back to article Microsoft buries Sinofsky Era... then jumps on the coffin lid

The path to redemption in 2014 lies not through baptism, but blogging. Former Merrill Lynch dotcom analyst Henry "what a POS*" Blodget agreed to a lifetime ban from the finance industry – but bounced back with a blog. And perhaps one-man wrecking crew Steve Sinofsky, former President of the Windows Division at Microsoft, can …

COMMENTS

This topic is closed for new posts.

Page:

  1. Alan Bourke

    A common API is definitely a must.

    Also clarity as regards to the path forward for developers.

    1. Bob Vistakin
      Holmes

      Re: A common API is definitely a must.

      For microsoft, it's also a first.

      1. Anonymous Coward
        Anonymous Coward

        Re: A common API is definitely a must.

        Windows has always had a common API. It is called Win32. As of today, you can run a Win32 app written 10 years ago in any machine without any issues -except perhaps some DLL hell- Nobody wants to write raw in Win32 nowadays, however.

        1. Yet Another Anonymous coward Silver badge

          Re: A common API is definitely a must.

          But it's hard for Microsoft to do this.

          It's not like they 10,000s programmers, complete control over the critical app and server infrastructure and a common language runtime that insulates the apps from the details of the machine

        2. Richard Plinston

          Re: A common API is definitely a must.

          > Windows has always had a common API. It is called Win32.

          You are obviously new to this thing called computers. Win32 did not exist on several versions of Windows. It started as Win32s, a limited version, as an add on to Win3.1. It developed in Win95, 98 and ME. A _different_, but almost compatible Win32 was written for NT but it wasn't until XP that the incompatibilities were mostly resolved.

          Windows RT does not have the full Win32 API, it does have a limited subset and probably many incompatibilities.

          > you can run a Win32 app written 10 years ago in any machine without any issues.

          Simply not true for _any_ program. That is why Win7 has 'XP compatibility mode'. Basically an emulation of the WinXP Win32 API because there are differences.

          1. Anonymous Coward
            Anonymous Coward

            Re: A common API is definitely a must.

            <<You are obviously new to this thing called computers>>

            Perhaps what is at play here is that I've been longer than you? I can dig my ZxSpectrum and MS-DOS 1.0 installation disks if necessary.

            <<Win32 did not exist on several versions of Windows>>

            Mmmm... I suppose it was pretty hard to write Win32 apps when Win32 did not exist yet. So yes, the word "always" is perhaps not accurate because it should say "since Win32 was created" . I'll take care of saying "I read The Register since the Register was created" instead of saying "I always read The Register" in future communications to you.

            However your nitpicking on the precise wording is surprising considering the generalizations on my comment that you (incorrectly) assume in the rest of your reply.

            << It started as Win32s, a limited version.... >>

            Oh yes, and evolved in a lot of iterations over the years. My point is, whatever you wrote for Win32 following MS guidelines, it will still work with W8 today.

            <<Windows RT does not have the full Win32 API, it does have a limited subset and probably many incompatibilities>>

            Yeah, that's why it is called Windows RT and not Windows version number X . Same with Windows CE, or other variations of Windows. My comment was about Windows, not RT, CE, embedded, etc.

            <<Simply not true for _any_ program. That is why Win7 has 'XP compatibility mode'. Basically an emulation of the WinXP Win32 API because there are differences.>>

            As said above, the point is, will it work or not? You say that is not true, but then explain how W7 has a layer to allow make it happen? So are you saying Win32 apps work or not? For which programs this is not true?

            There are whole categories of software written for a specific version of Windows that don't work on the next one. Off the top of my head: device drivers (if/when MS changes the driver stack) and programs that take advantage of undocumented or undefined API behavior.

            If you have an app making use of undocumented features that doesn't run in the next version of Windows you should address your rant to the app vendor, not to Microsoft Windows. Even for those apps, Microsoft usually (some exceptions applied in the past when they want to took out a competing app off the market) makes very notable -sometimes epic- efforts to be backwards compatible even if it means reproducing undocumented/undefined/plain wrong behavior.

            You see a Win32 app not working from one Window version to the next and blame it on changes in the Win32 API. And that is just __not_true__ . I'd suggest you educate yourself a bit in software development and Win32 history before your next rebuttal. Popping in installation disks and clicking on icons does not tell the whole story behind why something works or doesn't.

            I stand to be corrected: if you wrote an application using the Win32 API, it will still work today. It may not work if the app uses undocumented features, but that is hardly a problem created by "Microsoft not having a stable API" The Win32 API has remained stable since its inception and programs written for it still work.

            1. Richard Plinston

              Re: A common API is definitely a must.

              > I can dig my ZxSpectrum and MS-DOS 1.0 installation disks if necessary.

              I can dig out my 1975 Polymorhic and 8inch CP/M disks.

              > Yeah, that's why it is called Windows RT and not Windows version number X . Same with Windows CE, or other variations of Windows. My comment was about Windows, not RT, CE, embedded, etc.

              Windows NT, Windows XP. You seem to be able to claim 'all the ones that you mean', and if it doesn't then 'I didn't mean that'.

              Windows 1.0 _is_ 'Windows'.

              > explain how W7 has a layer to allow make it happen?

              Explain away why it is _needed_.

              > You see a Win32 app not working from one Window version to the next and blame it on changes in the Win32 API. And that is just __not_true__ .

              Many Win32s/95/98 programs did not work correctly on NT because the API implementation was different. Even though the _documentation_ was similar the behaviour was different enough to cause problems. Such as:

              "Windows 95 RegOpenKeyEx Incompatible with Windows NT"

              1. Anonymous Coward
                Anonymous Coward

                Re: A common API is definitely a must.

                Please, Mr. Plinston go back and read the comment before downvoting/commeting again.

                "Windows 1.0 _is_ 'Windows'."

                Again, to reiterate the point, the comment referred to Win32 and it is next to impossible for something to be compatible with Win32 if Win32 did not exist at the time that something was created.

                "Windows 95 RegOpenKeyEx Incompatible with Windows NT"

                Please see http://support.microsoft.com/kb/137202/en-us "This code is incorrect, but it works under Windows 95 because OldKey and NewKey refer to the same key. However, the code fails under Windows NT, because NewKey is not valid after it is closed. "

                So you're using an example of an invalid use of a Win32 function to support your point? Please read again my post, you cannot blame Microsoft for your reliance on an undefined behavior. Thanks for reiterating my point by the way.

                1. Richard Plinston

                  Re: A common API is definitely a must.

                  > Please, Mr. Plinston go back and read the comment before downvoting/commeting again.

                  Well I have, and I found that it actually did have:

                  """Windows has always had a common API."""

                  What you are belatedly attempting to do is redefine the word 'always' to exclude anything beyond a dozen years ago or so, and re-invent 'Windows' to only mean the ones that you were thinking of at the time. 'RT' is out, 'XP' is in, 'CE' never happened, and numbers less than 7 or greater than 8 should be ignored. 'Vista' of never should have happened, but that is a different story.

                  > it works under Windows 95 because OldKey and NewKey refer to the same key. However, the code fails under Windows NT, because NewKey is not valid after it is closed. "

                  You failed to continue to the 2nd example. The problem is not that the code was wrong. The problem is that the implementations are different, yet the documentation is the same. What is perfectly correct code under 95 fails on NT.

                  """

                  RegOpenKey(OldKey, NULL, 0, MAXIMUM_ALLOWED, &amp;NewKey)

                  RegCloseKey(OldKey)

                  RegQueryValue(NewKey,...)

                  RegCloseKey(NewKey)

                  This code works under Windows 95 because NewKey is still valid after closing OldKey, but the code fails under Windows NT. Code that is correct for Windows NT (don't close the handle until both OldKey and NewKey are no longer needed) leaks registry handles under Windows 95.

                  """

        3. JEDIDIAH
          Devil

          Re: A common API is definitely a must.

          > Windows has always had a common API. It is called Win32. As of today, you can run a Win32 app written 10 years ago in any machine

          How a bout a Rasberry PI? Where's the version of msword to go with it?

          Talk is cheap. I can recompile stuff for the PI and go on my way. Microsoft seems unwilling or unable and their lack of comittment sets the stage for the rest of the industry.

    2. yossarianuk

      Re: A common API is definitely a must.

      Like Ubuntu is doing.

  2. Anonymous Coward
    Anonymous Coward

    What we want to know is...

    1) Will the modern UI/TIFKAM/whatever still be there on our non-touch desktops

    2) Will the majority of us still have to install a third party product to make it usable with some form of menu system that we can recognize and understand.

    IMHO, everything else pales into insignificance.

    Now I must get back to important things like figuring out why MSMQ queues are not visible from the Cluster IP address. Yes I know it is a toy compare to other queuing system but that is what the customer wants.

    1. Ragarath

      Re: What we want to know is...

      1) I assume / Hope it will be an option (I actually like it)

      2) Yes if you cannot learn new things.

      1. nexsphil

        Re: Ragarath

        Nice try Sinofsky. Actually, shit try. You sound exactly like a grubby, butthurt astroturfer.

        1. Ragarath

          Re: Ragarath @nexsphil

          And you just sound rude. Do you not see the irony in your post?

      2. Nigel 11

        Re: What we want to know is...

        @Ragarath - can I interest you in a car wirg a square steering wheel, the brake pedal where the accellerator used to be, and the throttle on the dashboard? I assure you, with a little practice it really is possible to drive the thing.

        1. Anonymous Coward
          Anonymous Coward

          Re: What we want to know is...

          car wirg a square steering wheel, the brake pedal where the accellerator used to be, and the throttle on the dashboard?

          Too late, Apple has already patented that.

          1. Someone Else Silver badge
            Coat

            @AC 16Jan14, 13:29GMT Re: What we want to know is...

            car wirg a square steering wheel, the brake pedal where the accellerator used to be, and the throttle on the dashboard?

            Sounds like a 70's-vintage Peugeot diesel to me....

        2. NumptyScrub

          Re: What we want to know is...

          quote: "@Ragarath - can I interest you in a car wirg a square steering wheel, the brake pedal where the accellerator used to be, and the throttle on the dashboard? I assure you, with a little practice it really is possible to drive the thing."

          I have upgraded your strawman:

          Can I interest you in a car new IDE with a square steering wheel context-sensitive parser, the brake pedal where the accellerator used to be type-casting checks, and the throttle on the dashboard a keyword auto-complete feature? I assure you, with a little practice it really is possible to drive code with the thing.

          Also note: your strawman doesn't take into account that some cars have the controls on the other side of the vehicle (e.g. UK or AU RHD vs US or European LHD models), some manufacturers put the stalks on opposite sides (Uk market RHD vs Japanese market RHD, where indicator and wiper stalks are reversed), and some cars have specifically modified controls (e.g. an accelerator / brake lever on the dashboard and no pedals) to allow differently abled people to drive them. That is also completely ignoring the whole manual vs automatic gearbox thing, which is a whole other debate.

          1. Getriebe

            Re: What we want to know is...

            "quote: "@Ragarath - can I interest you in a car wirg a square steering wheel, the brake pedal where the accellerator used to be, <snip>

            people to drive them. That is also completely ignoring the whole manual vs automatic gearbox thing, which is a whole other debate."

            And of course if you drive dump trucks, bulldozers, back hoes or even Russian trucks, or much much yummier pre WW2 race cars, all of the oddities that the person of very little world experience mentioned.

            I know of a 75year old that can handle a central throttle pedal and get around Oulton better than me - so I guess it's down to free thinking and not wanting it all common.

        3. jgarbo
          Black Helicopters

          Re: What we want to know is...

          You've described almost exactly the Cessna 172 I fly at weekends. Squarish control stick, throttle in the dash, brakes tip-toe on rudder pedals. Then I drive home in a car. No big learning curve.

          1. Anonymous Coward
            Anonymous Coward

            Re: Cessna 172

            Funny, when I read that comment I was sure it was going to have come from Jake...

      3. Anonymous Coward
        Anonymous Coward

        Re: What we want to know is...

        "2) Yes if you cannot learn new things."

        I learned pointing and swiping when I was one, however, now I have 28 open and active applications spread across seveal monitors and need to work, not watch cat videos.

        1. JDX Gold badge

          Re: What we want to know is...

          So switch to desktop mode. You don't need to see all 28 applications when you're launching a new application. Even in WinXP the start-menu was by definition a modal experience.

        2. Anonymous Coward
          Anonymous Coward

          Re: What we want to know is...

          What are those 28 applications that you need to use the gui on simultaneously, out of curiosity?

          1. Anonymous Coward
            Anonymous Coward

            Re: What we want to know is...

            Currently:

            4 terminals

            10 ssh sessions

            3 kvm sessions

            2 vnc sessions

            2 text editors

            opera

            firefox

            evolution

            calc

            2 java apps

            doc viewer with pdf

            2 caja

            and a spreadsheet i-in open off--ice.

            1. John 172

              Re: What we want to know is...

              No wonder you're so uptight and self appointed, that lot would stress anyone out!

          2. Getriebe

            Re: What we want to know is...

            "What are those 28 applications that you need to use the gui on simultaneously, out of curiosity?"

            Me too!

            That is some mutherfukka multitasking going on there dood (or doodette)

            1. This post has been deleted by its author

        3. Ragarath

          Re: What we want to know is...

          WoW you do more work than me (assuming you don't have some cat videos hidden in there), I have no problem switching windows. How come you do?

      4. Wanda Lust

        Re: What we want to know is...

        Ah, Ragarath, it's a brave man who speaks favourably of Modern UI and its start screen and denigrates the legacy 'Start' menu. You have no care for downvotes.

        @Nigel11: I, too, have used the analogy of the car and its controls many times, however, I don't consider the technology industry is at a place where the steering wheel and the pedals are yet defined.

        As for the idea that its not possible to be creative or productive without a keyboard and a mouse: we were (mostly) all born with fingers to point and touch, pencils (yeah, yeah, and chalk and crayons) were the first tools for recorded communications. The keyboard will probably have its place for a long time but its my experience that pen like stylii are developing fast and should replace the mouse.

        I am surprised that Microsoft is screwing up here, maybe their transition is misplaced and they should've followed a path with less radical desktop OS changes and brought together a merged 'mobile' strategy with RT and WP faster. But, hey, somebody else did it that way.....

        1. John Sanders
          Linux

          Re: What we want to know is...

          """As for the idea that its not possible to be creative or productive without a keyboard and a mouse"""

          You lack the clarity of having produced real content with a computer other than the odd email or comment in a forum.

          I have worked as an IT trooper in the printing industry, engineering, travel, advertisement, law industry, etc.

          Trust me the keyboard and the mouse will still be here for a long long time, the day the day any of those industries do not use a mouse and a keyboard to produce stuff is because either they have been made obsolete, robots have made us obsolete or the more impossible one: we have figured out how to connect our brains to the computer to control it directly.

      5. Dave K

        Re: What we want to know is...

        Many people I know used it and learnt it - myself included. I can navigate around Windows 8 from TIFKAM, I can get into the control panel, shut down Windows etc. The problem is that I find TIFKAM to be less usable and less optimal than the Start menu. The Start menu was smaller, quicker and easier to use IMO.

        TIFKAM is like Marmite. Some do admittedly love it, but too many people out there do not like it. Too many for a company like MS with around 90% of the desktop/laptop market to just force it unconditionally onto everyone without creating a backlash.

        1. Nick Ryan Silver badge

          Re: What we want to know is...

          While a lot of the comments around the Win8 UI revolve around the Start Menu, it's more the braindead manner in which it was implemented that was the problem.

          As many people have already noted, the Start Menu in previous versions of Windows was a Modal interface - which for those that don't understand what "modal" means, it basically pops to the front and blocks access to anything else. Modal interfaces are generally modal with respect to either the application (or in tabbed browsers / applications, sometimes per tab) or the operating system (graphical shell). A modal popup within an application will force your attention to that popup window within the application when you try and use the application. A modal popup at the graphical shell level will prevent you from doing anything else in the graphical shell until you've dealt with it...

          Having a full screen modal Start Menu (effectively a Start Screen) or a partial screen modal popup window as the previous versions of Windows had shouldn't make that much difference. The previous incarnation of the Start Menu had serious deficiencies... install more than a few applications and before you know it you're having to navigate scroll lists, nested menus and all kinds of usability horrors. To help with this when Microsoft transitioned away from the "Classic" start menu to whatever the hell they called it in Windows XP, it was possible to pin favourite or commonly used applications, Most Recently Used application documents linked with these and the other most recently used applications were automatically listed while still allowing the user access to the full, nasty, tree of applications if they needed it. One downside of the TIFKAM start menu is that it removes the control that the user had and introduces an (subjectively) ugly and unusable lists of icons in place of the useful things that were in place previously. My main system runs Windows 7 and some of the most application launches I work with are, for example, opening the Start Menu, and selecting a spreadsheet out of the most recently used documents listed by the Excel link that I pinned to the Start Menu. This is far quicker than opening Excel and finding the document by either opening it through the file system or performing the ghastly operation of finding the most recently used document list in Excel and eventually locating the document I wanted. I also have the option, if I had a spreadsheet that I used all the time, of pinning common spreadsheets so they don't fall out of the Most Recently Used list (it's also possible to pin a document link directly in the Start Menu, but it's not simple: http://blogs.technet.com/b/deploymentguys/archive/2009/04/08/pin-items-to-the-start-menu-or-windows-7-taskbar-via-script.aspx has the details).

          The (short) point of it is that the new Start Menu removes the functions that were steadily added that made the older Start Menu actually useful. As a result the Windows 8.1 Start Menu is a huge step backwards in usability and, while it can be customised, it can't be customised enough to replace the useful functionality lost and to make the interface itself actually usable on a non-touchscreen device.

          ...and that's just the content of the Start Menu / Screen. There are many other very serious user interface (user experience) deficiencies in Windows 8.1

          1. Pookietoo
            Windows

            Re: full, nasty, tree of applications

            Hey, I like those deeply-nested menus. I agree with the rest though. :-)

          2. weathertop

            Re: What we want to know is...

            If only there was a way to customize the organization of your Win 8 native start menu to allow you to quickly launch your most frequent programs. Oh wait, there is...

            And if only there was a way to access programs you don't frequently access by simply typing the name of the program. Oh wait, there is...

            Click for click, navigating the new UI is not difficult. It is actually a lot more streamlined.

            But we are talking about learning and adapting, something that a vast majority of the population seems to have great difficulty with. "Must... learn... new... things! BAAAH! Just give my some damn Crayolas! I've been using them since I was two! They're good enough for writing this check or completing this form!"

            "Sir, blue or black ink please."

            "But, but, PEN! NEW! BRAIN! DOES NOT WORK! Crayon work fine! See, it write!"

            "Sir, here is a pen."

            "See, it doesn't write. Crayon does."

            "You have to click the button to expose the ball point."

            "Butt.. button... button? Crayon no had button."

            I tend to think this was part of the shift from command lines to GUI.

            "You mean I have to click here, here, and here to launch that program, I can't just type "run myfavoriteprogram.exe? BALDERDASH! I'm taking my baud modem and going HOME!"

            "You can put a shortcu..."

            "I SAID HOME! CRAYON! WAAAH!"

            1. asdf

              Re: What we want to know is...

              >But we are talking about learning and adapting, something that a vast majority of the population seems to have great difficulty with

              People don't want to learn and adapt to a solution that is not considered better by the majority of the market. Also Enterprise has to spend money to get employees to learn and adapt and again unless it increases productivity they aren't going to. As the article correctly points out Win8 has been a lead balloon in the enterprise. Face it all the butthurt in the world isn't going to save Win8 or Metro. The market has spoken.

            2. Pat 4

              Re: What we want to know is...

              The very SECOND you start making fun of very legitimate user complaints about a new UI by just responding " Well ALL you have to do is LEARN something new!" Or "Well I guess your not smart enough to learn it because I've done it, and it's SOOO easy..."

              You have :

              1: Lost the argument COMPLETELY because people do not give a crap how easy YOU think it is.

              2: Become a complete douchebag because seriously... you're not that special, or smart, or better than everyone else.

              Nuff'said.

          3. This post has been deleted by its author

            1. Piro Silver badge

              Re: What we want to know is...

              Ohh! You mean 9x style. Not 9.x.

              I honestly didn't know what you meant first read through.

        2. Decade
          WTF?

          Re: What we want to know is...

          Quote:

          TIFKAM is like Marmite. Some do admittedly love it, but too many people out there do not like it. Too many for a company like MS with around 90% of the desktop/laptop market to just force it unconditionally onto everyone without creating a backlash.

          And many people do not like the Start menu. It seems like a major waste of space, to have the entire screen in front of you, and you have to scurry into the corner to reach important controls. People are just used to the Start menu, because the Start menu was the main interface since 1995.

          Note, I rarely use the Start button. I vastly prefer to use the keyboard to open the Start menu/screen.

    2. Spearchucker Jones

      @AC - Re: What we want to know is...

      All queuing systems are toys - expensive, unnecessary toys: http://www.infoq.com/articles/no-reliable-messaging

      1. Anonymous Coward
        Anonymous Coward

        Re: @AC - What we want to know is...

        While I don't agree with the first line of your post - "All queuing systems are toys - expensive, unnecessary toys: " - the linked article is excellent, and my own experience bears out everything the author says - in context.

      2. Steve Davies 3 Silver badge

        Re: @AC - What we want to know is...

        Oh deep joy. That link points to the clusterf*k that is a pure webservices solution. That just don't hack it in the real world.

        A decent queuing system offers stuff that webservices fans can only dream about.

        I totally agree with the original post that things like MSMQ is a toy when it comes to proper reliable queuing systems. Even a pure JMS solution is better than MSMQ.

        1. Roo

          @ Steve Davies 3 ~ 2014-01-16 13:22 GMT

          "Oh deep joy. That link points to the clusterf*k that is a pure webservices solution. That just don't hack it in the real world."

          The webservices aspect is tangential to the the main point the guy is making, ie: that you don't need reliable messaging and it doesn't actually help you because you really need to handle duplicate messages and non-delivery at the business layer to make sure the system behaves reliably.

          I was using the design principles outlined in that article before webservices were invented, so there is nothing stopping you from applying the same design principles to raw sockets, pipes, WCF, RabbitMQ or whatever else you want.

          FWIW I haven't seen a system built on a discrete "reliable messaging" layer that is actually reliable and fails safely. They usually choke on restart, run out of storage or deadlock if comms are stalled for a significant length of time. :(

          1. Roland6 Silver badge

            Re: @ Steve Davies 3 ~ 2014-01-16 13:22 GMT @Roo ~ 2014-01-16 19:21 GMT

            Whilst I large agree with the points being made, it is obvious the guy who wrote the article was confusing 'reliable messaging' with no loss of messages and 'guaranteed delivery and notification'.

            Everyone who passes a 'message' assumes the service to be 'reliable' - namely it will take the message and will try and deliver it to the correct recipient applications/systems and inform you if it succeeds, fails or doesn't know. The problems arise when things go wrong and how both the messaging service and the source and destination applications handle both the initial failure and then the subsequent escalated failure (eg. messaging system running out of storage).

            By the way back in the 90's I built a system with a 'reliable messaging' layer for handling international financial transactions and reporting. However the 'guaranteed delivery' component was achieved through an escalation process that culminated in an engineer visiting the site, downloading the message store on to portable media and couriering it back to the UK hub for processing...

    3. Bod

      Re: What we want to know is...

      While the Metro 'Start' may not be great, I've never quite got the love for the old Start menu. It's always been annoying deeply nested method of finding the apps you want to launch and much of it is grouped by vendor named folders which means you have to try to remember who made the app you're trying to find. That and if your mouse drifts off by a few pixels the sub menu disappears.

      Then you end up pinning things you want to find to the root of the menu or in modern versions to the start bar, along with sticking thousands of shortcuts all over the desktop in a right mess.

      I'm unsure about Metro Start but playing with a bit more I see it simply as a launcher much like iThings and Android. Also having found all the menu items are still there and can be dragged about into the main menu as you like, it's quite flexible, and it keeps the desktop from being cluttered.

      Okay you have to click/tap into it which brings up the menu full screen to launch something. But then most of us probably have to close a dozen windows to expose the desktop to launch various apps anyway.

      I'm a bit Meh about the whole thing. Only thing is yes Microsoft should have offered the classic menu for the luddites, who I admit are most of us. If it asked on first run if you wanted the classic experience and gave you the desktop with old Start, then few would be complaining. What you've got not is not far off, just lacking the old Start menu... which you can get 3rd party apps for.

  3. Anonymous Coward
    Anonymous Coward

    Using Sinofsky As A Scapegoat

    We all know who the real fools were in all this calamity for allowing an inmate to take over the asylum, the MS board. In a state of panic about the opposition they allowed a very questionable strategy and execution plan to be put into action that has resulted in a very costly series of mistakes.

    Vista should have a new OS entirely written from the ground up that didn't worry about backward compatibility whilst addressing some of the fundamental issues of the underlying architecture and was designed to use the latest hardware features of the time as a minimum. Sadly a missed opportunity for MS to have completely re-written Windows which I don't think will come around again.

    Maybe W8 should have been Vista but it's too little too late.

    1. Jonathan 29

      Re: Using Sinofsky As A Scapegoat

      I always thought that a major rewrite was one of the goals of Longhorn with WinFS, but ultimately they couldn't keep postponing the release and had to scrap the whole project.

      In parallel universe Bill Gates was forced by this failure to leave Microsoft. He formed a new company where he rediscovered his passion for building software and developed WinFS into a fully functional OS. Microsoft ended up buying this company and Bill Gates returned as CEO, successfully incorporating WinFS into all future Windows releases to much acclaim and success.

    2. Lars Silver badge
      Pint

      Re: Using Sinofsky As A Scapegoat

      Exactly, perhaps. As I remember Gates was very much for one Windows for every device. To think that Sinofsky did something without support from Ballmer and Gates does not sound believable. Who knows, perhaps he just did what he was asked to do (and hated it).

Page:

This topic is closed for new posts.

Other stories you might like