Slashdot Mirror
Vista Followup Already in the Works
DesertBlade passed us an InfoWorld article, which has the news that Microsoft is already hard at work on the next version of Windows ... and we may see it as early as 2009. Possibly codenamed Vienna, the next Windows iteration will be coming a brief two and a half years after Vista's launch. This is the same timeframe Microsoft claims it would have utilized for Vista, had they not put Longhorn 'on the back burner' to deal with security issues in XP. Corporate Vice President of Development Ben Fathi is already discussing features for the next OS: "We're going to look at a fundamental piece of enabling technology. Maybe its hypervisors, I don't know what it is ... Maybe it's a new user interface paradigm for consumers. It's too early for me to talk about it ... But over the next few months I think you're going to start hearing more and more."
The power switch?
"It's too early for me to talk about it"
Translation: "We haven't figured out who we're going to rip off yet. Probably Apple."
I am not too impressed by the name of "Vienna", especially since I happen to like the place.
I think something along the lines of Windows Hindenburg would be more appropriate. Or does anyone have a better name?
They do. After Windows is finished, the dev team proceeds to work on the next version, while a team called Windows Sustained Engineering takes over the released version. From the link:
Another Windows in two years, why bother upgrading?
http://en.wikipedia.org/wiki/Osborne_effect
There are places where the networks are not touching,and there are places where they are-Boeing's Lori Gunter
"we got nothing, someone think something up quick so we can steal it."
If you mod me down, I will become more powerful than you can imagine....
There is no shortage of manpower at Microsoft. There is a severe shortage of vision, and managerial competence.
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
I just don't understand why they are announcing this new version so soon after the release of Vista. The reviews I have been reading about Vista already make me think twice about wanting to upgrade; now that I know they are bringing in another OS in a few years' time what is the incentive for a typical MS customer like me to upgrade? Surely it is better to wait and see what they come up with next.
For those that do want to upgrade there is already a built-in lag before doing so anyway (at least for the sensible ones), either because they need to buy new hardware or because they will not install a new OS without some of the early bugs being ironed out and a service pack being released.
If we assume that MS actually delivers this new OS on time (which is a big if) there is not that long a wait between the time after lag for people to upgrade to Vista and the time this is released. Won't this reduce uptake on Vista? After all, if we are already happy with XP, why not wait?
Anyone already using Vista care to comment?
To be talking about this now. If this story gains traction then it will just hurt business adoption. Two years is nothing to wait out Vista and XP still works fine. Many small businesses I've personally heard from have not heard great things about Vista, this will scare them off even more. To take a page from Huggy Bear word on the street is...Vista is OK, nothing special and not worth upgrading to. News of Vista's early replacement certainly isn't the method I'd use to try and win people over.
If you wanna get rich, you know that payback is a bitch
There's a diminishing return on manpower. There's only so much the operating system can be fragmented, and each group can only be so large. That was part of Vista's problem - too many people having a say.
The more people you put on a project the more managers you require, the more meetings, the more decisions, more designs etc...
Larger code base means more bugs, more test time, more bug fixing teams etc..
You can't put twice as many people at a project and expect twice the work to result from it.
We walked in the cold air.
Freezing breath on a window pane,
Lying and waiting.
The man in the dark in a picture frame,
So mystic and soulful.
A voice reaching out in a piercing cry,
It stays with you until
The feeling has gone only you and I.
It means nothing to me.
This means nothing to me.
Oh Vienna
Disabling enabling? No way.
They've probably been fleshing out the feature list for Vista's successor since the first day a developer copy of OS X 10.5 reached the grubby mitts of a Microsoft employee. Don't expect the real work to start until spring, though, when it's released with its 'top secret' features.
Go ahead and mod me down, bitches, but after this tasty tidbit you know I'm probably right. And they did the same thing to Go Corp, BTW.
~Philly
Most likely because dispite Microsoft's reality vortex they still at least have the balls to admit to themselves that software still has not been realized as an engineering discipline. It would be nice if a large software project could be broken out into little modules with clear specifications that any coder could go off and make but it usually can't. Lots of development is very iterative, which means everything is changing. Lots of time stuff just has to be built to see how workable or unworkable it really is in practice; but when I change my interface it breaks your module. Maybe that is a minor problem easy to fix or maybe its a show stopper, how can I know.
Most large projects seem to work best with a few core team people who know basically how everything works at least at some level and can then farm out small clearly defined tasks to others. Their total bandwidth is bound to be limited though and so more 'others' does not always help. Growing the core team won't help much either because communication between them has to be total and constant, that is going to take longer the more specialed and nemerous those guys become.
Look at the Linux kernel for instance. You have Linus and pretty small core team that has different specialties. I know all those core team guys have some familiarity with the entire thing and Linus absoultly does. You can tell that from reading LKN. Maybe Jens is a block layer wizard but he know s how the network and VM layers work. He has to know inorder to mange block layer development well. He then has lots of other people submitting smallish patches and fixes to what is primarily his project.
I think we can reasonably assume that the Linux kernel and core GNU stuffs, includeing things like Gnome, have more developers.[qualified] contributing then M$ can put on windows even if they wanted. While those projects do seem to progress more rapidly then Windows its not by any means in an earth shattering way.
Repeal the 17th Amendment TODAY! Also Please Read http://www.gnu.org/philosophy/right-to-read.html
"After Windows is finished, the dev team proceeds to work on the next version, while a team called Windows Sustained Engineering takes over the released version."
And therein lies the problem. There is zero incentive to do it right the first time. After all, once its' out the door, its someone else's problem.
The people who actually wrote it should be responsible for fixing it - not writing the next-gen fuckup.
10 reasons
A perfect example of someone who should be kept locked away from the media until they have something concrete to say.
I mean, really ... Ben Fathi is supposed to be the guy overseeing everything, and he says "I don't know what it is" about what's next, and this is news????
So, he says he doesn't know what the next big thing in Windows is going to be ... here's a suggestion - new graphics and artwork to make it look more like OSX, a new startup sound that cost a billion instead of a few measly million to "enhance the user experience some more", a Duke Nukem Forever interactive screen-saver, and ribbons with dropdowns with flyouts with popups with menus, so that the user has at least 10 different ways to get to any particular option. And not one, not 2, but FOUR new programming languages - D minus (to replace C sharp), DOT NOT (a .net replacement that is ultra secure by refusing to do ANYTHING), J-Script/XML+J-Script/CSS for those who want to continue to build non-standard web sites, as well as Internet Explorer 9 - will only allow you to visit microsoft-signed sites, and a revamped cmd.exe and windows kernel that will only allow access to 640k of ram per process so that no application can ever be a resource hog. This last spec will be known as "Microsoft Dynacode Operating System 1", or MS-DOS 1.0. Plans call for an optional text interface sometime by 2012, and the removal of mouse support by 2015, because they can sell ms keyboards for more than mice.
Oh, and their engineering slogan will be "Windows ain't done until Wine won't run."
Hypervisors are the way to go for the OS of the future. Microsoft has had this vision for years. It was the foundation of their Next Generation Secure Computing Base, NGSCB, aka (ominous music here) Palladium.
Palladium got embroiled in the whole DRM controversy but there are good reasons to go this way independent of DRM. The idea is that you have a regular OS running, a Vista type OS, and then you launch your hypervisor. The hypervisor digs its way under the OS, takes control, and the OS is then packaged up and is running in a virtual machine. This is what they call "Late Launch" and is the key to one aspect of the technology I will explain below.
Now, here is the big win. You can create a new class of software, "applets" (maybe "virtlets" would be a better name) which interface directly to the hypervisor instead of the big legacy OS. These run in separate VMs so are immune to corruption of the big OS. They are simple and use a minimal API from the hypervisor so the chances of getting the code right and bug free are much greater. You can now use these for security oriented features you'd never dare to dream of on a monolithic OS. Think of Internet voting as a good example of what kind of security we are talking about. A more prosaic example is ecommerce - in a future world where people get their credit card numbers stolen all the time by malware there will be a real need for a secure way to shop online. Hypervisors and virtlets give developers a chance to start with a clean sheet of paper on the security front, while still maintaining full legacy backwards compatibility.
Then there's the kicker. Part of the goal of Late Launch is to use the TPM chip to measure (hash) the hypervisor and each VM separately. It means that each VM has an identity that it can securely attest to using a certified key embedded in the TPM chip. That Internet voting app? It can connect to the voting server and the server can verify that it is running in a clean state. Any corruption would be detected and show up in a bad hash report from the TPM chip. Malware can't fake that report because nobody can fake it, not even the user (meaning, he can't be fooled into faking it either - this is the flaw in EFF's "owner override" proposal, but that's another story).
This is all happening, folks. Intel's Lagrande Technology, now called TXT or Trusted Execution Technology, is rolling out as we write. This was the gating factor for all this technology and is probably the real reason it didn't appear in Vista - the hardware wasn't ready. But it's going to be there and it will be ubiquitous in a couple of years (at least, as ubiquitous as Vista-ready PCs are today). The next OS will take advantage of these features (and analogous ones on AMD, code-named Presidio) and will provide a whole new paradigm for security. This will leap beyond anything Apple can do and they will be playing catch-up, unless of course they start heading in this direction themselves.
To me as a security person, this is the obvious, inevitable path of OS development and is the only plausible thing Microsoft could be talking about. It should be very exciting to see these ideas brought to market in real systems.