First Details of Windows 7 Emerge
Some small but significant details of the next major release of Windows have emerged via a presentation at the University of Illinois by Microsoft engineer Eric Traut. His presentation focuses on an internal project called "MinWin," designed to optimize the Windows kernel to a minimum footprint, and for which will be the basis for the Windows 7 kernel.
But what about all that legacy crap in the Bios motherboard, when can we expect that some company will actual create a board without 15 year old technology or other obscure settings that is no longer used by anyone except maybe a 386.
The os might load fast with a bare minimum but what about the excess baggage of hardware?
has mac done this or is it just that the OS on a linux bas system is just plain faster.
now i know linux fans and mac fans will say that they already knew that but can someone provide hard facts
So Microsoft tells something about the next version of Windows not long after the people have noticed that their current version isn't all that it's made up to be?
I suffer from attention surplus disorder.
I can't help but wonder if this is a reaction to OS X being used on iPhone and iTouch(mySelf). Maybe they're trying to consolidate windows/windows CE. Or maybe this is just another feature that will be cut in favor of demanding a DNA sample before allowing you to access the internet.
Do you even lift?
These aren't the 'roids you're looking for.
This seems to coincide directly with some recent patents filed by Microsoft. It seems what they're truly after is an al-la-carte style OS where DRM is used to control the subscription of such "base OS" additions. Read more on the patent here, http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PG01&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.html&r=1&f=G&l=50&s1=%2220060282899%22.PGNR.&OS=DN/20060282899&RS=DN/20060282899
Basically, you purchase the base-system and tack-on additional subscription based modules. My concerns are how the subscription model will function, the subscription pricing, and the potential for removal of prior features such as 3D acceleration on the 'base' system.
It also appears that DRM will be used extensively in this model and will not be solely limited to music/video as previously thought.
Honesty, and I'm not trolling here, but this looks pretty scary. This reminds me of driver-signing gone awry. I don't see the potential for open-source/free modules due to item #3. Arbitrary application, memory, CPU, and process limits are also concerning.
The whole "add-on" 3D support as well as "don't limit my desktop to 5 open applications/processes" seems incredible. I imagine the base system will be usable to about 3% of the population and the subscription-based add-on modules may be pricey. I can't imagine a DRM style approach for 3D gaming/enthusiasts being acceptable. Imagine having to pay $20/mo for 3D + multiple core CPU + 2G RAM and the minute you stop paying all those modules expire and are no longer active until you resume payment; like Napster and other DRM based music models work.
-evilghost
If you're going to mod one of these posts up, pick this one.
;)
Also, notice that (with consumer releases), Windows seems to be following the even-odd rule? 3.1, meh. '95, good. '98, meh. '98SE, good. ME, ai f'thangan! 2k/XP, excellent. Vista? Pfft. Windows7? Good things to come.
Rampant carbon sequestration destroyed the Dinosaurs' tropical paradise. I'm here to help repair the damage.
Take it from a former Microserf - this "internal project" will be taken to the nearest corner and shot (and maybe also mutilated and spat on). When you have a huge turd of a codebase dating back 15 years in some places, the last thing you want to do is dramatically rehash it. Projects like this are DOA at Microsoft after the WinFS fiasco.
Good. Small kernel is a good start. Now make it open source and let me install whatever the hell I want for a desktop manager and applications on top of it.
I've been saying it for years now. Windows should either be an open standard for operating systems to be built or be a desktop manager built on a Linux kernel. Of course, then what would the diehards bitch about on slashdot?
The game.
At one point I had full XFCE desktop and latest (at the time) 2.6.xx kernel running in under 35 MB. That was a few months ago.
But I'll be the first to admit that it wasn't a typical install. I was going more for speed, but I compiled the kernel with exactly the set of drivers/modules I needed; and compiled X, XFCE, and most "important" system libraries myself. Base distro was Slackware.
I'm running a fairly standard Debian install right now, and with no apps running it'll use about 150 MB with X, Fluxbox, and some fairly "standard" background services.
I'll also point out that the 35 MB Slackware was running on a 32-bit Pentium 4, and this Debian install is running on AMD64. Doesn't make much difference, but enough that I thought I should point it out.
Maybe not
This is modded funny, but how about 'Reality'?
I honestly wonder if some of these posts aren't printed and used internally at Microsoft as either: cubicle decorations, motivation to make better code or ammunition to convince managers to improve the development process.
Please elaborate on the amusement?
I once had a crazy friend who was upgrading his Pentium III era PC, and got fed up with getting certain peripherals to work, so he started yanking things out of the PC with the power on. We then found that the sound card was hot plugable, it would dissapear and reappear in device manager every time.
I lost me sig.
HIRE SOME FUCKING UI EXPERTS.
Sorry to be shouting and all but I'm a Windows guy, I always have been a Windows guy, sure I have that slashdot bone in me, wanting OSS to be huge, great, free and out there for everyone to share and love but let's be realistic now, for some people it's not an option, myself included.
Honestly I have been really quite satisfied with XP (after becoming accustomed to its own issues)
However after having recently tried Vista (multiple times) it's a disgrace, PURELY from a look and feel perspective, it's like 500 people designed it around a board room table but consistency and ease of use just aren't even considered.
I'm definately NOT an apple man by any means, yet having now used OSX for a week and an ipod for a year, they just get (most) stuff right, logical and simple - just how it should be.
Vista is wrong, it looks wrong, some of you can whinge it sucks under the hood or perhaps DRM ate your babysitter, maybe it has poor performance copying files and playing MP3's (doesn't bother me) but that UI? Good lord if you can't make it better at least give us back the XP one as an option.
It's time that MS made some RADICAL changes to the user interface, crazy out there stuff, which is actually USEFUL! rather than just re-hashing the same old thing, stapling on some stuff (poorly) and expecting us to enjoy it.
looks like Mistersoftie is up to their old hype the vaporware [wikipedia.org] tricks to dissuade buyers from going with attractive alternatives.
:P
Because, of course, you can't wait to have MinWin on your machine - the Windows that does only one single thing: publish your tasklist via HTTP.
Hmmm, so much better than Leopard
Come on, it's just a tech demonstration, Microsoft in fact closed themselves solid after the release of Vista. Management thinks part of the bad reception of Vista is because they were so open about the whole process for the entire 5 years.
For some part they are right. We'd never know about the dropped features if they were never pre-announced. Most products plan various features that get dropped or deferred in the process of development.
We'd also be surprised at the Aero Glass UI, and the new security features.
What we'd be most surprised about though, is the lack of consistency in the UI and stability/performance issues. So I'm not sure Microsoft has the right strategy right now.
NT-OS/2 actually.
The way I understand it, the progression is as follows:
1.x (on DOS kernel) -> 1.01, 1.02, 1.03, 1.04
2.x (on DOS kernel) -> 2.03, 2.1 (Windows/286), 2.11 (Windows/386)
3.x (on DOS kernel) -> 3.0, 3.1, WFW 3.1, 3.11, WFW 3.11
3.x (new kernel, based on code originally slated to be OS/2 rewrite - known internally as NT-OS/2 for much of it's life, later simply WindowsNT) -> 3.1, 3.5, 3.51
4.x (on DOS kernel) -> 4.0 (Windows95), 4.1 (Windows98), 4.9 (WindowsME)
4.x (NT kernel) -> 4.0
5.x (NT kernel) -> 5.0 (Windows2000) 5.1 (WindowsXP)
6.x (NT kernel) -> Vista
7.x (presumably NT kernel) -> Windows7 (whatever it's final name is)
The story of Me is semi-interesting (actually not so much interesting as it is tedious). Basically, MS had intended NT 5.0 to be what XP actually became; they planned on merging the consumer and business platforms to a unified (NT) codebase (codenamed Neptune). NT 5.0's codebase was not finished with time enough to add all the features consumers needed, like compatability with 4.x and 3.x applications. The reason for naming NT 5.0 Windows2000 was to create a clear naming scheme for home users to upgrade, but by the time it became clear that 2000 would not be for home users, the name had already caught on. To prevent confusion, they revamped 98, adding Windows2000's icon set and interface enhancements, as well as a few odds and ends such as UPnP, to call it Millenium Edition, or Me.
Me really should never have existed, and it was quickly slapped together to avoid a marketing catastrophe. Unfortunately for Microsoft, Me is their Pinto. OTOH, it made XP that much more appealing.
Side note: NT starting at version 3 was a multi-reason decision, slightly influenced by it's relationship to OS/2, but much more due to marketing. NT started at version 3.1 because that was the version Windows was currently sitting at. Also, MS didn't want customers to think NT was less mature than OS/2 (then at version 2.1). NT got its name from the fact that they totally rewrote the OS/2 kernel, opting for a microkernel that was designed from the start to be portable and adaptable, with multiple "personalities" sitting on top of it e.g. a Unix personality, an OS/2 personality (IIRC there were even plans for a NeXT personality as well as a few others, though those never happened), etc. The concept was fairly new at the time, hence the name New Technology. NT 3.x included a 16-bit OS/2 "personality" and a basic POSIX "personality" both to ease migration to the new platform, but also to showcase NT's capabilities.
I'm an OS/2 fanboy turned Linux fanboy, and I've never seen the NT sourcecode, but from what I've read the NT kernel itself is a marvel of software engineering. It's the userland crap piled on it that makes Windows the lumbering beast we all love to hate.
We built an early 3.1 server soon after the release date, just to compare it with the Netware servers we used.
Since the system took 10 minutes to get to the login prompt, and the file and print services were atrociously slow, we soon abandoned the exercise.
It's not that long ago, though - only 15 years :P
Get off my lawn, you damn kids!
One swallow does not a fellatrix make
I should probably go one step further to point out that NT is probably the best (and most modern) general-use kernel in widespread use today. I'm no kernel developer, although after talking to people 'in the know', I get the general concensus that NT was one of the few things Microsoft got right and nailed on the head.
Although there's nothing fundamentally wrong with the Linux Monolithic Kernel (especially since it's behaving more and more like a microkernel these days), Linus has admitted that were he to start from scratch, it wouldn't be monolithic.
I don't know too many specifics of the OS X (Mach) kernel, although from what I understand, there are some fundamental performance and latency issues holding the entire system back that have existed in Mach since the beginning.
Although the software on top of NT is often less than stellar (ruined by the businesses execs, and trashed by the requirement for backward-compatibility), the NT kernel is generally regarded as being the most solid part of the operating system.
-- If you try to fail and succeed, which have you done? - Uli's moose
No, he's right and you're wrong. DOS was the underlying architecture and loaded before Windows 9x. Many (older) drivers needed to be loaded there (often soundcards) and Antivirus programs always embedded them in that part. The only thing that was different with WfW 3.11 was that you didn't need "win.exe" in the autoexec.bat.
However, that was easy to change and you could make Win9x boot in CLI by adapting a config file, I just don't remember which one. Also creating a bootdisk with 9x, gave you a 9x bootdisk that went straight to CLI and typing VER didn't say aynthing about DOS, it talked about Windows 95 but for all intents and purposes it was DOS. Heck it even said "Starting Windows 95" as first text upon bootup.
Apple: 15,810 Microsoft: 61,000 Microsoft has a little under 4 times the number of employees Apple has.
They both do Hardware: xBox vs Apple Line (I think apple probably has more employees on their hardware than Microsoft.)
They both do MP3: iPod vs Zune (It should be a wash in employee #'s)
They both do Office Suite: iWork vs Office (Office has obviously more employees than iWork)
They both do "Family" apps: iLife vs Microsoft Movie Maker, etc. (iLife probably has more)
They both do an OS: OS X vs XP/Vista. (With out a doubt XP/Vista has more employees on it than OS X)
You'd think that they'd be able to do something right. Heck AppleMaybe it's bureaucracy collapsing the whole thing. Maybe what Microsoft needs is a Steve, a dictator, someone that says what goes and no questions from above. Back in the day Apple wasn't run like this and we had Copeland and all other "Next OSes" there were some iffy products (OpenDoc). Then Apple bought NeXT. Steve came back and the rest is history. (And about 3000% in the stock market).