Slashdot Mirror
Longhorn to be Released in 2006, Sans WinFS
skillio writes "Everyone's favorite OS maven, Bill Gates, announced a release date for Longhorn on Friday. He confirms what many had suspected - Microsoft will attempt to complete this release in calendar year 2006. The most notable element of this announcement was Gates' admission that WinFS, Microsoft's next-generation file system, would not be complete in time for this release - surprising, since this was the most hyped component of the next iteration of Windows."
Actually, it might be a blessing. The pressure on IT to roll out new versions puts a real burden on us. We just got XP and 2003 server rolled out everywhere and I have a feeling we are *way* ahead of most other places.
He actually said 12006. Quite a few service packs away.
I doubt it if they are going to be putting it out in 2 years. So this is basically going to be Windows XP with a new UI, Avalon the new DirectX, Indigo a program "to allow software and services to work across networks and different devices." and some new programming tool WinFX that supports both XP and Longhorns UI.
Nothing special.
-----
Yep another Free IPods Link
hmm Be eningeers did not need several years to come up with a similar filesystem..what is taking MS so long?
Don't Tread on OpenSource
Pretty soon Gates will come out and say that the newly designed Kernel is not going to be complete, and they'll be selling XP Longhorn Edition. This is almost as bad as ID.
gShares.net
-------
artlu.net
It's not as bad as "Mozilla", which sounds like a kids' cartoon or breakfast cereal... especially when you consider "Mozilla" is the final name, and Longhorn is just a development code name.
I don't respond to AC's.
Microsoft has been doing this for too long for my taste now. Promising all remarkable and amazing things that keep us on our toes and when the product hits the shelves it's only ever so slightly different from its predecessor.
An interesting dilemma for Microsoft- Do they continue their typical practice of changing technology to keep people upgrading, or do they recognize that this one might actually delay (or even reduce) revenues and not change current compatibilities?
I agree with you - definately bloatware. But when has that stopped Microsoft?
I might not be a wit, but at least I am more than half way there.
Call me a luddite if you will, but for the life of me, I cannot see the reason for a new filesystem. I'm all for metadata and so forth, but why rip up the tried and tested file and directory structure for this magical, cure all, search based filesystem. Search works well in Google because web pages are connect. My files aren't connected, so I don't think search on my filesystem will ever be half as good as search on the web.
As far as I can tell, MS (and GNOME 2.6 it would seem), seem to envision a filesystem where every file is simply dumped to one / or c:\ directory and this uber search finds all the files I'll ever need for me? Is this a joke? In this senario, ~50% of all the metadata will be the same for every file. I made it, with my privilages, with my settings etc... . After a while, even the simplest of searches will bring back a dozen matches. I can't see this working.
The reason given for this is novice users, who don't know where to put their files. they rely on their default program settings and just dump their files anywhere and then complain when they cannot find them. Fair enough, they are novices, but essentially hey are keeping a messy hard disc. WinFS would help these people only in the initial stages. As soon as too many files named 'Picture of Aunt Tilly' are present, the system will fall on its ass.
Metadata/Search based filesystems are based on the assumption that users do not know where their files are. I do, you do and for those who don't, no amount of programming wizardry is going to help them in the long run. Ultimatly they will have to learn how to organise their files, just like they have to learn to type,use the mouse and browse the web. And in reality, most people do eventually learn how to organise their files, if they use computers enough. And if they don't, our regular searches will be of use to them with only minor improvements. It's tough, but consider the search results that 'Find my Accounts for Acme Corp. for the third quarter of last year' brings up on the shared drive for even a medium sized accounting department after only a year.
Give me nested directories 30 levels deep!! And no spatial browsing please!
I did wast an entry in my journal on this stuff. maybe now someone will read it?
May the Maths Be with you!
You don't seriously think that Microsoft had any intention of shipping WinFS with Longhorn did you? That's one of their standard reasons why you shouldn't switch to an alternative operating system - because [x] fancy feature is coming out Real Soon Now. Once they've held onto you long enough to get over the hype surrounding their competitors, and once the release date looms nearer, they drop the pretense that they are going to ship with the fancy new feature. WinFS is vapourware.
"In other cases, vaporware is announced by companies in order to damage the development or marketability of more real products by competitors"
Remember when Windows 95 was supposed to be uncrashable because of 32-bit memory protection? Did Windows 95 actually deliver on that promise? Did the half-dozen or so operating systems that Microsoft released after Windows 95 deliver on that promise? How long do you realistically think it will take them to deliver WinFS?
Microsoft, and in particular Bill Gates, have stated numerous times that Longhorn is the most expensive and time intensive project MS has embarked on and would be as complicated as the Apollo space program.
Would you want to fly into space on a shuttle that runs Windows?
"Uhhh Houston, what the hell is a pagefault in kernel32.386?"
I am no MS lover at all, and WinFS is just one of multiple implementations of the concept underlying it, but I think the point that it is getting constantly bashed is because it is by Microsoft. Either that, or our tech savvy slashdot crowd just developed into a bunch of retards.
Why can't you just accept new technology and first take a look at it? It's not that the good old tree like filesystems are the best ones. They do indeed have problems. Have you ever read about data modelling? Have you studied some papers about that? Obviously not, otherwise you would know what the current state of data representation is. Hierarchical systems are indeed some of the worst if it comes to expressive semantics. There are new concepts outthere, trying to solve those problems, and, yes, they come along with a new paradigm.
If you are too lazy to learn and accept new paradigms, you don't deserve to be a geek.
Then why is it called "FS" ??????????
... I was actually interested to see what WinFS would be like. From what I understand, it is supposed to be different from the traditional heirarchical filesystem. If the filesystem worked like a database, then folders would be the equivalent of tables and SQL statement results, if it actually used folders.
I know that Apple's upcoming release of Spotlight with OS X "Tiger" is probably what WinFS would appear to be like from the GUI perspective, but its underlying filesystem is still heirarchical since they're not changing it. I presume it would work similar to the way iTunes displays libraries and playlists like a database, yet stores the actual files in a heirarchical arrangement only visible to a user who manually browses the filesystem. Data displayed from WinFS would be a direct representation, rather than indirect one of data stored heirarchically.
For that point, I wouldnt like to fly into space on a shuttle that runs Linux. Neither Linux or Windows are stable enough to be used in that circumstance, you want a proven realtime OS to handle it.
Marketing. WinFS sounds sexy. It will get the Windows geeks talking about the revolutionary never-before-seen features, and when the Windows geeks are lusting over the next version of Windows, they 'ain't convertin' to Linux.
It's the old bait and switch. Now that WinFS has served its purpose, it's being moved back to the _next_ version of Windows after Longhorn. But don't worry, Longhorn+1 will be the best version of Windows EVER! It has this revolutionary new filesystem, WinFS. It will also be faster, easier to use, more compatible and more secure! Why risk changing to another operating system when the next version of Windows will be everything you have ever wanted, AND MORE?
Isn't all information potential file data? Is Microsoft really doing something different than has been done before?
The article also states "WinFS uses a direct acyclic graph of items (DAG)."
The math goes back to the 1970's, as referenced by MathWorld Old math can be used in new ways. Is his a new way when it's used in the FS that Microsoft is attempting?
The articles also says: " the WinFS data model provides the following concepts to describe data structures and organizations: * Types and subtypes. * Properties and fields. * Relationships. * Constraints. * Extensibility. "
Does the new Reiser4 file system support any of these concepts? -- Is WinFS really as new and exciting as the marketing and media says it is?
Thanks.
Unfortuantely most companies will say, "your job [apt-get] is now handled automatically?! You're gone! Wee, we saved 0.025% of our IT budget!" Six months later when the connection fails because some idiot messed up DNS, they have no one left who can figure out how to fix it, and no one who can do it manually...
Yeah, right.
The one area that I would say Microsoft is at a disadvantage is the very rabid and outspoken communities that other OSes have. Linux has an advantage due to sheer numbers of how many programmers work on various components at any one time. Apple has users that are creating some very cool software (Konfabulator) and then Apple takes that idea and runs with it (yes, i know the story behind the new widget system). Microsoft is very seperated from the user base and what it's users want/need. Apple listens to its users, and Linux is the users. I would say there is a clear advantage there.
This smells a lot like the failed WindowsME. As i recall it was supposed to be the next grand thing in computing. A step as big as the one from Win3.1 to 95. It ended up as a mere add-on for Windows98 with more crashes-prone features than you can point a "shrug and reboot" attitude at!
:-)
If they keep droping ground breaking feature like that, in 2006 they'll be releasing a "Windows XP longhorn edition"!
-- If you actually say LOL instead of laughing, maybe it's time to go outside! --
Either that, or our tech savvy slashdot crowd just developed into a bunch of retards.
The retards overran the techies here long ago. There's hardly any geeks left that think technology is cool or interesting by itself. Slashdot has become a bunch of MP3 Pirates who got bitched by Windows 98 and hate everything new because it might have a patent on it.
WinFS is an interesting, bold, and novel take on a file system
Bullcrap! It's been done before. One of the other replies to this post has a lot of links to seminal works on things like this.
Microsoft has higher demands on it
Now this I agree with. They do have higher demands on it. But they have nothing to do with functionality.
From one of the articles:
If users are joining the network remotely, for example, a network administrator will be able to ensure the system has up-to-date virus protection and patches, and if it doesn't the administrator will be able to kick the user off the network, said Muglia.
Wanna bet that earlier versions of Windows will NOT be allowed to participate, regardless of the administarator's wishes? This will be the final nail in the coffin of win98, winNT and any other OS that Microsoft wants to deem obsolete (obsolete = unregistered, uncontrolled by Microsoft).
From one of the articles:
Allchin said the moves will keep the company from having to scale back WinFS and will enable Microsoft to implement it on PCs and servers at the same time, something internal Microsoft testers said was important.
Has to be rolled out on both the clients and servers, eh? Bye, bye any kind of mixed network! You can be sure that this will not play with Linux clients or servers in any way! Companies will soon have a choice: all Microsoft or NO Microsoft; nothing in-between.
Could it be that they want to adapt their applications to use the new features before they release it?
Well, yes! Given the comments I made above, I'll bet that virtually every application will need to be rewritten, not just to use Longhorn's features, but to run on Longhorn at all. This is intended to be an upgrade bonanza for Microsoft; new OS, new Office suite, new Front Page, new Visio, etc, etc.
I, for one, am gonna pass.
But they didn't have to integrate it with the legacy Windows code base. Apple did OS X more or less from scratch. Windows never had a foundation for this type of thing.
.sig: Open Source, Open Mind
Very interesting analysis. From what you're projecting, it's not too far a leap to a subscription based model, which is where MS has wanted to head all along. (Remember the hue and cry when they announced moves in this direction before?)
It's also somewhat similar to the way Apple rolls out OS X updates. (That has also caused consternation among a small vocal minority of OS X users that don't want to pay for upgrades but don't want to be "left behind".)
Red Hat, Sun, Covalent, and others are embracing subscription models.
So it wouldn't surprise me to see MS try and put a subscription model under the radar. And it might not be such a bad thing, now that MS is facing pricing pressure from OSS.
It's not offtopic, dumbass. It's orthogonal.