Windows 7 is NOT a “Vista Service Pack”

July 27, 2009


It’s hard to argue with the 127K subscribers to Jeff Atwood’s blog, so when I first read his post titled “Windows 7: The Best Vista Service Pack Ever” I wasn’t immediately sure how to respond. Nonetheless, I would like to try and offer my opinion on some of Jeff’s points, in the hope of giving a different perspective on what Windows 7 really is.

Jeff writes:

Now that Windows 7 has reached its "release to manufacturing" milestone, I had the opportunity to install it for myself and see.

Jeff, I’m sorry, but where have you been for the past 8 months? Microsoft released the M3 build at the PDC in October ‘08, the Beta has been out for several months, the RC has been out since May… If it’s really only now that you first installed the OS, how can you possibly write something as definitive as… (brace yourself…)

Windows 7 is the best Vista Service pack ever.

This one has really got me. Just a quick reminder to my dear readers: There is a whole lot of shiny new features in Windows 7, just to name a few off the top of my head:

  • New taskbar
  • Native multi-touch support
  • Support for up to 256 logical processors
  • Native Ribbon UI
  • Sensor and Location platform
  • Windows Web Services API
  • Trigger-start services
  • Shell libraries
  • …and so many more!

Saying that a major release of the OS, which contains dozens of major new features, is a service pack is not just a slight inaccuracy: It’s borderline irresponsible to the 127K blog readers and the thousands of others who are going to perpetuate this claim. Sure enough, Windows 7 offers significant performance and reliability improvements compared to Vista—but it doesn’t make it a service pack. For this matter, Vista SP2 is almost to par with the performance and reliability of Windows 7. That’s the real service pack.

Jeff writes:

The core of the operating system isn’t that different.

That’s simply not true, regardless of what you define as “the core of the operating system”. The kernel has undergone significant changes, including major scalability improvements; the user interface has been revamped and the initial desktop experience is completely difference; the fundamental Windows applications have a different look and feel; there’s no area of Windows 7 that you can say “isn’t that different”.

Jeff writes:

I think Windows 7 works well as a de-facto Vista service pack. I guess that’s not surprising if you compare the version numbers. […] 6.0.6002 […] 6.1.7600

This is another example of poor fact-checking. In October ‘08 Mike Nash wrote on the Windows Blog that the 6.1 version number reflects only the deepest care for application compatibility. There are too many applications out there that break at the first sight of a OS major version change, and keeping the major version number will likely have significantly reduced the number of applications that refuse to run on Windows 7 for no apparent reason.

If you’re still unconvinced that Windows 7 is an incredibly large, major release, take a few minutes to skim through the archives of the Engineering Windows 7 Blog and see the massive amount of effort that has gone into the making of this OS.

Add comment
facebook linkedin twitter email

Leave a Reply

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>



  1. Duncan SmartJuly 28, 2009 ב 12:39 PM

    “…irresponsible to the 127K blog readers”: stop taking him that seriously. His point is that Windows 7 is what Vista should have been.

  2. RobzJuly 28, 2009 ב 4:01 PM

    I think the answer to the question lies in whether you think Windows XP was a service pack to Windows 2000 or not.


  3. Greg SuvalianJuly 28, 2009 ב 5:40 PM

    The main reason anybody sees OS as a “service pack” if you have to have dedicated drivers for this OS if drivers from previous OS will work (as they do with Vista) then he has a valid point which means kernel did not change that much to be considered as new OS if all drivers from previous OS will work on new version of OS.

  4. EBGreenJuly 28, 2009 ב 10:41 PM

    While I happen to agree that W7 definitely is not a Service Pack, in Jeff’s defense I believe he has been using W7 for some time now. I’m certain that I have heard him discuss it with Joel Spolsky on their podcast. I think it was just poor wording that made it sound like he wasn’t.

  5. EBGreenJuly 28, 2009 ב 10:56 PM

    Ok, so I went back and checked the transcripts. Jeff has discussed W7 on the podcast, but he explicitly decided not to install any RC versions.

    Knowing that it does seem a little fast to be making pretty broad statements about the OS.

  6. NJuly 29, 2009 ב 12:21 AM

    Well, Atwood can be at once very smart and a complete and utter idiot at the same time — my thoughts were largely the same as yours when I read his post too.

  7. Paul CoddingtonAugust 8, 2009 ב 2:04 PM

    One should probably take into account that for the average “Joe”, many of the changes are “under-the-hood”, involve technologies they do not use, or are of longer-term theoretical benefit – average people are not using touch-screens or have 256 processors, for example.

    Some Vista Ultimate users see their sparse extras being uninstalled during upgrade for no apparent reason, so they feel they have gone backwards on some points.

    Still, looking at all the changes listed, it is a remarkable achievement. I think the Windows Team has done an excellent job and the future of Windows is looking more exciting than it ever has in the 15 years I have been developing software.

    Jeff: “I guess that’s not surprising if you compare the version numbers. […] 6.0.6002 […] 6.1.7600”

    Yeah – that one surprised me as well. Still, it is very easy to blurt out something silly, especially at the end of a hard day. We all do it sometimes. Almost as surprising as the logic behind it (how much should one bend over backwards for incompetent programmers when designing an OS?).

    Many of the critics do not have the courage to use their own names, but Jeff writes under his (as I do mine) and chooses to deal with the consequences (just like real life in the off-line world). I would rather have one Jeff than 10,000 anonymous critics any day.

  8. wolvesdreadAugust 22, 2009 ב 6:38 AM

    The real question of whether or not something is a service pack.. (the implied difference in contrast to a major release) pretty simple from the perspective of Joe & Jane “End User” Doe:

    If you sit an average user that’s had Vista for a bit down in front of Windows 7 (booted / post splash screen / with nothing blatantly announcing the name), will they know they’re not sitting at another Vista box?

    Answer: No.

    If you take a Windows 3.x user and set them in front of Windows 95, would they? Yep. Windows 9x user before NT4? Yep. NT4 user before XP? Yep. XP user before Vista? Yep.

    Regardless of under the hood tweaks, new feature adds, modified Kernel.. AND relative to the historic difference between Microsoft upgrades: Windows 7 is solidly a service pack in the eyes of the general population.

    If you give credence to Microsoft flipping to the Apple OSX definition of what constitutes a major release, then you can call it whatever you want.. because you’ve change the rules out from under your userbase (not something Microsoft ever does lol..)

    If their new OS release paradigm is the Apple paradigm, then you can happily call this a major release.. but the population used to what Microsoft formerly defined a major release to be is going to think they’re being forced to pay for a service pack no matter how much geek tech nuance you throw at them.

  9. Sasha GoldshteinAugust 25, 2009 ב 4:51 PM

    @wolvesdread: Go ahead and perform that experiment. Do you seriously think the average user won’t notice the new taskbar?!