Inside The Development of Windows NT

mrpuffypants writes "Winsupersite has a 3 part series this month about the history and development of Windows NT all the way up through Windows Server 2003. The author goes fairly in-depth describing how Windows is developed, managed, and how all 50 million+ lines are compiled daily. Part One covers the history of NT from its early days at Microsoft and Part Two discusses how the deployment of the forthcoming server version of Windows is coordinated daily." *shiver*

Comments

hmmm...

[garcia]

The stuffed mascot in the background looks an awful lot like someone else we know ;)

Re:hmmm...

[Anonymous Coward]

Sorry, but I think that's Waddle, the Beanie Buddy (a larger version of Waddle, the Beanie Baby). Penguins are cool to a lot of folks - and not all stuffed penguins are Tux.

Re:hmmm...

[Anonymous Coward]

don't you find it sad that you know that?

Plus, I think it was pointed out b/c it was the "war-room" and there was a penguin in it.

Re:hmmm...

[MaufTarkie]

Actually, I think it's "Tux Rux" (second entry under Penguin -- compare the grandparent's photo with TR's photo). They were selling these two years ago at our local Safeway supermarket, and I got one as a Christmas present then. When you push his belly, he will alternate with "Hi! I'm Tux Rux" and a very disturbing growl/purr that I can only assume is the sound that a very irate poked penguin would make.

NT compile script

[Limburgher]

First line:

#!/bin/bash

Re:NT compile script

[pyrros]

Nah, my money is on:

@echo off

Re:NT compile script

[csguy314]

First line:

#!/bin/bash

Line two:
#By compiling this program you agree to the following terms:

the book 'Show-Stopper!' Ready to Buy? Sign in to

[Anonymous Coward]

This book, from around '95 I think, was a great read. I don't know how much overlapit has with the articles mentioned above, but I enjoyed it.

http://www.amazon.com/exec/obidos/tg/detail/-/00 29 356717/103-3105313-8496655?vi=glance

Also available at your local library!

Lines of code != quality

[JPDeckers]

Quote from site:
There are 5000 developers on the Windows team generating over 50 million lines of code for Windows Server 2003. It's an enormous task, the biggest software engineering task ever attempted. There are no other software projects like this

Accordig to this, we should be getting them in a year or 5

NT == VAX OS?

[peter303]

I thought the initial NT "heavily borrowed" (MS tradition) from the Digital Equipment Corp (now part of HP) VAX operating system. Then it gradually incorporated parts of the evolving Windows/DOS OS.

Re:NT == VAX OS?

[guy-in-corner]

No. Dave Cutler, who was lead developer for NT, was previously one of the lead developers for VMS. I don't think that MS actually took any of the source code from VMS for NT, however.

Re:NT == VAX OS?

[scrytch]

I thought the initial NT "heavily borrowed" (MS tradition) from the Digital Equipment Corp (now part of HP) VAX operating system. Then it gradually incorporated parts of the evolving Windows/DOS OS.

That would be VMS (some VAXen ran Ultrix, poor things). IBM and MS started a collaboration called OS/2, then later decided to part ways. Whatever MS's other motives were in the split, MS was staking its entire future on what was to IBM a toy project, so MS wasn't entirely enthusiastic about development at IBM speed. IBM kept the OS/2 name, MS hired Dave Cutler from DEC, Cutler dubbed the new fork WNT: that's the letters after VMS, and any expansion is entirely a backronym.

NT does include some of VMS's heritage, including strong async I/O support throughout. The DOS stuff is really a matter of emulating the interface -- a whole lot of work went into making drive letters and backslashes work everywhere, believe it or not. Not surprisingly, it tends to share more in common with OS/2, with the supervisor design and the object manager for starters.

Re:NT == VAX OS?

[Marsala]

That's a pretty big oversimplification of the matter. Basically, David Cutler left Digital after DEC decided to ax a project he was working on to build a new chip and an OS to go along with it, Microsoft offered him a job to work on the chip and the OS, and Cutler managed to recruit most of his dev team from DEC for the new MS squad.

So saying that NT is just VMS part II isn't really accurate, but the same guilty parties are involved. If you can find it, there was a book called _ShowStopper! The Breakneck Race to Create Windows NT_ that does a pretty good job of chronicling the history of NT during its early days.

Know thy enemy, and all that. :-)

Re:NT == VAX OS?

[Bob Abooey]

Right.

One can't deny that NT was "based" on VMS.

In fact I recall reading somehere that MS was initially promoting NT as VMS on the desktop, or something along those lines.

Re:NT == VAX OS?

[SerpentMage]

What I am amazed at is that in the history of Windows NT Dave Cutler was cut so little space. He was mentioned twice. Once at the beginning of the article and then in the timeline.

And from what I remember about the hoopla, etc. Dave Cutler was fairly important in the first version of Windows NT.

But you know it shows just how little MS remembers the people who got it there in the first place!

Re:NT == VAX OS?

[leandrod]

> IBM and MS started a collaboration called OS/2, then later decided to part ways.

Actually MS stabbed IBM in the back by breaking the agreement (MS-W16 never to get 32 bits or a DOS virtual machine, OS/2 forever) without warning by launching MS-W3.1. This effectively killed OS/2 and made going from OS/2 3.0 NT to MS-WNT a no-brainer for MS.

> MS wasn't entirely enthusiastic about development at IBM speed

You mean IBM quality, not speed. OS/2 development speed wasn't so bad; in fact, they decided to go for a command-line only, i286 version initially in order to ship early.

The real problem is that MS wanted to push forward the stupid MS Win16 API inherited from MS Windows 1.0 for the IBM-PC/XT, and IBM knew it was garbage. Also MS realised it could get more money by selling a product without any interoperability or standardisation, while IBM would have made it interoperable, documented and stable -- as they still do with the current IBM OS/2.

Re:NT == VAX OS?

[JonK]

IBM didn't need Microsoft to stab OS/2 in the back: Big Blue did a perfectly good job of that itself.

IBM quality

You never tried running DOS apps in OS/2 1.0, did you. To say it was buggy was an understatement...

OS/2 development speed wasn't so bad

As long as you weren't waiting for applications which ran on your shiny new requires-4MB-of-RAM-costing-to-boot OS/2? Whatever happened to OfficeVision again ;-)

The real problem is that MS wanted to push forward the stupid MS Win16 API inherited from MS Windows 1.0 for the IBM-PC/XT

No, the real problem was that the marketplace didn't want OS/2: even when Microsoft were fully committed to it (the period thru' mid-89) and pushing it as the best thing since sliced bread, no-one (outside IBM fiefdoms) bought it. If you want to blame anyone for this state of affairs, blame the clone makers who had no interest in handing control of the PC marketplace back to IBM. IIRC, at the time and with all of DOS's flaws, and with both IBM and Microsoft telling the market that DOS was dead and OS/2 was the way forward, DOS was outselling OS/2 by a ratio of about 200:1.

Oh, and targetting the 286 was a bad idea from the get go, but IBM didn't want to eat into sales of their (expensive) AS/400 line.

while IBM would have made it interoperable

... as part of SAA - so it would have interoperated with... ooh, let's see... that'd be the IBM AS/400 minicomputer and the IBM S/370 mainframe and... do we all get the idea here?

documented

As long as you were prepared to fork over $3000 for said documentation

and stable

I get it: you never used OS/2, did you. Because if you did, you'd know perfectly well that when OS/2 shipped it was so flakey it made Windows Me look like a beacon of system stability. Maybe by OS/2 3.0 it had become moderately usable, but OS/2 1.0, 1.1, 1.2 and 1.3 sucked in so many different ways and had uptimes measured in minutes. Datapoint: a customer at the time had about 500 PS/2 Model 60s running OS/2 1.3 w/ Presentation Manager, one in each store. At the time, this was an investment, in hardware alone of about $3 million, plus about another third of a million in OS licences (yes, only ten years ago $700 was considered a very reasonable pricepoint for a desktop operating system licence - if you were IBM...) They were shockingly unusable - as in the store manager would be entering that day's sales into a dBase app and running the numbers and they'd be able to enter stuff successfully about one time in three - the other two, OS/2 or one of the apps running under it would crap itself, lock the queue and once that happened it was goodnight Vienna.

NT & VMS

[syr]

Here's more background information including VMS data.

GameTab - Game Reviews Database

WinNT development cycle.

[grub]

0) CVS checkout the latest net stuff from freebsd.org
1) Look at code and scratch head until "A-ha!"; enlightenment.
2) Merge code into Windows source
3) go to 0

Re:WinNT development cycle.

[ImpTech]

> 3) go to 0

Dijkstra is rolling over in his grave...

Re:WinNT development cycle.

[jeriqo]

> 0) CVS checkout the latest net stuff from freebsd.org
> 1) Look at code and scratch head until "A-ha!"; enlightenment.
> 2) Merge code into Windows source
> 3) go to 0

Damn! They use gotos in the development of windows?!
I know understand why it keeps crashing..

Re:WinNT development cycle.

[MaxQuordlepleen]

[---snip---]
tanzarian:/$ grep -r ' goto ' /usr/src/linux | wc -l
1543
[----snip---]

that's from 2.4.19

Re:WinNT development cycle.

[IamTheRealMike]

There are quite a few in Wine also. I think they're primarily useful as an equivalent to try {} finally {} in languages that support exceptions (obviously C doesn't). If a condition fails just goto the end, where the cleanup code is guaranteed to run.

Used properly, gotos are no more harmful than any other construct.

Re:WinNT development cycle.

[Anonymous Coward]

Microsoft C actually does support structured exception handling with _try/_except (along with termination handlers using _finally), which is vital to the design of the NT kernel. I think it was added to the MS C compiler specifically for NT, but I'm not sure. Anyway, the proper use of C exception handling in kernel-mode NT code (e.g. device drivers) is explained in the NT/Windows DDK docs.

A related point is that I do recall that MS tried to get exception handling added to C some years back (for the C89 standard, IIRC), but failed. It didn't make it into C99 either, unfortunately, although I don't know if MS bothered trying to get it in again. (And people wonder why MS so blithely ignore standards.)

As for gotos, I agree there's nothing wrong with them. Even when exceptions are available as an alternative, gotos can be considerably more efficient in certain cases (e.g. where you know exceptions will never be thrown).

Re:WinNT development cycle.

[nathanh]

tanzarian:/$ grep -r ' goto ' /usr/src/linux | wc -l
1543

I'm not surprised. Dijkstra's "Goto Considered Harmful" paper isn't gospel. Lots of programmers disagree with it, including core developers of Linux. In fact, I only ever find first year uni students quoting the paper and that includes myself back when I first read it :-/

There we have it

[Anonymous Coward]

We thought, 'How hard could it be to build an OS?' and scheduled 18 months to build NT. But we had forgotten about some of the important stuff--user mode, networking, and so on."

Either this means that the NT team were actually fairly clueless...or incredibly cocky. Either way, that seems like a pretty stupid thing to say.

Re:There we have it

[chrisseaton]

_All_ developers are cocky - very cocky. It's not just a Windows thing.

Re:There we have it

[Guppy06]

It could mean that NT was the first time they ever wrote their own OS...

Hmm

[CoolVibe]

Both articles feel like "feel-good" articles. There is little mention about IBM and OS/2, and the relationship between the two in the beginning of NT.

It's just a big advertising piece about how NT is the greatest thing since sliced bread. Sure, it has some entertaining facts, but I'm still not buying it.

Re:Hmm

[mrpuffypants]

Winsupersite is, for the most part, a very pro-microsoft website...however, even if their reviews and previews may be slanted a bit they still get very early releases of different products and write decent reviews of them...with lots of screenshots!

Re:Hmm

[tshak]

NO, it's a piece made by developers, you know, people who care about code, not all of the politics and conspiracy theory's around them.

Re:Hmm

[IntlHarvester]

The article pretty much hits on the main point of contention between IBM and Microsoft -- IBM just wasn't interested in their Windows project.

There was a simmering fight over whether OS/2 should be "Protected Mode Windows" or whether Windows should be "Presentation Manager for DOS". Since neither platform had that many users or developers at the time, it could have gone either way.

The key characteristic of the new API, eventually named Win32, is that, though it was a new API, it looked and acted just like the 16-bit Windows APIs, letting developers easily move to the new system and port their applications. "We made it possible to move 16-bit applications to NT very easily,"

While they make this sound like a Gee Whiz revelation, but in fact Microsoft wanted Windows compatibility in OS/2 from the beginning and IBM wanted a unique API.

Since IBM wore the pants, they won the day originally. However this really bit them in the ass with the the subsequent popularity of Windows 3, because it was difficult to target OS/2, so software was either missing or dismissed as a poor Windows port.

Not that Win32 was a huge success in the early years either -- most software had to be run in the Win 3.1 emulation, and even MS themselves only belatedly produced a 32-bit version of Office and not much other software.

Alpha

[deacent]

I see a lot of complaining in the article about how some architectures were not ready for NT on a timely basis (Intel i860, PowerPC), but I see no mention how they were so slow to bring NT to the Alpha. I recall that DEC actually ended up porting VMS to the Alpha because they were waiting on MS for their promised NT release. I'm a bit curious to hear from the developers about their perspective on that.

I've used both NT and VMS on the Alpha (as well as a Unix varient). NT is sooooo slow.

-Jennifer

Re:Alpha

[garcia]

I have actually heard that NT ran better on the Alpha UDB than other OSs that run on it. The Alpha UDB was designed to be a small NT workstation.

Linux runs fine on mine but from what I hear NT also runs on it fairly well. I guess we would have had to be running it when the UDB first came out to make any sort of educated descision on that.

Anyone have first-hand experience w/NT on the UDB?

Re:Alpha

[4of12]

how they were so slow to bring NT to the Alpha.

Really.

I was surprised to learn from the article how the early NT was so non x86 centric, shifting from i860 to R3000, etc. They even boast of the portability to different hardware because they weren't tied down to the x86 instruction set so tightly as were the 16 bit Windows developers at the time.

So, why, then, did the Alpha port of NT take so long? And, from what I understand, it relied heavily upon the ability of the early Alpha chip to run in some FX!32 compatibility mode to emulate the x86 instruction set.

The Alpha/NT story just doesn't seem to add up to me. There's some missing dark matter.

Re:Alpha

[Elbereth]

Yeah. NT on the UDB was actually pretty tolerable. I ran NT4 Server on a 166 MHz 21066 (first generation Alpha). I found it to be quite usable. I didn't keep NT4 on the Alpha for all that long, as this was just an experiment. I had the NT4 disc, I had a UDB, and I had some time to waste.

You have to remember that NT4 was a 32 bit operating system, even on the Alpha. Therefor, you didn't really gain much by going to the Alpha, except for some nice speed boosts (it was definitely the fastest CPU on the market for years).

It was similar to running NT4 on a Pentium Pro 166 or 200.

The biggest problem I had was finding software. However, everyone's favorite telnet app, putty, comes compiled for NT4/Alpha.

The Register previously offered Windows 2000 for the Alpha, if you asked them for it. I never did, since my UDB was seriously underpowered (128MB RAM, 166 MHz).

I can see the round table discussion now. . .

[bplipschitz]

"You compile it today."

"No way--*you* compile it!"

"No way! Hey--let's get Mikey, he'll compile *anything*!"

VMS + 1 = WNT

[jjga]

It is interesting how incrementing each of the letters in VMS gives WNT. It is something similar to IBM and HAL.

Security?

[elliotj]

"By late 1989, the NT group began growing. They added a formal networking team and expanded the security team beyond a single individual who, incidentally, had also been previously burdened by file system and localization development."

You mean they've got more than one guy working on security for Windows? Oh come on, who's gonna believe that?

Re:Security?

[t0ny]

I think if you compare NT with the security of a lot of platforms that came out around the same time, you will see a very similiar set of holes.

I still remember all the programs all over the internet you could use to grab NetWare passwords, so its not like MS was the only one with holes.

What is a difference, however, is that people are still using NT4, whereas none of its contemporary OSs are around anymore (or at least, not around in a significant installed base).

Re:Security?

[stratjakt]

Thats a point many (purposefully) ignore.

NT was designed in a pre-internet era, where security for the office PC's was still mainly the guns-and-guards model. Users and passwords were there (from your average managers POV) to keep Jim from accidentally deleting or overwriting Sally's spreadsheets.

None of the 'holes' can be exploited if there's no access to the system whatsoever. It was a non-issue until the .com explosion when everyone decided that every digital device needed to be plugged into the interweb.

Re:Security?

[SuiteSisterMary]

Actually, it's more like 'the fact that your vest isn't bulletproof doesn't matter until somebody invents the gun.'

UNIX had the exact same problem; the entire point behind UNIX was that it was MULTICS with a bunch of the security stuff REMOVED. Go look on some old security lists; many daemons, such as sendmail or lpd, would give you root just for the asking.

Compiled?

[patvan]

I thought it was forged deep within Mt. Doom...

Re:Compiled?

[moosesocks]

From a rec.humor.funny posting (slightly edited) by dated 19 March 1996.

What you did not know about Windows NT

Recently one of my friends, a computer wizard, paid me a visit.

As we were talking I mentioned that I had recently installed Windows NT on my PC. I told him how happy I was with this operating system, and showed him the Windows NT CD. To my surprise he threw it into my microwave oven and turned the oven on. Instantly I got very upset, because the CD had become precious to me, but he said: 'Do not worry, it is unharmed.' After a few minutes he took the CD out, gave it to me and said: 'Take a close look at it.' To my surprise the CD was quite cold to hold and it seemed to be heavier than before. At first I could not see anything, but on the inner edge of the central hole I saw a inscription, an inscription finer than anything I have ever seen before. The inscription shone piercingly bright, and yet remote, as if out of a great depth:

12413AEB2ED4FA5E6F7D78E78BEDE8209450920F923A40EE 10 E510CC98D444AA08E1324

'I cannot understand the fiery letters,' I said.

'No, but I can,' he said. 'The letters are Hex, of an ancient mode, but the language is that of Microsoft, which I shall not utter here. But in common English this is what it says:'

One OS to rule them all, One OS to find them,
One OS to bring them all and in the darkness bind them...

History of Windows

[Toasty16]

Bill Gates: "We need an OS that doesn't suck."

Engineers: "No problem, we'll release betas every year and you can sell them to the public for the price of a finished product."

Bill Gates: "Good idea. What do you think Steve?"

Steve Ballmer: "Developers, developers, developers, developers, developers, developers, developers, developers, developers, developers, developers, developers...*wheeze* *hack* *cough*...."

Bill Gates: "It's ideas like those that will make you CEO in 10 years."

Why do Microsoft reviewers always sound...

[defile]

...so full of shit?

To step around the topic for a second:

Paul Thurrott's SuperSite for Windows is dedicated to providing all of the information you need to evaluate Microsoft's current and upcoming Windows operating system technologies. These exciting products include Windows XP Service Pack 1 (SP1), Windows XP Media Center Edition (code-named Freestyle) Windows XP Tablet PC Edition, Windows Media 9 Series (code-named Corona), and Windows Server 2003, which will launch in April.

Sounds like it'll be an EXCITING, unbiased, hard hitting, honest review to me!

Maybe that's not the best example. But even when you read technical treatises on Microsoft technologies the authors always manage to pack in gushing, surrealist praise.

Wasn't there even a book? THE AWESOME POWER OF DIRECT3D? Amusingly enough, it was released several months after John Carmack and the rest of the gaming industry started bitching Microsoft out for pushing Direct3D over the clearly superior OpenGL.

I'd hate to be all conspiracy here, but damn it's either that or believing that all Microsoft reviewers/writers are really stupid.

Re:Why do Microsoft reviewers always sound...

[Anonymous Coward]

>> John Carmack and the rest of the gaming industry started bitching Microsoft out for pushing Direct3D over the clearly superior OpenGL.

Rest of the gaming industry? From my viewpoint it was Carmack alone.

Clearly superior OpenGL? Depends what you're using it for. OpenGL certainly wasnt (and still isnt in many cases) faster on consumer level cards. Direct3D was developed alongside consumer level hardware supporting features that actually exist, OpenGL was designed on paper.

By and large 3D gaming was being written for glide, and developers absolutely loved an open api specifically targetted for game development.

Re:Why do Microsoft reviewers always sound...

[Alric]

I have had this thought myself, more often lately. I have come to the conclusion that it's probably not an active conspiracy by MSFT. Instead, I think it is a passive effect of a monopoly system.

My example:
Reviewer A writes a technical summary of some new MSFT product. Reviewer A invested months learning this new product and how it fits into MSFT's overall strategy. Reviewer A runs a consulting firm that specializes in MSFT products. That firm has invested time in training its people to know the new MSFT product. Reviewer A is probably not conciously being unethical, but he needs people to use this new MSFT product so his firm can make money helping companies solve the new problems that this product created. He writes a review/book that highlights the good points and downplays the bad points.

So, his review is biased, but it's not exactly a conspiracy by MSFT.

Re:Why do Microsoft reviewers always sound...

[kfg]

This isn't a conspiracy. It's an effect of what my granny would have called "knowing what side your toast is buttered on."

It's endemic in the entire "review industry." In fact, it's rampant in the media industry. Do you think The Filthy Critic gets invited on junkets, or out on the yacht with all the hot, willing little starlets?

Many reviewers are nothing more than karma whores. That's the reason for the founding of Consumer Reports.

The ones who are not get "modded down" to the fringes and you're less likely to even come across them.

That's the way it is, that is the way it shall be.

KFG

Punted to longhorn

[dracken]

"On the day I attended, one feature group had four of its bugs punted to Longhorn because they had failed to shown up for War Room. When someone argued that they should be given another day, Wanke simply said, "F#$% 'em. If it was that important, they would have been here. It's in Longhorn. Next bug."

Did one feature group have its *feature* postponed to longhorn or the *bug-fixes* postponed to longhorn ? hmmmmmm interesting.

Re:Punted to longhorn

[thoth]

It is hard to tell. Bugs are filed for the obvious (software errors) but also things like new features desired, performance enhancements to be had, and occasionally things like rearchitecture needed (say, redo horribly confusing UI to something better).

It is true though, when the war room meets you have to have somebody there to vouch for any fixes checked in to resolve bugs. Mostly the war room wants to hear about impact, if the fix was tested, any issues that arose from the testing (regressions and/or new problems), and if the fix is really needed.

best quote from the article

[babycakes]

"For Windows Server 2003, the War Room is run by Todd Wanke, who we eventually found to be an amazingly likeable guy. However, in the hour-long War Room sessions, Wanke rules with an iron fist" :)

Re:best quote from the article

[Anonymous Coward]

Wanke rules with an iron fist

That's gotta hurt...

Incremental build?

[BenjyD]

"...compiling and linking it into the executable and other components that make up a Windows CD is a 12 to 13 hour process that is done every day of the week

So they rebuild Windows from scratch every day? Somebody send them a copy of make, please.

Re:Incremental build?

[BZ]

If you look at other large projects of this type (eg Mozilla), both clean builds and dep builds using make are done on automated build systems. The two types of builds will find different types of issues.

Re:Incremental build?

[SnowDog_2112]

In all the software groups I've been involved in, it's considered good practice to do a full clean build nightly. Doing incremental builds is fine for developers, but when you want to make a drop that goes into an automated testing suite, etc., you do a full clean build each time, "just to be safe."

Re:Incremental build?

[sheldon]

"to me rebuilding libX throws another wrench into the fire."

Exactly!

If rebuilding libX is going to cause you problems, then you want to know about that NOW and fix it. I don't see any benefit to waiting to address build issues. Do you seriously think you're going to improve your productivity?

Re:Incremental build?

[thoth]

They use build.exe which winds up calling nmake, Microsoft's make.

The build process is divided up as well, each major group has a build lab (e.g. COM, Networking, kernel, shell) where they build their part of the product daily. Builds mean generating free, checked, for all SKUs (server, advanced, datacenter, etc.) and all CPUs (x86, ia64, amd64). When each group's build achieves a level of stability/quality, they integrate their changes into the master build lab, and eventually there is a reverse integrate from master to group to pick up other group's changes which might effect your area.

There's a morning and evening build, but we just started to ignore one of them because it was way too much time to be installing and/or upgrading twice a day.

So "they" aren't just building from scratch every day, the master build lab is building 5+ SKU's, free/checked, for 3 CPU's every day. And the group build lab's (called virtual build labs) are doing the same, but just for their area ;)

Re:Incremental build?

[SnowDog_2112]

"my software is being updated, but libX isn't, then i wouldn't want to recompile libX"

I guess it depends on exactly how much you trust the software that decides what depends on what and what's changed. Can you honestly tell me you've never had a problem that doing a "make clean" fixed?

All I'm saying is that for the groups I've worked in, the cost of having the automated script do a "make clean all" as opposed to "make all" each night is considered acceptable for the peace of mind that knowing that absolutely every change is accounted for and there's no possible dependencies that got dropped in by make.

Re:Incremental build?

[sharkey]

So they rebuild Windows from scratch every day?

That explains why MS tech support thinks simply reinstalling Windows every day isn't a big problem.

Did someone say cocky?

[Anonymous Coward]

...the War Room is run by Todd Wanke...

Oh dear. Poor Todd.

The NT Kernel Is Good

[DakotaSandstone]

I know this is the equivalent of Flamebait on /., but the NT kernel (borrowed though it may be from other OS ideas) is actually darn good.

Passing IRP's (IO request packets) between drivers creates a much more well-defined interface that a bunch of globally namespaced functions just calling each other (like some other OSes we all know). It also lends itself to a layered driver model (Bus Driver, Physical Driver, Functional Driver) much better.

I really like the NT Kernel. What driver developers do with it isn't the kernel's fault.

Re:The NT Kernel Is Good

[Elbereth]

The NT kernel was good. Then Microsoft moved the GUI into the kernel. We all know what happened after that. Okay, to be fair, the NT kernel is still pretty nice, but it's deeply annoying that Microsoft is so willing to sacrifice stability for a little more speed. I find it difficult to crash Win2k and XP, but it does happen... mostly from PC games.

Re:The NT Kernel Is Good

[stratjakt]

I partially agree. I would have preferred to see it implemented both ways, then let the user/admin decide.

If I want to run the GUI and video drivers in userspace for stability on a server or workstation, then I could. But the kernel mode is definately a huge speed boost for gaming, multimedia (remember how impossible it was to do anything like that under NT 4.0?)

All in all, I probably like it better the way it is. I've never had a problem so long as I stick with WHQL certified drivers. It'd be nice to be able to choose.

Re:The NT Kernel Is Good

[DakotaSandstone]

Good point. That was a mixed blessing. But then again, so is the fact that any 3rd party can write a kernel driver (although MS is trying to assuage this with things like driver signing)

I've read some interesting defenses of moving GDI to the kernel. Some of the rationale was:

• GDI crashing, be it in the kernel or user mode, is basically a fatal system error. As designed, it is not clear how NT could "restart" a crashed user-mode GUI subsystem. Even if GDI is user mode, if it crashes, you'd probably have to reboot anyway.
• There are many other complex subsystems that exist in the kernel, and have been made pretty bulletproof (scheduler, disk subsystem). What makes the graphics subsystem any more dangerous? Yes, moving code to the kernel requires less buggy code, but we're Microsoft, and we're up to the challenge. It can be done.

Personally, I would love to see an OS take advantage of more than 2 of the 4 "rings" an x86 processor has. In such an OS, one could theoretically have a driver crash, and could still recover.

Until that day, though, I agree - GUI subsystem code is hard to make bulletproof, and moving said code into the "sacred" kernel is pretty gutsy.

Re:The NT Kernel Is Good

[dildatron]

I have crshed Win2k Advanced server quite a number of times. Here are the latest few I can remember:

Had three downloads coming down with Mozilla, nearly maxing out my connection (each about 700KB/s). Then, the machine bluescreened and said TCPIP.SYS (or something similar to do with networking... i forget exactly what it was) had dumped.

Another instance was when I accidentally started 3 instances of outlook. The machine just bluescreened and rebooted. these are two different machinines. They are dual boot Linux/Win2k, and Linux works fine doing the same (type of) thing.

I have also crashed linux, so don't think I am just trashing the NT kernel. Linux and NT could both use some work to get where some of the commercial unixes are. But, I think Win2k and Linux are fairly solid, but there is definately room for both kernels to improve, and we don't need to get started on the older VM problems either.

I remember one day about a year ago when we managed to crash an N-class HPUX server. I thought the world might end - it is that stable. We found out it was a hardware failure that didn't manifest until several months down the road. Still, HPUX and Sub boxes are a lot more stable than windows or Linux, luckily that gap is getting smaller though and we get more stability for less money.

Re:The NT Kernel Is Good

[IamTheRealMike]

I really like the NT Kernel. What driver developers do with it isn't the kernel's fault.

Well....... maybe.

I seem to recall an MS employee claiming that it was entirely Microsofts fault Windows was so unstable, even though crashes were normally caused by faulty drivers. His theory was that if MS were more open with the kernel code, driver manufacturers could work more closely and easier with them, and the overall stability would go up. Instead what happened (they claimed) was that they would investigate a crash, find that some dodgy driver was screwing about with the kernel and so they'd tighten up the interfaces, get even more secretive with the code. The driver developers, faced with a brick wall, would then invent even more elaborate (and fragile) hacks to do what they want, so the stability went down, not up.

So, you can't really blame the kernel as a thing per se, but perhaps you can blame the management of it. Linux is now facing a similar problem with the growth of binary only drivers - they tend to hook into the ksyms and cause extremely hard to track down bugs, which is why they are no longer allowed to use those hooks.

Re:The NT Kernel Is Good

[Billly Gates]

"The NT kernel was good. Then Microsoft moved the GUI into the kernel",

Boy, its a good thing that no frame buffers are in the Linux kernel.

argument clinic

[Mr_Silver]

If there are one or more bugs in IIS, for example, a representative of the IIS team needs to be present to not only explain the merits of the bug, but whether customers are affected, how the fix might affect other parts of the system, and how soon it will be fixed.

To be honest, I don't see why they just don't hold these bug fixing meetings around the IIS guys desk :o)

Interesting...

[tshak]

There are 5000 developers on the Windows team generating over 50 million lines of code for Windows Server 2003.

I think it's safe to say that they're most defniitely _NOT_ using VSS!

50+ million lines of code?

[grub]

Remember a while back when Bill mandated a whole month of nothing but bug fixes? I find it highly unlikely that their people were able to go through 1.667 million lines of code per day (assuming a 30 day work month. I'm generous)..

Re:Not stupid. . .

[Bastian]

. . . just naive and inexperienced.

You know how to an 8-year-old boy, his dad's favorite sports team is the greatest thing in the world, able to turn lepers to supermodels and bath beads into geltabs? It's basically the same phenomenon.

It stops being amusing after a couple years reading the /. GNU/Linux crowd do the same thing.

Re:Finally...

[StressedEd]

Off topic but....

Give a man a fish, he owes you one fish.

Teach a man to fish, you give up your monopoly on fisheries.

My favorite...

Build a man a fire and he will be warm for the night. Set a man on fire and he will be warm for the rest of his life.

Michael Landon Is My Cousin

[Acidic_Diarrhea]

I think that is because the articles are trying to give a very general overview of the life cycle of NT. The various corporate dealings of Microsoft aren't of consequence when discussing how a huge project such as NT gets off the ground and how new demands cause new solutions to be found. And the article did mention that the choice to go to Win32 rather than OS/2 helped to sour the relationship between IBM and Microsoft. What more did you want?

And while you're right that this article is a very happy view of NT, it's interesting from the standpoint of how a project grows and new versioning control systems are added to handle such growth. Sure, the articles are heavy on fluff and light on details - but Microsoft is closed source. They're not going to give you much more. I honestly am not sure why you're so upset with these articles.

Re:Michael Landon Is My Cousin

[AndyElf]

> Windows NT 3.51 had an OS/2 2.x subsystem in
> there.

So did NT4, so does (seems to at least) 2K. When you start the installation, the message about loading it into still flashes on a screen.

> what I don't understand is WHY Microsoft abandoned
> all those other platforms (MIPS, Alpha et al) and
> decided to go with IA32 only

Agree with you here 100% -- especially considering that at the beginning they spend quite some time praising modularity, etc. -- how easy it was to port to MIPS/x86/Alpha! Then they seem to have realized that maintaining all of that is not as easy as porting...

Re:Michael Landon Is My Cousin

[SN74S181]

Linux zealots tearing at Windows all the time is just as stupid and as much FUD as Windows zealots that tear at Linux all the time.

There's a history behind this phenomenon in the 'Linux community.'

In the early days of Linux, everyone involved was a Unix enthusiast. There was a snobbish tendency at that time to dismiss anything Microsoft, but there wasn't the anger and hostility that there is now.

As the Microsoft Batallion rolled on, the expatriates from all the losing platforms (Amiga, Atari, OS/2, some Mac people) crowded onto the Linux 'ship.' Now there is a solid subsection of the 'Linux community' that actually represents the majority of vocal Linux users, who are bitter 'Anything-But-Microsoft' naysayers.

These people really foul things up. They bring a strong taint of negativity to the community. Much of what they constantly push Linux to become (these are the 'We need ONE UNIFIED DESKTOP to DEFEAT MICROSOFT' people) actually ends up hurting Linux and Open Source.

It sure used to be a lot more fun to run Linux.

Re:Michael Landon Is My Cousin

[Locutus]

"Microsoft wasn't in the mood to expend the effort to support platforms that couldn't carry their weight."

In 1994, they sold less than 400,000 copies of NT and had less than 700,000 total market for NT. IIRC. Not a very large market but they had just finished putting 100's of millions of dollars into MARKETING Windows 95( aka Chicago ) and IBM said it was going to push OS/2 into the server market. 100's of millions of dollars were dumped into MARKETING Windows NT into the server space. They are very much like vultures. They feed on the technology and markets of others. Go throw $50 million at a market and you'll see Microsoft throw $100 million at it while they figure out what the market is and how it make it only work on MS Windows.

Yup, the reason a company like Microsoft left the MIPS/PPC/Alpha/etc market was because there wasn't anybody in that market to kill. They weren't very large and Apple killed the PREP & CHIRP platform for the IBM/Apple/Motorola partnership and the last hope of an alternative DESKTOP hardware platform. When the $$ behind that market dried up so did Microsofts interest outside of x86.

There were very large companies putting $$$ behind those platforms and Microsoft had to be sure they'd be there to crush any other OS on those platforms. When the big $$$ leaves, so does Microsoft. They are not going after Linux because a bunch of developers like messing with it in their basements. It's because there is big $$$ behind it now and $$$ being paid for it as a solution base. IMHO.

LoB

Inside the Development of Windows NT?

[labratuk]

"My god, it's full of crap!"

NT source

[ptaff]

Oh, so now we learn that NT is not from "New Technology".

So in a couple of years we'll learn that:

• ME: Miserable Everytime
• CE: Cramped Environment
• XP: Xor Performance
• Office: Other File Formats Imply Collaboration: Encrypt!

Re:NT source

[DeadBugs]

And of course when you combine their server (NT) with the Desktop (ME) and portable OS (CE) you get predictable results

CE-ME-NT --- Cement.

Developer Count.

[Grendel Drago]

You forgot the five thousand developers.

Each person would need to review 50,000,000/(5000*30) = ~333 lines of code per day. Not quite so intimidating.

--grendel drago

Re:Developer Count.

[delus10n0]

[chanting] Developers, developers, developers, developers!

Scariest quote:

[pmcevoy]

"The first two weeks of development were fairly uneventful, with the NT team using Microsoft Word to create the original design documentation...Finally, it was time to start writing some code."

Does anyone else design an OS in two weeks?

Microserfs find solid products "unrewarding".

[Lethyos]

The most amusing bit I found was this:

"NT 3.51 was a very unrewarding release," Thompson said, contrasting it with Daytona. "After Daytona was completed, we basically sat around for 9 months fixing bugs while we waited for IBM to finish the Power PC hardware. But because of this, NT 3.51 was a solid release, and our customers loved it."

How horrible for a monopoly software company to have its programmers sit around and do bug fixes! My God, how ever did they survive? Fixing bugs at Microsoft must be like... Hell.

Re:Microserfs find solid products "unrewarding".

[mccalli]

C'mon though - bug fixing is dull. That's what they meant by unrewarding - personal satisfaction. They weren't quibbling the necessity.

Cheers,
Ian

Re:Microserfs find solid products "unrewarding".

[kawika]

Yes, unfortunately, a lot of programmers seem to believe that scratching their own itches is much more important than scratching their customer's itches. If they could enjoy the satisfaction of coding the perfect algorithm AND the satisfaction of making their customers happy, they would be more likely to succeed.

Is their focus security/stability or marketing?

[Mostly a lurker]

A year ago, Micro$oft claimed the future focus was security and stability. According to Part 2 of the article, the biggest issue now is the name change (from Windows .NET 2003 to Windows Server 2003). So, is this change for greater stability or something to do with marketing?

Thank that fscking IIS guy...

[siskbc]

...for not showing up to the goddamned meetings with his bugfixes.

Re:Dave Cutler's "Vision"

[nemaispuke]

I remember reading an article in 1992 where Bill Gates said "Windows NT will be a better Unix than Unix". So I don't think the vision was limited to Dave Cutler. I am still wondering how Bill could say that when Windows NT/2000 has minimal Posix support (1000.3 System Calls), is single user multitasking (unless you use Terminal Services, another Microsoft product you pay through the nose for), and has the worst scripting language ever! I guess that is why MKS (MKS Toolkit), Interix (OpenNT), and Cygwin are around, to fill the gaping holes in the "vision" of Bill and Dave and bring Unix to Windows because Microsoft can't or won't do it!

Bug handling

[vrt3]

"NT 3.51 was a very unrewarding release," Thompson said, contrasting it with Daytona. "After Daytona was completed, we basically sat around for 9 months fixing bugs while we waited for IBM to finish the Power PC hardware. But because of this, NT 3.51 was a solid release, and our customers loved it."

Does that mean they only solve bugs when there's nothing else to do?

Linux?

[13Echo]

"Getting all those people going in the same direction, cranking out code, is an enormous task. Building the results of their work, compiling and linking it into the executable and other components that make up a Windows CD is a 12 to 13 hour process that is done every day of the week. It's the biggest software engineering task ever attempted. There are no other software projects like this." And Microsoft compiles the whole thing--all 50+ million lines of code, almost every single day, he said. "We're evolving the development environment all the time," Lucovsky noted."

I think that they were forgetting about that orher large-scale software engineering task, in which thousands (tens of thousands?) of people crunch out code and compile software every day.

What's that called?

Linux.

Re:Linux?

[SuiteSisterMary]

Yes, and they quite clearly say that they're not compiling the NT kernal, they're compiling the contents of the CD; everything from notepad, calc and solitare to the admin tools and everything in between.

Damn lightweights

[YAN3D]

There are 5000 developers on the Windows team generating over 50 million lines of code for Windows Server 2003.

50,000,000 lines of code / 5,000 developers= 10,000 lines of code for each developer/

Spread that over a 3 year development cycle. Thats only 9.13242 lines of code a day per developer! How much are they getting paid? Sign me up!

It lost its independence with 4.0

[idiotnot]

In the same article, it mentions the changes with the graphics subsystem in NT 4.0. IIRC, NT 4.0 only supported i386 and Alpha, while NT 3.51 those, plus ppc and mips.

They moved the graphics subsystem into the kernel, and it ceased to be a microkernel. When pretty much everything lives in userland, portability is pretty easy. In fact, you can essentially write a new kernel (with the same external interfaces) for each architecture if need be. You also get neat features like being able to restart networking or the graphics system if they crash, without bringing down the system.

The problem that you have on i386 is that context switching is expensive (read: slow as a dog). On other platforms (sparc, ppc), it's not that big a deal.

Now, Windows doesn't look like a microkernel at all. And it's not at all portable, either. From what I understand, the Itanic port is giving them big headaches, and Intel is none-too-pleased about it.

Tux, undercover...

[Capt_Troy]

Looks like one of our guys is on the inside. Caught him on film. He's infiltrated the war team. Check him out, in the background by the TV...

See it here

Secret .NET language leaked

[jsse]

When someone argued that they should be given another day, Wanke simply said, "F#$% 'em. If it was that important, they would have been here. It's in Longhorn. Next bug."

C#, J#, S#....now we have F#....shall we pronounce it "Fuck Sharp"?

Competition

[sql*kitten]

It's really interesting to see how Microsoft actually relate to their competitors. They wanted to run on PPC, but IBM messed them around. They wanted to work with Novell, but Novell weren't interested. Even Intel failed to deliver on the promise of i860.

Given that, is it any wonder that MS would rather do things "in house" than rely on third parties?

Only the Linux KERNEL is 5 mln lines

[stevenp]

Only the size of the Linux KERNEL is currently about 5 mln lines, while the size of the WHOLE Win NT is about 50 mln. lines. This includes the whole GUI, OLE/ActiveX/COM+/InternetExplorer/OpenGL/DirectX and so on shebang, and also a myriad of printer/camera/scanner/etc drivers included in the default install.
The size of the WinNT kernel is nowhere near the 5 mln lines of code, I believe it is well below 1 mln. lines.
The WinNT is also only compiled for Intel platform, so it does not include code for other platforms.

Where's the profit?

[YetAnotherName]

That "goto" in line 3 prevents lines 4 and 5 from working:
4) ???
5) Profit!

(Not that Micro$oft needs anymore of that.)

don't be happy

[newsdee]

That mascot is probably reserved for voodoo rituals :-) Geek or not, it's still MS... :-)

Bug fixing is "unrewarding"

[jetmarc]

> "NT 3.51 was a very unrewarding release," Thompson said, contrasting it with
> Daytona. "After Daytona was completed, we basically sat around for 9 months
> fixing bugs while we waited for IBM to finish the Power PC hardware. But
> because of this, NT 3.51 was a solid release, and our customers loved it."

I wonder why I think so bad about Microsoft products?!

Re:it's tux, troll.

[frenetic3]

Waddle does not have a red eye. [iluvcollectibles.com] At least not on the TY site and he was discontinued in 98. You must have been paid to make that silly remark, that or very stupid.

Where is my "+1, Huge Nerd" mod option when I need it... :P

-fren

the sheer numbers of developers involved...

[davejenkins]

I am astounded at the sheer number of developers involved-- well, let me clarify that-- the number that are being paid, overseen, and managed by a single entity for this code. Clearly we know where all the money from those licenses is going, but it's structurally flawed: as software evolves, it will take an increasing number (linear? geometric? exponential?) of developers to build and maintain that OS. However, by trying to maintain them all under one roof, with one management structure, one 'political system' if you will, will always either make the process needlessly inefficient or horribly expensive.

I am reminded of the massive engineering projects the Soviets used to do just because they could-- it wouldn't make sense in terms of feeding their people or making their lives any more secure, but they did it because the central planners knew they could plan it.

This seems similar-- NT will become such an incredible beast that the bureaucracy to maintain it will suffocate it, or they'll start taking shortcuts.

Branding Issue Bugs?

[RoboLobster]

On that day, WinServer 2K3 had just a few active bugs, and at least a quarter to one-third of those bugs were simple branding issues.

What is a branding issue bug?

Re:Branding Issue Bugs?

[odaiwai]

A 'branding issue bug' is when you have morons who hard code the name of the OS into the source files instead of referring to a specific variable or a fixed file.

As a quote from the interview says: "I went out and handpicked the three best developers on the team and said, 'just go and fix it.' One developer fixed over 7,000 references to [Windows] .NET Server. Let's just say that there are people I trust, and people I don't trust. I told these guys, 'don't tell me what you're doing. Just do it.'"

So clearly a lot of the developers are hard-coding certain things into the code rather than relying on a solid design document. Sloppy, very sloppy.

dave

ReactOS

[dcuny]

Why not have a look at ReactOS, an open source clone of NT?

Unlike some doomed attempts to make a "better" Windows clone *cough*Freedows*cough* that degenerated into a puff of vaporware, the fine people at ReactOS have been keeping their noses to the grindstone and quietly worked away at getting an NT clone working. It's still a long way from replacing NT, as this screenshot of the one and only GUI application shows.

But if you want a free and open look at Inside the development of [a] Windows NT [clone], ReactOS is a good place to look.

They've done a number of things right:

• Shut up and coded...
• Picked NT as a target instead of the more glitzy Win9x
• And coded...
• Focused on core features instead of the GUI
• And coded...
• Borrowed from Wine where it seemed sensible

Did I mention they spend thankless hours coding?

How MS "punishes" bug meeting truants

[dpbsmith]

Let's see if I've got this right:

"This late in the development process, bugs are often passed along, or "punted," to the next Windows release--Longhorn--if they're not sufficiently problematic."

"The atmosphere in War Room is intimidating, and I spent most of my time in the room, silent and almost cowering, praying that Wanke wouldn't turn his attention to me or my group.... The most virulent treatment, naturally, is saved for those foolish enough to blow off a War Room meeting. On the day I attended, one feature group had four of its bugs punted to Longhorn because they had failed to shown up for War Room. When someone argued that they should be given another day, Wanke simply said, "F#$% 'em. If it was that important, they would have been here. It's in Longhorn. Next bug."

So... in this macho atmosphere, reeking of testosterone... the punishment for not being that the bug meet is that... YOU DON'T HAVE TO FIX YOUR BUGS UNTIL THE NEXT MAJOR RELEASE?????????

Words fail me...

Re:How MS "punishes" bug meeting truants

[MeanMF]

So... in this macho atmosphere, reeking of testosterone... the punishment for not being that the bug meet is that... YOU DON'T HAVE TO FIX YOUR BUGS UNTIL THE NEXT MAJOR RELEASE?????????

The NEW FEATURE that the group was working on gets moved to the next release.

Re:How MS "punishes" bug meeting truants

[BigJimSlade]

It sounded to me like this was more of "if you're not here for the meeting and there's a problem in your feature, your feature is not going to be in the next release". And I would imagine that if their compensation stems partially from getting those features in, that could be quite a blow for the team.

Re:BS

[SN74S181]

Marketroid driven closed source software does not work.

I feel ridiculous for even having to say this, but the first line in your comment was 'Blanket statements are stupid....'

Don't you feel like an idiot now? You should have clicked 'review' and looked closely at your message body before submitting.

Bugs

[sstory]

If they're really that fix-crazed, how come outlook has had the same bug (overly persistent and reappearing send/receive dialog box) for years now? I first got that bug on Outlook Express on Windows 95. Fast forward 5 years, and I have that bug on Outlook on Windows 2000.

I disagree - not spam, just a great book.

[Ndr_Amigo]

Load of crap, I doubt anybody is spamming slashdot for a book this old.

It is simply the BEST look at the insides and development of NT around. The article on Paul's site is dry and doesn't even remotely express the truth behind NT.

For example, ShowStopper reveils the amount of disgust the NT team (or at least Daves side) -really- felt towards Win16 compatability. The canned articles says how 'easy it was' and what a 'good idea' it was.

Show stopper explains how rejected it REALLY was from most of the team and how much of a complete and utter headache it was to implement the Windows 'personality'.

The articles linked are not bad when talking about the latest version of Windows, but it's very much preaching and definatly doesn't tell much about how NT started out and it's effect on Microsoft as a whole.

The whole GUI thing was actually pretty much a laugh to the DEC people Dave dragged over with him, and the practice of 'eating dogfood' at Microsoft was really first imposed by Dave himself. And there's intresting background on the IBM/MS OS/2 debacle (Windows NT was originally going to be OS/2 NT!) and how Windows very nearly didn't have anything approaching usable graphics until Michael Abrash himself came in with his 'new trick' that actually allowed NT to show graphics at a usable rate and was the first step to removing the console.

Anyway, I don't want to do a book review here, let's just say that the above articles are far too preachy, the book shows the REAL story and I recommend it. And I'm sure that's why the other three people have plugged it also :)

Still and always.

[Richard Steiner]

You've been living in a cave, perhaps? ;-)

7000 "bug fixes" for one developer

[gosand]

The speed at which the team was able to fix all of the branding graphics, text, and registry entries in the system is a testament to the company's dynamic process for fixing bugs, Wanke said. The problem was that several thousand changes needed to be made, and that would normally require several thousand new entries in the product's bug tracking system. "I went out and handpicked the three best developers on the team and said, 'just go and fix it.' One developer fixed over 7,000 references to [Windows] .NET Server. Let's just say that there are people I trust, and people I don't trust. I told these guys, 'don't tell me what you're doing. Just do it.'"

Ahh, good ol' sed. I wonder if he used the Windows version, or if he booted up the Linux box? :-)

This just goes to show that even the biggest software developers have to deal with "simple" requests like name changes that are very inefficient uses of engineers time. I want to know what super-duper advanced bug system they use.

Sounds like typical Micro-FUD to me, bud

[ausoleil]

Microsoft's record of "innovation" has sunk to a new low -- now it looks like they are going to embrace-extend-exterminate Tux. These bozos can't even invent their own mascot...but then again, a furry, squishy bug (the animal most reminiscent of Windows, IMHO) isn't the most inspiring marketing tool.

Bet they claim they had a penguin for a mascot all along and it was those hippies, foreigners and un-American freaks that stole their idea and made Tux the mascot for that mean ole' Linux.

How typical.

wow...

[SirHalcyon]

"We should be able to reproduce that [build] three years in the future, using the various tools, compilers, and scripts we used at that time."

amazing... they figured out CVS, aren't they special.

Flamebait??

[Bishop]

The NT kernel was good. Then Microsoft moved the GUI into the kernel.

This is pretty much spot on, not flamebait.

To be fair to MS they had to do something to improve GUI performance in order to move NT onto worstations, and to a certain extent compete with other servers. NT3.51 was fine as a server, but painfull as a workstation. Win3.1 and later Win95 both provided much better user response. It is hard to sell your "high performance" OS when it feels slower then the old tech.

History is written by the losers

[Anonymous Coward]

(of the anti-trust case)

Microsoft is pushing awfully hard to sell their version of NT's history. The fact is, that even if someone was on Microsoft's payroll to tinker with some "new technology" since 1989, there was no NT product until 1996, when Windows NT 3.5 came out. It was a flop, because it still had the old Windows 3.1 GUI, and everyone was enarmoured with Windows 95's look and feel. Their official "NT 3.5" was actually just a bolted on netware client to windows for workgroups (16 bit.) NT 4.0 was just a GUI upgrade, but wasn't really usable for a couple of service packs. NT started having "server" functionality around 1997, and that was really just back office -- whose main feature was a network printer driver!

12 hour compiles!!!

[zackbar]

Yikes.

I figured someone here would post about MS recompiling 50 million lines of code every day.

Does this mean they update each program source every day? The kernel, as well as each of the dlls? It seems like overkill.

Or at least it seems like major code bloat. Perhaps it's too much in the OS itself.

Doesn't linux separate the file system from the user interface? Seems there has to be a better way than the way MS does it.

Re:12 hour compiles!!!

[Procyon101]

No, it means they run a clean automated daily build, which is common practice for most projects. Incremental builds are great for devs, but the official build is most always done clean.

Even as a dev, I try to run a clean build of my systems when I leave Friday. I can't tell you how many times a weird bug has been caused by some missed dependancy that the make system wasn't rebuilding.

50+ million lines of code!

[vena]

all brought down routinely by 5 line scripts :(

Source Code Control System

[dbk25]

I was interested, a frankly a bit bothered, by this:

"The source code control system we use now is new, because we really pushed the scale of the previous version with Windows 2000. Mark [Lucovsky] personally lead the development of the new system and introduced it post-2000."

The process described in the article depends heavily on branching and migrating changes between branches. Microsoft's version control product, Visual SourceSafe (VSS), has some serious and well known weaknesses in precisely those areas, which is why we are looking at a replacement.

VSS development has been minimal for years. Since Microsoft advertises that use the same development tools (Visual Studio, Visual C++, etc.) that they sell, I find it frustrating that for version control they've gone forward internally but left us VSS customers behind.

Re:Linux Redneck Geeks! Look ma! I compiled my OS!

[g_goblin]

Hmm... I didn't know I was a Redneck... maybe I should go buy some tobacca and start spitting in your eye... f'ing coward.

The truth of the matter is both OS's, *Nix and Windows, have their strengths and their weaknesses. You have to be able to identify how they affect your environment and work around them.

I like the article because I like to see what goes on behind the green curtain. I just wish Bill would let us see if it is a giant or a little person doing all the work.

Pa Ting!!!!

Microsoft and the Empire State Building

[esarjeant]

What I don't get is why it takes 10,000+ developers to develop an operating system. Granted, there are a few nifty utilities included, but it seems like a case of an awful lot of cooks.

For comparison, the Empire State Building took a little over a year and had at most 3,400 workers on the project at any one time.

WHY dont people want to work with m$

[geeklawyer]

"..They wanted to work with Novell, but Novell weren't interested.."

You have this inverted. It would've been rather more insightful if you'd asked why people dont want to work with Micro$oft. Think stacker patent, DOS, Sendo, utility apps, etc etc etc etc. People who do joint ventures with MS wind up getting f**ked up the arse by them. This is why phone and embedded device manufacturers have the often expressed attitude "we'll work with anyone but M$".

Asking why no-one wants to work with M$ is a bit like Saddam asking "no-one wants to hang with me. Why is that?"

Re:no the kernel build still works on alpha

[Guy Harris]

the NT kernel is a microkernel much like the mach kernel in MacOS 10.x

Yeah, both of them keep a ton of stuff in kernel-mode code, such as file sysstems, network stacks, etc.. They're not microkernels much like, say, QNX, however.

That's a PUFFIN [was Re:hmmm...]

[beerman2k]

What are you people blind?? That is not a penguin--it's a puffin.

Availability

[Door-opening Fascist]

Windows Server 2003 hit 99.995 percent availability at the Release Candidate 1 (RC1) stage last summer

Correct me if I'm wrong, but doesn't this mean that they must have had something running a year before that? 4.5 9's of availability is what, a day of down-time per year?

Re:Availability

[ajv]

The statistic is a form of mean time beween failures (MTBF).

I hate percentage uptime stats as they're useless and as you've just proved, hard to calculate correctly if you're a moron.

If you have a large server population of say 50,000 RC1 boxes all reporting back to MS (and it does), you can easily determine the MTBF of the fleet. But more importantly, the error reporting tells you about the bugs your customers are seeing and how many are seeing the same bug.

Andrew

SAA/SNA

[IntlHarvester]

Are you malicious or just ignorant? Obviously SAA was part of the picture, But IBM OS/2 also has interoperability with Unix and the Net ...

The question was if IBM malicious or just ignorant.

His facts are correct. IBM tried to bury the i386 because they were worried about their midrange. This torpedoed OS/2 1.x and gave the entire project a "Loser" reputation long before Win 3.1 even shipped.

IBM also heavly marketed OS/2 as part of a Mainframe integration strategy, with it's special "Extended Edition", etc. And you would be hardpressed to find any significant OS/2 deployment that wasn't in a IBM mainframe shop.

This gave customers the impression of lock-in and that their Unix and Novell machines were second-class citizens in IBM's world -- Which they were. Windows NT shipped out-of-box with TCP/IP, IPX/SPX, AppleTalk, DECNet, you name it. IBM OS/2 2.0 shipped out of box with jackshit, and then made you pay for NetBEUI and pay some more (a lot more) for TCP/IP.

Not to mention that IBM themselves FUDded the idea of OS/2 on non-PS/2 hardware ... If you haven't come to the conclusion that OS/2's technical marketing was a clusterfuck from day 1, something's wrong.

Now I suspect at this point you are going to mention something about Warp 4.0 and so on, but trust me, OS/2 was already dead in the grave by then even at IBM.

Re:sick company

[Richy_T]

It's compiled on some M$ encumbered x86 every night from scratch

And when finally that CPU had come to the end of its working life and was finally retired, it's package was broken open and was found to be filled with some kind of organic matter. DNA analysis found its origins were from three men; Judas, Brutus, and Cassius

Rich

12 hours to compile?

[lilbudda]

I guess it's true: What Intel giveth, Microsoft taketh away...

Very Enlightening on Microsoft

[gottabeme]

Reading this article illuminated some of the things that IMHO are wrong with Microsoft's methods of developing software and handling their employees.

Heated argument and cursing are a given in War Room, and the penalty for not being on top of your bugs is swift and cruel ridicule from the other team members.

If that's true, that's incredibly unprofessional. Now I know many coders prefer to work in a relaxed, casual atmosphere, but to me that is taking it way too far. It can't be good for morale either. Devs should be expected to do their job to the best of their ability, get help if needed, and keep their bosses appraised of their progress. If they don't do all of that, they aren't doing their job. If they aren't doing their job, why are they working there?

The most virulent treatment, naturally, is saved for those foolish enough to blow off a War Room meeting. On the day I attended, one feature group had four of its bugs punted to Longhorn because they had failed to shown up for War Room. When someone argued that they should be given another day, Wanke simply said, "F#$% 'em. If it was that important, they would have been here. It's in Longhorn. Next bug."

That is a completely wrong attitude. If the rest of Microsoft is like that, that explains many of the problems with Microsoft software. That attitude would be fine if the only people affected by a bug were those responsible for fixing it. But the people that are affected by bugs are the people that buy and use their software: their customers, the people who pay their salaries. It is irresponsible of a person in charge of a project like that to not fix a problem he is aware of just because the subordinates who are supposed to fix it didn't show up for one of their three-times-daily meetings. It's starting to sound like they're only concerned with bug counts and severity ratings, reaching only for the goal of Release To Manufacturing to satisfy their shareholders that Product X is out the door and on sale.

In addition to the main War Team, each of the feature teams have their own War Rooms, so there could be as many as 50 such meetings each day, each going over a specific component of the system. These other War Room meetings occur at 8 a.m., every day. When a bug fix passes the local War Team process, it's introduced at Wanke's meeting. "They can't come into War Room unless they're fix-ready," Wanke said. "They must be fix-ready." Because there isn't a single person making decisions, there is a system of checks and balances through which each bug fix passes before it's introduced into the build.

So, if people don't attend the War Room meeting they get chewed out, but they aren't allowed to attend until they have finished what they're working on? Maybe some details are missing, but if it's that simple, no wonder MS software is such a mess. "Be here or else." "Don't show your face unless you have your fix finished." Do you see a conflict here?

We've sent out calls at 3 a.m. when the build is broken, find the developer that broke it, and get him into work right then and fix it immediately. The developers are on call 24 hours a day.

Again, irresponsibility. A dev should not leave for home until he's finished with his work. Finishing his work should include having tested his work that's been submitted as complete. If a dev submits broken code and leaves, he's not doing his job. Mistakes can happen of course, but a 24-hour on-call policy for programmers?! That can't be good for morale either. They should put in their hours, do their job right, and go home and (try to) forget about work. If they mess things up so badly that they have to get up at 3 AM to fix it, they are not doing their job right.

"I'm responsible for 8000 to 10,000 developers, program managers, and testers..."

Another problem. 8,000 to 10,000 people is like a division of a company, or a whole company in some cases. A person in charge of that many people should be in charge of general operations, not a single product. In other words, that's way, way too many people working on a single product. If you compared the number of people who work on Windows to the combined number of people that work on Linux or other open-source software with equivilant functionality, would the numbers be anything close? I don't know, but I doubt it. But even if the numbers are close, the way the software is made (independant programs that work on their own) avoids the problem of having to coordinate so many people.

On the day that we attended War Room, on January 21, 2003, Windows Server 2003 had hit an "absolute historic low" for bugs, according to Wanke. "We're shutting down the project this week," he said. "It's done. We're going to ship it." On that day, WinServer 2K3 had just a few active bugs, and at least a quarter to one-third of those bugs were simple branding issues. "So let's say there are about 150 outstanding issues to address," Wanke told us. "Of that, we'll fix about 100. All of the bugs are severity rated from 1 to 3, plus they get a priority rating. We have [a few] severity-1 bugs left to fix, and those all have to be fixed for us to ship."

Wanke said that the server team had already fixed all of the known security vulnerabilities.

So, Microsoft will ship software with known, fixable bugs in it (not news but making a point). That again demonstrates their attitude towards the quality of their software and their attitude towards their customers. They should but a sticker on the box that says "This software contains known bugs. A list can be found at www.microsoft.com/bugs." Of course, I guess their attitude is that their huge, legalese EULA discounts any responsibility and disclaims any fitness of purpose or quality. It also doesn't indicate much pride in their product (maybe they do have pride in it, but still...). Yes, all software will have some bugs in it I guess, but to decide to not fix already-known bugs and go ahead and sell their software seems irresponsible to me.

My two bucks. Should be interesting to see other opinions on this.