So, Microsoft has finally adopted the Linux development model?

Every single organization seems to follow this exact same path. Lean and mean at first, to fast and nimble second, to large but feature, to slow and bloated. The next step after this tends to be, jump at any and all projects to see if anything will stick progressing slowly down a spiral with a large change either acquisition by another company or dramatic slashing of middle-management workers and projects to focus on their core. Unfortunately I have yet to see a large organization that doesn't seem to go down something similar to this path.

Because it had to move through the digestive tract and on through the large intestine.

Doesn't "the approximately 24 people who worked on the shutdown menu" already tell you everything you need to know?

...uninstalled Vista instead? Now that would be a simple way to solve the matter.

People would want Vista if it were revolutionary. But you can't just sit down and say 'let's make something revolutionary' and then set up a timeline and claim to be able to create a revolution within that timeframe. Revolutions happen by accident if at all, not on purpose.

So why hurry? For money? In my experience hurrying to make money never works out.

TLF

I don't get the new cult of never turning your PC off. If I'm away from my computer, it's usually for an extended period (IE - a night I'm not downloading crap, or a full day of work). Doesn't it make vastly more sense to not have the power supply fan running for those 8 hours? Or the HD randomly going idle and then spinning up again? When I'm done, I shut the machine down and turn off the power strip. Interested in why others don't, however.

From the article:

"Windows has a tree of repositories: developers check in to the nodes, and periodically the changes in the nodes are integrated up one level in the hierarchy. At a different periodicity, changes are integrated down the tree from the root to the nodes. In Windows, the node I was working on was 4 levels removed from the root. The periodicity of integration decayed exponentially and unpredictably as you approached the root so it ended up that it took between 1 and 3 months for my code to get to the root node, and some multiple of that for it to reach the other nodes."

Monotone, BitKeeper, git, bzr, and so on would all handle this situation efficiently and gracefully; all the repositories can sync to each other and none need be more than a few minutes out of date. Amazing that Microsoft's solution is so poor by comparison

I've read other blogs in regards to Windows Vista, and from what I am gathering the primary reason why Windows Vista took so long to complete was because of management. Philip Su argued how the gargantuan amount of code included in Vista slowed it development dramatically, however I think that this strengthens my point and the point made in this article.

However, I'm not terribly surprised that this occurred for Vista. The higher execs at the company wanted Vista to be a revolution and had a clear and concise goal that they wanted this operating system to achieve. In order to do this, from what I've read, they needed to form many more separate divisions inside of the Windows division to concentrate on small parts of the operating system. This probably sounded like a good idea, but the problem was that none of their work was in sync with each other. Some had more work completed than others. Furthermore, rifts within divisions such as the one present here spurred disagreement after disagreement, that including the decision to switch the codebase of the OS to the one present in Server 2003 (something that from what I understand should have been decided from the beginning). With all of this, it was only inevitable that confusion and miscommunication would occur.

All in all, while I think Windows Vista is definitely more capable than Windows XP and warrants itself a much needed upgrade, I feel that the actual improvements of the operating system [wikipedia.org] do not warrant a five-year delay. Okay, so the compositing manager, networking stack, and audio stack may have needed some time to complete, but five-years? I am not a programmer, so my impression may not carry a lot of weight, but being that Linux and UNIX based systems have already included some of these "future technologies," it becomes naive to deem this delay as acceptable.

Ok. I've been running vista on one machine or another for a while.. since early beta.. and am now running the release version on my main machine. There are quite a few headscratchers in here. I often tell my colleagues I'm like the little kid from the 6th sense.. except instead of dead people I see bugs. Things that annoy the crap out of me that have been there at least one maybe two versions of windows ago.

In the past days of clicking through endless options and dialogs to configure things such as encryption certificates, etc I often wondered if this was really better than editing a single line in an easy-to-find text file.

Start menu? Hardly ever used the damn thing. Shortcut keys with and I put the quicklaunch bar off to one side with the 40 or so frequently used programs I use.

Vista doesn't support dragging the quicklaunch bar off of the stat menu and off to one side because it was "confusing to end users." No one seems to have found a registry override as yet.

Vista doesn't handle symlinks properly. It used to be "c:\documents and settings" but now in vista it is c:\users. I see a clever little "C:\documents and settings" shortcut on my C drive. OOOOoo is this a symlink? No? I get Access Denied when trying to double-click. Opening the path via an API however works fine. Go figure.

BUGS. Features? Half-Features? Call them what you want. I think most technical folks that have to work on this know these problems exist but architecturally or bureaucratically they are hard or impossible to fix.

Often on XP, 2000, NT and 95 I would hit control-esc then R for run and type frequently used programs into run. I would say this is just an odd quirk about me and how I think menus take too long and too much work to do something, but now the run area has been replaced with a little place you type in stuff and through the magic of windows desktop search it finds whatever you type in the area above that normally occupied by program icons. The bug? You have to let it search. No matter what. Yeah, WTF? This works great on a home PC where you maybe have maybe 10,000 files. Network drives? Oh no. You can't just type n:\ then hit enter. You have to physically wait a sec for it to pull up n:\ in the list of programs above the start menu THEN hit enter. WOW, WHAT A GREAT FEATURE. No more control-esc n:\ enter for me. It is nowctrl+esc n:\ wait..wait..wait.. enter. Otherwise I get some random program like Notepad. Or Flash. Or Firefox.

On the one hand I can see how the start menu splaying itself all over your screen as you "drill down" to whatever the hell obscure program you need might be unappealing. On the other hand confining the entirety of all programs available to you to a 400x600 pixel window doesn't seem like a good fix.

This is just the start menu. Don't even get me started on the new file explorer, which is the least half-baked area of Vista in my opinion. Does Slashdot have an option for submitting a rant and getting comments? I'm sure I could go on all day.

I take all this as evidence that a lot of new features in vista are based on good ideas.. new paradigms in UI design.. it just seems that the vast majority are implemented poorly at best and implemented recklessly at worst. I would not expect this in 2006 when others are able to produce such polished and solid OSs. I would have to agree this seems like code-rot from the inside out probably due to the megalithic internal structure at MS

Does nayone have any info on how the OS X team works? I mean in a few years Apple did a complete paradigm shift from OS 9 to OS X on the OS level. I would be interesting to see what, if anything, they are doing better. Links or experiences would be nice.

And while I am at it, the start menu requires input from the kernal team. WTF? This is violating some very basic software design principles. The OS should just be basic services, then the applications, including the UI, should ride on top of the kernal without really caring much about how the kernal works.

I can see integration with the shell, but the kernal? It looks like MS policy of tight OS integration with the applications is biting them *hard*.

Wait until you read about the development of the "About" menu item!

IBM is terminating the final remnants of their OS/2 staff at the end of December, 2006 as OS/2 takes its last few agonized dying breaths. What's interesting, though, is that over the last 5 years, there has been a skeleton crew of OS/2 people at IBM to support the last few OS/2 customers and this tiny crew was able to accomplish a lot of stuff to keep OS/2 updated and running on current hardware that a much larger crew probably could not have. They were even able to add a lot of stuff that was never even included in the last 'official' Warp 4 release such as the logical volume manager, journaling file system, updated kernel for multicore AMD, USB 2.0 support, UDF DVD support, etc. In this case, a small crew could do a lot more than a large staff and the final dying remnants of the OS/2 business at IBM became more like the original tiny Windows group at Microsoft.

The UI isn't all that terrible. Joel Spolsky is making a mountain out of a molehill. Look at the screenshot he gives in his article. Here's what I notice:

1) There's a power button. That shuts things down fully. ("I am going away from my computer now, but I'd like the power to be really off.")
2) There's a lock button. That leave it running, but keeps others out of your stuff. ("I am going away from my computer now.")
3) There's a menu of choices if you care to look at it, and the button is much smaller than the other two and has a nondescript arrow icon on it which makes it much less attractive to non-techie users.

Yes, his suggestions for combining lock with switch user and sleep with hibernate are good, but I don't think what they actually implemented is all that difficult to understand. His problem is that he's "one of us" and went looking for all the extra options. Most people will never click that arrow to make that menu appear. Ever. It's kind of unfair, even to Microsoft, to rag on something for being unfriendly to non-techies when non-techies are never going to even see it. Usually Joel Spolsky's observations are spot-on, but this time I'm going to have to give him an F for eFfort.

Would you have a 30 page argument on the merits of sleep vs. hibernate...

There was one quite interesting post on Moishe Lettvin's blog (emphasis mine):

disclaimer - I was a manager at Microsoft during some of this period (a member of the class of 17 uninformed decision makers) although not on this feature, er, menu.

The people who designed the source control system for Windows were *not* idiots. They were trying to solve the following problem:
- thousands of developers,
- promiscuous dependency taking between parts of Windows without much analysis of the consequences
--> with a single codebase, if each developer broke the build once every two years there would never be a Longhorn build (or some such statistic - I forget the actual number)

There are three obvious solutions to this problem:
1. federate out the source tree, and pay the forward and reverse integration taxes (primarily delay in finding build breaks), or...
2. remove a large number of the unneccesary dependencies between the various parts of Windows, especially the circular dependencies.
3. Both 1&2
#1 was the winning solution in large part because it could be executed by a small team over a defined period of time. #2 would have required herding all the Windows developers (and PMs, managers, UI designers...), and is potentially an unbounded problem.

(There was much work done analyzing the internal structure of Windows, which certainly counts as a Microsoft trade secret so I am not at liberty to discuss it)

Note: the open source community does not have this problem (at least not to the same degree) as they tend not to take dependencies on each other to the same degree, specifically:
- rarely take dependencies on unshipped code
- rarely make circular dependencies
- mostly take depemdencies on mature stable components.

As others have mentioned, the real surprise here is that they managed to ship anything.

Now I'm not a Microsoft employee, but even as an outsider I've seen some hints that it might be the "promiscuous dependency taking" that has delayed Vista.

1) Integration of Internet Explorer.
Microsoft claims that IE and Windows are inextricably linked together, and at least for Windows 2000 and newer this seems to be true. For instance, if you type a URL into the address bar of the Windows Explorer, it will show you web pages. IMHO a stupid design, the web browser should be an application, not a fixed part of the GUI.

2) The RPC service being responsible for things a "remote procedure call service" has no business handling.
In August 2003, a worm called MSBlast spread by exploiting a buffer overflow in the DCOM RPC service (see Wikipedia, http://en.wikipedia.org/wiki/MSBlast [wikipedia.org]). At that time me, trying to be clever, thought:
"I don't want anyone remotely executing stuff on my PC anyway. I'll just switch the service off and be fine".
Lo and behold:
After turning off the RPC service, various local functions were dead as well. Including the Services menu in the control panel. I was lucky that I could reactivate the RPC service by manually editing the registry, else I would have spent a day reinstalling.

So it seems quite believable that Microsoft is choking itself by lack of discipline in designing Windows ;-)

You can tell something is very wrong when the lamentable Zune software doesn't work properly (or at all) in Vista beta. I mean what the hell is going on? How could they be this far wrong?

From RoughlyDrafted's Leopard vs Vista 5: Development Challenges [roughlydrafted.com]

"In an almost spooky series of events, Microsoft has shadowed Apple's brush with death, making the exact same set of moves exactly ten years after Apple:

• In the mid 90s, Microsoft rapidly built upon its past success with MS-DOS to establish Windows as a vast empire ...just as Apple used the success of the Apple II as a stepping stone to launch the Mac in the mid 80s.
• From 1995 to 2001, Microsoft rapidly delivered advancements to its desktop Windows product ...just as Apple rapidly advanced the Mac System Software from 1985-1991.
• In 2001, Microsoft began announcing technologies that would be released as part of Longhorn and later Blackcomb ...just as Apple described new technologies intended for Copland and Gershwin a decade prior.
• From 2002-2006, Microsoft dropped features, changed plans, and started over several times in protracted efforts to ship Longhorn ...just as Apple had fumbled around with Copland ten years earlier.
• By 2006, it was obvious that Microsoft's Longhorn was not going to live up to the hype, and would really be just a refresh of the existing Windows XP ...just as Copland had been gutted in 1996 and its salvaged remains delivered as the optimistically named Mac OS 8.
• Microsoft outed Blackcomb as vaporware ...just as Apple admitted that Gershwin had never been anything but a list of deferred goals ten years earlier.

What's Next? The only difference between Apple and Microsoft is that today, in the final days of 2006, there is no equivalent to a 1996 NeXT waiting in the wings to swoop down and fix Microsoft's mess. Leopard vs Vista 5: Development Challenges [roughlydrafted.com]

And I thought that thing was complicated enough just with just the Log Out/Switch/Sleep/Shutdown options! No wonder it's taking so long!