Measuring the scale of a release


Thanks for all the feedback that we have been getting. That much of it is positive is certainly appreciated. I’ve been answering mails as best I can and along with members of the team we’ve been having the discussion in the comments. Everyone has done a great job sharing their views on specifics, wishes, and requests. I love getting these mails and reading the comments. It is fantastic. I just want to make sure folks know I can’t answer each one! What we are going to do is look to the emails and comments as a way of suggesting posts we should write.  The team overall appreciate the warm reception from all those that have joined us–we know we have lots of energetic discussions ahead of us and we’re genuinely happy to start.


With this post, I am hoping to continue the dialog on the way we think “inside the Win7 team” so to speak—in a sense this is about expanding the team a bit and bringing you into some more of the discussions we have about planning a release. This conversation about major or minor releases is very much like the one I have with my boss as we start planning :-)


When we started planning the release, the first thing some might think we have to decide is if Windows 7 (client) would be a “major release” or not. I put that in quotes because it turns out this isn’t really something you decide nor is it something with a single answer. The magnitude of a release is as much about your perspective on the features as it is about the features themselves. One could even ask if being declared a major release is a compliment or not. As engineers planning a product we decide up front the percentage of our development team will that work on the release and the extent of our schedule—with the result in hand customers each decide for themselves if the release is “major”, though of course we like to have an opinion. On the server blog we talked about the schedule and we shared our opinion of the scale of the releases of Windows 7 client and server.


Our goal is about building an awesome release of Windows 7.


Across all customers, there is always a view that a major release is one that has features that are really the ones for me. A minor release is one that doesn’t have anything for me. It should then be pretty easy to plan a major release—just make sure it exactly the right features for everyone (and given the focus on performance, it can’t have any extra features, even if other people want them)! As engineers we all know such a design process is really impossible, especially because more often than not any two customers can be found to want exactly opposite features. In fact as I type this I received sequential emails one saying “[N]obody cares about touch screen nonsense” and the other saying “[Win7 needs] more advanced/robust ‘touch’ features”. When you just get unstructured and unsolicited input you see these opposites quite a bit. I’m sure folks are noticing this on the blog comments as well.


Let’s explore the spectrum of release magnitude across a couple of (but not all) different types of customers: end-users, developers, partners, IT professionals, and influentials.


End-users are generally the most straight-forward in terms of deciding how big a release is going to be. For an end-user a release is a big deal if they want to go out and buy an upgrade or buy a new PC. We could call that a major release. Seems simple enough and a major release is good for everyone. On the other hand, one could also imagine that a release is really cool and people want to buy it, but they also want to use their existing PC and the release requires more memory, updated drivers that might not be available, or maybe some specific hardware to be fully realized. Then it seems that a major release goes from a positive to a bit of an under-taking and thus loses some of its luster. Of course we all know that what folks really want is all the things they want that runs on the hardware they want—then that is a great product to get (whether it is major or not).


Developers look at a release through a different lens. Obviously for developers a release is a major one if there are new APIs and capabilities to take advantage of in their software—again straight-forward enough. It could also be the case that a previous release had a lot of new APIs and folks are just getting familiar with using them and so what they really want is to round out the APIs and maybe improve performance. So one might suspect that the first release is a major release and the second type is a minor release. But if you look at the history of software products, it is often these “minor” releases that themselves become the major releases – Windows 3.1, Office 4.2, or even Windows XP SP2. In each of these cases, the target for developers became the “minor” release but in the eyes of the market that was the “major” release. The reason developers want to use new APIs is to differentiate their products or focus their energies on domain expertise they bring to the table, not just call new APIs for the sake of calling them. In that sense, a release might be a major one if it just happens to free up enough time for an ISV that they bet on the new APIs because they can focus on some things that are a major deal to them.


Partners represent the broad set of folks who create PCs, hardware, and the infrastructure we think of as the ecosystem that Windows is part of. Partners tend to think about a major release in terms of the opportunity it creates and thus a major release might be one with a lot of change and thus it affords the opportunity to provide new hardware and infrastructure to customers. On the other hand, incompatibilities with the past might be viewed in a less than positive light if it means a partner needs to stop moving forward and revisit past work to bring it up to the required compatibility with a new release of Windows. If they choose, for any number of reasons, not to do that work then the release might be viewed as a minor one because of the lack of ecosystem support. So again we see that a big change can be viewed through the lens of a major or a minor release.


IT professionals are often characterized as conservative by nature and thus take a conservative view of change. Due to the business focused nature of the role, the evaluation of any software product is going to take place in the context of a return on investment. So for an IT professional a major release would be one that delivers significant business value. This business value could be defined as a major investment in deployment and management of the software for example. Yet for end-users or developers, these very same features might not even be interesting let alone worthy of being a major or minor release.


Influentials are all the folks who are in the business of providing advice, analysis, and viewpoints on the software we make. These folks often look at releases through the metric of “change”. Big changes equal major release. A big change can be a “re-architecture” as we saw in the transition from Windows 9x to Windows 2000—even though these products looked the same there was tons of change to talk about under the hood. So for reviewers and analysts it was definitely a major release. Big changes can also be big changes in the user-interface because that drives lots of discussion and it is easy to show all the change. Yet for each of these, this definition of major can also be viewed as a less than positive attribute. Re-architecture means potential incompatibilities. New user-interface can mean learning and moving from the familiar.


We’ve seen a lot of comments and I have gotten a lot of email talking about re-architecting Windows as a symbol of a major release. We’ve also gotten a lot of feedback about how a major release is one that breaks with supporting the past. If I could generalize, folks are usually implying that if we do things like that then a number of other major benefits will follow—re-architecting leads to better performance, breaking with the past leads to using less memory. It is always tricky to debate those points because we are comparing a known state to a state where we fix all the things we know to fix, but we don’t yet know what we might introduce, break, or otherwise not fix. So rather than define a major release relative to the implementation, I think it makes sense define the success of the release relative to the benefits of whatever implementation is chosen.  We will definitely continue to pick up on this part of the discussion–there’s a lot of dialog to have.


The key is always a balance. We can have big changes for all customers if we prepare all the necessary folks to work through the change. We can have small changes have a big impact if they are the right changes at the right time, and those will get recorded over time as a major release.


We’ve talked about the timing and the way we structure the team, so you have a sense for the “inputs” into the project. If we listened well and focused our efforts correctly, then each type of customers will find things that make the product worthwhile. And if we do our job at effectively communicating the product, then even the things that could be “problems” are seen in the broader context of an ecosystem where everyone collectively benefits when a few people benefit significantly.


From our perspective, we dedicated our full engineering team and a significant schedule to building the Windows 7 client OS. That makes it a major undertaking by any definition. We intend for Windows 7 to be an awesome release.


I hope this helped to see that perspective is everything when it comes to deciding how big a release is for each type of customer.


–Steven

Comments (111)

  1. Andre says:

    I don’t see how it is a problem to bring the different camps together. If someone don’t needs the touch screen feature he should be able to turn it off, or better choose not to install it at all.

    Many end users might like Aero, I want my classic theme back. Where is the contradiction?

    I’m developing on and for Windows since 10 years but lately I’m getting more and more annoyed by MS products. Just an example, I’m switching quite often between documents in Visual Studio with Ctrl+Tab, but VS2008 added the "feature" to show a preview of the documents which really slowed things down and is annoying. I have no understanding for such useless gimmicks. Can I turn it off? Don’t think so. Want to sell me/us licenses? Stop annoying me.

    Some ideas for Windows: <Windows key> + <1> … <9> to switch between windows, enable keyboard navigation in Windows Flip. Actually pressing <Alt>+<Tab> and then key down sends the key event to the last active app rathen than to Flip.

  2. Mr. Dee says:

    I hate to say it, but Windows 7 client is beginning to sound like a minor release indeed. With both the Server and Client expected to RTM the same time, it pretty much adds up that Windows 7 will actually be version 6.1. The reason I am hearing for the code name is because you Steve likes whole numbers, but at the same time, it just does not add up why you would call the codename ‘Windows 7’, unless the Windows Team is considering it a 7th release of the Windows product, not technically a 7th ‘version’ of the NT kernel itself. We must take into account, Microsoft stop using the NT version in its branding with the release of Windows 2000 which was 5.0, XP 5.1, Server 2003 5.2.

    Here is the problem I just discovered after writing the above, Microsoft could not use that logic, since it would mean that XP was the 6th release of Windows, Vista the 7th and 7 being the 8th.

    Microsoft needs to explain themselves. If it continues with the 6.1 version by Beta 1, its definitely a Vista R2 and Windows Server 2008 R2 releases. Am I correct?

    During the early parts of the Longhorn development, when the OS was at Alpha, Microsoft christened it version 6.0, I am talking builds 3xxx and later. The leaked Windows 7 builds we have been seeing earlier this year have been using the version 6.1 for the kernel. Some said that was because not all of the product had matured enough to become a part of what at Microsoft is called the ‘winmain’ build. Persons in the enthusiast community assumed that by PDC 2008 Windows 7’s kernel would reflect version 7, but with PDC only a couple months away, its looking unlikely at this stage.

    Based on what you have posted and my interpretation, major or minor is in the eye of the beholder. But, I wish I could get a better idea of the significance of this release of Windows in terms of should everyone upgrade to it, who upgraded to Vista, or is it just a release that you can skip and those who are still on XP in 2010 will see some reason to to take the upgrade.

  3. horphus says:

    Right after I started reading the post, I came up thinking about how awesome would it be if Windows 7 had a bunch of new APIs and programming resources for us, developers, to play with on. But as I read on and got to the Developers-related paragraph, I ended up amazed by the way you had succefully pointed out how difficult it is to meet everybody’s expectations at once.

    I’ve been reading the blog since its very beginning but that’s the first time I write a comment here, so I wish you all the best of luck, and I’m pretty sure we’re going to get a great release here.

    Greetings from Brazil :)

  4. Yert says:

    I have to say that every release of Windows is a major release; each and every one has added a new feature, or expanded on one and improved it.

    My main suggestion would be to say, "Just do a good job!" as I am not a dev, just one of the many customers who point in a direction and hope you go the right way.

    P.S. Still, with my little voice, I hope to remind you (not just the Windows design team, but ALL of Microsoft) to get closer to the program with 64 bit. Offerings now are okay, but could be better!

  5. isimmons says:

    From my perspective, if Windows 7 has a faster boot time, uses less resourses, and is available in a business version without all the fancy graphics but with all the features and functionality and support for 64 and 32 bit hardware drivers (even older hardware that worked with XP) It would be a major release.

    This is from a IT professional and a home users perspective.

    You guys at MS have an oportunity here since you prove that you are listening to the customer instead of telling the customer what they will use which has been the feeling with Vista in my opinion. It was a major bone of contention with many to not be able to buy a new computer without Vista. I think that way of doing business in what appears to be monopolistic needs to be a consideration when 7 is ready for release. I made the switch to Linux and now use XP on a VM more as an occasional tool instead of an OS because of that.

    There are a lot of great features in XP and I hope Windows 7 will restore my faith in MS in terms of the OS itself and MS business practices.

  6. SoCalCreations says:

    Jon and Steve obviously you are credited greatly for taking our input, so for the record one more thanks from a previous beta team member, who would love to be on the W7 Beta.

    It seems to me a recurring theme 64bit OS only and with WinFS or at least an add on that supports better meta tags to find what people need on their gigantic hard drives of late.  With the Internet becoming a vast resource of large files (pictures, videos, Zip/Rar files) the speed increase of a larger files system over the default 4k limit would help to increase the speed, so that is a suggestion.

    One of the biggest things I read was not only go to a 64bit system only, but loose the backwards compatibility.  Why not? You guys have VMWare so why not for a small additional charge throw in VMware with which ever past OS they want.  This way technology can move forward but emulated backwards if the need arises.

    Another common theme kept reoccurring as I read through ALL the posts to date (yes my day sucked at work) the idea of finally removing the registry.  Now I am not 100% certain this is a great idea, but I can see the benefits and cons for both.  A big con for keeping it is, the morphing Trojans love the registry and some like the Vundo!R finds comfort there in drones.  I suggest if you keep the registry that it have a serious anti-malware scanner attached to it and also that it can perform a proper clean up and defrag.  It seems many manufactures still can’t write a good software removal routine and the add/remove programs won’t make the effort to clean up the registry.

    While on the subject of Defrag, why not partner with O&O Defrag to license their software.  It Defrags your hard drive in numerous ways and in the back ground while you work.  It does a fantastic job in keeping your system running much faster.

    Of course it might be a great idea to auto partition off space to create the Virtual Memory in as it will help prevent fragmentation, especially when updates are done.  Most people complain how much slower their system is after a ‘critical’ or ‘express’ update was done on their system.  Once I defrag their drive and clean their registry they are golden and happy again, then again so am I cause I just got $100, but that’s not the point.  Microsoft’s image would be better if the Defrag was better overall for both the Registry and Hard Drive.

    Less rebooting was a vague if not rare suggestion, but more dynamic updates with less reboots would be a BIG plus.  I guess if the registry goes away and dynamic application or service modules are developed this might be very possible.

    On the subject of dynamic control, I am reminded by the old Macs (I am not familiar with OS X) but the older Mac OS you could disable services, applications, plugins, etc to help find the trouble makers or improve the speed of the system.  This would also greatly improve the focused use of the system.

    If a profile system was used in conjunction with ‘plugin services’.  One profile could be set to games, which would turn of the services for Bonic, Antivirus, Defender, Acrobat, iTunes, IM clients, Outlook, etc.  Another profile could be a social profile or a media profile or a computational profile….the point is profile services done through dynamic loading and unloading will bring a better experience to the user at that time.  Now the trick is to make it work under multiple logins.  You would want your gaming profile rocking your system and in comes the wife who must send out an urgent email, the system should handle switching user logins and dynamic service loading for her and then dump them when you get back to your profile and Unreal. In any event to the end user this would have to be, dare I say it, Wizard driven for the type of performance or situational profile you want of the computer.  Again it must be done dynamically without rebooting the machine.

    The subject of Multiple desktops was also brought up often and I think this is a great idea!

    On the subject of parental controls, a previous poster hit the nail on the head when it should email the parent of the child what new sites the child went to, because face it the rating system isn’t perfect yet.  It should also capture all IM conversations and archive them during shut down of the application or system and then email them to the parent on the next boot up or if just the IM is shut down.  Not to mention a dictionary for the parents of the IM’s, just as what does POMS mean? (parent over my shoulder) things like that so that the parent can work with their kids….this would be the biggest marketing tool to bring TRUST in a product back to Microsoft for parents.

    I offer a new suggestion of a full Image of the current system and its files for when the system is upgraded to OS7.  This way if people have the same ‘feeling’ from XP to Vista they can down grade from OS7 to their previous OS in less than 20min from achieved DVD’s. A huge bonus would be for the restore system to note the date of upgrade and then ask the user if they would like to back up all new documents, music, etc preformed or created since the upgrade.

    My last suggestion is OS on the run.  This is where you take a 2 or 4 gig memory stick and your OS ‘lite’ gets loaded on it with your applications (not games) to take with you to another system and have it boot from the memory stick (USB drive) so you can continue your work else where in your environment.

    I am sure I could come up with other ideas, but after all that reading I am done and I am sure you preferred this to be shorter, but if you read through it all, my sincerest thank you for taking the time to listen to an IT professional, system builder and end user all in one. And send me the email link for the beta team, I am dying to do something productive once again with Microsoft.

    William

  7. steven_sinofsky says:

    @SoCalCreations —

    THanks so much for your thoughtful suggestions.  I just wanted to use this as a chance to remind folks we really are reading all of these.  We’re trying to figure out how to keep up with the feedback.  There are definitely many common themes and we will address these in posts for sure.  

    I just don’t want folks to think we’re not listening.  And of course the developer in me is hoping we don’t get in the situation where not doing every suggestion makes folks think we don’t listen–the challenge as every developer knows is that you can’t do everything and often suggestions from customers are opposites and you really can’t do both.  We are listening for sure.

    Please keep the discussion open!  We love it.

    –Steven

  8. eduardvalencia says:

    Hi steven

    i’ve been sending my suggestions here but none has been published i don’t know why you guys are blocking it.

    Regards

    User: Eduardvalencia

  9. phschmidt says:

    "And of course the developer in me is hoping we don’t get in the situation where not doing every suggestion makes folks think we don’t listen–the challenge as every developer knows is that you can’t do everything and often suggestions from customers are opposites and you really can’t do both."

    That just takes me back to my comment on your previous post. Well, keeping all that modular approach in mind, while it is true that you can’t do everything as a developer there is really no need to, at least not at once, just deliver the bare foundation upon which everything else is based, then go on releasing those other modules deemed as interesting. Like that those who are demanding something can wait believing their wish/suggestion will be delivered in turn while those not related to such particular issue can keep on going.

  10. horphus says:

    Steven,

    Thank you for being so worried about making us sure you’re actually listening. That makes us feel like you guys really care, and encourages us to keep telling you what we think.

    William’s comment recalls me of someone (maybe Mark Russinovich, but I’m not sure) from Microsoft saying about Windows 7 being much more focused on backwards compatibility than Vista was, so I think it’d make things much easier for both of us if you could expose some of the design goals for Windows 7 that won’t and can’t change.

    For example, if you guys are really going to work on backwards compatibility it’s not worth it to keep asking you just to drop it from the system, right?

    I mean, of course many of us still thinks it’d be the right thing to do for the system, but we also know that there are things that are just predefined, and I think they are worth mentioning, making things clearer. After all, if people know what can’t be changed, I’m pretty sure people will come up with new ideas on top of the predefined features and goals you already have, instead of keep asking you guys to change or remove them completely.

    Ivan

  11. im.thatoneguy says:

    I think the most important focus for Windows 7 should be consolidation.

    The Mojave experiment already (via publicity stunt) that people actually do like the features of Vista.

    What needs to happen is a Office 2007 style consolidation of the features and presentation.  The UI needs to be made more consistent.  All of the existing features need to become more obvious and more simple for power users and basic users.

    I don’t think there is a single new feature in Office 2007 I use.  But I find it the best upgrade since Office 97 (the last time I upgrade office.)  

    The only way this is going to happen is with leadership and organization from someone who oversees all the projects and puts polish and refinement ahead of new glitz.

    I think the Zune team is another team which is doing a great job to make a consistent, simple and attractive product. I love my zune.

    And maybe in order to appease both the family PC and the UMPC business it would seem prudent to create a "Performance Center" where services that aren’t necessary are important are listed and power users can easily tailor their experience for more snap in exchange for understanding what they will lose as a result.

    If Windows 7 were a Office 2007 style upgrade I would be more than pleased.  Oh yeah and the 64 bit’ers could use some serious love.

  12. aterphasma says:

    Steven and associated team,

    It’s very refreshing to see the openness of how things are being handled in the development of Win7. It’s also quite heartening to hear that comments are actually being addressed.

    I’ve been looking around on various forums and sites recently, and have noted a number of people who swear that Windows Server 2008 can be converted, albeit somewhat incompletely, to a workstation. There still remain the issues of some software refusing to install on a server OS, driver incompatability, etc., but for the most part, people who have done this have reported a much improved Vista-like experience.

  13. Kosher says:

    Thanks for the reply to the comments.  It does sound like this is going to be what you mentioned (a minor release).  The suggestions that people have made are certainly not along the minor scale.

    Rewriting the registry, active directory, dns, and the filesystem schema, are all huge undertakings that require major testing and development.  One thing is certain though, .NET 3.5 SP1 is tested and solid.  C++/CLI is amazing.  But that’s just what users don’t see and that’s like talking about organges instead of apples (maybe a good punn).

    Users want to see unification in the UI and less hassle.  This is going to be a large task for any team.  I find that many of the Microsoft strategies for security, file system, user management, etc.  were all developed upon existing solutions in order to build on the existing code.  I am all about code reuse but sometimes, the code isn’t worth keeping and ends up being more of a headache than its worth.  The UI frameworks and style guides in Vista look like they’ve come from different teams.

    The end-to-end of it doesn’t mean you can’t have standardization across various domains.  That’s what’s lacking and painfully lacking in the UI.  If you have to write WPF as a framework in native C++ for performance reasons, do it!  I for one would love it.  This is the kind of paradigm shift needed to go big with a solid release.

    Much like Jobs did with Apple, you guys need to work on an entirely new OS from the ground up (like ‘BeOS’) and take the best of Windows with you.  Put the registry into the Entity Framework or allow applications to contain their own settings like Mac OS X.

    It would be great to see improvements to the services manager.  Like many others have posted, I would like to have the ability to really understand what "needs" to be running on my system and disable what doesn’t.  I am not talking so much about disabling UI based features but things like the DNS Client (why do we need it)?  A fancy UI that treats services as applications that can be updated indipendantly.

    Rework the core, modularize applications/services/features, unify and revamp the UI (keep the fancy UI stuff optional), tie it all together in a new OS that’s written with UX in mind first.

  14. Kannan Balasubramanian says:

    I agree with Ivan, if it is possible for you guys to provide the design goals then we can either restrict asking for features which atleast will not be possible on the first release or may be we can provide better  or alternative suggestions so that those features can be implemented in the first release in a different way.

    Kannan

  15. aterphasma says:

    (Continuing my interrupted post from above)

    I know Windows Server 2008 is a product that costs a great deal more than Vista, and it’s also aimed at servers, but with much of today’s hardware being so powerful, many of those server features also useful on a desktop. Windows Server 2008 supposedly has the same kernel, network stack, and countless other features in common with Windows Vista. Thus, I ask the question: why does enabling the desktop features of Windows Vista on Windows Server 2008 result in a more stable, speedier system? Shouldn’t it be pretty much the same?

    Just a few thoughts I wanted to put out.

  16. steven_sinofsky says:

    @eduardvalencia — there is no moderation of comments, just a slight delay in them showing up due to the service implementation.  As you can see your comments do show up :-)

  17. stalepie says:

    Kids should be given some freedom to play by themselves

  18. sciguy14 says:

    The last time I checked, Linux has been doing this exact thing for years.  Windows needs to be customizable based on the user.  If the user wants touchscreen support, then they should check it during the install.  If they don’t want it, they should be able to uncheck it.  And, by putting the disc back in at any point, the user should be able to COMPLETELY add or remove system packages.

    If you are worried about this being too complicated for very basic users, then just make some easy presets, like (customize this computer for touchscreen use, or make this a media center computer).

  19. LCARS says:

    I would just like to add my two cents to the "feature requests".

    I echo what others are saying about ensuring that Windows 7 has a clean and consistent UI. I would like to see close attention to little details (like icons). I would also like to see widespread usage of the Fluent UI introduced with Office 2007 in Windows 7 and its bundled applications.

  20. Brian in WV says:

    Wanted to add my kudos on the new blog.  Great posts and comments so far; it’s quickly become a tech blog "must read" for me.

    Also, a topic suggestion: I (and lots of others, I bet) would love to hear more about how integrated (or non-integrated) the Windows 7 and Windows Live development processes are/ have been in this development cycle.  In the past, we’ve learned a bit about plans for greater integration from some of your leaked internal MSFT memos, and at the appropriate point it would be great to get more detail about whether and how true integration between the efforts has come about.

  21. TehPenguin says:

    I know that in Vista, once installed, adding and removing Windows Features is reasonably easy.

    However, maybe a Windows 98 style installation system should be implemented, that is allowing people to select what they want to have when installing.

    Additionally, this could tie into Windows Marketplace and the different SKU’s very well. For instance, I could buy (using Vista as an example) Home Basic and then buy "Aero" and "Fax and Scan" duringafter the install – thereby having a cleaner install, paying less and yet with the opportunity to get new features as I need them.

    Having a "defrag and cleanout" for the registry is also recommended. Perhaps this can be integrated into a better "System Cleanup" that removes unneeded services, programs, registry keys and startup items. While this "System Cleanup" may require ISV to provide better descriptions for their software (and may end up being run by IT Professionals anyway), it’d be nice (as an IT Pro) to be able to sit down with a client and go through want they do and don’t use, without having to Google to find out what each application does.

  22. im.thatoneguy says:

    Oh and even closer 360 support would be great too.

    Xbox Live Marketplace is fantastic but why can’t I take my movies and TV Shows from one place to another?  

    It all goes back to everything talking to everything else.  That’s the Microsoft dream.  You  have a vertical market.  Let’s get it connected.

  23. RotoSequence says:

    If there are more opportunities provided to decide which elements of the Operating System package make it onto the end user’s machines, the ‘major vs minor’ release debate will be irrelevant. When people can choose what goes onto their personal boxes quickly, efficiently, and easily, Windows 7 will be a warmly welcomed release.

  24. Jaruzel says:

    The trouble with Vista is that it is first and foremost, an end user platform. All the extra features that made it into Vista after XP were squarely aimed at the end-user; Photo and Video work, the Aero interface, Flip, Simplfied file navigation…

    In the corporate arena, these features are undesired; we just wanted a sleeker, faster, newer OS. Something that does what an OS should do, and that is namely provide a stable foundation for applications. We don’t want all the bells and whistles that come with Vista, and nor should we spend time and money hacking with the build to remove them.

    I had high hopes for Vista business. I saw it as a baseline OS, with none of the bloat. However, what did we get? Vista Ultimate without the media center layer. That’s hardly a ‘business’ OS is it ?

    So, lets make Windows 7 nice and modular shall we (as mentioned in other comments). Enable an installation system to _easily_ check or de-check modules. Don’t want all the over-the-top eye candy? Uncheck. Need drive encryption? Check. Don’t want Media Center? Uncheck. Need domain functionality. Check. For sys admins, or build engineers supply a decent tool for profiling a build type, with a set of standard profiles such as ‘Core’, ‘Basic’, ‘Media’, ‘Everyting-plus-the-kitchen-sink’, which can be then tailored even further.

    Simple really. It’s a lot of development overhead I know, but at the end of it, you will be able to satisfy all the users, all of the time.

    -Matt

  25. daveli says:

    Here are some of the things that irritates me in Windows today, that I would love if you could fix:

    o Inactive windows that steal focus. If I switch to another window after starting a program, I do not want the started program to repeatedly steal the focus.

    o Inactive windows that steal focus. If I am typing in one window, I do not want another program popping up with a dialog that I might accidentally dismiss while typing.

    o Inactive windows that steal focus. Yes, it is that irritating!

    o Explorer taking away my control over my files. Explorer should not hide files from me, or lock files I want to delete because it tries to analyze the files content, or hang for several minutes because it tries to search my network (without any way to make it abort).

    o Limited support for multiple/large monitors. The taskbar should span all monitors and it should be easy to move windows between monitors and arrange multiple windows on a large monitor.

    o Frustrating to use for professionals. Settings is buried in multiple levels of "user friendly" dialogs and guides. No way to select what is installed. It feels that I have less control and ways to customization Windows for each new version.

    o DRM. It does not work, and adds nothing of value for the customer. Rather the opposite is true.

    I also think that Windows 7 should be 64-bit only, since this will force all hardware manufacturers to produce better 64-bit drivers.

  26. Wunderblume says:

    I really liked Vista from the beginning and see that it’s a great improvements compared to XP.. However considering the years of development I just feel like it just wasn’t enough..

    Anyhow,I really hope that I’m able to contribute ideas to this development.

    I’m certain that everyone is going to give his best to make Windows 7 awesome.

    (Ever thought about changing the codename to ‘Windows Awesome’?) :]

    I’m really curious to see how Windows 7 is going to perform and wish you all achieving your vision of Windows 7.

  27. fux says:

    Are you going to use flash instead of regular HDDs in Windows 7? Boottime has to come down to 2 or 3 seconds, max… like when you turn on your television set, or your receiver or any other home appliance! HDDs feel really outdated, prehistorical. And yes, the new icons, the 16×16, in Vista are really non-aesthetic. They’re really nice in big scale but the smaller ones.. you really cant see what they’re supposed to be in some cases! Hm, what else.. yeah all driver related problems. There has to be a way to put and end to all of that. Isnt it the drivers fault 90% of the time when you end up with a BSOD? Btw, I’m still having issues with my Realtek sound controller and its driver… it seems that whenever I install a driver from windows update Vista makes some crazy things to the system and I end up with a corupt sound system. One last thing; whenever I unrar an archive so that Windows creates a new folder in the folder where the archive is, that newly created folder always ends up sorted amongst all the other folders compared to XP where it always ends up at the bottom of your folder/files list! /greets from sweden

  28. stryqx says:

    +1 on the window stealing focus issue. This would have to be the most annoying problem I encounter. Closely followed by baloons.

    When I’m working on my PC I want to focus on the task at hand and not be distracted by the endless events that occur in other programs. A "do not disturb" action would be greatly desired.

    As for defrag – fix the problem at the source. Create NTFS v4.0 (or 7.0, dpending on your idea of NTFS version number) or even another FS that just doesn’t fragment and also performs smart placement of FS objects.

    Completely separate the UI from enumeration tasks. I really couldn’t care if an NFS mount/SMB share/whatever isn’t there. Do what the various BPAs and prerequisite checks in the installers do – drop in a progess animation and move on. UI freeze waiting for network timeout is last century.

    There’s an incredible amount of metrics recorded. Use it. Provide health check reports and guidance on solving immediate and impending problems. Provide optimisation profiles based on usage patterns. Preload/precache based on these patterns.

  29. NoEternity says:

    I want to back first for the windows focus stealing problem. Indeed VERY annoying.

    Two suggestions I have for the system:

    – at installation allow to put "Usr", "Programs" folder on different locations then C: e.g. allow to put them on different partitions to allow clear separation of user data and system. Maybe have an export mode installation for that…

    – Allow UAC to remember a setting. I have one program running at start-up that needs admin rights. I want to tell UAC please allow it once and then forever to run this program. NO NEED to ask EVERY start-up. UAC is useful but just half done.

  30. anonymuos says:

    One of the reasons I hated the Longhorn a.k.a Vista release was the extremely low level of customization and powertoys that the earlier releases offered. Please give us as much choice as possible, whereever classical old features are changed such as in IE and Windows Explorer, give users the option to use the old style behavior. And please don’t drop existing features completely assuming users never use them.

  31. Nidonocu says:

    Here’s my own personal feature list:

    1) Faster out of the box: The most impressive thing by far to those ‘Influentials’ (aka, Critics) who bad mouthed Vista is if they installed 7 and it was faster in most respects on similar hardware. Quite a feat to achieve, but one that would be worth its time to do.

    2) More polished UI: The Aero UI is certainly nice, and gave Windows a much needed clean up, but there is still some way to go and dusty corners that are showing things up. Having the team find and kill every little bug there would also work wonders on those critics (I’ll plug the Aero Taskforce here too who are finding and identifying these: http://www.aerotaskforce.com/

    3) Even more .net support: There are lots of neat Vista specific features that were introduced at launch, along with the brand new .net 3.0 framework. But two releases later and things like glass and TaskDialogs still require p/invoke! If you introduce any features, make them easy to access from native and .net with the same level of ease and sensible experience downgrades for those on Vista or XP without those features.

    4) Educate users about UAC: UAC is a great idea, but users still don’t know what clicking Continue means. Provide more information in the dialogs and explain to the user how to make a good trust choice.

    Thanks!

  32. mn4az says:

    @daveli

    "two customers can be found to want exactly opposite features"

    IMO DRM does have it’s place.  The ZunePass is a perfect example.  For $15/month I get all the music I want on my Zune.  Without DRM this wouldn’t happen.  For music lovers that used to spend way more than $180/year on music this is the greatest thing since sliced bread.  

  33. joewood says:

    Great blog, good to see it so frequently updated.

    This comment is about a technical issue relating to how your teams work together.

    The Windows Engineering team seems to work on the platform independently to other ‘towers’.  By that I mean that there doesn’t seem to be that much co-operation between Windows and DevDiv and even the Office tower.  I say this because the managed platform being built out on DevDiv doesn’t seem to be propagating over to the OS team.  For example, we recently had an issue with Remote Desktop and WPF.  I found out that the WPF primitives are not now supported in RDP and not likely to ever be.  This means that the Windows Team and auxillary apps don’t seem to be planning to leverage the ‘managed platform’.

    Rather than addressing this point, could you comment on how the teams interact and plan strategy?  How do you see the value-added apps in the OS fitting in with the equivalent ‘Live’ apps.  When will the engineering team start buying into .NET strategy?

  34. There is another possible category that I think you may be overlooking, especially with regard to which features to include or not include.  It is the one of Enthusiasts.  This specifically applies to the Windows Media Center ecosystem, which, to many of us, is one of the key components of Windows Vista.  Often it appears features are left out or “dumbed down” to minimize the support requirements for the masses.  However, there is a very strong community of Windows Media Center Enthusiasts that “want it all” and have been waiting for certain capabilities for years.  I feel safe in saying that Windows Media Center is the one feature that has garnered the strongest community of enthusiasts and supporters over any of feature of Windows (client or server) ever.  Please do not forget this group of loyal followers as you decide what features to include or not.  Although there is a huge list of “what we expect to see” in several blogs, I would like to point you to what I consider a great condensed version of feature requests at http://mediacenterguides.com/feature_requests.  As expected, some are simple modifications and others are complete architectural changes.  We enthusiasts would love to see both encompassed in addition to your adding some that will surprise and astound us when the product is released.  We look forward to the next generation.

    Regards,

     =D-

    Derek R. Flickinger

    Interactive Homes, Inc.

  35. Pendragon says:

    The "Must Have" for me is more like a "Not Have" that is I would like to see an end to the 32bit version of Windows. The 3 and a bit memory limit is the elephant in the room here. It can’t be avoided, My first PC has 640K and wow I though it was easily going to be enough. The more features we want will take more memory, it’s inevitable. Would the omission of a whole version of Windows class it as major release? not sure, but I believe it’s needed.

    For those that are asking for stripped out OS for business or speed. My question is why? If a £399 laptop can handle Vista without any issues with all of the eye candy what’s your problem? If stripped out OS is what you are really after there are other options out there like Linux.

    Windows has always been a heavy weight Errr sorry "Feature rich" OS needing(driving)the the ever increasing speed of hardware to get a move on. This is what Windows is, big, really big with lots of whipped cream and a cherry on top. If you want it to be fast then invest in the hardware and it flys along.

    As for this site, there are a lot of people here with lots of ideas. I’m sure any one of us could fill a couple of pages of ideas and suggestions of what we want to see. How about a section of the site where ideas for features can be posted and votes cast. A dynamic "Top Ten" wish list. The suggestions can be from us the users or from the Dev team themselves.

  36. Nacimota says:

    I think it would be cool to see a moar customizable GUI, even if its just little things like being able to group pinned start menu items and stuff like that. Organization and productivity go well together

  37. Hino Musouka says:

    Hello Steven, and greetings to all members of E7Team,

    For starters, I would like to turn your (and this blog readers’) attention to http://www.aerotaskforce.com/

    It’s a page I enjoyed very much recently as it allows users to promote the glitches in Vista that disturbs them to be seen by others and allow it to be assessed by all the community. Windows steeling focus, old icons, it’s all there. I, more than once, were pretty amazed to discover UI inconsistencies I haven’t had idea of.

    It’s a page of Long Zheng, who, not so long ago, with his team won 1st prize on Imagine Cup. I presume it’s a quite good argument to look at some suggestion placed there. Moreover, already at least hundreds of users made their vote and still keep announcing new discoveries.

    However, I understand that this is only a part of the problem your teams has to deal with. UI is a part addressed mainly to end-users, as businesses tend to keep their GUI smooth and clean (at least that’s the image arising from comments I’ve read here).

    Although I’m just a hobbyst developer, I can imagine to some extent the amount of job that you need to do to make a new version of system running. And moreover, I admire your efforts and the efforts of the people before you. I believe computer programming is a wonderful, though sometimes upsetting activity, and as many people here wish you success. Using the opportunity, I’d like to express some of my disappointment with Vista development and nowadays some negligence of this product in Microsoft policy. This is, I suppose, caused by a mostly disapproving reaction from the community after serving Vista RTM. And now everyone is counting days to Windows 7, guessing its release date, and hunting for any scrap of information concerning. Whereas Vista, by some, has already been pronounced dead. I feel it’s quite saddening and maybe embarrassing for all the people that gave everything to write Vista. Though we had to wait 5 years to see the Aero in its full glamour, and another one for SP1 to be able to copy our files at reasonable speeds, I still express my highest admiration to all the people who prepared ASLR (or what’s that randomization on DLL loading called), multiple enhancements in the Kernel (here I’d like to advertise Mark Russinovich’s  presentations  on Spotlight if there’s anybody who haven’t seen that), WPF and all the WXOXFs (I mean WCF, WF and whatever else). It’s a great deal of job made for improving usability and security of the OS.

    Still there are IT guys and big important companies that are concerned in manageability of product. I have no idea of how has that been improved or not so I cannot even make assumptions. However I had a little experience with small school domain based on Windows 2000 Server and later migration to Windows Server 2003 (BTW How can I get the address of the guy who decided about that change in naming?) and I was very positively surprised by an enormous, IMO, leap that was made between this two releases. Making it work on not so brand new computers was a challenge (I’ll remorsefully abstain from talking about my futile attempts of acquainting Windows domain with Linux which served that institution’s web page) but a very educating one.

    Furthermore the problems of compatibility and performance persist. But I’d like very much that you made a very secret applet, hidden from the eyes of the small, which would allow the more advanced users to configure their system accordingly to their needs. Even if it meant whipping out Aero and similar concepts. In this place I express more admiration for your patience when reading all these ‘do that’, ‘don’t to this’-like comments. For mine I’m very sorry, but I’m more lame than a developer and the desire is stronger than common sense.

    To end my tirade I’d like to express my support for your team and wish you more luck than the previous team has had. Now when Vista’s dying in agony (at least to some of the press I read) and Seven ‘s bar to jump over is placed quite high it may be of much assistance, this unpredictable luck. And do know  your posts are like sweet cookies, for us lamas, the most enjoyable kind of dish.

    I hope that comments in this blog remains close the merit of constructive discussion that maybe will help you with some implementation doubts you may have. I believe closer the user you are, the better result you’ll get. And though Romans used to say ‘de gustas non est disputandum’ I’d like to see this place a location of exchanging ideas that may follow to the product at least 50% of users are happy of.

    I also hope you can and will open the doors to Windows 7 (or let’s say open the window in Windows7 a little bit more). Some lame guys like me would appreciate some teasing screens from time to time as well;)

    –Chris

    PS And I beg you, please oh please delegate one esthetically bright person to revisit old icons. As you’ve probably already noticed, there are users that just love how their system looks like;)

    PPS May I ask if you could at some time give some hint of your position on implementing ‘open source’ ideas? E.g. OGG, FLAC in music. I mean why you can/can’t will/won’t think of that kind of interoperability between platforms?

  38. Charles Boykin says:

    64 Bit

    Rock Stable Core

    Drivers that work on the release disk

    That would be a major release

  39. Jalf says:

    Oo, just saw this in the very first comment, and thought I’d add my $0.02:

    "I’m developing on and for Windows since 10 years but lately I’m getting more and more annoyed by MS products. Just an example, I’m switching quite often between documents in Visual Studio with Ctrl+Tab, but VS2008 added the "feature" to show a preview of the documents which really slowed things down and is annoying. I have no understanding for such useless gimmicks. Can I turn it off? Don’t think so. Want to sell me/us licenses? Stop annoying me."

    What on Earth is Visual Studio doing reinventing this in the first place? Doesn’t such tabbing functionality belong in the OS? Why doesn’t the OS expose *usable* API’s for this? Why doesn’t it expose API’s that are good enough for Microsoft’s own products to use them? Why is it that Visual Studio and Office pretty much reinvent *everything*? I doubt it’s because they’re idiots. Perhaps they just want to achieve some kind of consistent presentation. Why doesn’t the OS give them that in the first place?

    I read on one of the VS blogs recently that they’re working on better multi-monitor support for VS10. Great. Wonderful. But why isn’t this the responsibility of the OS? Once again we’re going to see a dozen independent implementations of the same feature, just like the infamous Office Ribbon.

    About windows stealing focus, as mentioned in quite a few comments above, that too is one of my most hated "features" of Windows.

    Please, let 7 act like an operating system, rather than a shiny toy.

    The OS is responsible for presenting windows, tabs, tabbing order and giving focus to windows. It’s responsible for doing that *well*.

    A toy is responsible for entertaining, and nothing else.

    It’s time Windows started living up to this responsibility of being an operating system, because it matters. It matters far more than all the transparency and animations in Aero. I don’t care how shiny the ‘ok’ button on a dialog is if it pops up while I’m typing and I accidentally press it because I typed a space.

    I don’t care how gorgeous windows look, if they’re able to render themselves on top of the full-screen game I’m running.

    And I don’t care if you do the most advance realtime raytracing when rendering the system tray baloon tips. They shouldn’t pop up and interrupt me in the first place.

    Repeat after me, Windows is an operating system, not a toy. Its goal is to work for me, not entertain me.

  40. rokhoe says:

    Here I´d like to share some suggestions for the new Windows 7:

    1) Release versions:

      – Windows 7 Home Edition

      – Windows 7 Business

      – Windows 7 Home Edition 32 Bit

      – Windows 7 Business 32 Bit

      As you see, I think you should emphasize more on the development and use of a complete 64 bit OS, as most of the newer processors support it. Remove the 64 bit suffix at the end of the OS brand, as it should already be the standard. (64 bit is getting older: It´s been already 5 years since AMD released their 64 bit processors if I´m not wrong, and when "7" gets released it will be almost 7 years!) Make some kind of pressure, so that hardware manufacturers are obliged to develop 64 bit compatible drivers for their own stuff. Older applications should run on a virtual machine, so no one would complain about compatibility issues. Remember: Restricting progress on the new operating system, due to compatibility issues would be a very wrong decision.

  41. rokhoe says:

    2) GUI:

      Now I begin with the interface, as it is the first thing you see on an OS. Taking as example Mac OSX wouldn´t be in my opinion bad at all if it´s about interface aspects. What I´m saying is not copying 1:1 features of another OS, like what some guy claimed two years ago at the Macworld Vista was doing. The first rule to follow is don´t try to be something you´re not (you are Windows, not Mac OS, it´s that simple). But: There are several aspects and features which are good under Mac OSX, and which could even be better if ported to Windows. Remember: As I said before, it´s not wrong to take other OS´s as an example, it´s wronger if you just ignore them all.

       Now to the key features I´d like to see on "7":

       a) Try to find something more useful than Windows Flip 3D, because it´s not much more helpful than the old windows switcher (Alt+Tab). You see, that´s the kind of eye-candy we don´t need at all. A good alternative would be something like "Exposé" (I hope you know what I mean, hit F9 under Mac and see what happens ;), which immediately resizes all open windows and sorts them nicely on your screen. Click on one of these highlighted windows and you bring them to front. Until now I´ve not seen any better tool than that for managing your windows, and I don´t really know if there´s still something to improve (maybe you have some more creative ideas!).

       b) The "Show desktop" feature could use some transition effects (I know, like F11 under Mac)

  42. Kosher says:

    If we don’t switch to a different Microsoft OS, we will be switching to something. *hint

  43. Kosher says:

    hrrm… you didn’t post my novel? :(

  44. Antoine Dubuc says:

    Hi Steven, again,

    Thank you for this blog as is it entertaining and informative.

    I am sadly skeptical however as to the impact all the posters have. As far as I know, and I do admit I don’t know a lot about your processes, I thought that there was a group within Microsoft that ran user studies and determine from *there* what was getting in the feature list for Windows 7.

    Hence my question is really what concrete action is your group mandated to take concerning the feature requests everybody post in this blog?

    I mean, you say its a discussion. Hence its a two way street. You post, we post, you reply, we reply, etc. This is entertaining and informative. I am happy with just that.

    In the end, does this blog has any impact on the feature list of Windows 7? Are you mandated to report to the user study group or whatever it is called what the people here say and want, and is it taken into consideration?

    I would presume not. For many good reasons. The first being its not how its done.

    Hence, I am seing this blog not as a feature request entry tool, as most people seem to believe it is, but more as a line of communication with the Windows 7 team, who essentially, is mandated to implement features decided mostly by others.

    Please tell us who gets to decide what feature goes in, how, and who writes them!

    I hope you can enlightened us on this issue, as I believe that it may create disapointment and even resentment in many posters (who write long feature request lists and even websites!) who seem to think this blog is a feature request entry tool where actually it may be nothing like that. You tell us :)

    Thanks again for this blog, as it is a step in the right direction I believe, and thanks for your time.

    Antoine

  45. I understand that backwards compatibility is a major bugbear and O/S Bloat but is a necessary one, but with the advances in technology and virtualisation (soft grid, Virtual PC, VMware) surely all api calls to pre NT could be removed, and to speed up windows greatly, the application compatibility layer for pre vista could be a removable component.

    By the time Windows 7 hits RTM, I would have thought that most driver manufacturers would have caught up with x64, and that user’s hardware will have been refreshed to be x64 compatible (well, over 80% anyway) thus, X64 should be put ahead of the x86 version. Also, giving developers more incentive to make programs and drivers x64 compatible now will greatly increase the reception that windows has in the future.

    I would love to see Windows more modular, and especially think the business version should be. In my opinion the best way to achieve this is to allow IT Pros to rip it apart and rebuild it how they want (just like Windows XP Embedded).

    Last but not least, I really want native support for ISO’s in the file system, even if they look like read only folders.

    Regards,

    Robin Jones

  46. mtapman says:

    "End-users are generally the most straight-forward…a release is a big deal if they want to go out and buy an upgrade or buy a new PC." I disagree. The end user’s metric is whether or not they lose current functionality (e.g., some programs or hardware become unavailable), and the length of time/amount of energy required to perform the upgrade.

    Windows Vista SP1 was a great example of a technically major release that was insignificant to every end user I know. It happened virtually without them realizing it therefore it was minor.

    The other metric for an end user is a paradigm changing feature. For example, Apple’s backup/recovery tool, Time-whatever, changed the way people worked because they weren’t worried (as much) about losing data. Another example, Vista’s instant search capability changes the way in which people navigate their computers, once they know to use it ;-).

    I’m confused about the difference between Developers, Partners, and IT Professionals. Aren’t IT Professionals a super set of the other two? Perhaps a more defined set of stakeholders would allow a more sophisticated discussion. Off the top of my head I’d propose: Software Developers, Hardware Platform Developers, Hardware Peripheral Developers, Network Developers and Administrators, Security Engineers, Casual End Users, and Professional End Users. Each of those groups views a Windows release in a very different way and requires a different approach in order to satisfy. For example, the Casual End User is likely to be very interested in GUI upgrades and multi-media enhancements, which are virtually inconsequential to most Security Engineers.

    "If we listened well and focused our efforts correctly, then each type of customers will find things that make the product worthwhile." Please, please, tell me this was a misstatement. I don’t want Windows 7 to be worthwhile to everyone, I want it to be great in and of itself. Build a spectacular foundation and let it be worthwhile to whichever groups really need that foundation. Over time it can become more useful to other groups but trying to satisfy everyone at once is a recipe for disaster. Look at the development of Linux, for example, starting with a solid server platform has enabled the migration to a desktop platform. It’s the solid base that makes Linux so attractive, even with the GUI problems that persist with X Windows.

    Finally, what about security? Several long posts now and very little, if any discussion of security. I was particularly disappointed that the team structure mentioned earlier didn’t include a security architect.

  47. Nehemoth says:

    "We intend for Windows 7 to be an awesome release."

    That all what I was thinking when reading the article. In the end the only think that matters to me (as an IT and End User guy) is the experience, my experience would be a nice balance between performance (change under the hood) and a better interface (a lite one as Windows XP Classic, no Vista).

    Indeed, the another thing that I want from this team is that you listen us a little more

    For example this Youtube video tells a lot of to me

    hxxp://www.youtube.com/watch?v=B5D43p4_qcY&feature=PlayList&p=5F82CEDB9A6B9643&index=30

    Thank you

  48. teho says:

    What I would like to see from a new version is an OS which does what it is supposed to do. Run the programs the user wants to run when the user chooses to run them, nothing more. I mean don’t preload everything into memory at startup. I don’t need word, excel, IE, outlook, adobe-programs, itunes etc. to  use memory in case I should want to run them. Often I chooses  not to run them.

    I want programs to load and run when I decide to run then. If they loads in 5 secs instead of 2 sec is ok by me.

    Make an option to load only what the OS needs, diskdrivers, network, virus/spam checks etc., nothing else. No program preloaded, no blinking gui, just a simple gui.

    This would be an OS for me. Today I am running Vista 64 business and it feels like driving a truck when I only needs a little fast car.

    – Terje

  49. mike_abreu says:

    I think a sizable portion of users are completely neutral to an OS release, be it major or minor. Many users are only interested in changes to the software they use. So for the most popular tools that are well supported across many OSs (Office, Adobe, iTunes, Java, web applications, etc), the user’s choice of OS or choice to upgrade might merely be one of convenience, personal taste, or performance.

    From that mindset, I give a +1 that a key motivation for Windows 7 ought to be consolidation. Optimize the user experience for the common case. Some people don’t want anything from the OS; which really means they just want it to be fast, stable, and unobtrusive. Make it consistent and compatible. The best new features will be those that solve existing problems.

  50. MortenJ says:

    Hi W7 team

    Thank you for sharing the insights.

    My company are really looking forward to Windows 7, we are longing for upgrading our Windows XP client base.

    I was wondering whether a higher degree of efficiency is part of the coming Windows 7?

    To avoid the huge criticism which hit Vista in my country among others, I would presume the resource requirements for Windows 7 would have to be say 5-10% less than that of Windows XP.

    Releasing a new OS, which requires more resources than the one it is trying to replace would, given the current focus on the global warming and our limited resources, be like releasing a version of Windows without support for the internet.

    I think you have done that one. No ever so smart new features could make up for that, so give us something more efficient, please!!!

    Kind regards,

    Morten Junkstöm

  51. PimpUigi says:

    I doubt the resources requirement would go down.

    Many people would probably like it to go down, but I highly doubt it will.

    It’s most likely being based off of Vista’s initial code.

    While I wish it would instead be based off of one of the longhorn betas.

    They still had the XP start menu, and the Windows Picture and Fax viewer.

    They could still view animated gifs as well.

    Here’s hoping Windows 7 has both of those things included.

  52. TechWhiz1337 says:

    What MS needs to do is make a new version of Windows. I’m sick and tired of Windows 2000, its been almost 9 years now!

  53. aaronsteers says:

    One of the most commonly requested features, ironically, is a feature that is already in WinXP.  If you type "msconfig" into the run dialog, you get a fairly straightforward start-time optimizer.  It doesn’t get used because no one knows that it exists.  This gets back to what another commenter says about Office 2007.  Without adding features per say, Office 2007 was a huge benefit because of the consolidation affect.

    Similarly with Windows, please work to consolidate, especially configuration.  I don’t want big pictures OR appletss when I go to the control panel: I want a seamless configuration panel for EVERYTHING (like a fully-loaded mmc console with better navigation).  When tools and features grow organically, they often appear in odd or difficult to find places.  Advertise the features that are already there and make them easy (easier) to find.  

    Another suggestion: incent people on building up their own skillsets.  Charge them an extra ten bucks for the OS license (who will know?), but then give them twenty back (in cash) if they complete the installed training course (perhaps incrementally per training module).  If you incent laypeople to get some basic training, I am positive you will have a more satisfied user base.  This would also allow you to "change more" of the interface with less backlash ("I can’t find this, I can’t find that").

  54. PatriotB says:

    Steven, thanks for a good post.  However, a blog post about "measuring the scale of a release" just wouldn’t be complete without adressing the numbering of Windows "7" vs. its apparent version number 6.1.  Was Windows 7 originally intended to be version 7.0, and then scaled back?  Are the underlying version numbers going to become less relevant as time goees on, as releases become more consistently "medium" (major to some audiences, minor to others) over time?

    From what I’ve read, allegedly the Windows Server folks have already come out and said that Windows Server 7 is the same as Windows Server 2008 R2 and can be considered a minor release.  However since Server 7 and client 7 are the same codebase (presumably), and share so many common files (e.g. the shell), how can the codebase changes be both major and minor at the same time?

    The answer is exactly as you described, that different audiences have different perceptions — even when looking at the same product.  The difficulty is that Windows has a single version number for all almost all its components (except for things like IE and WMP), yet some components undergo more changes than others.  It’s hard to gauge an "overall" change level for a product as broad as Windows.

    Overall, I’m a fan of minor versions — what’s the point of having that field in the versioninfo struct if it’s always 0?  (Like in Office — other than Outlook 98 (8.5), when was the last time Office had a non-.0 release?)  The only danger with minor versions is that sometimes you never escape them — Mac OS has been doing 10.x’s for years; Java had 1.1-1.4 until it jumped up to 5.0.  A series of minor/medium sized releases may over time add up to a "major" change, but without any single one being large enough to do it on its own.

  55. steven_sinofsky says:

    Many folks have done the math to explain why we chose the name Windows 7 – this is because Vista is version “6” of the Windows product line (Windows 1, 2, 3 on 16 bits, Windows 95 was version 4, then Windows 2000 was 5, XP was 5.1, Vista was 6). So we chose “7”. (This doesn’t count 98, 98 SE, Me, and of course NT 3.1/3.51 but they all fit in).

    What we learned along the way was that one of those “major/minor” decisions that is material is changing the version number. We learned that many software packages and drivers (particularly setup programs) *hard code* the version check and fail to operate if the major version number changes. There are many reasons for such a check—some are because we might have made an architectural change to Windows creating the deliberate incompatibility, or we might have inadvertently created an incompatibility, or it might have been a defensive coding practice against either of those, or it might be a version checking “bug” in the third party code. Of course it our job to keep products running and compatibility, but when a developer does this check we don’t know if there is a reason or not and so we can’t just “fool” the product into thinking the version is older than it is. We’d love developers to check the minimum required version to run on, rather than exact or greatest.

    As we started Windows 7 we chose to keep the major version number at 6 so as to maximize compatibility for third party developers. This is really about our commitment to compatibility. And as you have seen with past releases, the major/minor nomenclature for the qualitative aspects of the release don’t necessarily line up with the numeric designations. What you’re seeing with Windows 7 is just a deliberate focus on compatibility over version number vanity (just as you saw with the “major” release of Windows XP).

    It is worth considering that in all these releases we probably could have kept revising the minor version number just as easily as the major version number regardless of how much of Windows we changed (and for which customers). While we make architectural changes and our share of mistakes, as a general rule we bring forward a very significant amount of software.

    There’s a fun story about this with respect to Office and it shows the wacky world of version numbers. Office 2007 was codenamed Office “12”. That was because it was the 12th version of Office. Really? There was Word 1, Word 2, but then we changed to Word version 6 to compete in retail with WordPerfect 5.1 (speaking of version number vanity!). At the same time we had Excel 1, 2 (the first version for Windows), then Excel 3, 4, and 5. Office for Windows was version 4 (the one most people started to use) and had Word version 6 and Excel version 5. Then we did Office 95, which the official name was “Office v 7.0 for Windows 95” which was named that because our support teams insisted on having a version number that was the same across all the “programs” in Office. So we picked 7.0 as the one that would be compatible even thought it was not “5” (the version of Office) but simply the next in the sequence of the highest numbered product. Then what followed was Office 97 (8.0), Office 2000 (9.0), Office 2002 for Windows XP (10.0), Office 2003 (11.0), and Office 2007 (12.0).

    –Steven Sinofsky

  56. EnergyResearcher says:

    I think it’s quite obvious that Win7 should be faster in its operation than previous Win versions, so I’m not going to add my redundant vote to that request. However, something very important in terms of speed is the UI.

    There was a study ( http://www.sciam.com/article.cfm?id=tough-choices-how-making ) recently about how the brain fatigues after making lots of small decisions. Well, I think that Vista makes my brain fatigued because of its poor UI. I have to make lots of clicks in order to find the feature I’m looking for.

    I would appreciate if the engineers behind Win 7 could think of each click as "tax" that the user has to pay. Naturally, the user wants to pay as little "tax" as possible which translates into an OS with as few clicks as possible in order to operate it, and ultimately a faster OS.

    I have used Windows since the 80’s and never made the switch to Linux or Mac (although I have tried them). Win ME nearly drove me to a switch, but Microsoft killed WinME quickly (thankfully). XP is currently Microsoft’s lifeline, Win 7 (without SPs) has to trump XP (not Vista) right off the shelves when it gets out. Otherwise, I’ll consider Linux/Mac one more time.

  57. ITP says:

    Me too I take for granted, that the W7 will be faster by all means. But the performance I’m thinking is not about "better MM page locking scheme","new working set trimmer algorithm" or like this.

    What W7, IMHO, should do, is put the user’s wishes and actions above all.

    I mean, when I boot the PC, I don’t care if "Remote Registry service","Network list","SSTP" or other things are running or not, or that system needs to build network discovery, or that search is doing updates. I just want to launch my browser as quickly as I can to check out things. Anything else must wait as much as possible. It cannot take 7 minutes from the start to be able to launch my favorite app, just because system wants to do "x and y in the same time". The system should wait the user, not otherwise. When I’m done, I’m idle, do whatever you need.

    The I/O and MM prioritization in Vista should be extended in W7 to override much more system activity to allow the user actions be carried out first. I launch an app. Halt, do whatever is needed to launch, and then resume.

    Of course there are dependencies, things that must be done before and so on, but I think much of it should be deferred when user is making actions.

  58. davidtan says:

    Great to see microsoft recognizes the categories of people your product is targeting. Keep it simple. Cheers!

  59. lyesmith says:

    1.) More responsive file system. – It should be possible to rename or even move a document (any file) while it is open.

    2.) Multiple desktop – swithcing between desktops with Windows+[0-9] keys.

    2.b.) maybe a consol available on [W+0] I love that in Linux.

    3.) Quick overview of the desktops that belong to the user.

    4.) Should be possible to log-in to diffrent desktops with diffrent user account.

    5.) Microkernel, dont load modules that you dont need.

    6.) Easy and quick (1-2 click) way to connect any number computers and share files between them. Easy way to set up a local ad hoc network. Even if the computers are in differnt workgroup or domain.

    7. Easy way to set up secure peer to peer conection between computers from anywhere.

    8.) Better safer sleep and hybernation mode. No more BSOD. Stricter rules for third party devs.

    9.) Consistent UI. It should be supersimple but with a button to reach advanced features and settings. I like Office 2007 but I dont use the 99% of the featueres. Those features should load only on request.

    10.) Spellchecking though the entire OS. Not just in Office and Outlook.

    11.) Thesaurus and dictionary available all time in the entire OS.

    12.) Wikipedia search available all time. (maybe online service but reachable from search bar);

    13.) Geo search, maps available all time. (maybe online sevice but reachable from search bar)

  60. thecolonel says:

    i work as a 3d artist in a small office. apart from our render farm we have 10 win XPx64 workstations and one windows vista business 64 mobile workstation. XPx64 is a joy to work on, Vista64 a nightmare. every single tiny thing about it is just so slow, which i simply can’t understand as it’s the newest hardware we have. navigating in explorer is like wading in mud. installing any programs always seems to take longer whilst we work out why they don’t work like they do on XP. getting hardware to work, like our totally standard HP laser printer is torture.

    we just want to work fast and efficiently. we don’t care about bling.

    please make Win7 work as well as XP, simple as that

  61. bpaddock says:

    lyesmith –

    Thanks for the comments, some good ideas in there :)

    A couple notes:

    Being able to rename an opened document isn’t a matter of the filesystem, it’s a matter of how the application opened it.  If the application *locked the file* the OS (no matter what OS it is) cannot do anything about that.

    By locking the file the application is specifically telling the OS *not* to allow other changes to it, because it is in the middle of doing something with it that might change it’s content / size / location, or even its name.

    The Windows+[0-9] keys already have a function, and changing that would be very annoying for people that are used to having them launch QuickLaunch items.

    Windows already uses a microkernel architecture.

  62. TaylorVista64 says:

    Kudos on making the dev process public, also for taking feedback.  I would make a feedback form where users can collaborate on suggesting new features.

    Here are my suggestions:

    1. Put all system functions in one place:  Control Panel.  Currently, Windows has Control Panel which holds most system functions.  However, there is also the Administrative Tools folder on the Start Menu.  Or, if I want to go to Disk Management, I have to open Windows Explorer and right click Manage on My Computer.  Or, to ping another computer on my network, I have to type “cmd” in my Search window.  Or, to defrag a drive, I must right click the drive and click on Tools.  Put all these functions in one place to make Windows easier, cleaner, and more efficient.

    2. Include antivirus software in Windows.

    3. On the start menu there is a frequently used program list.  How programs are added and subtracted is a mystery.  If I use the program once, it should appear on the list.  As I use others, they appear, with the oldest used program dropping off, as is done with every other program that has this feature.

    4. Get rid of UAC.  If I am the only user and an admin on my machine, why do I need permission to delete a file?  How and why is this permission then denied?

    5. Similarly, don’t put folders on my machine that I can’t delete.

    6. Make different versions for different levels of user.  Make a Newbie version with all UAC delete protections in.  Make a Power user with newbie stuff out.  Make an Elite user version with extra system and network utilities installed.

    7. Get rid of hidden files and folders.  It is irritating to have to go to Folder views and turn on hidden files to go into hidden folders to make changes, and then go back and turn it off again.  If you don’t want noobs to mess with them, put all hidden file/folders buried somewhere where they can’t reach them.

    8. Beef up Windows Explorer.  There used to be custom buttons for simple operations, like Cut, Copy and Paste, but those were removed.  Add those back and allow for more file management features in the interface.

  63. bpaddock says:

    TaylorVista64 –

    Thanks for the comments as always.  A few responses:

    3.  As you mentioned, the Start Menu includes a "Frequently Used" list.  Using something once does not make it frequently used.  That would be a Recently Used list, which is not what the Start Menu has.  If you want something to always be on the start menu, you can right-click the icon and click "Pin to Start Menu."

    4.  If you are the only user and an admin on your machine, then UAC gives you control over what privileges you allow programs to run with.  It isn’t YOU asking the computer for permission, it’s the application asking you.

    If an application runs with lower privileges, then it is less of a risk to your system.  I am NOT talking about running a malicious application.  UAC isn’t built to stop that.  UAC is built so that Outlook and Firefox and AIM and all the other internet-connected software you use doesn’t have permission to do things *they never need to do*.

    That way, when you browse to a web page that exploits Firefox or open an e-mail that exploits a bug in Outlook, the malicious code injected into that process is restricted in the damage it can do.  Some people call this "sandboxing" although that’s become a bit of an overloaded term.  But the idea is that you build a security wall around your applications, defined by the application, which it promises not to try and escape.  If it does, it’s probably been exploited by an attacker.

    In the case of IE, UAC offers quite possibly the most powerful protection mechanism available, by running IE in "protected mode," which means it has even fewer privileges than your normal non-admin applications.  This is a fantastic security advantage and is likely the #1 reason that Vista machines have far, far fewer malware installations and viable attacks than even XP SP2.

    5. I make the opposite argument – why would you be allowed to delete something that you will never want to delete?  That only opens up the possibility of you accidentally destroying your system, and then complaining to Microsoft that if something can’t be deleted it shouldn’t be deleteable.

    6.  You probably don’t mean "power user," yuou probably mean "you."  Unfortunately it’s very difficult to make a "power user" version of Windows.  For example, you mentioned leaving out UAC.  UAC is often most effective and most appreciated by "power users."  The same people who appreciate the least-privilege security model in other OSes and have long requested better access to it in Windows.

    7.  This is just silly.  You say it’s too hard to get to hidden files because you have to toggle ONE SETTING just ONCE.  Then you propose a solution that would make them harder to find and get to?

  64. lyesmith says:

    bpaddock,

    As far as I know the Vista and WS 2008 uses hybrid kernel essentially a monolithic one.

    For some reason in OSX they manage not to  lock a file when you open it. It is free to rename or edit with other application.

  65. whizzmo says:

    I am resigned to the fact that "Windows 7" will probably just be Vista/2008 R2 and may bear a version number of ‘6.1’ .  With that in mind, here are a couple of good ideas that I think should end up in the "true" (v7.0) Windows release:

    *  64-bit only.  Several other people have mentioned this, but I just wanted to add my vote.

    *  A new per-program permissions architecture.  I envision an environment where the install file specifies a mfr/publisher("Adobe"), program name ("Reader"), and version ("10.0").  The OS then creates "C:programsAdobeReader10.0" and "HKLMSoftwareAdobeReader10.0" and then gives RW permission to this application (hash-based?).  This could be implemented in a number of ways, including local security groups ("MYCOMPUTERAdobe Applications" gets RW access to HKLMSoftwareAdobe).  This has the benefit of limiting damage that one mailcious/exploited application can do to other applications.  There’s no good reason for Word to be *writing* to c:programsAdobe, anyway, right?

    *  A pluggable codec pool for Compression and Encryption algorithms.  This would allow Explorer (and all other applications) to open archives based on methods or schemes not present when the application was developed. ("Notepad can’t open that .PAQ10o file?  Let Explorer query $search_engine to find the appropriate decompression/decryption DLLs!")

    *  Speed of .Net applications.  This is probably outside the scope of Windows 7 engineering, but I have to bring it up.  One of my biggest complaints with the Exchange 2007 and WSUS 3 management interfaces is simply the time it takes to load up and perform simple(?) tasks.  The admin consoles from previous versions are speed demons by comparison.  I don’t know if this is code bloat or something else, but it definitely detracts from the user (IT staff) experience.  

    Hope this helps,

    Whizzmo

  66. kizkoool says:

    Aero glass is so dull and boring looking. Black and transparent shades. It’s quite frightening and repellant if you think about it a little.

    To have a pleasant experience with Windows 7, the end user need colourful and vivid patterns that bring us joy. Something along the Windows XP skin would be perfect.

    You have to make us feel that the OS is not targeted at rocket scientists.

    Vista with all it’s technical glory has lost what I would qualify as happiness and conviviality.

    It’s quite hard to describe the feelings given by a User Interface but I hope that you understand my point.

    Wishes you the best with Windows 7

  67. Michael_D._Barber says:

    I wanted to say two things about Windows 7.  First, I hope the performance is much faster than Windows Vista.  I expected more and was disappointed when I didn’t get it.  Second, as a blind computer user, I am asking that you please provide the option for a talking install of your operating system, much the same as the Leopard Operating System has.  It’s time for Microsoft to step up to the plate and really get serious about accessibility for blind computer users.  I hope this will be integrated into the new system.  There are experts at the National Federation of the Blind headquarters who are ready and willing to work with you.  Might I also ask that you continue to include the screen reading software developers such as GW Micro, Freedom Scientific, and Serotek corporation as you develop this new system to ensure the greatest accessibility for us.  Thanks much.

  68. erik233@hotmail.com says:

    Like many, I’ve been using Windows since the beginning (well, no, not Windows 1.0, but close.. 2/386).  I’ve seen many iterations of Windows and followed it’s course from application to subystem to OS and beyond.

    I use Vista every day.  I choose to use Vista.  I could be using XP, but I don’t because I believe in many of the advances made in Vista, and the longer we have the world hanging onto old versions of the OS, the longer it will take for applications to begin using new features.

    Having said that, I still find myself annoyed beyond belief quite often with Windows.  I have contemplated switching to Mac and Linux many times, but haven’t for a variety of reasons.  However, those reasons are slowly becoming less and less important.

    If you want to keep me as a customer, then a lot of changes will have to happen.  I’m not unreasonable, and I’ve thought through both sides of most of these issues.  I understand that all this is a balancing act, but that’s small consolation when I as a customer end up on the wrong end.

    1) Reduce complexity.  I know that’s a vague comment.  However, the one thing i’ve noticed in Windows over the years is that it gets more and more complex in it’s architecture.  It’s a good thing to build layers and build on what has come before, but all too often those features appear more bolted on and cosmetic rather than integrated.  Aero is a great example.  All the time I notice quirks and gotchas that seem to stem from it’s "bolted on" nature.  It seems like the old Windows USER is there, drawing the old UI, but then it’s somehow intercepted and AERO redraws it the way it wants to.  I’m certain that this is threason AERO uses so much memory when similar implementations on Linux and Mac use significantly less.

    2) Speaking of Aero, it needs to be fine tuned.  The monster thick borders are annoying, and it needs to be optmized to use far less resources, particularly memory.  I know memory is cheap, but I prefer to use the memory for my applications, and not the OS (buffers and caches not withstanding, because those can be discarded when applications need the memory).

    3) DWM needs improvement to make it more programmable.  It would be really nice to see API’s to allow third parties to easily create new effects (similar to Compiz/Beryl).  The current API is quite limited in this regard.  I want to see spinning cubes or cylinders, warped displays, wobbly windows, new transitions, etc..  And please, don’t hard code these API’s to specific effects, make them generic enough to allow third parties to use their creativity, rather than locking them into yet another paradigm.

    4) Speaking of resources.  Vista uses way too many of them.  I know that too is generic.  But seriously, the OS is unusable with less than 1GB of memory.  It’s barely usable with 2GB, and really requires 4 if your’e a power user.  Even so, with just a dozen applications running and several instances of IE with multiple tabs on each, i’m already constantly using 2.5+GB, usually rolling into 3GB quite often.  There has to be ways to reduce application and memory footprints.

    5) Improve 32 bit compatibility on 64 bit.  That includes device drivers.  I wouldn’t mind a 32 bit WOW layer for this.  There still a lot of very functional hardware out there that doesn’t have 64 bit drivers.  And yes, I realize this may conflict with #4 above, but I consider backwards compatibility to be something for a VM layer, which means that if you have all 64 bit hardware and software, you can jettison the VM layer and recover the memory.

    6) Rethink UAC.  I love what UAC does, just not the way it does it.  Both Mac and Linux have a timer associated with their SUDO implementations so that it doesn’t keep asking for permission for a small period of time.  Also, reconsider combining UAC with file permissions.  This is particularly a big problem on things like removable hard drives where there might be permissions from another OS.  It simply prompts too often for basic tasks.

    I know there are a lot of people who complain about UAC being there at all.  They want to be able to do whatever they want without annoyance.  But there has to be a middle ground.  Perhaps you can create a "level" of protection, where some actions do not require or automatically raised.  Perhaps even make this a selection of items in a check-listbox.  "Do not prompt for non-system file permission elevation" or some such.  Also, it’s fine to make this a group policy item, but there needs to be a normal UI for this as well, as group policy is hard for many to figure out.

    7) Reduce the waits… or at least give the option to reduce the waits.   As an example, when I go to delete or copy a folder with a lot of subdirectories, it takes forever for the copy to begin as it has to scan the whole subtree just so it can generate a semi-accurate time estimate (and other things, like determining if there’s enough room on the destination drive).  I really don’t care about the time estimate.  I’d rather have the option to turn that off, and have it just start the copy immediately.  If there’s not enough space on the target drive, then that’s my own fault because i turned that feature off.

    There are similar time sucks all over the OS.  Scanning subfolders for other folders so that the + can be shown in explorer, for instance.  This is particularly bad over network drives where it can take minutes for the UI to become responsive.

    8) Above all.  Make the OS more responsive.  Make it "FEEL" fast.  it doesn’t matter if it is or isn’t, if it feels fast users will complain far less.

    I like to give a good example.  When Windows 98 came out, it introduce a transition in the start menu and right click menus.  This transition made it take longer for the menus to appear (even though if you moved your mouse pointer right away, it would cancel the transition and make the menu draw right away).  People complained that the new OS was "too slow" because of this.  It wasn’t, it just felt like it.  And, when people think something is slow, they assume it’s because it’s buggy or bloated.

  69. mmind says:

    It’s great to read your post but it worries me somehow as it seems you guys are thinking more about small details like major/minor than the product itself.

    I think it would be best to simply try to create an awesome OS and let the users decide if it is major or minor. Considering the disaster called Vista any release would be major when you do things the right way.

    Regarding this I second most of the comments above and in the other posts:

    – make W7 rock solid

    – make it fast and efficient. Honestly, when an OS sucks most of your hardware power something is terribly wrong. I don’t buy a computer to run an OS. I buy a computer to run applications.

    – make W7 customizable in every way – at least give us the chance the be in charge when it comes to our computers. This way you can comfort any customer with any taste. Offer those options during install and anytime later.

    By the way: would that last wish possible for you to realize? For example: are you able to rather easily design the Windows system in a way that I can remove IE or any other "tool"? Or is that not possible because those apps are so tightly integrated into the system that it would take years?

  70. Happy-Dude says:

    I’m glad to see that Windows 7 is coming along really nicely.

    From what I’ve been reading at ShippingSeven and Engineering7, the future of Microsoft looks bright and the *refreshed* attitude is in the right direction. (Like IE8 and standards; glad to see that MS was gone the *correct* way.)

    I just have one really big request:

    –Please make Windows 7 (32-bit) friendly with single core processors (like P4’s).

    **Reason: many people’s main reason for not upgrading to Vista (or mine at least) is the demand of the hardware. Sure, I have 2 GB RAM, a nice video card (GeForce 7800), and others.

    But in order to run Vista well out-of-the-box (tweaking aside), people should at least be running Dual-Core processors. I, unfortunately (and like many other veteran power users who do not like upgrading right off the bat) still use a P4 for my system.

    I truly hope that Windows 7 will be more resourceful with the other hardware too (the new driver model and RAM management works well). The jump from Windows XP 32-bit to Vista (any) was unlike any other for most people.

    I believe this was because the jumps from Windows 2000 and Windows XP wasn’t that great. And given the time in between Vista and XP, people weren’t expecting that (Longhorn builds were near XP specs).

    Sure, 5 years and hardware evolves (hence Microsoft’s expectations for Vista and dual-core and 64-bit and graphics). But … You guys forgot one thing:: You ARE the computing industries STANDARD. Windows XP IS the standard and perhaps will be for the next year at the very least.

    People got by those 5-developing Vista years with enough to run XP (enough, well, or superseding). Those people who ran XP with Dual-Cores and 64-bit and 2GB of RAM (and almost, by default, a great video card; it wouldn’t make sense for people to have these types of systems without a great video card to do with it) were prepared for Vista.

    Those unlucky to discover that their old hardware (perhaps legacy compatible with the XP standards) ran undesirably or even were incompatible. That makes people to stop wanting to move … That makes eveolution (or understanding) a little harder for some.

    So, my only request: make 32-but Windows 7 more resourceful with single core processors. I ask for only 32-bit because 64-bit is the future (hench, high clock or multicore) and ready almost out of the box (like 64-bit P4’s with hyperthreading to Pentium D’s to Core 2 Duo’s). Please make 32-bit Windows 7 efficient and resourceful with single cores. Please allow it to run and process at surprising and desirable speeds. Please … Make it efficient and yet, powerful. Please, optimize the code and make it flexible. Please don’t just give us another Vista and expect it to run due to time. XP is still the standard, and hardware will continue to supply it as a minimum.

    –I wish to note that since Vista, manufacturers have been adding into the hardware to compensate for it. That’s good … For people who can take advantage of it (purchase or upgrade to fairly new components). Those who have come from the 98/ ME/ 2K/ early XP ages aren’t as lucky … Everything’s going out of reach for those users now. And one of those users is me.

    I want to experience the visuals … I want to take advantage of the services Windows has to offer … But I do not want to give up on my PC.

    This will make single-core users happier for the given time (before technology evolves to beyond their reach, again forcing another upgrade or PC purchase).

    –MS must realize the people who choose UPGRADE instead of PURCHASE are the people who are running older hardware, but still want to keep to date. These people are resourceful users (power users maybe) and sometimes choose that they rather not give up their systems.

    **Yes … My PC has sentimental value to me (2001 to today, XP, best computing experience for me by far, and the only computer in my lifetime I can call my own). I’m in my teens, if you wish to gauge how long I’ve been using computers.**

    The processor, unlike RAM and Video, is a difficult part to upgrade. There’s only so much that a motherboard of the XP SP1-SP2 ages can take (2002-2005). My motherboard is maxed out with a 2.8 GHz processor, and no choice to upgrade to dual-core (trust me, I would if I could).

    It’s impractical for people, who got by with  XP (2001-today) in addition upgrading their system overtime, to actually purchase another system. I love my computer … And I ABSOLUTELY do NOT want to give it up simply because the operating system is not resourceful with its given hardware.

  71. erik233@hotmail.com says:

    Here are some more comments I’ve been thinking about.

    1) Paged Back Executables and Page Backed VM has become a bottleneck with todays multi-core processors.  I find my processes are often disk I/O bound even when there is plenty of free/available memory.  In addition, with the trend of moving to non-volatile flash based hard disks, the old VM is proving to be more of an impediment.  I think the VM needs to be rethought to make disk I/O less of a bottleneck for virtual memory performance.

    2) NTFS is also showing it’s age.  It’s aged well, mind you, but I think it’s time for a radical change.  The filesystem has to become more flexible.  Reparse Points and what not have greatly expanded it’s functionality, but it’s time for a change.

    A big change would be to effectively drop the concept of disk drives, ala Mac or Unix filesystems.  Yes, there’s a lot of software that depends on drive letters, but i’m sure you could come up with a compatiblity layer where drive letters could be mapped to folders in a more advanced way than subst or drive mappings.  All the parts are there to accomplish this, just pull the trigger.

    Also, the abiity to replace in-use files without rebooting (similar to Unix fileysystems) and a fragmentation resistent filesystem would top things off.

    3) As someone else mentioned, we need built-in virtual desktops.  There are two ways to accomplish this, and I think both need to be implemented at the OS level.  The first level is Virtual desktops, however most third party tools are just window hiding hacks.  This is not acceptable since it’s too easy to confuse these solutions.  A better solution is to create Virtual Displays, that act like extra monitors, with complete OS level window management.  These can then be swapped into the "real" displays as needed.  The second solution is more like Fast User Switching, and allows multiple users to be logged into the console, however this should work with computers joined to domains as well.  Further, it should be easier to switch between virtual consoles than FUS is.  There should also be API’s to make for third party display managers.

  72. vinicius.snts says:

    Looks like I can be categorized as IT Professional.

    Some features I’d like to see at W7:

    >>>>Since in most Corporations the machines are branded ones, it is interesting to have more hardware administration features via AD. -I am using today 3rd-party software to administer machines boot-up configs, SMART and temperature monitoring; Wake-on-LAN and hardware inventory would be an interesting addition.

    >>>>Same with software.

    -Installed software and configuration parameters report would be an interesting addition.

    >>>>Non-Microsoft software configuration via AD.

    -Common software like Adobe Reader and JRE are a pain to configure. You need to make logon scripts to mass-configure them.

    Custom Software configuration would be nice too (without the need to make it in logon scripts).

    Actual case: I needed to configure Jinitiator, and make memory cache to go from 128 to 256. I spent almost an hour to write a script to overwrite the files, change registry keys, reboot the machine and change all the logon scripts (i have more than 80 different scripts, with non-matching patterns in the names to do a bulk editing). Also, some machines with the same logon script were not supposed to run this script.

    It would be awesome if I could filter from some OU with the Find command the machines I wanted to do this, not from the logon script field.

    I think that’s it for this time.

    But please… think of the IT pros not only in Server OSes..

    Remember, Servers are to Serve Clients. So, Clients need to have support for remote administration just as they were being operated locally.

  73. jlchereau says:

    There are good suggestions in this blog but Windows is now a very mature operating system and there are very few need-to-have, although many nice-to-have, new mainstream features that can be added to the operating system.

    Actually, in the v6 release, I can only think of virtualization as a break-through new technology integrated into the OS.

    The Vista "mission statement" about a better GUI and improved security has proved more confusing in its implementation than useful to most users. The GUI is certainly nicer although often inconsistent and the learning curve has proved too steep in most organisations. The UAC is resented more as a productivity drainer than a useful and efficient enhancement to the operating system especially since it asks repeatedly "Do you really want to allow this?" without learning from the user experience or considering the level of risk that an application presents. In other words, UAC is good in principle but it could be much cleverer in its implementation.

    Also Vista has been released with far too many bugs and although SP1 has improved on this, Vista is still too unstable in my opinion. Most businesses that I know are very happy with Windows XP SP3 and are still reluctant to upgrade to Vista just to run Office and web-based applications. Adding cluttering applications like sidebar gadgets into the operating system certainly defocuses from what a good operating system should be: robust, secure, fast, modular, and as far as client operating systems are concerned, user-friendly.

    My point is Vista is unfinished and before considering adding any new feature to the operating system (especially features like gadgets), please simply finish Vista and make it the robust, secure, fast, modular and user-friendly operating system that it should have been.

  74. RicLewis says:

    Steven, can you talk at all about the release intended release rhythm for Windows moving forward?  Windows was humming along at a release every couple years before it got off-track.  Mac OS X was moving at a yearly pace but it has slowed down to every couple years.  What’s the goal with Windows?

    Additionally, this post was a good opportunity to help frame Vista as a major release.  I’m guessing you chose not to do that to keep the Win7 message clear?  I’m anxious to see Vista finally presented in it’s proper frame (as a major architectural release:  whole new audio, graphic and networking stacks, security, .NET and WPF by default, componentization, etc.)

    Lastly, I may suggest that when we do finally see the release of Win7 features that you do a post which maps actual blog comments/community requests to implemented features.  It’s one thing to give the appearance of listening, but showing a tight coupling between real user feedback and feature implementation would build massive goodwill (at least among those whose comments you choose!)

  75. Carlesa25 says:

    Gentlemen: I am user of Windows from their beginnings and I can say the same of Unix; In some of the comments that I have read I believe that one of the main keys is so that new SO completes that that of the it is expected…

    To listen the end user that is not needed to change the computer to run new SO, in summary that it doesn’t seem that the final desire is to SHINE before the users teaching the complex and sophisticated that aser can arrive.

    The Client that pays at the end is the modest user that requests potent solutions but that he finds reliable, simple and quick of use although they are not so spectacular.  Greetings.

  76. AndiG says:

    I agree, people want a lot of different things. From a developers point of view I know what you’re talking about – ask three people and you will get at leat five opinions.

    So maybe it is the best to focus on what people really need. It makes me laugh, when I think about what our customers need. Well – they need windows XP in the end.

    At least they need absolute backwards compatibility, caused by their software infrastructure (talking of what companies need).

    That was, what makes me think of a win32 (xp?) virtualization layer im my first post. IMHO, I know no other (better?) way to bring old drivers and software to the new platform thus making the transition to the new system smooth.

    This would give you the possibility for a cleaner architecture and you could go on fixing all the little annoying problems without having backwards compatibility in your mine all the time.

    I’m interested in more details about windows 7.

  77. hardon says:

    The problem when talking about scale is, "OS" is too broad. It should be narrowed down to visible and non-visible changes, maybe split this blog accordingly. All i know is, the visible changes in Vista was MAJOR, and mostly for the worse. If the visible changes had been none (excluding bug fixes and improvements), with major non-visible changes, Vista had probably been more successful, at least for me:-)

    I would seriously consider splitting the shell from the rest of the OS. If I could run the XP shell on Vista, I might actually have discovered there is some interesting changes under the hood, but since I can’t stand the GUI, it really doesn’t matter: tried it for some days and instantly went back to XP, with my classic Win2000 look:-) If its not possible to add the XP shell as alternative shell, then you have some work to do: the shell and the "OS" is obviously too tightly integrated.

  78. stalepie says:

    Regarding the desire for UI design unity, I thought I was above such things until I realized tonight that I enjoy using Windows XP in classic mode, even using IE6 over 7. I like especially how Explorer looks the same as Internet Explorer and you can type a URL into the address blank, or a local address, just the same.

  79. AndiG says:

    Reading through the comments, I noticed the following (I call it the microsoft point of view)

    You were talking about users, developers, pros and so on. I realized the parallelism to Vista Home, Professional, Business, Ultimate, Ultimate Home and so on…

    And I read a comment and I thought, that could be microsoft – it was about introducing a complex rights system for keeping the registry from beeing damaged.

    But we have to remember kis "keep it simple". So maybe all of your problems minimize when you focus on one windows version and try to keep things simple.

    Imagine just two user groups, power user and normal user.

    The first user group can do everything with a shell – this is for the professional users, administrators and so on.

    The second user group gets a nice frontend that mirrors the professional superset and just needs a few mouseclicks.

    For the registry I suggest a layered approach, that means every programm has its own registry layer. Then the system assembles all the layers at startup. Simple, powerful the initial system layer never gets touched. To reinstall your system just delete all of the registry layers.

    The same is for the user interface. KIS again. Here it means reduce. Less colors, less effects, less transparency, Focus on usability, decorate it with nice effects and some round corners.

    Keep following interesting ideas. Display pdf or display ps is such an interesting idea. Introducing a new ms resolution independent format is not…

  80. consumer4beta@hotmail.com says:

    Just to get a rough idea: If I buy an Extreme model of the upcoming Nehalem processor from Intel and all-round balanced fast hardware such as ATI/NVIDIA’s current generation GPUs, fast HDD, high-end chipset/MB, can I be rest assured that Windows 7 will simply fly/run at the fastest possible speed on it? I think not. I think it will run but it won’t fly.

    I expect MS to create a performance optimized OS.

  81. burgesjl says:

    As another poster mentioned above, there are things that are visible to the end user and things that are not. You need to analyze why Vista was a market failure, because it has been. Several themes spring to mind:

    (i) I don’t believe anyone in their right minds at this point will try an in-place OS upgrade: too many things can go wrong and there are bound to be compatibility issues. There’s no guarantee the existing software will work with the new OS, and there’s no easy way to migrate it. So, take this off the list of things you are trying to achieve. Assume a user will have new hardware or will start from scratch. Its not a solveable problem, so don’t try, and get it wrong.

    The only possible exception to this rule is a headless server upgrade. And even then, there’s enough issues with servers to make that a non-starter as well unless they are simple, single function modules, which most aren’t.

    (ii) The Vista shell was a bloody disaster – badly designed and badly implemented. Throw in UAC on top, and it was truly awful. First, you are trying to dumb-down the user interface for novices all the time. All this does is make things more difficult and time-consuming for power users. It takes several more clicks to achieve the same results as it did in XP, and the things aren’t logically structured. There’s no excuse for such obvious design failures.

    (iii) There’s one thing guaranteed to tee off an end user: when he can’t use his existing software, or even more importantly hardware, with a new OS. Again, that speaks to really bad design when you can’t find a way to ‘wrap’ existing drivers to make them work with new underlying plumbing. I’ve got an HP multi-function printer/scanner/copier, I know full well won’t work under Vista. Thats the original $300 investment down the tubes, as well as another $300 for a replacement. All for some dubious new features in an OS? Its not going to happen.

    Finally, your assertions about who classes a release as major or not are pretty pointless. Who cares what ‘influencers’ think. And as for corporate IT, clearly you show no understanding of that market. I’m an IT consultant, and its pretty bloody obvious you aren’t listening to what is needed. Corporate IT is badly broken, and its because of you and the fact that you try from a marketing perspective to force a big-bang change, and that is massively expensive and can’t be supported. Its ESSENTIAL to have peaceful co-existence and a rolling change. There’s NEVER been a business benefit to an OS change; just in the same way there’s never a business benefit to changing carpets or desks. They simply get worn out and need to be replaced. An OS is simply a vehicle to run applications, what the end users actually do, transactions or documents. Microsoft have been the worst for ‘architectural innovation’ that then gets dropped when someone comes up with a slightly different idea. How many file, document, object and database elements have come and gone in the last 10 years??? COM, DCOM, and all the alphabet soup. Hundreds of different configuration tools. And a central concept of a Registry that is a single point of failure and completely unable to be corrected by an end user, no matter how you try and dress it up.

    Someone else said keep it simple. Thats true. There’s an awful lot of legacy stuff in Windows that needs to get retired. Instead, you’re adding more, and increasing complexity every time. It can’t continue. Windows is a dogs breakfast of code and features. It needs a ground-up rewrite, not more lipstick on the pig.

  82. burgesjl says:

    How do you simplify and retain backwards compatibility? Thats the 500 million dollar question that needs answering.

    The answer has to lie in virtualization somehow. Whatever is ‘real’ Windows 7, I’m assuming it will be largely new kernel code, and it will have a new desktop shell. It’ll retain some existing APIs for developers to work with, but some will have to change. However, this clearly won’t work for everything. So, you need a concept of the ‘OS Box’. You’d install a REAL copy of XP, 2000, whatever, over the top of Win 7. When you’ve got an app or hardware that only works with XP, you install it into the ‘XP Box’. Now, some of that ‘XP Box’ may have to get patched up somehow to point at underlying Win 7 code for efficiency purposes, but really, that needs to be as little as possible and only for published and supported APIs i.e. well behaved apps. The file systems need some level of compatibility so that a document on one ‘box’ can be visible or usable in a different ‘box’ or should fail gracefully. Any one ‘box’ should however be protected from any other ‘box’ so that weaknesses in the ‘XP Box’ can’t be exploited in the ‘Win 7 Box’.

    This has to be possible. If you’re not aiming for this, then you’re just adding more complexity and lipstick to the pig. And from your statements so far, I see lots and lots of lipstick, rouge, hairpsray and mascara. If you do this, you’ll have another Vista.

  83. whitenoiz says:

    I will try very hard not to bag Vista along the way, but please excuse me if I do slip a little. I am an IT professional running an all XP network, I continue to purchase computers with downgrade rights, and I will continue do so until there is something better than Vista available.

    I work for an architectural firm where performance of the OS and software that sits on top is of the utmost importance. We run very demanding and memory intensive applications and I can’t have the OS wasting system resources. Vista runs like a dog and a Vista rollout would be a massive performance downgrade.

    So, all I would like to see is an operating system that outstrips XP two fold on performance, not more (to quote another post) "lipstick on the pig".

    If windows 7 is just Vista R2, then I guess this will not happen. It’s a great shame, it all seems to have gone horribly wrong at Microsoft. I can see the balance of power starting to tip another way.

    Thanks.

  84. Cartman05 says:

    This might be a stupid question and off topic but is it too late to get invited to be a tester for Windows 7? I am interested in participating in testing early releases as I am sure many of us are and was wondering how one goes about that or if it is too late.

    Thanks

  85. spivonious says:

    "Lipstick on a pig" – that is a great phrase.  It described Windows ME and I’m sorry to admit that it describes Vista.  The focus of the OS has been lost and instead of really improving the base functions all we see are GUI changes and trend-driven additions.

    I think virtualization is the answer to maintaining backwards compatibility.  If someone needs to run a Windows 95 app, then open it in its own virtual machine.  I think people would accept a slower startup of an old app in exchange for much improved startup of Vista and newer apps.  In any case, it’s time to dump the legacy code.  If someone’s app from 1992 doesn’t work anymore, then tough.

    I also am interested in testing.  I’m a developer by trade, but at home do pretty standard stuff (browse the web, run some games, record TV, play music, etc.).  Where can we sign up?

  86. CezaryK says:

    Just wanted to propose: a service-oriented paradigm to apply in 7. That means a microkernel surrounded with systems services, encapsulating a complete pack of functionality, running in their private spaces, minimizing dependencies, and maximizing loose coupling.  That sort of model of OS was implemented in Tanenbaum’s MINIX. Although it’s just an academic project, there are many reasons to follow its way. The benefits of that approach could be high modularization, better fault tolerance (minimizing impact of a crash of a particular service to a whole system), user-space drivers (no more BSOD), and finally: less lines of complex monolithic code – that subsequent versions of Windows blew up to 50 millions! Gee, I wonder how you folks can understand how this OS really works.

    As for legacy apps, I believe that there is always a way to provide an environment for them to operate the old API. In other words whatever re-engineering you’ll do – having principles of design patterns in mind – you’ll be able to hide the new inside architecture behind an old API facade.

    Really, it is high time to make OSs simpler. These days complexity rises, and some time it will kill us. It will kill the progress, if do not stop that trend.

    Greetings,

    Cezary

  87. thecolonel says:

    here’s a suggestion – hire burgesjl and mystere from these forums. their comments absolutely hit the nail on the head

    ‘lipstick on a pig’ i’m gonna have to start using that expression, an absolute classic! (and true btw)

  88. boss_kevin_boss@hotmail.com says:

    1. Sry about my english

    2. I have read in a older Post about older OSs in VMs and i think this is a good idea. You have relased with Server 2008, Hyper-V(i dont know a lot about but i think it works same as good then VMwares software) and its a great peace of software. I think you can run older OSs in a Lite Version of Hyper-V under Windows 7

    3. I hope you will change the GUI, i dont mean the Taskbar or something like that but make Menues easy to use. In the Control Panel under Vista( its better then the CP under XP but) you need to go trough a lot of "Levels" to come to the menue points. Also there are some menue points double or more times on 1 Place, you have 4 or 5 ways to come to the System Preferences,it think it would be mutch easyer if you can come to am menue point with just 1 or 2 klicks in the CP. Also put the Start Menue over the hole screen (wy is it so small?) also you can work with ribbon in the startmeneu…maybe the first for Programms, second for the search, 3 for the Control Panel…

    So put all together to the Start Menue…so you just need to open the Start Menue and not a windows for the CP a window for window update, a window for something else…

    I like Windows Vista very mutch and i hope that 7 will blow me away 😉

  89. dosulliv91 says:

    First of all I feel compelled to provide some praise for Vista. Despite all the negativity surrounding it, I’ve found it to be stable, functional and as fast, if not faster than XP was on my laptop. So you’ve got a solid base to start from IMO : )

    That having been said, I do have two requests for Windows 7:

    1) Allow re-ordering of application buttons on the taskbar (Drag and drop preferably) – It’s purely an aesthetic thing, but it would help keep applications organized in the same vein as grouping.

    2) Provide hooks for applications to use windows explorer sorting preferences when opening explorer windows (i.e. During ‘File -> ‘Open’ operations)- It’s absolutely illogical to have to constantly resort explorer windows (Using the ‘View’ -> ‘Details’ menu, then clicking on ‘Date Modified’ for instance), that are spawned by applications when you have already saved your sorting preferences in Windows Explorer! There should at the very least be an option for users to enable that would allow applications to use the pre-defined sorting preferences set by the logged in user.

    Thanks for listening : )

  90. CezaryK says:

    As I went through the post and comments once again, I came up with a fundamental question: what does "re-engineering" in this case suppose to mean? Would 7 be a turn in architecture or in a specific set of features? Well, in case of adding/polishing a set of features, we can expect the same complex, unstable, developer-unfriendly OS.

    Please, make a clear statement what sort of change you plan. I moved to MS side because of a really great .NET 3.0, that beats Java in many aspects. But as I explored internals of Windows, its design in many areas, I ended up with simple conclusion – it’s a pile of nonsense. It definitely needs better architecture, one that’s as good as .NET’s in terms of quality.

  91. demian says:

    I would like to see all the networking tools displayed in its own area like the network and sharing center on vista but with more information/options like accessing the local firewall, system information, security logs, etc

  92. marcinw says:

    Hi,

    Before I will write comment to topic I would like to say something more generic:

    1. it’s very good, that this blog has been created and it’s good, that comments seems not to be censored

    2. I don’t like Vista

    And now my other comments:

    In my opinion system needs architecture changes.

    1. part responsible for running processes and supporting hardware (win32, .net, dos runtimes + drivers) should be totally separated from client applications (explorer, help, web browser, task manager, control panel, etc.). First should be put in windows directory (and not available for applications) and second in program files.

    2. applications must be totally separated from each other. it means, that application X can’t have access have access to files from application Y put into other directory in Program Files…unless user will set differently.

    3. there can’t be used one central registry for everything – core settings should be put in one place and settings for each application should be put into separate registries.

    4. filesystem should be more simple – no more hidden data assigned to files (or at least standard applications should show them)

    5. applications in Add/Remove programs should be separated at least into few categories – drivers, runtime subsystems, applications. Many default parts (web browser, wordpad, etc.) should be uninstallable.

    6. system services should be more logical, more intuitive and loaded only, when required. for example why Task Scheduler is required for working prefetecher, why wifi service is run even when wifi card is not found ? maybe some current parts (performance logs and alerts) should be available as services too and some removed…

    maybe it should be available one central place only, where user defines – this and this process will be run one, this will run even without logging, etc. it will join services and startup.

    7. system can’t make too many automatical actions – no more automatical defragmenting, indexing, etc. (or et least it should be possible to disable it)

    8. it would be good to have XP GUI with Control Panel, Network COnnections, etc. with known options. Themes service should be run automatically and unloaded, when not necessary. Users should have more info, why some theme or option is not available (with tooltips or descriptions).

    Applications shouldn’t have ability of moving into top – for example, when you have application X, Y, Z run, you run A, you switch to X, A can’t be displayed on top after running.

    9. installer shouldn’t install all drivers into hdd – for example shouldn’t install AMD CPU driver, when Intel is available…

    As summary I will say: Microsoft programmers have very difficult task now. I think, that they should start from small, easy core and prepare such architecture, which will not allow for damaging it and which will separate applications (for example there will be no ability of setting Registry entries/files from other applications without special actions). And now the most important point: checking integrity of many components 100 times per second can make even the fasters one computer very slow… Maybe DRM should be implemented in application layer only ?

  93. acquatile says:

    Sorry to be negative, but this recent post gives the impression that you don’t know what Windows 7 will be.

    If the scale of a release is completely subjective, then how does Microsoft know what sort of ad campaign to launch?

    Is it possible to determine whether XP and Vista were major releases, or does that depend on the client?

  94. marcinw says:

    Some addition to my previous comment. System should have such parts:

    1. core (kernel + drivers + runtime systems + own Registry part) in Windows

    2. applications (explorer, GUI, web browser, system help, etc.) in Program Files

    One of applications should check on startup/after logging one central place for applications, which should be run. Some applications can’t be deleted from this place (they will be equal to some current system services).

    And nothing more…

    Additionally, what I would like to see is ability of defining routing for each application on low level. For example application X will be able to use first connection A, later B and C will be unavailable for it.

  95. Hairs says:

    Thoughtful and reasonable commentary all round. You mention there’s a "focus on performance" coming from the comments. I think basically every release of Windows since 95 has gotten some sort of flack for "reduced performance"/"enforced upgrade", and that’s to be expected when new features are coming in.

    I do think we’re now getting to the point in hardware and software however when the return on investment in software features is getting lower and lower. By that I mean that while Win 3.1 > Win 95 involved a massive jump in hardware requirements, there was a corresponding jump in features and usability. The jump from 98 > 2000 wasn’t as bad in terms of increased hardware requirements, but provided a great benefit (better code base). 2000 > XP provided a much lower ROI in terms of features. XP > Vista was an absolutely horrendous jump in hardware requirements, and the ROI in "must have" features was the lowest ever.

    So: my suggestions? (which have been made already I’m sure)

    64 bit only. Vista’s hardware requirements mean that any PC capable of running it was using a 64bit chip. Bite the bullet.

    Only two versions: Home for the Vista-equivalent "shiny Mac users" with all the visual bells and whistles, and hand-holding "this is a keyboard" walkthroughs enabled by default.

    Pro for the people who can run an installer and let them switch off what they don’t want!

    End-user focus! Vista felt like something that was created for the benefit of hardware vendors rather than users. While it was pretty and modern, it didn’t feel like that was in order to make our lives more productive, easier, or more pleasant.

    Intel adopted a speed rule with the Core2 and Nehalem. 1% increase in TDP for 2% performance increase. Obviously that’s a meaningless metaphor in software creation. But Windows needs to start getting more ruthless with what it adds in. Anything that goes in has to justify itself on its performance ROI.

    The performance gulf between hardware and software is getting more and more vast. Even the most tech-unsavvy user is starting to notice that all the hardware under the hood is getting wasted by larger and larger software that’s doing less and less for the user as time goes on. Nothing highlighted this for me more than testing uTorrent vs. Azureus. Both do the same job. Both have the same features. One is a 250k download that requires no install and manages its own memory. The other is a 20meg (+130megs for Java) that installs junk all over the place, hogs and leaks memory, and slows everything around it to a crawl.

    Return On Investment. That should be the mantra. From that, usability and performance will follow.

  96. Hairs says:

    Why is it that there is always a lag opening the Control Panel? It’s a list that never changes, and yet, the first time you open it in a session… you get a nice big inexplicable lag while it loads the icons.

    That’s bugged me for years now.

  97. marcinw says:

    Hi,

    I think, that this is very good topic.

    Let’s run Windows NT 4 on modern computer (of course, I assume, that it will be possible to run it) and Vista on the same computer. And compare them – in Vista there are many modules for increasing performance and NT doesn’t need them…

    I understand, that creating something bigger in plain assembler is today maybe not possible. But maybe techniques and technologies used for creating Vista were not good…and you should start from changing them…

    And once again: people want small core, which will separate applications and which will not be slower after year or two because of a lot of missed files and entries (after uninstalled applications). Remove all stuff, which is not liked by people and which doesn’t work like expected (like DRM) and which will make core (much) slower.

    Let’s start from it and various prefetchers and other things simply will be not required. Without it you shouldn’t go into next part of Windows 7 project.

    Additionally – do you remember how slow was Windows 98 when compare it to WIndows 95. People were speaking about integrating IE with system in many bad words. When this and other integrations will be removed in current system, performance will be definitely better :)

  98. magicalclick says:

    I think Vista already spent great deal of time on stuff under the hood. Like PC backup without rebooting to some other environment. BitLocker. Parental Control. Other stuff I can careless.

    I hope it will be a release with cool casual user experience.

    For example, WPF applications. You spend a great deal of time on WPF and yet no build-in app demo it? That’s just weird.

    Really hope you can upgrade Paint to "Paint .Net" as originally intended. Why not buy it and change it to Live Paint?

    Animated Windows. This is pointless, but I am sure people would love Jelly Windows and other pointless stuff.

    Again, what is your objective for Windows App and Live App? Live Mail is way way way better you know.

  99. hitman721 says:

    What Windows 7 really needs is many different options. I think the classic Windows UI interface (Windows 2000 and XP both should be in there.) should be one of many UI options. Seven needs to be speedy and responsive. Quick start up and shutdown. I really think Seven should be 3 times faster than XP. Graphically, it should make Leopard and SnowLeopard look like a 24 year old OS platform thats out of date. Featurewise, it needs to achieve true parity with Leopard plus some new things thats exclusive.

    For the business users, I agree they need a special business oriented version of Seven that takes care of security, business interoperbility, ease of use for transitioning users, and the ability to uninstall unwanted features.

    On the consumer side, I need the ability to remove Microsoft created applications like Windows Media Player, Paint, Calender, and Internet Explorer. Unless the new WMP and I.E. are significantly advanced to counter both Firefox, WinAmp, Opera, and iTunes; we should have the right to completely take them out of Windows. This should be easily done within the Control Panel, with a few clicks. Plain and simple.

    Also, we need an answers to iLife. I think Office is way superior to iWork. It would be nice to see a separate suite that counters iLife bringing a digital suite for all areas of music, movies, photos, podcast, etc. That should be related to the Windows Seven launch.

    My idea would be that users would add to Seven what they need. At the install screen, you decide what you need. One of my favorite ideas, would be to allow you to install third party apps during install of Windows using an Internet connection. Imagine having all third party apps available as you’re installing the OS? Being able to substitute Firefox instead of Internet Explorer? Being able to install WinAmp and leave WMP out? It would be revolutionary to allow competitors the ability to put their apps in Windows during install. It would do a lot to counter anti-trust issues and give competitors a chance to have a fair shake on the Windows Environment. It would make Windows the truely open to third party applications suite and the counter to the Apple platform.

    If anyone needs those apps later, they can whip out their Windows disk and put it back. Plain and simple. Or you guys could include it as a Windows Update, just incase the disk is damaged or missing.

    Finally, DRM and WGA needs to be pulled out of Windows once and for all. Find a new way to verify legitimate copies of Windows. Also, open things up and keep intact individual users rights a top priority. I understand the need to protect Intelectual Properties, however the broadcast flags in Windows Media Center has got to go. We should be able to record television without NBC or any other network blocking our rights. Guys, Good luck.

  100. itcore_andrew@hotmail.com says:

    Wii looks kewl with its wii remote

    Ipod touch kewl with its touch

    want your windows 7 looks kewl, try this one…

    touch have problem, its not good for working, not efficient….

    so a better one should be "Glove"

    today technology should be possible to build it

    say i dont want to use keyboard and mice anymore, glove can be used to simulate at least my 10 finger on windows…

    so the first one is, how you control 10 cursor on windows, when i want to type, i use virtual keyboard on windows, maybe windows in 3d space, instead 2d…..

    when i play game…….

    while the glove is default, it can be combine with webcam and speech recog

    cheers !

  101. steveninchrist says:

    Hi,

    I have two suggestions on Windows 7:

    1. It’s a good idea to have more desktop space. Sometimes, we may have a bad habit that put all the icons, documents, etc on the desktop. Before cleaning up my desktop, I would really want more space.

    2. As a IT guy, I would like to increase the experience in using perfmon. Hopefully the time filtering function can be better, the graph can be more detail. And even better if some major event (like PageFile size changed, software installed) can be displayed on the graph.

    Thank you.

    Regards,

    Steven Yeung

  102. lyesmith says:

    An idea for Date & Time.

    Use UTC. It should be possible to set the OS clock to UTC. It should be possible that different users log into the OS and set there clock to different time zone. (especially on servers but on PC as well) Personally I need to know and work with UTC all the time, and it is a big problem with all the Windows OS at the moment. I like the UNIX aproach in this.

  103. lyesmith says:

    An idea for Date & Time.

    Use UTC. It should be possible to set the OS clock to UTC. It should be possible that different users log into the OS and set there clock to different time zone. (especially on servers but on PC as well) Personally I need to know and work with UTC all the time, and it is a big problem with all the Windows OS at the moment. I like the UNIX approach in this.

  104. graham.lv says:

    It may be unpopular, but from a business sense if you want to move another 500 million units of Vista before Windows Se7en release on 3rd June, 2009, then make it an update ONLY from Vista.  You could discount and offer incentives (to business) to accomplish this. (Piss me off but – AUS$180)

    What will sell Windows Se7en is ->speed ->connections (everything plugs in & works) ->it looks different -> reliability ->ease of choice (one CLEAR version for mum & dad) -> No putting ‘click everything to proceed’ mum & dad don’t know what to do – just because you can’t secure it.

    What will sink it (as of this date) is -> Internet Explorer 8

    It’s a stinker – it’s just a damn shame that you hate Google.

    If Windows Se7en had burst out with something fresh, like Google Chrome -> as Internet Explorer 9, then WOW!  I would say you could claim the WOW! factor.

    As long as I don’t have to stand in no damn sheep paddock looking up waving to a Microsoft satellite!

  105. cirurgia plastica says:

    From my perspective, if Windows 7 has a faster boot time, uses less resourses, and is available in a business version without all the fancy graphics but with all the features and functionality and support for 64 and 32 bit hardware drivers (even older hardware that worked with XP) It would be a major release.

    This is from a IT professional and a home users perspective.

    You guys at MS have an oportunity here since you prove that you are listening to the customer instead of telling the customer what they will use which has been the feeling with Vista in my opinion. It was a major bone of contention with many to not be able to buy a new computer without Vista. I think that way of doing business in what appears to be monopolistic needs to be a consideration when 7 is ready for release. I made the switch to Linux and now use XP on a VM more as an occasional tool instead of an OS because of that.

  106. mikehudson says:

    It was a major bone of contention with many to not be able to buy a new computer without Vista. I think that way of doing business in what appears to be monopolistic needs to be a consideration when 7 is ready for release.I would like to increase the experience in using perform

  107. sasa says:

    thanks, nice post, keep posting

  108. Machine vision says:

    Great writing it is such a cool and nice idea thanks for sharing your post. I like your post very much. Thanks for your post.

  109. colour sorter says:

    I read your blog. That was fantastic. I like your blog. Thanks a lot.

  110. GigE says:

    I like your blog. I enjoyed reading your blog. It was amazing. Thanks a lot.