Slashdot Mirror
WinFS' Demise Not a Bang Or a Whimper
Shadowruni writes "The Seattle-PI confirms with Mircosoft what MS bloggers and pundits have been saying all along. WinFS simply isn't going to happen. Some of its features have been 'merged' with other projects." From the article: "WinFS was dropped from Vista in what company executives described at the time as a trade-off to get the operating system completed in a timely manner. The release of Vista has since been delayed again and is now scheduled for November for large customers and January 2007 for the general public, though some observers say it may be out even later." Final confirmation of a story from last month.
I would lay even money on Spring 2008. How long did Win2k take to stabilize? Granted XP went a little quicker, but the explosion of mal-ware made this an almost impossible, and some say unachievable task. I am sorry, but 10+ million lines of code just do not strike me as reasonably predictable and thus stable. At some point, the combinatorily explosion even might give the code sentinence...
really this could have been an inovation worth looking at from microsoft. to bad it's not coming.
With more and more announcements like these, does anyone else think it is inevitable that Linux will overtake Microsoft on all bases one day? I mean, it's starting to show Microsoft is only one company devoting a portion, large but just a portion, of it's resources to its OS while Linux is an entire industry with a bunch of diverse people working on small parts seperately.
I wonder if the Vista's voyage is any kind of vindication to the Linux side, who was always ballyhooed as having "too many distros" earlier, but at least we could depend on someone, somewhere releasing some small update with some type of progress (small but frequent steps) rather than the monolothic approach of large but infrequent steps.
So MS has was founded just over 31 years ago. Wouldn't a company that has spanned that many decades have a better understanding of software engineering and have a better grasp at making deadlines? I just don't get it. I'm not a fan of MS, but I'm trying to look past that: I just don't get how they can keep underestimating Vista the way they are.
:wq
I would love to see an OS released for the market that combines all of the research done within the past 10-15 years in kernels, file systems, HCI, application development, programming languages and APIs, virtual machines and virtualization, etc. However, look where we are at now. We're still using (for the most part) monolithic kernels, old file systems, old development tools, etc. There hasn't been any radical improvements in commerical OSes for quite some time. (One could say that OS X is a dramatic improvement, but much of OS X is based off NeXTSTEP, which had existed for quite some time before Apple bought them out).
I would like to see a new NeXTSTEP (technologically, not in terms of business success). NeXT was able to look at all of the current CS research of the time and integrate that into their operating system. NeXTSTEP was far ahead of its competition and, if it weren't for hardware support and the need for modern software, I'd probably run it as an everyday OS. Mac OS X is still ahead of its competition because of its NeXTSTEP roots, as well as Apple's improvements to the OS since 1997. Imagine if there was a new OS that took advantage of all of today's CS research, was very easy to use, and was compatible with existing software. I'd be the first person in line to buy it.
Until then, I can dream about my ultra-secure, exokernel OS with a database file system, flexible yet safe programming language, very easy to use UI, "boxes" to run Windows and *nix software....
The house of cards that is Windows is finally unmanageable and on the verge of collapse. We've known it was coming all along, as they kept bolting stuff on and attempting to patch a crufty old OS in the interest of backwards compatibility. Their inability to integrate a new filesystem into it is a sign that their 2 decades worth of nasty spaghetti code has finally reached critical mass and simply can't be futzed with anymore.
They're just going to have to bite the bullet sooner or later and do what Apple did-- drop the old OS in favor of a new one, and ease the transition to it by allowing the old one to run as an application.
The billion dollar question is, will they be able to manage such a huge transition? Based on how terribly their OS projects are mismanaged, it's extremely doubtful.
Overtaking Microsoft is not enough to become the dominant OS, for example see OSX.
Philosophy.
Microsoft has not yet finalised plans to make the most commercial success out of WinFS. Making it part of a highly pirated OS doesn't make commercial sense. Lack of features in a rebranded OS doesn't imply loss of sales / profits either. Improved features doesn't imply more profits from the OS business as well.
.Net or Active Directory or the Aero interface or BSOD... we'll have to wait and see.
And so, until MS dcides whether to package WinFS as part of SQL or
If you keep throwing chairs, one day you'll break windows....
So what you want to see is an compilation of immature academic technologies into a mature stable production system. Why not just wish for a gold house?
How we know is more important than what we know.
Odd...
While Vista does have steeper requirements, and by default runs a bunch of GUI effects that aren't really necessary--I don't see it as a disaster at all.
I just removed Beta 2 from this very laptop, and my reasons for doing so had nothing whatsoever to do with the OS (My laptop's DVD reading software decided not to work, and ATI's beta vista driver didn't support OpenGL). For the entire time I had it on, there was exactly one crash -- and it was caused by the aforementioned laptop software, not Vista. For the past few weeks, every time I've seen someone's computer problem, I've remembered part of Vista that would make that problem easier to fix.
Say what you will for the cathedral, but the book coming down the pipe isn't one that's going to fall apart at the seams.
Meanwhile, other marketing people are looking at the feature set of distributed link tracking.... And another set of marketing weasels are looking at DRM respect... and attributes for near-line storage management... and (name any competitor's advantage, and expect Marketing to want to add it to the feature set).
The failure isn't in Engineering - it's in Management. Someone promised too much complexity.
Given a year or two per feature set, done incrementally, with product releases that allow the code to be tested and refined, WinFS probably could be engineered into a fine solution.
But the deadline is too close now. They need to cut their losses and bug-check what they have, now, so that the file system that does ship is stable, and not a huge disaster.
Interestingly, the open source solution of file systems is far better at trying out new ideas and making progress. It may take longer to make the features integrated - but that integration hasn't been a defining requirement for success or failure.
"The most sensible request of government we make is not, "Do something!" But "Quit it!"
I call BS. What "large customers" are clamoring for all the great updates that Vista will bring? Every major corpration that I've ever seen is bent on stability and steady IT deployments. NOT shoot from the hip, let's jump on this tech because it's "the latest per our IT department". Executives care about the bottom line, not the latest software release - unless it's MSeBaysoft(with the Paypal module).
I'd love to see one large customer named.
Don't think that a small group of dedicated individuals can't change the world. It's the only thing that ever has.
1) While the filesystem architecture is pretty horrible, there have been successes there by other companies.
There's are file system interfaces to NFS, FTP, EXT2, UDF, and a probably a few more that I can't think of right now. This has nothing to do with the previously badly written code.
The problem with WinFS is WinFS. It's got features in it that would make it unacceptably slow and easily corrupted. That won't fly. I think they thought that they could overcome these obvious problems through genius. Apparently its still hard.
2) Like every OS trailing back almost to the invention of the compiler, Windows is modular. And by that I don't mean "it has modules, or even dlls" I mean that the ideas within it are divided into real (and occasionally conceptual) pieces.
Some of the pieces are new and shiny and well written. Some are old and spaghetti-like. There's no reason to throw out everything to get one new piece. The fundamental design of the Windows kernel is neat even if the registry isn't. The network stack works pretty well even if the filesystem interface doesn't.
Along those lines,
I think they should stop selling windows as one thing. I'd like to know what new thing it is I'm getting in the latest version of Windows. Because they do occasionally throw out the old and replace it with something new and fresh that works great. But sometimes they only sell things that are exactly the same as the old, but with things I don't care about at all, or sell me lots of things I don't care about and only one that I do.
Mod me down and I will become more powerful than you can possibly imagine!
Why WinFS failed to deliver...
When the concepts of relational database FS were being thrown around back in the mid 90s, there was a need for this technology. WinFS was to be the next progression of this work, but in its new form a non-structure, non-relational database FS technology.
WinFS was designed to sit on NTFS, never to replace it. In fact none of the proposed MS FS technologies were ever to replace NTFS.
WinFS did develop several inroads in database technology to move past relational and object oriented database storage concepts; however, this was not enough for it to succeed, but rather for its technology to be used in database and data access technologies like MSSQL and the ADO models.
There are two big reasons WinFS was stopped before ever seeing the light of day.
1) Efficiency over functionality
2) Business & Networked File Systems
The first is probably the biggest nail in the coffin, but yet also the hardest one to get through to people.
In current computing environments, adding in a good indexing technology, you can provide 99.9% of the functionality of WinFS and the overhead in doing so turns out to be less than if a full WinFS was implemented.
For example, it is easier and more efficient to have a database indexing backend that references the standard FS and FS contents than it is to put the FS contents into a database. This can be witnessed in products like MS Desktop Search, the Vista Desktop Search, and Apple's Desktop Search as well. (Although the Apple incarnation at this point is a bit more poky than it should be.)
The second part of this is the added functionality. One of the promises of WinFS was the ability to tag and relationally add content to files and file listings. Again, this does not offer 'enough' of an edge compared to the current FS technologies. Most of these features are already supported in NTFS, so you can add tagging, and additional fields of information to the files stored on an NTFS volume, basically providing the same features as adding new fields as a database FS would offer.
The only portion that is somewhat left behind in current technology that WinFS would have provided is the 'relational' nature of items in the FS. But again, the database indexing engine that is used for searching can also provide a certain level of these relational aspects to the file and contents.
So when you look at just these basic issues, you can see why in the end MS pulled WinFS as it exists today, and instead has put the functionality of WinFS in the current technologies, as you find in Vista already. (Fast search, relations between files and file contents, tagging using NTFS, etc.)
It may not be the best PR move for Microsoft in the long run, as people here will have a field day with WinFS being abandoned in its current form as an add-on to NTFS. But if you were Microsoft and could provide 99% of the functionality of WinFS with the database indexing services in Vista (and XP) and do it faster than having to add on a new WinFS layer to NTFS, they why would you progress with a product that isn't going to offer what they can already offer with the current technologies.
If computing power was on par with 1995, then something like WinFS would have more viability as Hard Drives and Processors could more efficiently do all that Vista is doing in a Database structured storage. However today, the overhead of doing this outside a database store is fairly non-existent.
On to the second reason, which is business. Implementing localized database stores for files and documents and keeping these in sync with corporate stores is a rather big hurdle when you consider that businesses are not average Joe users and have tons of applications and infrastructure to coordinate Files spread across networks that are outside of existing MS technologies. WinFS would break many business tools and models rather badly.
As for WinFS and Database FS concepts being 'vaporware' or dead, simply is a myth for the MS haters
I do. Ubuntu is certainly making impact even with good Windows fans. Also they like the free (as in beer) part, and wonder when it will overthrow Windows as popular OS.
I am interested to setup Ubuntu support operations in India. Any leads will be very helpful.
Linux does have a ton of research-level development on it. You can do amazing things that you can't do anywhere else. Unfortunately the history of UNIX weighs in very heavily in almost all OS development. The fact is that the problems with UNIX are obscured by how horrendous Windows is. Think how much we could really move forward if we were to take some fresh ideas like Plan9. Unfortunately the software economy is too mature for a cutting-edge research OS to be able to get a critical mass of developers. No one wants to write software for a new OS when there's already so much open source out there for Linux/UNIX. If you could get paid to do pure research it would be pretty fun though.
I never had any problem surfing porn with my Linux box, and it shouldn't be a threat for you...
"Windows has the most problems simply because it is the most popular, and the biggest target for malware."
In all fairness there is more to it than that. The basic design kind of sucks from a security point of view. ActiveX is a security nightmare and there are many other problems as well. Not the least of which is the result of Microsoft s decision to integrate IE so even if you're not using it you're well.... using it...
Security has been Microsoft's top priority for how long now? They simply can't secure their OS.
I agree that no OS is completely secure. There is little protection for users who install questionable software but let's be honest, Windows has had MORE than its fair share of security problems.
The race isn't always to the swift... but that's the way to bet!
The problem is most likely that they couldn't figure out exactly what the product would do or who would use it. I think it's sort of like "That's great, but what do we do with it?" kinds of things. It started out as mostly a way to index content and metadata, but then you have the problem of how to get data into it. So it's better to make it an object store; since it knows about the objects and they have schemata, it's easy to understand the data and index it. But then you have the problem of how to access the objects.
By now, it has sort of evolved into a object-relational manager, which doesn't really belong as a separate product now. It makes more sense to integrate it into a database engine and/or data access frontends.
If you look at ADO.NET 3.0 it actually makes some sense. The LINQ features of the next C#/VB allow you to write queries directly in your language (instead of as strings containing SQL or XPath or whatever), and get objects as results. Right now, though, you can only get regular datatypes (string, int, etc.) from relational or XML data stores; only in-memory data structures like arrays and hash tables can actually return objects. The next version of ADO.NET (3.0) can return actual objects.
What's needed now is some sort of data storage engine that stores objects in a native form that can be returned by LINQ queries. Well, it turns out that WinFS is more-or-less the perfect thing to use here.
Now it starts to make sense that WinFS doesn't really belong as a stand-alone product, but as parts of other products like ADO.NET and SQL Server.
And look at Reiser4, for example. It's got great ways to store data, but any application that uses it is only going to work on that filesystem and any files created by that applicaton can't be moved to any other filesystem or even over standard network protocols. Additionally, it requires kernel integration, so is only available for Linux. While there are certainly uses for it, and application that requires Reiser4 is going to only work on Linux machines with users who have Reiser4 filesystems, and aren't using remote data. No application with those requirements is ever going to become very popular.
I mean, it would be cool to have all of my data in some advanced database-like storage thingy, but ultimately I don't want to have to reformat my drive for a new filesystem and I want to be able to access my files with FTP or send them as email attachments. This means that any application that uses something fancy like WinFS has to use it just as a data store, so it makes sense that it would ship with data access technologies.
dom
I am reasonably handy and so I prefer a car that I can tinker with myself. Sure sure modern cars are a lot more comfortable but they are a black box. Fine when everything works but when it doesn't you got a ton of scrap iron until some guy comes along with a laptop and charges you a fortune for replacing a part rather then fixing it.
It is not that my linux desktop runs better but rather that when something goes wrong I can fix it. Recently my fileserver after a much delayed upgrade from 2.4 to 2.6 (finally got rid of a promise software raid card that is unsupported in 2.6) I had a huge problem with reading some partitions. It stumped me for about a day of trying everything. In hindsight I think what happened was that extended attributes had been enabled without full support making all the files and directories on some partitions take bad values marking them as unreadable/mountable. In the end the fix was simple, set all the extended attributes to nothing with reiserfs tools.
This wasn't however easy to figure out. On several forums nobody knew what the problem was and I had to figure it out myself in the end although there were hints. Each time I was stumped there was still a tool to try, a parameter to look at, something to list so I could google it. Over the day I got more and more details out of the affected partition until I finally realized that the partitions that did not work had these exteneded attributes and then figuring that this was the cause.
A 8+ hour problem but solved.
Now my windows machine, one other task I use it for is reading fan-translated manga. cDisplay is a good tool for this (since been replaced by qcomicbook on linux) and I wanted it to appear in the right click context for directories so I could use it to also browse manga that is not zipped/rarred. Easy enough except it totally upset windows so I deleted it again and now windows opens a search window instead of opening the directory in explorer.
I no longer can edit the file associations and windows has been borked like this for months. I can't examine anything, there is no logging. Google returns nothing.
The problem is far more trivial. I can still right click to explore and by now I gotten used to it. The linux problem meant I could not mount several HD's causing me to fear I had lost close to a terrabyte of data.
But the linux problem I could fix, the windows problem I can not.
Why? Because windows is easier to use then Linux. Yes this sounds contradictive but think of it like this. A car with a handcrank is far less userfriendly then a car with an electric starter. But the handcrank won't be drained by leaving the lights on.
If linux becomes more userfriendly it will also loose that what makes it linux.
Take the constant complaint about linux using clear text configurtion files. These are indeed for the novice a bitch to maintain BUT they are what makes it possible to boot your corrupted system from a floppy and then just hand edit the system back into working order. Even the boot system (lilo) adds complexity for power. Am I the only one who sometimes forgets to replace hda with sda on their one machine with scsi causing lilo to be unable to find the root partition? The solution, give it as a boot parameter.
But this doesn't fit with a nice simple graphical screen with a nice animation showing that activity is going on. Hell on windows you would not even be informed that a critical drive can't be found. Windows just hangs. Easier when it works, useless when it doesn't.
Linux can be technically superior to windows. Some would already claim it is. It can never beat windows in userfriendlyness because if it did so it would no longer be linux but just a windows clone with all the problems that windows has. Lets not forget that a fair amount of windows famed insecurity
MMO Quests are like orgasms:
You may solo them, I prefer them in a group.
I think that calling Windows users scared grandmas and receptionists is completely missing the point. Most people fall into the same category as "scared grandmas and receptionists", whether they'd like to admit it or not. Hell, I think *I* fall into that category. I like things that are consistent, even if they offer me a little less in terms of customization. Every Linux distribution is different. They all have different default window managers. Different single/double-click behavior for a variety of tasks. Different file managers. Different help systems. Different installation procedures. Different methods for managing administrative tasks. Different bundled libraries. Different bundled applications. Different ways to change the desktop resolution. Different support for different hardware. I could go on, but I think you get my point.
I will be switching off Windows when XP becomes obsolete enough to be a hassle. Perhaps by then Linux will be in better shape as a whole. For now, I will just keep using Linux for my file servers since I think that's the only role it fills well right now.
-William Brendel
First of all, with any other OS you can surf all the porn you want without problems like that. Second, you obviously know nothing about computing history. If the reason Windows has so many malware problems is solely due to it's popularity, where are all the Unix viruses? Unix existed decades before Windows and the entire internet is based around Unix. Until this year, Unix always outsold Windows as servers -- and all the *nix OSes combined still outsold Windows. If Unix was as vulnerable to malware as Windows there wouldn't be an internet; it would have been impossible to maintain because all of the servers would have been infected with loads of malware despite anitivirus. The reason that there aren't so many malware problems on *nix OSes isn't due to lack of popularity, it's because they are designed in such a way that it's very hard to write a program that can even infect them, let alone affect them. I see this "every other OS would be just as vulnerable as Windows if they were as popular" every other day. I wish to God people would learn a little something about what they're talking about before continuing to spout this ignorant nonsense.
I dream of a better world... one in which chickens can cross roads without their motives being questioned.
Amen. Until I started playing with Plan 9 I never realized how silly some aspects of modern Unix systems really are. If you want to see a manpage you use "man", but because it has to run in a terminal emulator it needs a full-featured pager, and IIRC on some systems it even re-generates the pages from *roff source. The hideous complexity of autoconf. X11.
I just started using Plan 9 about a week ago on an occasional basis. Though I can adjust to the system, and admire the elegance that allows, for example, rio to be run within rio, ideas like process-specific file heirarchies and the lack of cruft, you have to realize that the developers of Plan 9 have it easy: they get to make a system for people that will learn to use it.
I wouldn't mind seeing how well Plan 9 would deal with having the system adjust to the user a bit. My first idea is to adapt vi to Plan 9; my goal would be not a straight port, but something that incorporates the familiar commands and modal nature of vi with the text-editing support that Plan 9 gives for free to graphical programs. Sam and Acme are fine editors, but I want to find out if a totally different editor can be written in a reasonable way on the system. Actually, I think a Plan 9 implementation should be cleaner than a Unix one, simply because terminal emulators are so damn wierd to interface with. But that may fall apart when I actually learn how to code on the system.
Another similar issue is in window management: you can use nested rio instances to elegantly get similar functionality to Unix-style multiple desktops (and you can do so much more, too), but it's just not as quick to switch between them using the desktop menus in rio as it is with a simple keystroke in fvwm. The basic question is, could you combine the adaptability of Unix (and particularly Linux) with the elegance of Plan 9? That would be a great environment to study *and* use.
The manager would have said, "good enough, now drop everything and do this super-urgent project before the deadline next month".
Whether or not you are willing to accept it, I suspect there are as many bad managers out there as there are lazy programmers.
http://outcampaign.org/
To add insult to injury once I installed xfs I copied some of the data back from a Win2K box that never lost a byte in the 5 years it's been running 24/7.
I am saddened to say it but it's going to be a very long time before I use Reiser fs again. I'd rather spend more money on faster hardware if I want more speed rather than have to deal with data loss.
Network effects economics. It is a lot less useful to society if it is not used. Filesystems are important, but nobody will use an OS just because of the FS.