS matters

[TheRaven64]

THe article misses something very important. Solaris and Linux are both the same fundamental design, so emulating one with the other is easy. Now throw in something like QNX. Where Linux/Solaris/J Random UNIX is based on a synchronous function call model, QNX is based on an asynchronous message passing model - very different. You can emulate one on the other, but you take a huge performance hit. Now, at the moment most people use the synchonous model, but in the future the asynchronous one is going to be more important for two reasons:

1. It handles latency much better - send a message, do something else for a bit, wait for reply - and so is better in a high-latency (read: network) environment.
2. It scales better. On an n-core system, you are going to be able to have one core for every busy thread you have. You are not going to want to do two context switches every time you do a system call. Instead, you will write some data to a shared memory segment and carry on - signalling the OS if you have run out of things to do. The OS will periodically (or when kicked) read the waiting messages and send them to other processes. On a system that supports SMT you will probably have one OS message-dispatcing thread per core.

Emulating a well designed, scalable, message passing OS on UNIX is a lot more complicated than just providing a different set of system call vectors.

P.S. QNX can run POSIX apps if they are recompiled, but they are much slower than asynchronous apps doing the same thing.

Platform independent software

[kevin_conaway]

I think the role of languages that don't rely on a specific platform will become much more important in the future. I write my software in Python and it works wherever Python works (well, not really thanks to GTK+, but its getting there!). As these languages and toolkits mature, I think we'll start to see less of a dependence on the OS

Linux Torvalds on x86

[Teckla]

Right around the time Linus Torvalds announced his employment with Transmeta, he said something to the effect that the world already had a portable byte code, and that byte code was x86.

Java?

[jbolden]

How is what you are describing not the Java -- bytecode, JVM model?

Noooo....

[LWATCDR]

"x86 platform become the critical things"
Why? If you are going to abstract the OS why not the ISA?
IBM actually has been doing this for years with the System38/AS400.
The came up with a "prefect" ISA. When a program is installed it is converted to the actual ISA of the machine it is running on. IBM went from a CISC to the Power ISA without a hiccup.
I have wondered why Linux hasn't come up with a similar system. When you install and RPM or some other style of package the system could "translate" from a perfect ISA to the native ISA of the system you are running on.
Sort of like Transmeta did but do it at install instead of runtime.
Think of it as a just at install compiler vs a just at runtime compiler.

RSN - Red Sox Nation?

[truckaxle]

For the those clueless people, like myself, RSN does not stand for the "Royal Singapore Navy" or the Religious Science of Nashville but for "Real, Soon now" which to the initiate could alternately mean "Real Soon, Possibly Never".

I guess you need to be a science fiction fanzine fanboy or a regular reader of "Chaos Manor" to know this. Tribal Knowledge...

Re:welcome to 1999

[KrispyKringle]

First, Java, while not as successful as Sun hoped (what is?) hardly "failed miserably." Prior to C#, what other options were there for object oriented, garbage collected high level languages? People used C++, which, while in some ways stronger than Java, is hardly comparable. Java (partly) filled a void that, hopefully, C# will fill in an even more convincing way.

Second, machine independent bytecode isn't really the crux of the issue, I don't think. I say this for two reasons; first, x86 is pretty much the standard at this point (I say this from a G4 Powerbook, but with Apple and Sun shipping x86 machines, the desktop, workstation, and much of the server market seems to be going x86); second, any language with cross platform libraries and compilers is "write once, run anywhere;" VMs are only interesting (in the context of compatibility) insofar as they are "compile once, run anywhere."

The OS is already a hardware abstraction layer; it allows you to ignore what kind of I/O devices someone has, what size their disk is, how much RAM they have, and so forth. The technology to write code that works across multiple platforms was realized decades ago, and we use it daily. The only reason we're still talking about this is essentially an economic one--while each OS is an implementation of a standardized hardware abstraction layer, there are simply multiple OSes, which means multiple standards. The obvious solution to this is to be able to a) run multiple standards on one machine (either by running multiple OSes or by an OS that is compatible with, or emulates, the syscalls and libraries of another OS) or b) use a higher level language (like Java) that has its own standard and its own abstraction layer for each OS it is compatible with.

In other words, standardizing platforms is easy; getting people to agree on a given standard is hard. Being compatible with multiple standards is a good bit less hard, and we've been doing that for at least a decade as well.

Re:welcome to 1999

[Billly Gates]

I just purchased a nice cs101 book for java 5.0.

Why?

Because my computer science department wants to standardize on java in addition to c++. Why standardize on Java? Because its the most sought after language in business.

Don't believe me? Go to www.monsterboard.com or some job site and look at jobs in your area. Java is the most sought after language with c/c++ second, and perl third.

Java is essential for any big ecommerce servlet. Php is not there yet and neither is c#.net in terms of scalability and maturity.

Java is a success.

OS Matters, and MS is in the lead.

[CDPatten]

The OS matters and MS is making big headway in that front. They are doing it right in the face of all their naysayer's, and the irony is those self-proclaimed geniuses are missing what's happening right in front of their face.

For example, the Apple crew touts being first to market with features like indexed searching as reasons why they will beat MS. All the while MS is quietly getting XP Embedded in ATMs and Cars. MS can ad a search in an update (e.g. Vista), but Apple isn't going to power any BMWs with OSX 10.5, but MS already does with XP Embedded.

MS is diligently working with GE (one of the worlds largest companies, 1st or 2nd place) to advance home automation, and integrate with household appliances. Home automation is the FUTURE of computing, period. MS is working hard to penetrate the Home Media market (media center, Xbox, IPTV, etc.), the phone market, and many other fronts. You can say they won't make it, but they are doing a hell of a job to date. Look at the next generation of television, IPTV. MS is starting to get so far ahead of everyone else it's getting sad. Take some time and watch the demos, they are very impressive. The zealots keep saying it isn't true, but they have been saying this since Windows 95. They were wrong then, and are wrong now. Apple and RedHat don't have any big exclusive deals with Verizon or SBC to power IPTV, but MS does. Those deals are getting fiber brought to everyone's doorstep. IPods are cool, but they are a novelty device and they aren't going to power the home of the future, but at the current rate, MS will.

Phones: Mobile 5 blows the doors off of all business class phones today with the exception of RIM's. With the exchange integration, RIM won't be able to compete... MS phones will support Push with more then a 100,000,000 people overnight. RIM is struggling to top 5,000,000. Linux phones are a nice idea, but they don't offer push, and the ones at present can't hold a candle to Mobile 5. Then there are PDAs. MS has crushed Palm, and Nokia's hail marry is neat, but won't beat Mobile 5.

The bottom line is if you like MS or not, they are growing in many areas that aren't being publicized. The naysayers are a sleep at the wheel. The platform of the future isn't going to come from Google, Sun, and certainly not Apple. MS is getting in at the ground floor of these industries and they have far more money to fight off the others.

The platform matters. I know so many of you are out of your mind pissed at me for writing this. I'm sure some of you will have some wiki-pedia posts to try and make your case, or some blog of an anti-ms zealot. And to you I say; it doesn't matter if you use a Windows computer for surfing the web or not, you aren't going to be escaping MS powered operating systems anytime soon. History will prove me right.

Re:OS Matters, and MS is in the lead.

[idsofmarch]

Your post is really interesting, and there are a few things I'd like to counter. no innovation? Hmm... Maybe you should start to research home automation; quite a large chunk of my little "spiel" was about that. Take a look at what MS and GE are doing with that... no innovation huh? Umm... I don't see linux or osx powering any cars other then some freak project by college students. BMW is a major premium brand using XP. But hey, that's not your kind of innovation. Shadows under desktop icons are. MS and GE are doing a corporate equivalent 'freak project.' Quit being so judgmental; Microsoft itself was once a student project. I see the point you're trying to make, but Windows XP is running the entertainment system, not 'powering' the automobile--that's for embedded systems, some of which run Linux. Shadows under desktop icons can be innovations for human-interface, which is actually really important even if Microsoft treats it like an afterthought. I have lots to say, but I fear its lost on a zealot like yourself. I'll try to be brief. You are just so blinded with hatred you have started to believe you own rhetoric. The problem is that's all it is, rhetoric.... Unfortunately for your side people have heard you, and they don't agree. When you say Windows is horrible they hear you, but think "well it does what I need it to". Your message does get out. The media is almost 95% Apple users; trust me your view gets out. When a user tries Linux they can't get software installed... try Apple and no software is available for what they needed to do. Windows market share is not losing to Linux or Apple despite your rhetoric. And its not because people aren't as smart as YOU! Rhetoric is rhetoric; Microsoft has plenty of paid shills and corporate pushers. Many in the media use Macs, many in the accounting business use Windows. What does this tell us? Not much, what's more interesting is how the virus/spyware/malware stories hit. Was it because the media is a bunch of biased-Mac users or is it because it became such a problem that the media--in their bondi blue towers--didn't notice until later? The idea that the media uses Macs can cut both ways.
Attacks on Windows are a real problem that needs to be solved by Microsoft, and I think many people would prefer they do that rather than jamming Windows Media Player into a BMW. Not that this isn't an interesting development, but many feel that Microsoft should solve problems they had a hand in creating. The desktop monoculture is their fault.
Furthermore, a recent study indicated that one million people switched from Windows to Mac OSX this year alone. It's a drop in the bucket, but between that and Firefox grabbing 10% of the browser market in under a year, we can see that Microsoft is mortal. Dying? No, but they're certainly mortal. OSX has had 4 paid upgrades to Windows XP. Annual OS upgrades are a nightmare for IT departments, bad for a company's balance sheet, and an irritation for end-users. Each version costs about $150 x 4 = $600. MS XP is $300. If you kept up with Apple, you spent double then going with XP. Never mind all the programs that break with the upgrades. OSX's upgrade cycle works on a different paradigm than Windows XP. Each 10.x has constituted a new OS with so many under-the-hood changes as to be equivalent to Windows 2K, Windows XP, etc. And, only in some cases did a 10.x break applications--we won't mention XP SP2, which broke lots of applications as well. But, this I think is tangential to the main thrust of your argument. MS didn't abuse any monopoly power to make Office number one. They didn't abuse or break the law to beat Palm. They didn't abuse anyone to make Server 2k3 grow by leaps and bounds, faster then any other server platform this year. They didn't abuse anyone to sell 20 million Xboxes, and didn't abuse anyone to get a 1 year jump start on the PS3. Xbox Live... no innovation? They have taken online game play to a whole new level. Their IPTV stuff is amazing, and that is why Verizon and SBC were so willing to lock

The Jargon File

[RAMMS EIN]

The Jargon File explains many of these terms, has some interesting notes on hacker culture, and is sometimes plain funny.

Already happening

[Junks Jerzey]

Slashdot is clearly behind the curve on this one.

Point #1: Embedded devices

Do you know what "OS" is running in your digital camera? Your DVD player? Your MP3 player? Your GPS system? In the majority of cases, the answer is no.

Point #2: Web applications

Google search, Google reader, gmail, Flickr, etc. They look the same to me whether I'm running Linux, Unix, OS X, BSD, etc.

Point #3: Cross platform apps

Python coding and development feels the same on Windows, Linux, and OS X. Makes no difference to me. Ditto for editing with vim. Quite a few other languages and applications are identical, too: Inkscape, The Gimp, etc.