WinFS Update


It’s been nearly a year since I wrote my entry about WinFS Beta1, but rest assured, we have been working furiously since then.  Today I have an update about how we are delivering some of the WinFS technologies. It represents a change to our original delivery strategy, but it’s a change that we think that you’ll like based on the feedback that we’ve received. 


As most people who read this blog know, WinFS has always been about many things – a new model to enrich how users manage information, rich storage technology, and sometimes also a packaging of technology.  The real change I am addressing today is in the packaging strategy. 


There are many great technical innovations the WinFS project has created – innovations that go beyond just the WinFS vision but are part of a broader Data Platform Vision the company is pursuing.  The most visible example of this today is the work we are now doing in the next version of ADO.NET for Orcas.  The Entities features we are now building in ADO.NET started as things we were building for the WinFS API.  We got far enough along and were pushed on the general applicability of the work that we made the choice to not have it be just about WinFS but make it more general purpose (as an aside – this stuff is really coming together – super cool). 


Other technical work in the WinFS project is at a similar point – specifically the integration of unstructured data into the relational database, and automation innovations that make the database “just work” with no DBAs – “richer store” work.  It’s these storage innovations that have matured to the point where we are ready to start working on including them in our broader database product.  We are choosing now to take the unstructured data support and auto-admin work and deliver it in the next release of MS SQL Server, codenamed Katmai.  This really is a big deal – productizing these innovations into the mainline data products makes a big contribution toward the Data Platform Vision we have been talking about.  Doing this also gives us the right data platform for further innovations. 


These changes do mean that we are not pursuing a separate delivery of WinFS, including the previously planned Beta 2 release.  With most of our effort now working towards productizing mature aspects of the WinFS project into SQL and ADO.NET, we do not need to deliver a separate WinFS offering. 


Be encouraged that we are able to get the underlying feature work into Orcas and Katmai.  It’s great technology and we are super-excited to be productizing this way.  And most importantly, it’s what people have been asking for – as we work with customers, we’re constantly hearing that they want many of the technologies to be more broadly available in the data platform products. That feedback was taken seriously.


Of course, there are other aspects of the WinFS vision that we are continuing to incubate – areas not quite as mature as the work we are now targeting for Katmai and ADO.NET.   Since WinFS is no longer being delivered as a standalone software component, people will wonder what that means with respect to the Windows platform.  Just as Vista pushed forward on many aspects of the search and organize themes of the Longhorn WinFS effort, Windows will continue to adopt work as it’s ready.  We will continue working the innovations, and as things mature they will find their way into the right product experiences – Windows and otherwise.  Having so much ready for SQL Server and ADO.NET is a big impact on the platform, and more will come.


That’s all for now.  I know people won’t be shy with questions and comments. 
– Quentin


Author: Quentin Clark

Comments (361)

  1. Dan says:

    So WinFS would be similar to Cairo?

  2. nesher says:

    Еще одна грустная новость от Microsoft. WinFS лавочку прикрыли :(Quentin Clark сообщил, что части…

  3. Kevin Daly says:

    I’m struck by a curious symmetry in this…I remember the announcement that ObjectSpaces was being rolled into WinFS (I was not exactly chuffed), and this seems almost like the converse (if you take the view that the Entities support in ADO.NET and LINQ have between them picked up where ObjectSpaces left off).

    Since some features that were originally to be offered as part of WinFS seem to be finding their way into Vista under other forms, I’ll have to have a good play with Vista before I have a feel for how much may have been lost.

    But I assume: There goes the relational file system. Shame.

    A message I got very early on (PDC 2003) was that file/object types could become meaningful in themselves rather than relying on the intelligence of a particular application designed to work with them: I hope that vision persists in some way.

  4. lexp says:

    WinFS is dead.

  5. Alex James says:

    I hope people see this for what it is: WinFS has morphed, rather than what the naysayers will argue: WinFS is dead.

  6. Jos says:

    When WinFS was delayed the reason given was they were waiting for the next version of SQLServer. Now WinFS is scrapped because the next version of SQLServer is coming

    Guys, please tell us whether we would get a relational file system in future versions of windows. We understand that the next version of SQLServer would support unstructured data. We understand that entities got adopted in ADO.NET. But could we expect a relational file system?

  7. lynn says:

    If it’s not a stand alone store system then it’s dead. Having these features in SQL are really nice, but we thought it was about a file system.

  8. OPC Diary says:

    WinFS Update@WinFS Team Blog はぁ?どこがアップデー…

  9. Kevin Daly says:

    Alex: Um, barring some clarification (especially touching on the relational file store issue), people could be forgiven for thinking that WinFS *has* morphed, certainly, but into something not unlike a corpse.

  10. Mohammed says:

    Sounds like they’re killing WinFS to me. It’s really amazing. MS has been trying to pull this together for what now, 10+ years? This was the only thing I was looking forward to in Vista, and then post-Vista, and now never…

  11. In the post that is sure to be slashdotted all to hell in the coming days, Quentin Clark of the WinFS…

  12. lynn says:

    What happens to "Project Orange"?

  13. WinFS is no more. At least not as a separate product to be released as an add-on to Windows Vista some…

  14. G.T. says:

    All what I wanted really from a file system is a smart lower layer, something you can call NTFS2, but a one that can handle very large amounts of files without knowing their contents.

    I wanted all the basic things, like hard drive snapshots, the possibility to truncate a folder, like deleing 1,000,000+ files in no time (similar to the database) etc.

    Try today deleting a folder with a million files, and tell me how many hours will it take 🙂

    I wanted a system that stores the file information itself to a database, a very simple database, not a database of the file contents.

    And as expected, all what Microsoft wanted is that damn email message, the one used in all samples since the beginning of the 1990s, how to split it to its smallest pieces and store it in a smart file system.

    But people are just repeating the mistakes, well; I think whoever writes a simple file system driver for windows, with a file allocation table similar to a database, a simple dbf table, and the source code is there on the internet, whoever does that, will be the next milliner.

    The files on my machine are just not going anywhere, they are growing and growing, and Microsoft (nether the competitors) has a file system capable of handling them all, well, ntfs can store them, but what about performance 🙂

  15. steve says:

    So the relational file store for Windows is dead while the product technologies live on for use in other products?  Everything sounds so rosy in the article with all the progress on the technologies, but yet there will be no SQL file store for Windows.  So that’s it then … WinFS is dead.

    For the guy who said it morphed… um, is there a product that will sit on top of Windows and do what WinFS promised?  No.  It’s dead.  Thanks for playing.

  16. PatriotB says:

    Wow.  Talk about spin.

    I’m normally a pretty strong supporter of MS, but I don’t hesitate to lay into them when they deserves it.  This blog posting is pure spin.  WinFS is dead.

    Ok, so you were able to salvage some of the years’ worth of work put into WinFS and apply it to other platforms.  But in this posting you are severely twisting what WinFS was.  WinFS was *not* a platform for developers building on SQL Server, it was a part of Windows.  Heck, it was even billed as an entire "pillar" of the (at-the-time) Longhorn OS.

    I remember reading about it, and there was an example about Outlook’s PST file.  Currently, Outlook stores its mail in its own database format; with WinFS, mail messages would be simply another type of "item" that could be in your WinFS "store".  You could search through your mail items just as easily as your files; you could create relationships between your mail, contacts, and regular files.

    By what this blog posting describes, this will no longer be possible.  This posting tries to spin it, however, saying things like "innovations that go beyond just the WinFS vision"–well, the WinFS vision (part of which I described in the paragraph above) is NOT being fulfilled.  Forget going "beyond" a vision if you’re not going to hit its core first.

    The Outlook PST example will not be possible with SQL Katmai–Outlook isn’t based on SQL Server.  Though, I suppose Outlook could be rewritten to work on SQL Express (or the equivalent of the MSDE desktop version).

    The bottom line is that WinFS was promoted as a Windows component that would enhance the file system and provide a new platform for data storage for Windows apps.  And this is now dead.  No amount of spin is going to cover that up.

  17. Tim Robson says:

    This is one of the most ridiculously managed project in microsoft. Teams that have taken dependency on WinFS are dead now. I remember MBF taking dependency on this dam thing for OR Mappling and is no where now.

    You guys don’t talk. First ship something and open your mouth! Never come to PDC and give preview of some prototype that never ships.

  18. Ryan says:

    "With most of our effort now working towards productizing mature aspects of the WinFS project into SQL and ADO.NET, we do not need to deliver a separate WinFS offering."

    Non sequitur.

  19. Kevin Daly says:

    News from Quentin Clark.

    I was quite keen on the idea of a relational file store.

    Oh well.

  20. Lalo says:

    It sounds so positive. But it’s like giving a speech in front of the coffin. You just keep remembering the guy inside, and the more you do, the more you remember he’s dead.

    Rock dead.

  21. Teis Johansen says:

    What does this mean for the vision of me having a central store of say contacts across all applications in Windows? By making WinFS a teknology in SQL server, it seems the whole "central store" idea is no longer possible, as it will be up to each application to utilize the functionality. That’s bad news in my book

  22. Teis Johansen says:

    What does this mean for the vision of me having a central store of say contacts across all applications in Windows? By making WinFS a teknology in SQL server, it seems the whole "central store" idea is no longer possible, as it will be up to each application to utilize the functionality. That’s bad news in my book

  23. Quentin Clark (Richard, un parent à toi ?  ) nous livre quelques informations sur WinFS (Windows…

  24. Tom Servo says:

    So I guess this pushes an object file system in Windows another 6-8 years into the future…

  25. davidacoder says:

    I think you are not helping with posts like this. Be frank and don’t try to spin news like this. Here are the reason people were excited about WinFs:

    – Relations between files

    – Common schemas for things like contacts that can be shared between apps

    – Access via "old" file system APIs for legacy apps

    – Incredible synchronization story

    All of these were features that end users would have witnessed directly. And, implicitly, all of these are gone for good now. Writing a post like "Oh, the exciting stuff will come in Katmai, this is really what you have asked for, this is good news" is just PR spin, be open and frank about this.

    Part of that would be to come forware and let us know WHAT the reason was this was killed. Did perf not come along? Did important other apps like Outlook not sign on? What was the real reason?

    I strongly believe that having a blog and engaging in an honest conversation can only work if you then don’t try to play the usual PR "every decision is great" messages.

    So, can you say anything about the four topics listed above? Are you still working on those for future Windows versions? Or have you given up completly on those? Let us know and talk to us!

  26. Tom Servo says:

    For something completely different, since things get refactored and put into Katmai, what’s the release date for it? Before Vienna? Is there a remote chance that Vienna will implement a relational file system?

  27. bob says:

    May 22nd you’re hyping it and June 24th it’s no longer? You guys are the keystone cops of software development.

  28. With the announcement of Bill Gates’ planned departure, there was speculation as to the fate of some his pet projects including the perpetually elusive WinFS (”Windows Future Storage”) file system.  Well, the other shoe

  29. Quentin Clark of the WinFS Team posts of the projects demise. Well he actually said was they broke up…

  30. Douglas Husemann says:

    At first I read this and thought, oh no another 10 year wait for the cairo file system. and then I sat back and thought about it.

    1:) the entity system has been embedded into ado.net and will probably will be the default framework (3.5??) in Fiji.

    2.) we have not heard all of the feature set in fiji, so it is possible that we will hear more after vista is released, but is highly likly that there will be a data platform concentration with fiji, in conjunction with the released data platform. I at least hope so.

    3.) we know peer to peer sync and castle was removed so that and the data platform would be a very good combination.

    and working with the ntfs team so unified schema platform in windows can be realized.  

    thanks

  31. Welcome to This Week with Windows! That means everything from Windows XP to Windows Vista. Here are a

  32. The WinFS team have announced that their product will effectively be absorbed into .NET 3.0 and into…

  33. Bilbo Baggins says:

    "productized" LOL

  34. rei says:

    So WinFS’ features are being absorbed into SQL Server. In effect, we won’t be calling it WinFS; we’ll be calling it SQL Server.

    You just moved the target audience of "WinFS" from the Windows end users (the majority of the computer users) to a small fraction of the server market.

    Revert the decision or WinFS is dead. To be brutally honest, you’re in deep shxt.

  35. Gus says:

    Damn, this was the only reason why I was even thinking of upgrading to Vista when it came out. WinFS was awesome. Now, Vista is dead to me. 🙁

  36. anders says:

    Seems that Microsoft finally gave up on one of their most promising ideas…

     

    Good thing for…

  37. Two interesting perspectives on what’s happening with changes to the WinFS release:

    http://blogs.msdn.com/winfs/archive/2006/06/23/644706.aspx

  38. Cathode says:

    Seriously this a huge blow to Windows Vista technology portfolio.

    Once considered one of the ‘pillars’ of Longhorn (wasn’t it?), considered by me the revolutionary and much needed way of organizing info on the desktop, is now, file-system wise dead!

    Please tell me this stance will be reviewed and reversed, if not for the next windows version at least!

  39. MikeP says:

    Unlike I believe the vast majority of your posters, I followed WinFS deeply and understood what you where doing with it. It was a brilliant way to address the issue of a whole new fs. But it also did occur to me how it wasn’t really much on top of other emerging technologies and looked like it should just ‘be there’ as opposed to a separate product. So this move doesn’t entirely surprise me. Everyone I discussed WinFS with didn’t get it. Anyhow, great work on getting to here, some of us understand what you’ve done.

  40. Jeremy Morton says:

    Alex James said, "I hope people see this for what it is: WinFS has morphed, rather than what the naysayers will argue: WinFS is dead."

    But whither Windows Future Store [or Windows File System]? That is gone. SQL will get some new features, but how does that help me organize my files?

    It sounds like some implementation advances will get to be applied to another product, but WinFS, as a product, *is* dead.

  41. Vivek says:

    Why Why Why ? cant you take a risk ? and come out with something innovative ? Something you can call your own ? What are you afraid of ?

  42. Anonymous says:

    How surprising…. 🙂

    Come on guys — it was written all over for at least 3 years!

    Yet another fine example of mediocre execution in a bureaucratic environment. Microsoft in his prime

  43. Translator says:

    For all people that don’t understand the MS marketing blabla here is the translation of this blog article:

    We have no idea how long it will take to complete the features we have promised in the last couple of years and get around all the problems so we have decided to rename the few things that work and put them into an other product. This way we will have some more years until even our most clueless customer will find out that we just can’t deliver what we promised.

  44. WinFS observer says:

    Instead of wrapping the death of WinFS in euphemisms, MS should admit it and instead share with the world the technical challenges (if any) that made the project infeasible. It’ll be helpful for other people who are planning to do stuff similar to WinFS.

  45. Caspar says:

    Woah, there Quentin. Take a writing course! Your text is full of corporate blabla and nonsensical words. What you are doing to the english langauge is a scandal

  46. MS-shareholder says:

    After multiple years of wasting our share holders  ‘s money, I hope to see someone to take responsibility for this WinFS failure.

  47. ToBeExpected says:

    Cairo went south and now WinFS, one of the three core features of Longhorn, advertised and partied about, declared abandoned.

    Vaporware, the second time.

    Hopefully, with Gates’ retirement, the visionaries grab some sense.

  48. Justin Broderick says:

    Boooo Hissssss

    WinFS was an innovative forward thinking project for the client. Basically you guys are telling me now that was all untrue and that it’ll serve me better as part of SQL server and ADO.Net.

    Now why do I feel like I was being sold a pipedream?

  49. Dan Kegel says:

    Cairo was a version of NT that was supposed to

    ship with an earlier incarnation of WinFS,

    OFS.  (See http://www.computerworld.com/softwaretopics/os/story/0,10801,69882,00.html )

    That was, oh, twelve years ago.

    I used to think that computers just weren’t powerful

    enough for a database to replace the OS’s main

    filesystem yet, but even though computers are now

    about 250 times faster, it’s evidently still not

    the right time.  Maybe it was just a bad idea?

  50. Callahan says:

    So when will we see some work to improve NTFS from a performance and reliability standpoint?

  51. Adam S. says:

    So, where does this leave us for filesystems for Vista?  Is WinFS the filesystem dead?  If so, then I consider WinFS itself dead.. SQL Server stuff is nice, I guess, but it’s nowhere near as useful or relevant (how many normal users use SQL server?)

  52. jonnie savell says:

    EXCERPT:

    And most importantly, it’s what people have been asking for

    COMMENT:

    Customers are NOT asking for this technology to be removed from Vista (even if they do want it available in SQL Server). Besides, if WinFS were available in the operating system, would it not also be available to both SQL Server and its clients? Something doesn’t smell good.

  53. Alex James says:

    Over the last 24 hrs I’ve had a fair bit of time to digest some initial reaction to the ‘no seperate WinFS’ announcement. My initial reaction was very ho hum, ADO.NET Entities is going to be pretty cool.

    But listening to other people’s reactions, I can see that I had kind of lost sight of the fact that it means we will have no relational file system in the forseeable future…

  54. I really tire of this.  Cario then Blackcomb then Longhorn now Vienna?  .NET technologies totally tossed out, leaving little more than "Indigo" which is sandboxed .NET Internet services atop of [polluted] Win32 (didn’t we already have that with Java 5 years ago)?  And no CairoFS — sorry, I mean WinFS — is dead.   "Vaporware and switch" strategy — why don’t you guys just stop promising what you’ll *NEVER* deliver?!

    I’m not just a "pundit."  I was a solid, NT 3.1+ supporter from day 1 in 1993 — coming from OS/2 and UNIX.  But there is not even a *BASIC* solution that addresses the SAM-SID NTFS store other than using a "domain" SAM or a "hidden area" in a LDM disk label (dynamic disk).  These hacks up on hacks upon hacks really make it *DIFFICULT* to maintain enterprises of Windows Servers and desktops.

    In 1993, I thought NT was the future and community-developed software like GNU/Linux was a toy.  But by 1994, with "Chicago" (DOS) getting the thumbs-up from Gates, forever making "Cario" (NT) it’s "bitch" — I knew NT was screwed forever.  Community developed has continually impressed me by underpromising and overdelivering.

    Making Microsoft should take a note and stop promising solutions that are "grand" and end up being vaporware.  You can start by addressing the real flaws in NTFS filesystems with just some _simple_ modifications — stuff you promised some 12 years ago in CarioFS.  It seems you keep promising more and still won’t deliver on just the basics we need!

    Sorry for the rant, but Longhorn has been a 100% repeat of Cario, only 10 years later.

  55. Mark says:

    When did programs become "product experiences"?

  56. Chris says:

    Ext3 filesystem for life! 🙂

  57. Microsoft is awesome at everything, except "Actually releasing products we can use".  They talked about ObjectSpaces back when I was a baby. They didn’t deliver.They talked about DLINQ, and came out with Entity Framework, which is still a good

  58. joe says:

    I love Microsoft, its such a wonderful company

  59. anonymous says:

    WinFS is dead.  God save the WinFS.

  60. modicr says:

    Hello!

    WinFS was baby of Bill Gates. Is death of WinFS related with his departure? I want to hear BG’s opinion regarding this news …

    Thanks, Roman

  61. Joel Jacob says:

    IT IS A SHAME. MS promised us this thing nearly two decades ago.

    More, it was in my opinion the only component that should give Vista its *real* value. Now this OS is just a revamped XP with candy gadgets to suck memory and waste CPU cycles.

    Poor management, crazy marketing and names changes : MS is clearly in the wild.

    Time to seek alternatives…

  62. PatriotB says:

    And… it seems that comments that aren’t pingbacks aren’t getting posted?

  63. KK says:

    Why doesn’t it surprise me? WinFS is a classic example of paralysis by over analysis. Lofty goals need lofty execution, which the WinFS team lacked.

  64. Doom says:

    Dead, dead, dead. Just when I though M$ would for once actually be able to innovate and produce anything. Oh well, I guess you’ll try again once Apple of Linux has this working.

    Remind me again what’s left in Vista except lots of pain from apps not working within their new sand box?

  65. Joseph says:

    It shows how Microsoft is still 15-20 years behind what Unix and IBM Mainframes can provide today.

  66. Anonymous Coward says:

    "With most of our effort now working towards productizing mature aspects of the WinFS project into SQL and ADO.NET, we do not need to deliver a separate WinFS offering."

    OH YES YOU DO! you promised it, remember? besides, WinFS was the "main" feature and selling argument of Vista. you stripped like each new feature in Vista, and it looks to me that Vista will be nothing more but a new GUI skin for WinXP.

    "After multiple years of wasting our share holders  ‘s money, I hope to see someone to take responsibility for this WinFS failure."

    LOL, don’t even dream about that. you cannot blame anyone, nobody is responsible, because everyone played by the rules (i.e. The Big Holy Process Defined By The Company), and the 12 levels of management singed along.

    now seriously guys: all the new features of Vista got stripped by now – so tell me at least one selling argument: WHY should i buy? cause i’m not going to pay for a new shiny XP skin alone. show me a single example of something, which is not possible to do with Windows XP, but will be available in Vista!

  67. Sundeep says:

    Quentin, from your posting and many of the comments above, it would perhaps be beneficial to clarify the following:

    1. Your posting implies that WinFS benefits will be released as they become mature. The corollary is that WinFS is being dropped because of technical immaturity – could you be more specific as to the technical issues you have faced.

    2. What are both the timescales and features which are scheduled for release over time, and in what products?

    3. After so much investment, and with the feeling that we’re so close to the prize – what was the rational behind shifting delivery strategy in this way? How much of the decision was based on technology, and how much was based on business/strategy issues?

    4. In terms of Windows Vista, what WinFS benefits are expected to be released and what are the expected timescales?

    I may be being naïve, but clarification of the above questions would help consumers of Microsoft products understand the implications of your announcement.

    Sundeep Sidhu.//

  68. Bob Lewis says:

    Reiser baby!

  69. frik says:

    The blog article is just a way to paraphrase the whole thing and make it look good. People were waiting for a Database Filesystem and some new feature for SQL Server. Lame. WinFS is dead, like Cairo. Microsoft isn’t able to learn from their flaws of the past.

    They should evaluate to hire some former BeFS/BFS (BeOS Filesystem) Developers.

    So are there any plans for a Database Filesystem or a successor for NTFS.

  70. Linuxrocks says:

    HAHAHAHA… HEHEHEHEHE…. So damn funny. Microsoft is so paralyzed; I think the proper mood is Windows is dead…Think GOD for Linux; Something predictable that JUST WORKS….

  71. Well 🙁

    I was interested always in IT future. For me it was amazing to play with Windows Longhorn since I’ve got 4008 build. The most interesting thing in it was the feature called Windows Future Storage. The whole idea to create the new type of data system (against file system), that would offer data relations, full-text search, smart data organization made me happy with it. It was very nice to play with 4074, and 5048 told me that the world is going to crash now. But, unfortunately, we’ve got WinFS Beta 1 by August 2005, and that was the most biggest surprise for all of us.

    I’ve started to learn WinFS from 4008, and have currently a set of applications that utilize WinFS Beta 1 Refresh, among them there are some Shell Extentions wrriten by me and wrapped into Windows Explorer itself.

    I’m really feeling bad no as WinFS is closed.

    Windows Desktop Search even with TxFS willn’t ever try to do the work that WinFS does.

    That is easy to understand.

    Windows Vista Search is so so, better than Windows XP Search of course, but it willn’t ever try to offer us other features of WinFS.

    RSS is cool, but WinFS Notifications Framework is somewhat.

    Windows Vista Sync Center is nice, but features of WinFS Sync are the things that developers need in real.

    The whole centralized store for all data, structured, related data system – that was the thing we all waited for.

    Dear Quentin, please tell me, will Microsoft try to recreate WinFS in future? If so when do we expect it?

  72. Porges says:

    So, the underlying technology (i.e. a relational database system) will be shipping with SQL Server (a relational database system)… Groundbreaking!

  73. Billy Bob Joe Bob says:

    I’ve just re-read the blog posting and have to hand it to Quentin – great Spin! He took hundreds of words to say one simple thing: "WinFS is Dead".

    Quentin has a great future working for MS markeing.

    Sadly,we don’t get WinFS.

  74. DanyX says:

    This is a very big mistake IMHO. WinFS, as part of Windows, could have been the one thing that would have set Windows, as a platform, years ahead of the competition (MacOS, Linux, …). It would have been the foundation for a new way to think about organizing (Desktop-)Users data. It would have enabled a whole new concept of data-sharing accross applications. But the only way this could ever have worked would have been if MS put all its weight together and just make a bold statement like: this is how data is going to be organized in the future. The sad thing is that it will be almost impossible for a 3rd party to develop something like WinFS and see it widely adopted. This stuff has to be a core element of the OS, because only then will enough devolpers pick up on it. I actually wrote a blog post on pretty much this topic back in Summer 2005: http://www.danyx.com/node/44

    I think this is maybe the one strategical mistake at microsoft that they will regret most over the next 5-10 years.

    Daniel

  75. Jimbo says:

    I was somewhat interested in seeing what would happen from a security perspective if MS were to slap a Relational DB on every client machine running Vista.  I am sure that would massivly increase the surface area for attack on client machines…..  That said, it had some fantastic functionality which was most intersting.

  76. Luka says:

    It certainly sounds like you are killing off WinFS. Now why on earth would you do that?

    Nobody can accuse Microsoft of being the leader when it comes to innovation, but WinFS was actually something fundamentally new. Without it the newer versions of Windows seem like minor cosmetic upgrades. OS X is already there, in some areas perhaps even ahead of Vista. Linux will catch up.

    With Web 2.0+ applications, which OS you are running is becoming less important from a compatibility point of view. So if Microsoft wants to maintain its dominance in the field it can’t produce mediocre products. It must take the lead and actually give people a reason to buy a Microsoft OS.

  77. Brent says:

    What a shame and disappointment.  Two comments and lessons Microsoft should seriously take to heart:

    1. Don’t promise what you can’t deliver.

    2. Serious consider what REALLY matters and add values to your products.  

    WinFS had to potential to add true value to the Windows OS.  WinFS was a true consumer need.

    Vista is filled with "glitz" and eye-candy that in the end is not focused on important end-users needs.  Take a close look at the Vista feature page: http://www.microsoft.com/windowsvista/features/default.mspx

    What does Vista really have that you can’t get from using XP (or other OS), with a few other (not to mention, freely avalable) add-ons?  Priorities at Microsoft need to be "realigned".

  78. Spock says:

    Cut the crap.  You knifed the baby.  WinFS is dead, and Vista is a DRM-infested, bloated, resource-sucking, eye-candied yawn.

  79. Karim says:

    "The key here, I think, is to not think of death as an end, but think of it more of a very effective way of cutting down on your expenses."

      — Woody Allen, "Love and Death"

  80. Jonathan Rothwell says:

    Basically meaning that if you want a relational file system, develop it yourself or use Linux. Thanks.

  81. Sho says:

    Huh? This is the worst spin I’ve ever heard. You’re "productizing" in response to "customer demands"? I have never, ever heard anyone say "boy .. WinFS sounds great .. wish it was only available in SQL Server!".

    What a load of bollocks! WinFS was supposed to be the grand new Windows FileSystem. That’s what it stood for before you renamed it "Future Store". Now, it’s neither, now it’s some proprietary DB thing no-one’s going to use? Guess what guys. We already have databases. I can name a billion ways to store data of any type you want in any DB you want and it’ll work great.

    We don’t want a new DB, we want a new filesystem. This project is nothing but a TOTAL FAILURE and as you can easily see, no-one who read your silly "put on a happy face" post was fooled at all.

    Seriously .. can MS do anything big anymore? Enquiring minds want to know. This is just failure after failure. What the hell’s gone wrong up there?

  82. noitsnot says:

    Yes you do need to offer WinFS as a standalone product:

    Some of us novice developers don’t want to have to go through the hassle of using an SQL server or dealing with ADO.net. We want a god damned clean and easy to use, STANDALONE system of managing files for our applications, ans WinFS was something I personally looked forward to until you guys killed it. I fail to see why I still beta test for you guys as a whole, because you guys drop features left and right.

    WinFS didn’t evolve, it was cut into pieces and left for dead.

  83. Kalervo says:

    So, I think you managed to achieve blowback out of not just one but TWO body orifices.  Congratulations, I thought that was a special honor only reserved for the White House press secretary.

    Let me see if I got the basic gist.  WinFS as a client side technology built into the operating system is dead but the smart technology bits will be folded into an enterprise level tool.  In short, the ability to have relational data schema is gone for the VERY forseeable future, or until the OpenSource community figures out how to get it to work.  Bad news, it looks like they may be close (<a href="http://www.linuxplanet.com/linuxplanet/reviews/5816/1/">Tenor Project</a>).

    I am a big Microsoft fan, I like your products.  Windows XP is my primary operating system, Office 2003 is my primary productivity environment.  Visual Studio 2005 is my primary development environment.  I have a special loathing for Mac, but I am forced to utilize Linux for some development and work with dedicated Medical technologies.

    From that vantage point, I think that Microsoft has some serious trouble coming their way in the next year.  There is a convergence of several technologies, specifically: the Looking Glass Project, the Tenor Project, and KDE 4.0 which provide a compelling reason for me to move from a Windows development environment to a Linux environment.

    Specifically, I want and need the relational file system that WinFS promised to be.  A revamped SQL server isn’t going to give me the functionality I want to have at the Operating System level.

    I want the ability to treat individual file types like objects and have universal relationships which is the core functionality of WinFS and why it even matters.  How precisely is a re-vamped ADO.net going to allow me to do that?

    In addition to the vaporware claim, Microsoft just made a huge strategic blunder.  You are choosing to abandon the largest consumer market for a niche market.  While I understand WinFS was to be a free add-on, you had many application developers convinced it was end-all and be-all.  With WinFS, the added functionality at the OS level would have been huge.  I may have finally be able to convince my higher ups to port our Linux Apps over to Windows.  Now I think I will be making the pitch that we port our Windows Apps to Linux once Tenor and other technologies become available.

  84. linux.com says:

    Kinda expected it do die a slow death.

  85. EACC says:

    nice job MS!

    you are the best company!!!!

    ur making Vista the next ME!

    … time to look somewhere else, perhaps OS X?

  86. Well this news sucks. The WinFS team at Microsoft has posted a new blog entry that sounds so positive,…

  87. bubble memory, voice recognition, WinFS, QED

  88. Jim Gay says:

    WinFS is dead.  I had viewed Vista as a prerequiste to the eventual release of WinFS.  To me this means, there is nothing much of interest on the horizon beyond Vista and there is no real improvement to Outlook coming either.  So, if Office 2007 runs on Windows XP, tell me exactly why I want to switch to Vista.  Every day the idea of switching to a Mac that can run the few Windows apps I can’t do without is looking better and better.  Furthermore, desktop Linux probably isn’t as dumb an idea as I once thought either.  I’m even going to have to go back and check on how Chandler is doing.  When a die hard Windows nut likes me begins to see a horizon filled with Macs and Linux desktops with open source core applications instead of the promise of a better Windows, I think that might be something for Microsoft to be concerned about.  Ray Ozzie, who I admire, may be Microsoft’s last hope.

    Ray, you got one shot.  Make it a good one.

  89. PatriotB says:

    "We’ve worked on advanced file systems before, and our customers should be happy to know we don’t give up on something that’s important to them even if we didn’t get it right."

    — Steve Ballmer, re: WinFS, 10/19/2005  http://www.microsoft.com/presspass/exec/steve/2005/10-19Gartner.mspx

  90. Abubakar says:

    Damn it !! its so depressing !!!! BillG is leaving his empire at a very wrong time. Have they discontinued it on Ray’s orders ?

  91. JT says:

    This just makes me angry.  Usually I am the one sticking up for Microsoft, but this time they just went too far.  I don’t care if it takes until winter of 2008 – make WinFS happen as a standalong "file system".  I’ve seen beta 1 and I can’t understand why they are able to move forward with this.  I have to imagine that is internal political nonsense.

    I’m very disappointed.

  92. Oran says:

    I seriously hope this decision isn’t due to conflict of interest with bigger, more mature Microsoft products, but I’m guessing that’s what happened.

    WinFS was set up to be an MSDE/SQL Express killer.  But that lives in a separate product family.

    WinFS was also set up to make the Office family of products have to work harder on the front-end by making a level playing field on the back-end.  But Office also lives in a separate product family.

    Both of those product families are major "premium offering" cash cows, while WinFS would be free in the "non-premium" Windows.  It wouldn’t offer the "get them hooked on the lite version and force an upgrade when they need more power" upgrade path like SQL Server does.

    If you want to change the world with a relational file system, you have to make it absolutely ubiquitous.  This means giving it away, no ifs ands or buts.  This makes it a foundational platform service for Ray Ozzie’s services future, similar to the RSS Platform in IE7.

    But the cash cows get more love from Ballmer.  People like Gray "Premium Offering" Knowlton who came from Adobe and was put in charge of InfoPath in the Office family keep killing awesome products by refusing to open the doors that would make the product truly ubiquitous.  Acrobat wouldn’t have taken over the world if its Reader wasn’t free.  But no, they still believe they can take over the world, one enterprise silo at a time.  It won’t work.  You’re fundamentally stuck in the silos, no matter how many you can win over.

    I wish WinFS could have been another one of the Windows * Foundation pillars, but I understand the desire to "productize" instead of "foundationalize".  All I can say is good luck.

  93. rei says:

    How about we stop crying and organize a petition or something?

    I’d say it’d be worth doing.

    Anyone with me?

  94. radioman says:

    i’st bad! I’m very disappointed! Fuck…

  95. Nice Work says:

    I’ve always been a staunch M$ believer but today I’ve seen the light.  I think I’m going to take a look at the Mac OS X.  M$ you need to get your SH%$ together.

  96. Bob Thomas says:

    Wow! Such negativity! It shows just how much these readers do care for Microsoft products!

    It’s interesting to note that there isn’t a single other totally bug-free product available that has delivered on every single feature that WinFS envisioned!

    The good point is that some WinFS features Will be delivered! For this, the development team should feel proud. For those portions that aren’t being delivered at this point in time, a yardstick indicating their ETAs is definitely in order.

    Bob

  97. E.E. says:

    I think the key take-away from that Ballmer quote is that very few customers are in a big hurry to move away from the standard hierarchical filesystem.  Seriously, who is WinFS for?  As a geek I think it’s pretty neat, but as a software developer with bills to pay, I don’t see the point – are there any commercially interesting scenarios that demand WinFS?

    I’m sure somebody will demonstrate such a scenario eventually, and that will give MS some real customer-driven requirements to design around, and that’s what it will take to get the WinFS vision into production.  

    Probably Apple or some OSS group will do the proof-of concept, and MS will embrace and extend as with GUIs, spreadsheets, tabbed browsers, web-based email, etc, etc.

  98. I was going to write about this but it seems like everyone feels the same way I do.  I wonder which monkey at Microsoft made this decision.  Perhaps this is why Bill has decided to depart from Microsoft over the next couple of years.  His visions crushed by, oh I cringe when I imagine, the people that have other ideas of what a schema should look like.

  99. Hal says:

    What you have with WinFS is a chicken and egg problem.  Indeed, this is a key problem with all of Microsoft’s attempts at "integrated storage" going back to Cairo.  It makes no sense to have a WinFS unless the Microsoft utilities and applications make use of it.  When WinFS was pulled from Longhorn/Vista (for reasons unrelated to WinFS itself) its fate was largely sealed.  All of the teams that had dependencies on WinFS had to go to alternate strategies in order to make the Longhorn/Vista shipment.  They implemented as much of their planned functionality as possible this way.  They aren’t likely to turn around and redo all that work in the release after Vista.  I suspect it will take two or three releases for most of them to be ready to make another change to their storage strategies.

    So the question then becomes, who will REALLY use WinFS over the next 3-5 years?  The answer is ISVs.  Well, most ISVs (particularly bigger ones) currently build products on relational database systems.  So which is more likely, that you’ll get the ISV to use an extensively enhanced SQL Server or a new file system that the system itself isn’t even using?  Well, the answer is the former.  And that’s what I think this strategy change embraces.

    When Windows is ready there is nothing to prevent Microsoft from creating a special "System Instance" of SQL Server that can be used as a Relational File System.  Nor is there anything to prevent them from offering a more file-systemish API on top of that instance.  

    Nor is there anything to prevent those Microsoft applications that are already SQL Server based (and there are a lot of them, including portions of Office) from using and exposing the WinFS technologies that make it into SQL Server.

    So it IS fair to say that WinFS is dead, and I wish Microsoft had just said that.  I think Quentin tried to say it indirectly, by making the Data Platform reference.  Microsoft is less focused on creating a relational file system and more focused on evolving a next generation Data Platform that can be used by all applications.  And eventually that Data Platform seems likely to be materialized (in some variation) as part of the base platform.  That probably is a better strategy, but it would take days of explaining Microsoft culture before most people could understand why.

  100. Maybe the time is ripe again to have a go at an in memory database? 😛

  101. Weatherman says:

    I remember watching an MS guy in a video telling us that we should architect the apps while waiting for the next WinFS dev release. So I did. And wasted my time.

    When I tell my boss that WinFS is dead tomorrow, he’ll be willing to re-eval his vision of MS-only future of our company because that was the one thing we wanted to build upon…

  102. Joe says:

    First the very first time, just now, I seriously thought about getting a mac. Wow.

  103. "The only measure of a technology’s strategic importance and success at Microsoft is the extent to which it is adopted and integrated into other Microsoft products", Michael Herman (about 1990).

    Clearly Windows wasn’t buying into the promise of WinFS; neither was Microsoft Business Solutions.

    The fact that the WinFS technologies have been picked up by SQL Server helps insure they will eventually ship.

    It sounds like a great decision.

    p.s. I would also guess this is, in part, an example of the Ozzie Effect taking hold at MS.

  104. Neil Rieck says:

    The problem for our industry is the same as it has always been, to be on the cutting edge of technology, you walk softly on a rope bridge over a seething pit of molten code.  

    Engineers get excited by whats possible with the tools, and the marketing people want to build that hype to assist product flow,  so they want it out there as soon as they can get it.  

    So the beta products are pushed and pushed until one day the bean counters go…  enough…  this is just uneconomical…  and slash it out.  

    I actually am in favor of miscrosoft telling us what is possible.   i would rather know that than get to delivery and discover all this wonder stuff that i can suddenly use.

    what i have to accept with this attitude,   is that a key component of what i want to use MAY NOT BE THERE when i do get something.  In this instance, we have been given 6 months to get to our plan B…  or is it plan G….  

    from the blog a lot of people are really upset now

    ok…  so this would have been a really bold move to integrate data management into the file system.

    so we miss out on say "content-addressable files"

    "relational file system" is what it was billed as

    so it didnt get there…  so what…

    um…  arent there other things that give us that??? arent there other data management features…

    and who uses it in an OPERATING ENVIRONMENT anyhow???

    maybe, but it would have been built into the operating system

    again…   i think its quite bold to do that…

    data management is data management…

    and operating environment is just that….

    i know…  why dont we build an operating environment that does statistical analysis for us…

    that way every time we click on a data file it will just spit out the analysis for us…

    finally…  is this not an opportunity for someone to write this as a plug in… and sell it…   given the demand that has been generated… and MS have walked away from it…

    anyone can write an overlay for an operating system…

    so overlay = distribution hassles = low take-up  yeah… unless ms like what they see and buy it…

    in the operating system = a standard that ALL application developers can rely on

    in my world,,,   i dont look at what impacts ALL application developers…    

    i look at what impacts ME….

    and standards are things that currently exist – not that should exist.

    as developers,   we are actually quite selfish in what we develop.

    we tend to hoard our little bits…  and protect our domain like a ghobi fish…  occasionally stealing from others we encounter…

    once in a while a glittering jewel is dangled before us and we all swim madly to possess it…

    very occasionally one or two of us do… and are heralded as the king ghobi fish…

    when something threatens our territory, we react…

    and of course,  while we swim after the jewel..  other ghobi fish raid our turf.

    Then of course when we finally get back to the nest we have

    to evict the intruder and get back to being in control of

    our little territory…

    and the really really good thing about what i have said here…

    you dont have to agree with a single word of it.

  105. I find it interesting that there isn’t a single reply to the "WinFS is dead" comments from Mr. Clark.

    When I read this my take fell inline with what everyone here is saying, that WinFS is dead and that this blog entry was a lot of BS trying to pretty up a lame announcement.

    I had WinFS installed for a bit and didn’t really think it was ready for primetime, but the idea behind it is a no-brainer. The reality of a strongly-typed file-system would be huge.

    But then again, if MS is going to go to an Internet/Services based architecture, it might be argued that the future file system will be the Internet via services.

    Still sucks.

    So anyway – Hey Quentin, can you answer the question? Is WinFS dead?

    David C.

  106. Meio Bit says:

    Parece que a Microsoft decretou o fim do WinFS, o revolucion&#225;rio sistema de arquivos que a empresa pesquisa desde os anos 1990 ( ent&#227;o, sob o nome de &quot;Object File System&quot; ). Inicialmente anunciado como parte do Windows Vista,…

  107. WinFS or Windows File System, what was supposed to be a part in the WinFX has been declared dead. Or…

  108. This is very disappointing.  I think this is a huge strategic mistake.  I really hope someone at Microsoft is reading this blog and rekindles this project.

  109. First Cairo and now WinFS!  All I can say is that if you guys ever decide to tackle a relational file system again again, you’re better off not leaking a word of it until it’s basically cooked and ready to go.

    I have to tell ya that I’m feeling more and more like we were sold a load of bull@#$t at PDC 2003.  WinFX as the managed API for Vista never really came to pass and now WinFS got the final shot in the head.  For future PDCs I would suggest that Microsoft be a bit more careful about promising stuff they are not sure can be delivered.

    Come on guys, fess up and don’t try to white wash this.  It is only hurting your credibility which after the emasculation of Vista is already pretty bad.

  110. Check out Quentin Clark’s blog about WinFS and the article about ADO.NET Entity Data Model.

  111. Lex says:

    What a disapointment.

  112. Clemens Eisserer says:

    I’ve never heard anybody speaking about himself in glory after failing with a project.

    Indeed the last 2-3 years nobody honestly believed in winfs.

  113. In its blog, Quentin Clark (here) is announcing a big change in the availabilty of WinFS technologies….

  114. WizbangTech says:

    The biggest club Vista-Bashers have wielded up to this point just got a little bigger: …we are not pursuing a separate delivery of WinFS, including the previously planned Beta 2 release. With most of our effort now working towards productizing…

  115. TechBlog says:

    (Jim Thompson is blogsitting TechBlog while Dwight Silverman takes a 2-week unplugged vacation. Dwight will resume his usual TechBlog duties on July 1.) Here’s a summary of recent news from the world of Windows Vista: Ed Bott notes that a…

  116. c0t0d0s0.org says:

    The one and only feature that i would call more than a 10.4-rip off is no offically dead. Here is the report of an  microsoft developer in mabushi speak. Or as an independet view : The Fishbowl or Developing on the edge.

  117. In case you haven’t heard the news, WinFS is dead. At least, what it should have been is dead; the relational…

  118. Mark Sutton says:

    Just another nail in the coffin.  But then what do I care…?

    OFF TOPIC: I haven’t run Windows on any of my own PCs for over 2 years now, and have only one single support contract left that involves supporting M$ Winblows.  On the other hand my work supporting Linux (both on the desktop and in the datacentre) is maturing nicely and there’s nothing I haven’t been able to fix myself.

    So – for me and my clients its Linux and F/OSS all the way baby…

    The other day an old friend of mine asked where he could get copy of Office for less than £350.  I just laughed and told him to pay the tax or look for something else.

    In _his_ words he told me he didn’t want to be raped again.  So I told him that the days of _needing_ Office and Windows were well and truly over and that he should try Openoffice. Guess what, it worked great for him and he loves it, and he even managed to open some old .doc that his last version of Office couldn’t!

    OK, so it might not be quite as pretty, but it does work well, and it doesn’t cost £350.  He made a £10 donation to the OO project and told me he’ll never go back.  He even dusted off an old Laptop to try out Linux.  Email from Gmail, pdfs from OO, and he can even access his corporate Exchange server with Evolution.

    Fair enough, if you need Autocad or something else proprietary then Windows is your only real option at the moment.  But in time this will change too.

    Microsoft: are you listening?  The majority of people just need web, email and an office equivalent.  These people don’t need to bend over any more, and in time the old M$ habits will die.

    You have been warned, and not just by me. Software and computing will move forward /with/ or /without/ you.

  119. So, WinFS seems to have bitten the dust.

    It only seems like a fortnight ago, I was wondering what the…

  120. In this blog entry Quentin Clark writes about the future of WinFS. Apparently the amibitious goal of…

  121. Sentido Web says:

    Una de las novedades m&#225;s anunciadas que iba a tener Vista (anteriormente Longhorn), WinFS (Windows Future Storage) al final no saldr&#225; nunca a la luz, Microsoft dejar&#225; de desarrollarlo. Iba a ser una revoluci&#243;n dentro de los sistemas

  122. It seems like the WinFS team has tossed in the towel. None the less, the team and its offerings will…

  123. Jeffrey Whitney says:

    "productizing" is not a word!!! Why to tech people feel the need to make gerunds out of words? Sorry, it just makes you sound like a dope.

    Also, it sucks that this is happening. I’m hopeful, however, that someday the reality of WINFS will become a reality.

  124. Andy says:

    Microsoft is blowing it again.  Why do you think everything has to be rolled into SQL Server?  SQL Server is not the solution to everything.  People wanted a relational file system and one that performed fast and had advanced syncronization features.  And what about SEARCHING?  Google is going to kill Microsoft in this area with their google desktop.  Microsoft search today on the desktop is just awful.  With all the delays that have come with Vista this is just more bad news.  I’m sure the stockholders won’t see this as a rosy picture as you have portrayed in your post.

  125. What is going on at Microsoft these days? From my perspective, it seems just totally out of control.

    Their vision for (and support of) developers these days is just horrendous. MS seems determined to confuse and derail developers at every turn.

    The recent announcement about WinFX becoming ".NET 3.0" (even though it uses the .NET 2.0 CLR) is another fine example of recent absurdities and "spin oriented" communications:

    http://blogs.msdn.com/somasegar/archive/2006/06/09/624300.aspx

    Whoever is making these decisions really needs to go home.

    Seriously.

  126. Andy says:

    Goodby WinFS 🙁

  127. Alexey says:

    Maybe Longhorn/Vista should use ReiserFS 4.0, you’ll get most of what WinFS was supposed to be.

    http://www.namesys.com/

  128. Kirk M says:

    You were right Quentin. I’ll bet you knew what the response was going to be even before you wrote  it. I wonder if it left a bad taste in your mouth when you found out what you had to announce. I’m speaking strictly professionally, no offense intended at all.

    Productizing? Data Platform Vision? The post is so full of CompanySpeak any reader with an only mild case of awareness would know that it was nothing but so much air. WinFS is gone as far as the home/small business desktop platform is concerned and the user sadly loses out again because at least half of the management at Microsoft can never seem deliver anything on time or at all while the other half are rolling out half-baked  "Web 2.0" components. The only "obvious" exception seems to be the MS Office.

    team.

    Again, no offense intended but it’s not "super cool" and the public at large private and professional are getting pretty tired of all the empty promises that MS have dished out over the years.  

  129. Roger says:

    So at some stage with a future version of MS’s world all applications that wish to access a file on a shared file system will have to be coded to work against ADO.NET so that they can access files located on a MS SQL server, which in turn needs needs MS Server to provide it with a file system!!

    God is the open source community going to have a field day provding a unified file system that can also provide the required ADO.NET interface.

  130. John Topley says:

    Is "productizing" a word…I don’t think so.

  131. TIMSTONE says:

    Well, for my money either Microsoft has found something better or now thinks that todays desktop/laptop hardware could not hack running a real WinFS release.

    Either way – without Microsoft we would all still be wearing one leg of our trousers inside out – and I remember the pre Microsoft days of hell.

  132. It’s a shame.  It’s the only thing I was looking forward to in Vista.  What’s happening Microsoft ?

  133. mike miller says:

    Will this effect my 360, media or internet experience? nope… does the general public care nope… does it effect the techies and developers that wanted ms to provide something worthwhile, on time and not need a SP less than a year after its release maybe….  You need to get it together MS, Visual Studio 2005 is buggy and inherently flawed that notepad is a far better alternative.  If dropping WinFS is going to create a better environment to develop for (MSSQL) and on (.net) then carry on.  I don’t need WinFS and Aero to complete my work, but I do need a bug free SQL Server and Visual Studio 2005! 🙂

  134. Eberhard Schefold says:

    I was — shocked. Yes, only that would be the word.

    Really, after all thouse YEARS, the expectations, the hopes … What is going ON, Microsoft? This time I am getting worried, seriously. As far as Windows is concerned, I feel the end of an era is in the air.

    … and it’s the customer’s wish. Now how cynical can you POSSIBLY get?

    Devastated,

    — Eberhard

  135. szlevi says:

    This whole story of WinFS is so ‘Microsoft’… at the beginning WinFS was heavily hyped, for several years as the future filesystem of every Windows – and then it’s being constantly scaled back, a gradual process which finally ends with this log.

    To be completely honest it wasn’t unexpected, at least not by me after last year’s ‘fast deterioration’… of course, still nobody had the balls to come forward and openly say that the long-awaited, next-generation filesystem of all things Microsoft, KNOWN AS WinFS HAS BEEN KILLED – no, all we got this ridiculous explanation, let alone that laughable notion that they are "super-excited" about the canning of their project…

    Apart from the pathetic nature of this story and this pathetic try to sell it as some kind of ‘good news’, we should also remember that Microsoft DOES NOT HAVE ANY FUTURE PLAN TO REPLACE THE SERIOUSLY OUTDATED NTFS WITH A MODERN FILE SYSTEM.

    Even though MS already lost a lot of upgrade sales when they have announced WinFS won’t be shipping in Vista, I think we can conclude that BY CANNING THEIR NEW FILESYSTEM MICROSOFT JUST KILLED MOST OF THEIR VISTA UPGRADE SALES AS WELL.

    At this point I don’t see any reason why would I upgrade hundreds of our workstations – there’s NOTHING in Vista, nothing but a dumb facelift.

    This story just convinced me that Microsoft is indeed in very-very bad shape – too much Ballmer and alikes and not enough engineers…

  136. anon says:

    So bascially vista will be XP with a warmed-over interface and extra DRM ?

    I’ll be first in line to get a copy (not).

  137. John Daniels says:

    I am saddened by this. I was looking forward to WinFS – soon. The concept should come around some day. But it might not be Microsoft who does it.

    This creates a huge vulnerability for Microsoft. If a competitor would release an OS with this feature, it has the opportunity to gain significant market share. Like AMD’s gains on Intel (AMD has the integrated memory controller, Intel failed to do this)

  138. john says:

    Please listen to more of your customers then and live up to your commitment to delivering this on the client sans MSSQL.

    WE NEED THIS TO DO CROSS APP/VENDOR RELATIONAL FEATURES!

    If priority wise it makes sense to delay this 6 months to deliver pieces of it earlier in MSSQL fine but please reconsider and then bring it to the client as promised.

  139. Anyway says:

    Well, there are already some neat meta data layers to be used with different Windows platforms:

    http://www.m-files.com

  140. tecosystems says:

    Catching up on some feeds over the weekend – via the miracle of GPRS dial up networking over bluetooth (it’s one of the cooler things I’ve seen in a while) – I ran across the news, and then reactions to…

  141. BB says:

    >>The most visible example of this today is the work we are now doing in

    >>the next version of ADO.NET for Orcas.<<

    This isn’t going to work. Orcas, being marine mammals, do not even have opposable thumbs. I don’t see how they will be able to use ADO.NET, unless they come up with an interface for them that does not require use of a keyboard or mouse . . .

  142. woodgray says:

    Just another…………

    There are several high end packages out there that already provide functionality similar to WinFS (Documentum for one). Making WinFS essentially a freebee by being included in there OS would stop a huge money stream from the commercial market, document management etc., that they could become part of by making WinFS part of SQL server. They will introduce new products over the next year or two based on all the feed back from the beta testers that will give them new products they can sell in the commercial market. The OS market is now to slow and stable for them to make money or grow with. This is purely a marketing ploy not a technical challenge that can’t be delivered.

  143. Jon Strayer says:

    Wasn’t an object based file system promised for Windows 4.0?

    If you don’t remember Windows 4.0 that’s because it never shipped either.

  144. CynicalGeek says:

    Microsoft sold Software Assurance to many corporations whose sole aim was to buy a usable Search function for Windows.  Now we’re just out of luck for several more years on the desktop side.

  145. As Tim Bray says &quot;Wow&quot;. Here is the announcement post with a huge number of comments. This is discouraging. As…

  146. TechBlog says:

    (Jim Thompson is blogsitting TechBlog while Dwight Silverman takes a 2-week unplugged vacation. Dwight will resume his usual TechBlog duties on July 1.) Here’s a summary of recent news from the world of Windows Vista: Ed Bott notes that a…

  147. G.T. says:

    Some people can, some can’t

    Sun Announces Availability of Solaris ZFS, World’s Most Advanced File System

    http://www.sun.com/smi/Press/sunflash/2006-05/sunflash.20060502.5.xml

  148. bananahead says:

    With the demise of WinFS, I think that makes three major projects that Mr.Clark has been in charge of over the last 5 years that collapsed under their own weight.

    This one hurts.  WinFS was the major reason Vista had a chance of success.  Now Vista is reduced to the role of WinXP, SP3.

    Nice work Quentin.

  149. Richard von Mallesch says:

    I don’t understand this at all. WinFS made it to beta, isn’t it just a case of ironing out the bugs? I must be missing something… Seriously, if anyone can explain why they would decide to drop such as promising technology at the beta stage I would love to hear it.

  150. Kilz says:

    More Quality Vaporware courtesy of Microsoft, everyone favorite monopoly.

  151. Now that Bill Gates has announced plans to work less at Microsoft after two years, blogs around the world are jumping onto the wagon to announce how Microsoft is going to change and how Microsoft…

  152. Radix says:

    While at Microsoft I was asked to produce requirements for what was essentially an agnostic heterogeneous data store.

    Using XML based standards.

    Access via XQuery

    Cell level security.

    Versioning, branching, relationships capable of supporting any arbitrary structure or multiplicity of structures.

    The result, after close work with the SQLServer XML team, was a system with a deceivingly simple structure yielding systems of amazing complexity.

    Why?  

    Separation between physical and logical structures allowed both the Data Architect and DBA to achieve their operational goals without having to seriously compromise their goals.

    Updates to data structures no longer required long development cycles to implement.

    The basic objects need not be shredded to a relational model for storage, security and searching.  The  XML object in .NET could persist (for all intents and purposes) as an XML object.  No more Object Relational mapping (50-60% of mid-tier code) consumes developer time and computational resources.

    To fully implement the system however required SQLServer Next (post 2005).

    Furthermore numerous data stores with similar properties already in development, leading to potential confusion among developers.

    It seems to me that Microsoft has delayed the release of a technology in order to consolidate object data stores, unify developer experience, fully implement all necessary requirements rather than provide developers with an incomplete or evolving APIs.

    Traditional RDB models suggest this type of system will suffer from performance difficulties.  The prototypes we developed were substantially faster than an RDB.  Mainly due to the great work of the SQLServer XML team in supporting the XML data-type and clever insights into the the nature of predicate evaluation.  

    Will the new server fully support the original requirements – I can’t say.  However it seems Microsoft is making the right moves to support those requirements.

    This is just another point of view on what may be happening behind the scenes.  It is not gospel, just educated speculation.  If the SQL team continues to deliver on their track record of success, everyone should benefit from the current restructuring.

  153. It’s sad to see. I was hoping it to be the one thing that would set Vista apart from the rest…

    &amp;nbsp…

  154. Quentin Clark has posted an update to the WinFS update and answers some of the questions that came since…

  155. There’s been a lot of discussion about the recent decision not to ship WinFS as a distinct product, but…

  156. Malek B says:

    While I was developing a Document Imaging Archiving application, I came to a point where I had to decide wether I should store the "Image Files" in a databse BLOB or on save them on the filing system … I looked at newsgroup postings and found many people saying that we should never store huge file images in database BLOBS.. while others had a completely different view .. at the end I decided to leave such a decision to the end user ..

    Now that future version of SQL will support a filing system is the answer to my paradox .. So i expect that my new version of my application will be highly suited to use WinFS as a part of SQL .. So I think that I’m on the right track .. can’t wait for the release of such a product ..

  157. Malek B says:

    While I was developing a Document Imaging Archiving application, I came to a point where I had to decide wether I should store the "Image Files" in a databse BLOB or on save them on the filing system … I looked at newsgroup postings and found many people saying that we should never store huge file images in database BLOBS.. while others had a completely different view .. at the end I decided to leave such a decision to the end user ..

    Now that future version of SQL will support a filing system is the answer to my paradox .. So i expect that my new version of my application will be highly suited to use WinFS as a part of SQL .. So I think that I’m on the right track .. can’t wait for the release of such a product ..

  158. amunix says:

    Here is theReg’s follow-up: Microsoft’s future file system dies, again (http://www.theregister.co.uk/2006/06/26/winfs_axed/)

  159. konstantinos says:

    I strongly believe that Vista except the extra security is just a circus without winfs!!! I dont care if a cartoon dog or cat or lions helps me do my work by asking me sily questions. WHAT I NEED IS A SERIOUS OPERATING SYSTEM KNOWING WHAT I WANT TO DO AND DOING IT BEFORE I THINK IT. Thats what i think winfs would do. put my files in an order with their subject without looking in the filename. I DONT CARE ABOUT THE AMAZING GRAFIX. Please look to vista’s "what’s new?"! Is this a new operating system? This is NT4 with new interface.

  160. Dan Brickley says:

    Next time, try exposing it as a SPARQL-compatible  RDF quad-store so that developers and co-workers who accept dependencies on it aren’t burned when it gets escalated from shared infrastructure to "product" component…

  161. Daniel Björkström says:

    M$ can´t develop anything that someone else haven´t. Always been this way.

    New features will come but in the ordinary way of accquiring the features from small brilliant developers who spend their life developing cool stuff.

    You should change slogan to "If we can´t make it -well buy it"

  162. No surprises here – MS announcing a move of technology from Windows to SQLserver. Guess which one is the better revenue stream? Lots of people want what WinFS has promised, now you will need to pay out even more to get it.

  163. ken says:

    #define winfs(f,m) fopen(f,m)

    There’s your d*mn WINFS. Now quit your whining.

  164. Anyway says:

    Anyone interested in alternatives for WinFS should try M-Files (http://www.m-files.com/). It has a real meta data layer and fast searches (meta data & file contents) etc… Why would we even need WinFS when we have this kind of product?

    You can categorize files without needing any traditional folder hierarchy (dynamic views), it is usable from remote locations, searches are fast (indexed) and this product is closely integrated to Windows (you can use it just as any other disk drive etc, from command prompt and all the applications)

  165. Armén says:

    No, dammit, you bring it back!

    It’s by far the most aftersought, attractive, useful and innovative feature! I ve been cursing at them folders for years now.

    People, what about petition? Looks like the only way to affect the decision makers at Microsoft? They claim they listen to the customer feedback, so a petition would do!

  166. Anonymous says:

    "Vaporware"

  167. As of 6.23.2006 Quentin Clark made this post&amp;nbsp;stating in a nutshell that WinFS is going to be rolled…

  168. Am letzten Freitag wurde die Entscheidung bekannt gegeben, dass wir WinFS nicht als seprates St&#252;ck Software…

  169. Otis Wildflower says:

    Chiming in here, WinFS was the only interesting thing going on with Vista IMHO.  Taking concepts that BeFS (wtf didn’t MS buy Be, anyway?) demonstrated and running with it.  What else in Vista isn’t just some form of Featurespeak?  DX10?  Meh.  .Net without a big fat download?  DRM?  Are you kidding??

    Sad to say it’s only the promised security updates, and the only _real_ way to get security is to break compatibility with single-user assumptions, which will _never_ be permitted to happen.  

    WinFS was the only technically interesting piece of Longhorn, and it’s been snipped like the family dog’s youknowwhats.  Yipe yipe yipe indeed.

  170. This last weekend Quentin issued a press release via two posts on the WinFS blog announcing that WinFS…

  171. Riaan says:

    I can see the entry in the Encyclopedia of the Future – Microsoft is dead and failure to deliver WinFS killed it.

    Without WinFS how do you plan to effectively integrate the desktop and server versions of Windows and how do you plan on delivering a product that anyone wants to buy?

    The average computer has millions of files on it these days and until a relational model for file handling can be established there is no point in continuing to expand the capacity of computers. With Moore’s law comes the fact that the complexity of information increases – and that this becomes unmanageable. Quite frankly I struggle to see why anyone would ever buy Vista.

    Google aims at organising the world’s information. They reckon that it will take another 300 years or so – but I’m fairly sure that Microsoft will not be around anymore then. If we cannot find an effective filesystem on the desktop and server then the effective filesystem will be the web and search will remain the killer app.

  172. Brian Kelley says:

    A blog posting from the WinFS team

    caught me a bit by surprise today. Apparently I wasn’t the only one,…

  173. 4sysops says:

    It seems that WinFS as a new file system for Windows is dead. This is what Quentin Clark says on a Microsoft blog:
    Since WinFS is no longer being delivered as a standalone software component, people will wonder what that means with respect to

  174. Those of you eagerly awaiting beta 2 of WinFS should

    note that WinFS has been canned. Well thats my…

  175. Xepol says:

    Good spin, but let me unspin it for you.

    WinFS was meant to be the successor for NTFS as a revolutionary new file system, now the pieces are being gutted and what parts can be scavanged are being incorportated into other projects.

    The WinFS has failed utterly, don’t expect to see any more code from it.  Soon, the WinFS team will be officially disbanded, and those team members deamed worthy will be absorbed by other projects groups who find bits of the winfs code base interesting, others will be found responsible for the failure and cast out, and many team members are just as likely to slip between the cracks unnoticed as MS pretends WinFS was something it wasn’t rather than truely research what went wrong to prevent it.

    It’s a little like watching an episode of Invader Zim, except there is no Dib.

  176. D'oh says:

    What the hell were you thinking? That everyone would buy your bulls**t spinstering lock, stock and barrel, and not notice that a bomb had just been dropped on three years of promises and expectations? Were you hoping for "Oh goody, a database with my next database!"?

    It’s one thing to have the largest concentration of capital and engineering talent on Earth, with a luxury of time that an utter monopolistic market lock affords, and still report back with yet another breathtakingly spectacular failure.

    It’s another to do so in a blog post that is cowardly and insultingly sophomoric, does violence to the English language, and spits in the face of honest (look that word up), straightforward communication with customers which your own Robert Scoble has bravely championed.

    Tell us you don’t actually sit there and wonder why people are repulsed by Microsoft.

    PS And I’m one of your fans, developers, and long-time CUSTOMERS.

  177. It isn’t much of a surprise but apparently WinFS is pretty much being scrapped. You can just read through…

  178. jabits says:

    This is great…



    >>The most visible example of this today is the work we are now doing in

    >>the next version of ADO.NET for Orcas.<<

    This isn’t going to work. Orcas, being marine mammals, do not even have opposable thumbs. I don’t see how they will be able to use ADO.NET, unless they come up with an interface for them that does not require use of a keyboard or mouse . . .

    Anyway, I am sad as a dev and mad as a shareholder.

  179. Douglas J. says:

    ’nuff said.  I was hugely disappointed when they started to pull Vista apart in order to get a decent implementation date.  But was mollified in knowing that a separate incarnation would at least produce WinFS.

    But, I’m happy – HAPPY I tell ya! – to learn that WinFS is well and truly ….well, pushing up the daisies…joining the choir invisible…you know the rest.  Because it means my fallback plan of going over to the DarkSide (Apple) is going to be a reality. And I can finally acknowledge that Jobs is my fahhhhther.

  180. Brandon says:

    It’s a shame the relational file system is not coming out in longhorn, but we will see these unstructured data in the relational data store and this "rich data" functionality and the non-DBA automation in other future products, namely Exchange Server and Active Directory.  Both products are built on top of the ese engine and will soon be built as part of the sql engine.  I would like to see multi-master replicated sql servers.  

    MSFT wants to pool the development talent with an understanding and expertise of a similar API set.  It’s getting to expensive to built, test, debug and support multiple technologies.  If they can focus on one "Data Platform Vision" Microsoft can inovate and build better products faster.  Seems to me WinFS isn’t ready for primetime and that’s good that microsoft isnt adding a version 1 product in their OS.  Microsoft can’t afford to ship bugggy products, just like Boeing can’t make products with defects.  The profit margins increase substantially with quality and costs also scale when people use microsoft bugware.  Unfortunately, when the stakes are higher and we hold MSFT to a higher standard, their costs increase as does the development cycle.

  181. richbl says:

    Marketeering speak:

    "It’s great technology and we are super-excited to be productizing this way."

    Engineering speak:

    "WinFS is dead."

  182. You have believed says:

    WinFS is dead. Long life to the relational database.

  183. Mjinga Wawa says:

    Why, G.T., I wasn’t aware you needed a hat!

    Stop by my office on the second floor and pick one up – I’ve even got some super-sized ones!

  184. Rich Bulan says:

    Oracle has some interesting file server type capability with their XMLDB feature in their database.

    http://www.oracle.com/technology/tech/xml/xmldb/index.html

  185. Jason Taylor says:

    HYSTERICAL.  What I wonder is, how do they suck the brains out?  Does the soul come out right then, or is that another machine?

    I hope I don’t get productized at my next job review.  I’d then be super-excited to tell my family my company is using a new innovative packaging strategy, which is what I had been asking for, and therefore I would not need to provide a paycheck.

  186. Eric says:

    It would seem to me that only a few of the people posting here *have* installed a recent WinFS drop, tried any samples, or at least understood the most recent implementation of WinFS. Most are likely not developers either. (I don’t mean *you*, of course.)

    If you can’t understand exactly how all of WinFS can be phased in, maybe it’s because you (we) don’t work on the WinFS team. If you can’t *imagine* how it might be done, then maybe you haven’t been keeping up with the overall direction of things.

    Has no one seen how WinFS easily sits "beside/on top of" NTFS and plays nice with existing apps? How it already *is* a combination of the things described here?

    Why can’t SQL Server Express become an invisible part of Vista, delivered as R2 or SP3 or whatever they want to call it? I think they can work out a favorable OEM licensing agreement with themselves. Why can’t creation of WinFS objects show up as a simple [Attribute] on your class? Or, why can’t a new namespace yet emerge that makes "storage" a unified no-brainer?

    These things will come, and you will have "WinFS". Seems most of you will fail to recognize it, and you will just be in some other blog railing on some new perceived MS idiocy, but you’ll be using your slick new WPF/"WinFS" powered bloging dohickey that is seemlessly integrated with all the other apps and *the* data store on your machine.

  187. luckily we got SemFS for Linux.

    join the project

    http://semfs.ontoware.org/

  188. help.net says:

    Well according to the WinFS Team themselves, what was supposed to be the new milestone of file storage…

  189. linux user says:

    see… if you windbloze users had listened to us from the start, you’d be using a proper operating system and software (gnu/linux) and you wouldn’t now have to look for an alternative to the Microsoft Windows ‘Promises’ operating system…

  190. Full of I.T. says:

    WinFS: What&amp;rsquo;s up?

    Readers who follow Microsoft news and blogs already have probably seen the buzz…

  191. WinFS a fost anuntat cu surle si trambite ca fiind noul sistem de fisiere din viitorul sistem de operarea…

  192. As Microsoft blogger Quentin Clark eloquently put it, &quot;Wow.&quot;At the recent TechEd conference in Boston,…

  193. This is pretty pathetic.  I too had seen WinFS as the only substantial paradigm shift in OS design in a long time.  Vista seems now truly to be mere "Windows dressing” – another GUI overhaul.  My apps fly around!  Golly.

    WinFS promised radical improvements in the way people worked day to day.  Data is, after all, the crux of the thing. WinFS had the potential of making a truly data-centric OS, of exposing metadata to every app on the system and letting end users invent new value by connecting the dots in ways never imagined by individual vendors.  

    I’m getting tired of defending the value MS brings to the world to advocates of open source, MacOSX and the like.  When Microsoft makes promises so publicly and for so long it is astounding that there can be a viable rationale.  The damage to the good will among their core supporters is enormous.

    Was it as simple as this: Someone at MS (someone with more booked revenue than the WinFS crew) saw WinFS as a threat to the SQL business? That ISVs would provide SQLServer analogs using the WinFS plumbing – that WinFS would essentially cannibalize the SQL Server biz?  If so it’s a clue that MS sees their increasingly cloudy future revenue resting on server products and that they’re conceding inevitable defeat in the desktop OS market, at least in terms of competing on real innovation.  WinFS was a meaningful OS differentiator.  It was a way of “out Googling” Google and could have played a strategic piece in branding MS as relevant in search technology.

    Bummer.

    I’m going to go and download the Google Desktop search tool.  MS had their chance.

  194. GQAdonis says:

    I am starting to have little confidence that we will see the features of WinFS OR ADO.NET 3.0/Entities/ObjectSpaces.

    It also seems to be a confusing message to have ADO.NET 3.0 released apart from the .NET Framework 3.0, which includes WinFX, WCF, and all the other new technologies.

    How are we to keep track of all the technologies, their dependencies, and their compatibiities when it seems that one hand does not know about the other?

  195. Soy unos pecadores de la pradera y unos mintirosos. Ya sólo falta que el VISTA necesite 3450 GB de memoria para funcionar.

  196. I am sorry, but it is shame to use RDBMS as file system. This idea has no perspectives. It is better to implement some object-oriented database management system instead of file system. Some OODBMS like this http://www.shuklin.com/ai/ht/en/cerebrum/

  197. Laughing John says:

    One word: "Shame"

  198. szlevi says:

    Clark wrote it:

    "It would seem to me that only a few of the people posting here *have* installed a recent WinFS drop, tried any samples, or at least understood the most recent implementation of WinFS. Most are likely not developers either. (I don’t mean *you*, of course.)

    If you can’t understand exactly how all of WinFS can be phased in, maybe it’s because you (we) don’t work on the WinFS team. If you can’t *imagine* how it might be done, then maybe you haven’t been keeping up with the overall direction of things.

    Has no one seen how WinFS easily sits "beside/on top of" NTFS and plays nice with existing apps? How it already *is* a combination of the things described here?

    Why can’t SQL Server Express become an invisible part of Vista, delivered as R2 or SP3 or whatever they want to call it? I think they can work out a favorable OEM licensing agreement with themselves. Why can’t creation of WinFS objects show up as a simple [Attribute] on your class? Or, why can’t a new namespace yet emerge that makes "storage" a unified no-brainer?

    These things will come, and you will have "WinFS". Seems most of you will fail to recognize it, and you will just be in some other blog railing on some new perceived MS idiocy, but you’ll be using your slick new WPF/"WinFS" powered bloging dohickey that is seemlessly integrated with all the other apps and *the* data store on your machine. "

    Clark, you just convinced most of us that you don’t even understand what we’re complaining about, let alone your apparently fuzzy konwledge on filesystems…

    I suggest you to study the early WinFS history, the claims made by MS over the years, the original hype etc *before* you post *your next unimpressive yet offending* post…

  199. Cyberjudas says:

    No reason to upgrade now : ( . This was the one reason for upgrading. I may as well stick with sp2 and the addons . Its really sad that they put such a spin on what is a complete failure for windows vista. Please tell me how moving these technologies to ADO and SQL will benefit us Windows (Vista users). IT wont. So it is Dead. Atleast be honest and dont insult us with your bogus spin. I am a beta tester for Vista and there is little reason to upgrade now. Somthing that really could have been a Milestone in computing history has now become just another Windows ME. Sad really sad.

  200. Anne Martens martena@home.nl says:

    I know many People have NO Memory, but still.. Remember WINFX has been a demanded as the Ultimate GOAL By visionair Bill Gates. No one else wanted it within Microsoft, NTFS is Great and LOOKOUT and Sharepoint Portal server are great substitutes. Now Bill has left, so there is NO need to continue the path of developing another filesystem that only he wanted, which in the end might be farv less robust than the  Open Source filesystems(reisser) It was only a playing tool from Bill and he curiously just left the company. This ain’t a coincidence at all.

  201. With the user base, respect and popularity that iView has it would do a lot to evangelize the benefits of WinFS/Katmai.  Additionally, Microsoft could do what they did with the Creative House Expression and Lookout teams; use the experience gained from those apps to build application. With a little work a Katmai backed information explorer built by the iView team could become the next Windows Explorer.  The emphasis on virtual sets, cataloging, tagging, metadata, and visual file manipulation metaphors are already there.

    You heard it here first; the acquisition of iView Multimedia was made specifically for the WinFS/Katmai project.

    http://www.osreview.com/2006/07/01/winfs-and-project-orange-to-be-reborn-in-iviews-mediapro/

  202. MS, what are you doing???

    This -was- THE oppurtumity to get of one of the biggest problems of information management: say version control, author control, linking files to applications, autoupdate shortcuts, networking FS, and so on and so on.

    With WinFS we could get rid of all those stupid and too expensive DMS systems. I expect MS will have a DMS product in their portfolio one of these days….

  203. 想知道关于WinFS开发的一些内幕信息么?想了解WinFS开发的最新情况么?WinFS团队的BLOG是寻找这些问题答案的不二选择。

    以下两篇特别值得推荐:

    Monday, June 26, 2006…

  204. Pepito Grunwald says:

    There are other commercial and open source products for file management, collaboration, and document management that have databases integrated into their architecture. Many of these products have WebDAV support for desktop access, Java APIs for customization, and do other things that WinFS would do only moderately well, or not at all.

    Therefore, I’m not sure why this question is being posed as, "If Microsoft doesn’t provide an embedded database, we’re out of options." Is the core issue the expected integration with other Microsoft server products, like Exchange? The .NET APIs? Or something else?

  205. Stretch says:

    You know – we’ve just not learned the lesson from those companies that are massively successful.

    Bundling WinFS into a super massive "Data Platform Vision" is a good idea from a strategy perspective, but not from a delivery perspective.

    Tie WinFS into the strategic direction – absolutely – but GET SOMETHING OUT THERE.  Deliver stuff in small chunks and innovate rapidly.

    Putting WinFS into something like the bigger system areas will just slow it down to the snail’s pace that has afflicted Vista during it’s development cycle.

    First prize would be:

    – Look at the data platform vision

    – Put WinFS on there as one component, and keep the team driving to deliver this quickly

    – At the same time, put the other components on there as short-sharp deliverables (keep it small, and keep the pace fast).

    It would be great to see the first pieces (e.g. WinFS Beta 3) of this data platform vision landing by October, with follow-up drops of betas happening very soon after that – it’s great to have a long-term vision, but don’t try to boil the entire ocean all at once!

  206. SIKOVIT says:

    I comment as a SOHO user, so first some basics to set the context of my comment:

    1. A computer is just one of many tools I use to do what I have to do in daily life.

    2. A computer costs money.

    3. I make a simple "value for money" (VFM) consideration to determine whether the amount in 2. gives the required benefit to 1. before buying.

    4. We used to have shops where you walked in, gave someone a list of your required goods, that person collated the goods, you paid, thanked the person, and walked out.

    5. Now we have supermarkets where you walk in, wander around trying to cross-refer your list with a mental map of the aisles and shelves (and doesn’t get harder every time the supermarket gets bigger!) whilst pushing a progressively heavier trolley, stand in-line to check out and leave – no-body to thank.

    To me Windows is a supermarket. All we get are makeovers every few years – a paint-job and new shelving. But at the end of the day – I do all the work, you want to charge me an ever-increasing entrance fee and that makes Windows’s VFM head in the direction of zero.

    I already store in the order of 10E5 personal documents on my computer. This is growing fast as my PC deems to become the centre of all my information, communication and entertainment activities. I am soon going to reach the point where the FAT16->FAT32->NTFS "supermarket" has bloated into a giant mall that will be too big for my brain to manage – I just will not be able to find my way around!.

    WinFS promised me a "shop" with an efficient, fast, courteous person to store and retrieve my documents for me – regardless of the quantity/type/size/complexity etc. That person would give progressively increasing VFM. You would be willing to pay for this SERVICE, you would WANT to give a tip. Look at the phenominal success of Internet search-engines. "Great – thats the way computing is going" I thought.  Then I see the above.  Microsoft – you have betrayed the vision of your founders.

    Thats why I am so disappointed that WinFS is not an inherent part of Vista.  Thats why I will not buy Vista.

    Microsoft, please will you have the decency – from a customer-care point of view – to put your WinFS archive into the open software development community so that the work can be finished.

    Yours – despairingly . . .

  207. annoyed says:

    It never ceases to amaze me how many self-proclaimed "experts" on topics come out of the crevices when *ANY* technology company makes any sort of announcement like this, let alone when it is Microsoft.

    Half of you who chastise others for "not understanding filesystems" don’t understand them yourself – nor would you be intelligent enough to succeed where apparently MS engineers have failed.

    The other half of you are simply MS haters who have nothing better to do than to post ridiculous messages filled with unwarranted proclamations and accusations against MS or whatever technology company you happen to be putting down.

    If you have all the answers, why not help the cause rather than speaking out against it? If you don’t have all the answers… simply put – shut up!

  208. LinoX says:

    "now seriously guys: all the new features of Vista got stripped by now – so tell me at least one selling argument: WHY should i buy? cause i’m not going to pay for a new shiny XP skin alone. show me a single example of something, which is not possible to do with Windows XP, but will be available in Vista!"

    Well..There’s one thing in vista that’s not in XP.. it’s a techonogy called "DRMverywhere" 🙂

  209. Editorial note: there will be a certain type of Drunkard’s Walk feel to this post, but that is because…

  210. What a week. First Holland loses the World Cup match from Portugal and everything turned from orange

  211. A ra&#237;z de un post en el blog del equipo de desarrollo de WinFS, se est&#225;n alzando muchas voces que dudan

  212. In this blog entry Quentin Clark writes about the future of WinFS. Apparently the amibitious goal of

  213. Mario Briana says:

    Check out Quentin Clark’s blog about WinFS and the article about ADO.NET Entity Data Model.

  214. While the debate rages on between who provides the best extension for search beyond the web, and on to your local file system. In the near future, it is inevitable that the best results will be obtained by whom ever…

  215. I had the opportunity to hear from Mike (the one behind the Windows Speech Recognition language support

  216. YAIB — yet another IBOT blog — feel free to ignore if they aren’t your thing… Prior blogs in the