Slashdot Mirror
Five Ways Microsoft Could Change After Gates
Might Squirrel noted a perfectly mediocre story to chat about on a boring post-holiday weekend Monday. This one is a look at 5 ways Microsoft could change after Gates. From accepting Open Source to serious interoperability work, there are definitely 5 things on that list there. Nothing about my solid gold rocket car.
1,2,4: Wishful thinking
3,5: yes, of course
There's a fair amount of information on MS protocols and standards on MSDN - http://msdn.microsoft.com/en-us/library/cc216514.aspx
and their source-code is available for review at least @ http://www.microsoft.com/resources/sharedsource/default.mspx ... which 10 years ago would've been unthinkable; but yeah, although it's a far cry from the GPL, is still hugely better than it was.
throw new NoSignatureException();
Think about it - if you're making a clean break from Windows, would you choose a mature, well established alternative like Linux or MacOSX, or would you choose a completely new, unproven and completely incompatible and unstandardised operating system from Microsoft? Even if the new Microsoft OS is cleaner, being incompatible with EVERY operating system out there would absolutely kill it.
Microsoft have already done one kernel rewrite, going from Windows 9x/Me to Windows NT. They have no need to do another - the NT kernel is already more modern than a Unix style kernel. It's preemptible, reentrant and has fine grained locking, all the things you need for good SMP performance. User mode stuff has been tweaked over the years to add features and has probably been rewritten several times incrementally. But they aren't going to do a big bang rewrite of the user mode stuff or break compatibility because there's nothing to be gained and everything to be lost.
echo -e 'global _start\n _start:\n mov eax, 2\n int 80h\n jmp _start' > a.asm; nasm a.asm -f elf; ld a.o -o a;
There are another issue that I think is being overlooked is the 64-bit issue. This also adds to be backwards compatibility issues. Here's the way I understand it: The LP64 model (used by Linux and Unix) redefines long (32-bit) to 64-bit. The model MS chose is the LLP64 model which introduces a new integer type called long long which is 64bit. The effect of this is that a 32bit MS program will work in a 64bit Windows, but a 64bit MS program will not work in 32bit Windows. So companies who want to take advantage of 64bit Windows will have to develop 2 different versions of the same software. This hinders some companies from moving forward to 64 bit. In the LP64 model, a company would have to compile 32bit and 64bit versions but their code can be the same.
Well, there's spam egg sausage and spam, that's not got much spam in it.
Microsoft have already done one kernel rewrite, going from Windows 9x/Me to Windows NT.
Erm ... no. Microsoft had already finished the NT kernel when they decided to ditch the Win9x/ME "kernel" for the one in NT 4.0 and Win2K (NT came out ages before ME). It wasn't a kernel rewrite at all, just two different kernels running side by side until MS decided to kill the weak one and use the good one.
Erm ... no. Microsoft had already finished the NT kernel when they decided to ditch the Win9x/ME "kernel" for the one in NT 4.0 and Win2K (NT came out ages before ME). It wasn't a kernel rewrite at all, just two different kernels running side by side until MS decided to kill the weak one and use the good one.
The plan was to replace Windows 9x/Me from the start. In fact the original plan was that Windows 98 would be the last Windows based on the old kernel mode code and the transition to the NT kernel would be complete by Windows 2000. Windows Me was launched by popular demand. By the time Windows XP was launched the transition finally happened.
So the plan was always to kill off 16 bit Windows and replace it with an NT based OS. This wasn't quite ready as of Windows 2000 so they had to launch on extra 16 bit OS, Windows Me as a stopgap waiting for Windows XP.
echo -e 'global _start\n _start:\n mov eax, 2\n int 80h\n jmp _start' > a.asm; nasm a.asm -f elf; ld a.o -o a;
Well, let's think about how Apple just recently switched from IBM PPC to Intel processors. The old legacy PPC software still runs on newer Intel Macs thanks to Rosetta, and to most users there is no difference.
Agreed, an entirely new operating system would be a little bit of a bigger challenge, but if they can have an invisible Rosetta-like translation, they would allow users to slowly transition from legacy applications to the newer ones.
Microsoft would first have to make sure that all legacy apps run transparently on the new OS, and then release native versions of all their apps (Office, etc.). My guess is that most Windows users would keep using Windows in order to have the apps they're used to, and as a result, developers would have incentives to produce native software.
This space up for sale.
It appears the point completely missed you and apparently impaled an innocent bystander behind you. The XBox 360 has a long history which you're not properly accounting for. Before the 360 was the original XBox which did unexpectedly well especially considering the fierce competition of Sony and Nintendo. However before the XBox was Microsoft's work on the Dreamcast which did not do so well and sank Sega's hardware business. Before the Dreamcast was Microsoft's PC gaming division which only had a handful of real hits to its credit. Microsoft did what smaller companies could not do, fail repeatedly until they managed to get something working right. They were able to buy out game studios like Bungie and Rare in order to get some heavy hitting first party titles developed for their console. The XBox 360 is a good console because Microsoft has spent more than a decade struggling with a gaming business. XBox Live has a similar story, they bought out "The Village" which got rebranded Internet Gaming Zone which eventually became the MSN Gaming Zone and served as the conceptual basis for XBox Live. Again because of their size and money they could throw resources at a lackluster product and eventually make it stick. Other companies don't have that same luxury, look at Sega. Two successive market failures and it was lights out for their hardware division.
I'm a loner Dottie, a Rebel.
Seriously, are you going to trust a mission-critical program to run on a Windows box in the first place.
Sure. I have several in place right now. As do millions of other companies. Suggesting that it's a bad idea, or that it doesn't happen is FUD.
I don't respond to AC's.
You do realize that companies like Redhat and Novell(SUSE) exist right?
Well, there's spam egg sausage and spam, that's not got much spam in it.