Slashdot Mirror
WinFS' Spot on Back Burner Nothing New
osViews.com writes "Charles Arthur of Independant.co.uk has an interesting editorial which analyzes Microsoft's recently postponed 'WinFS,' the file system that Microsoft had been planning to implement in Longhorn. His editorial reminds us that this technology, previously referred to as the 'NT Object Filing System' was intended for a previous version of one of Microsoft's operating system's code named 'Cairo.' Microsoft first spoke of the 'NT Object Filing System' in 1992 and scheduled a beta release in 1996 and then a full release in 1997. But limitations cause it to continue being delayed."
HA HA!
This is my sig. There are many like it but this one is mine.
They should just license ReiserFS 4 and make everyone happy. Of course, then it'd probably be too easy for people to switch to another OS. Whoops.
...is a solution in search of a problem?
It reminds me of the old saying
Or for that matter the ORIGINAL goal of the Gnu project?
What's your point here? Why are you trying to bash Microsoft just because they decided to delay or abandon something?
Best Buy can have you arrested
duke nukem forever and winfs are fighting for the throne... of... stupid delays
http://ipod.fresh27.net/
To how many of Mircrosoft's MILLIONS of consumers, is a filesystem like 'WinFS' (theoretically) a feature to be desired?
Most people I know want eye candy, and things to work as they're used too.
Microsoft doesn't _need_ WinFS, therefore it's not a prime concern
Error 407 - No creative sig found
People COULD just use naming conventions and name their files according to the content. But I guess that's just too hard.
Vaporware. Microsoft is so famous for it, they are referenced in the definition.
Is there any project for a similiar file system in linux?
The idea itself is a good one.
Before you know it Windoze will be releasing things like 'windows classic' to make up for the time delays.
Windoze Classic- Remember dot matrix printers and phosphore monitors? We do! Now you can relive the 'good old days' with Windows Classic! For only $500 you can run the original Windoze 3.1 with three, that's right, three shiney new graphics! Act now and we'll throw in a free 3 month subscription to AOL 4.0!
Microsoft, I must applaud you. By delaying the best features of your operating system, and assuming you continue to do so in future versions of Windows, you wiil, one day, have the best OS to have never been developed.
From what I know of WinFS, it really won't be all that important anyway. It is supposed to provide a way for all files to be treated the same by the OS (roughly) right? Thus making it easier for users to search, browse, or otherwise find these files?
Well, I don't know all of the juicy details of WinFS but I have played with the new Longhorn build. The search tool that is in the Alpha release (MSDN) is much improved over the current WinXP search. It was pretty cool, although some of it can be chalked up to eye candy. It still had a certain ease of use to it.
I doubt WinFS will ever be complete, personally. But I am sure some of the innovation and development benefits will still reach us as consumers. I know where I work, we spend time doing things the customers will never see. But they will still reap many of the benefits.
So, in other words, would this mean that Microsoft has had this technology for at least 10+ years, yet are still working on it? Or perhaps it's because it may actually be useful that they've had to postpone it.
I'm working on a object file system right now, and it's really not easy.
It's a simple concept:
Store on a standard journaled b-tree (or similar) filesystem the binary data, and store in a database all sorts of meta-information about the data. Also if you want, store a reverse index of the textual info and maybe another 'index' of image features if it's an image.
Then if you want to get anything, no need to go through the filesystem's tree, you can hit the DB indexes and get info instantly.
The real problem is keeping all of this in synch, with almost flawless atomic operations. (of course it's pretty much impossible to be flawlessly atomic, but one should come as close as the current journaled filesystems are).
So if you're using 2 components, let's say, a filesystem and a SQL database, then you need to open a SQL transaction, do your inserts/updates/deletes, then do the filesystem operation, then do the SQL transaction commit. If anything fails, you can revert the SQL modifications and everything goes back to normal. But if the filesystem has problems, then you can't keep the damn DB synchronized, and at some point you'll have to resynch both.
On 100k files, no problem. On 200MM files (what I'm aiming for), you're pretty much screwed. Then you have to start thinking of a self-healing system with a constantly-running checker that must ensure that it's very resource-efficient, etc...
It's just a huge problem. Supposedly Apple is solving this by Q1 2005, but I wouldn't be surprised if we see a massive increase in filesystem corruption bugs for a while on OS X (unless the DB indexing piece is just that, an indexer that runs x times a day and isn't atomically joined to the filesystem operations).
IIRC, "Cairo" was what became NT 4... "Chicago" was Win95. Then there was the OS "Pink" by Taligent (IBM + Apple), but that never surfaced... And then there was BeOS and the BeBox... We can't forget the BeBox! It was... the precious. :^)
Slashdot's first reaction to VMware
Windows Bob NT :)
This is my sig. There are many like it but this one is mine.
Storage would be one example. I bet there are others.
Trust the Computer. The Computer is your friend.
Already posted my comments about Cairo here last time we talked about Office.
Moderators, please: +1, Psychic
And that's why it's taking so long. Accessing filesystems as SQL data has always been a dream of anyone who has had many files. They just never knew about it.
WinFS is the 'real' solution IMO to all things like iTunes playlist managers, and expensive Content Management Systems yadi yada.
Sure, no consumer is expected to actually use SQL statements, but that doesn't mean that user mode programs should *implement* SQL features. User mode programs should only be the 'translation' layer between the user's point and click GUI, and the OS' internal implementation of the db. Surely, anyone can see that collecting meta data from the file system, and duplicating it in usermode so that you can have search capabilities on it is wasteful.
This article wasn't news to me, I've actually been waiting for this damn WinFS since just about 1996... And by god, is it ever turning into Duke Nukem Forever, but you know what, it's such a cool feature that I still can't wait for it to come out... (figuratively speaking)
Maybe the real name of it in internal MS memos is in fact VWFS (VaporWare FileSystem) or will be the next codename for the promised object oriented database mumblejumbo fs for windows 2010.
Wow. Someone has a unique sense of what trolling is...heh.
This is my sig. There are many like it but this one is mine.
I'll pull out the link again: Storage (a GNOME project) uses some nice algorithms to let you look up anything from '1960s music' or 'films directed by Francis Ford Coppola' to 'pdfs from joe'. All in natural language and over a wide range of formats, although evidently it's still a work in progress.
Let's put this in perspective. In '92 MS was looking at the Sybase source code and thinking about building a new filesystem around a database engine. Chicago AKA Win95 was almost out the door and it seemed reasonable to shoehorn this into Cairo (NT4). They were absolutely the dominant and fastest growing player.
I commented to a collegue in '93 (paraphrasing Robert Heinlein) that I did business with MS for the same reason I obeyed Newton's laws.
What happened around 1995? The internet became a commercial entity. Suddenly, MS needed to provide new applications (like IIS, IE, Outlook Express, an SMTP aware Exchange server, etc.) not just dork with cool OS technologies. A few years later, they are comfortable again after playing catch-up and start thinking about filesystems again, this time in "Longhorn". Again, they started talking about the capability two OS releases into the future.
However, this isn't a feature that is going to drive sales. MS needs to keep developers of home and office apps happy so they develop yet another new graphics system to replace DirectX. The perception of Windows security has never been lower and is starting to affect sales. IIS is losing ground again to Apache/Linux.
It's time to focus on revenue streams again and the revolutionary, expensive, difficult-to-build features get axed. It's probably not a bad idea. Think about the problems they've had with MS-SQL and ask yourself if you want a similar technology built into every teenager's game and grandmother's email box.
Oh, no! Has Slashdot sunk to having sweepstakes in the headlines?
Hmm, maybe I am missing something</luser>
Don't waste your vote! Vote for whoever you want, unless you live in a swing state it won't matter anyways
if I had any modpoints left I'd mod this Insightful
605413? Yes, it's a prime.
Anybody use the search tool referenced in the article (Blinkx)?
Does it contain malware/spyware?
http://news.ft.com/cms/s/474450a6-01ce-11d9-8273-0 0000e2511c8.html
:)
Are you Americans getting all this, or is it all covered up!
ReiserFS version 4 is a database at heart. Its basic structure is just a table of FileName | Binary but it also contains a modular system where it can be expanded for many uses. There is a lot of talk of including meta data in ReiserFS for such a system.
http://www.namesys.com/whitepaper.html
If someone was to create a basic version of WinFS... good enough for the average Joe to think that they might not need the fancy features of WinFS, and was to make it readily available (open sourced or at least really cheep) and easily installed people might be even LESS inclined to upgrade to Win2008 or when-ever MS releases this technology. Add on top of this an easy integration of search technologies for OS X and Linux (desktop) computers and WinFS might be less desirable.
Take a gander over here.
Basically same article. And yup, d-e-l-a-y-e-d.
How many people read "NT Object Filing System" as "NT Flying Object System"? C'mon now, be honest.
God I must have dyslexia.
I can only say one thing: I miss BFS. Dominic Giampalo, where art thou?
Microsoft "solved" this problem for all intents and purposes by having every program save its files in the "My Documents" folder or a subfolder therein, and allowing for filenames that can be long and have spaces.
Sometimes I feel like Microsoft is rearranging the deck chairs while the ship is sinking. Anyone remember that cool "Tripping the Rift" movie? The ship is falling to pieces and the onboard repair robot repaired the machine that makes ice cubes first. The outraged captain smacked it with wrench and screamed "We're floating in space you decide to fix the stupid ice machine? Get to work on the fucking hyperdrive!!!"
Microsoft need a similar push.
Yeah. They also mentioned vaporware's early Atari history. It was really Atari that brought vaporware to the masses.
Anybody remember the Graduate keyboard for the 2600? How about the Mindlink?
The Atari 2700 with ergonomic wireless joysticks was ready for production then was killed. Let's see...what else? The 7800 keyboard was fully developed then killed. An advanced "Amy" soundchip for the 8-bit computers....yep! Oh yeah and then there was one of my favorites. They had an expansion cage ready to go that would let you add cards to the XL line of machines just like the Apple II. Come to think it, it was only a few odd ball third party devs that made use of the "Parallel Bus Interface" that Atari promised that soooo many nifty things were going to connect to.
Yeah, Atari got me salivating a few times back in the day before I finally learned my lesson.
The big "My documents" bag for everything is just a throwback to the dinossaurs time, before there was things like hierarchical filesystems.
And the search needs to be for more than just file names. OS X "Tiger" allows, for example, to find a reference to a name inside a PDF file.
http://en.wikipedia.org/wiki/Winfs/
If this is (was) supposed to be a key part of the OS, I don't know what they're thinking up in Redmond.
I know you can use lock-free algorithms to iterate through collections being concurrently modified. I haven't tried it with an indexable collection but you'd want something like that to avoid having rebuild the indices everytime you modify a table.
is laughing his a** off reading your post.
If you can't see humour in this one, don't bother to moderate.
It's better to be the foot on the boot than the face on the pavement. ~~ tkx Kadin2048
IIRC "NT Object Filing System != WinFS"
WinFS is supposed to be based on SQL Server, when NTOFS was announced, MicroSoft hadn't yet acquired SQL Server.
I thought NTOFS was what morphed into the fast-find thingie that shipped with Office.
I don't need no instructions to know how to rock!!!!
I want the finish to SwordQuest. Whatever happened to the jeweled sword someone was supposed to win, anyway?
I recently interviewed at Microsoft and one of the teams I met was the Windows team.. I can assure you it's NOT vaporware.
Yes, but SQL Server is the only Microsoft product that actually works in a half decent way (Don't laugh!)
"Who are in control, they are not in control of anything - they don't even control themselves!" - Glen Beck
WinFS is what they used to call Cairo?
Man, I remember that hype from the early nineties. I thought it had been quietly abandoned.
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
I agree the idea is good. /. did a story on a KDE version a day or two ago, its not the same, but it is a step in the right direction. It seemed to want developers, so if you wanna help out, hunt out the story
Of ALL the various computer problems I need to take care of for my clients on a daily basis, their ability to locate their "lost" files is NOT one of them.
What spolight does is let you get to the file in question really fast.
Already, I use it instead of the Finder to open apps.
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
ah, the old OS/2 can't do this filesystem from "Cairo".
it's lying in wait, waiting to lung out and kill penguins at just the right time...look out!
it's coming soon. really. no really. come on stop laughing!!!! it's going to come out any day now. yeah, in longhorn, that's the ticket.
Have you even seen the teletubby world of Windows XP?
and this is news?
e to the pi i plus one equals zero
Until the system can extract reasonable and meaningful metadata about the contents of files and documents, then it will still always be up to the user to do the bulk of this.
For better or worse, the most (only) "meta" tag 99.5% people use is the file name. Word has a very flexible metadata ability, but it is never used. It was turned on by default in Word 97, but probably quickly turned off by all users. Same for Excel, PPT, etc. This is one of the things that the Office Search (Find Office Files quickly!) keeps indexed. But of course, that is usually quickly turned off, also, because no one really uses it.
It's a user-space problem, not a system problem.
They should just instead include 'awk' or 'perl' (something a little more sophisticated than 'find -i') on the system, with some sort of natural language-to-regexp converter front-end for them.
No, "My Documents" is the windows quasi-equivalent to $HOME/docs.
Too bad MS needed to assume that computers only have one hard drive with one partition. While it is possible to hack one's registry to sort of do things the way Unix/Linux does, including creating symbolic links to different partitions rather invisibly, they just do what they can to make this a non-feature, unfortunately.
NT/XP system mamangement would be SOOOOOO much easier if the OS could be protected on a single partition, and ALL applications and their libraries be stored on another partition. Why? Since you probably can't move the Registry, a corrupt registry would, in an ideal world, simply necessitate recovering the Windows partition, either from backup or reinstallation. Then, an installation log on the applications partition could be consulted, which would effectively reinstall applications w/o having to find the original media, etc. Your documents (My Documents), spreadsheets, etc., would be stored on yet another partition (actually, C:\winnt\profiles), with the possibility of migrating user directories as needed, just like a Linux/Unix system.
But, no. $50 billion in cash, and arguably a good chunk of most of the best programmers and tech writers (PPT slide show developers...) in the world, it's not sexy or cool enough. Instead, one of the uber-geeks there comes up with something else, and gets SteveB and BillG to buy off on it.
Heck, they could probably even buy iFS from Oracle for some chump change and make 2006.
But iFS sort of has been a big flop, hasn't it, and it's NIH anyways.
That's not a filesystem, but a layer on top of it that tracks data in an SQL database. Someone will correct me if I'm wrong but it's kind of like locatedb with more features and data to track, that runs a cron job every once in a while.
This is not the same as a filesystem itself being implemented in an easily searchable fashion, getting rid of the static directory tree structures altogether, or just having it as an option for backwards compatibility as one of the properties.
this article is remarkably similar in many respects to the recent one of Joe Barr at linuxworld. But he makes a more linuxy point -- linux cannot/should not compete against the non-existent figment of microsoft's imagination.
Well a quick Google found quite a bit of information on NTOFS (with the suggestion "Did you mean: NTFS")
When a company releases promotional material on a new development and then doesn't release the development within say 6 months or a year, we call that "development" vapourware.. so what do we call a technology that hasn't been born after close to 10 years ? MS Extended vapourware ?
This is almost the first thing I do on my own machines, so that when I re-Ghost partitions, none of my documents get hosed.
Win2k finally delivered on the functionality Microsoft told us Win95 would give us. When I finally abandon Win2k I won't be moving to XP, Longhorn, or any other Microsoft OS. If an OpenSource OS doesn't meet my needs by that time I'm going to bite the bullet and switch to Apple.
You do realize WinFS is NTFS + a SQL layer, right?
One correction - filesystems (at least most UNIX filesystems) are not constrained to tree structure; the leaf nodes may have any number of parents, i.e. a file may be in any number of directories simultaneously. (Use the "ln" command). And using ln -s you can practically place a directory in any number of parent directories.
I use this to organize my music collection alphabetically by artist, by genre, and by the date I got the music simultaneously. (I tend to be most interested in music I got recently, because I'm not tired of it yet).
I know people tend to organize files and directories in a tree structure anyways. If you ask me that's because people are happy to maintain the analogy of a physical item that can only be in one place at a time - so what does that mean for WinFS?
"I recently interviewed at Microsoft and one of the teams I met was the Windows team.. I can assure you it's NOT vaporware."
Considering that WinFS has been dropped from Longhorn (soon to be renamed to Shorthorn), then it most assuredly is Vapourware. It is (and it will remain as) Vapourware right up until the point in time that it becomes an actual product that is available to be obtained and used by the great unwashed masses.
"You can't fight in here, this is the war room!"
I can only imagine how many patents must be researched in order to "embed" such a kludge like WinFS into the OS - you know, to avoid infringment.
Today, NTFS has only one primary index - the path/name components. Adding more indecies will significantly reduce performance since all indecies (system and user defined) will need to be updated with each change in a file's properties (i.e. name, times, attributes, etc.).
This will simply not fly, especially with heavy file writters/readers (backup, scanners, HSM, archive and the like).
How many IT admins have turned off their user's stupid, performance sucking volume indexing feature?
-MerkX
Glomming two related services into one blob of unmaintainable code is not necessarily a benefit. A database mapping has the advantage of being able to catalog distributed file systems, including those which don't have any object tag extensions.
The other problem is that it's not uncommon in the database world to spend far more disk indexing complex data for access than it actually takes to store the raw information itself. Do you really want the possibility that your inseperable all-in-one file system is using more space for the equivalent of directory entries than for data itself?
Remember this isn't about special cases like a user too lazy to sort their home directory or documents folder, but applying that overhead to the entire system. With all the tweaks people do to improve general FS performance and reliability, why would anyone think adding overhead is a good idea unless you need, and I mean need those features?
If you do indeed need those features so badly, why not just buy or use one of dozens of existing document storage and search facilities?
WinFS was just trying to find a way to make people think the two ideas were inextricably bound together and in some way unique to Windows. In truth that honour goes to hundreds of document database and repository products and the long-toothed AS400 (or so my cohorts tell me that work on the platform.)
I do not fail; I succeed at finding out what does not work.
Every filesystem is a database at heart. They already contain other attributes like permissions, create and modify date etc. The place to store this stuff is in the FS because the database is already there. All you need to do is add some more stuff like extended description, a few topic reference fields, and and slap of a query engine on it. The query engine does not need to be real complex either. You can get away with little or no formating/sorting/grouping support as the user space app which performs the query should take care of that. All you need is basic bool logic and string comparision. Most of this code already exists out there under a free license, I am not saying it would be a copy past job but there are examples of required algorithms which developers can look at safely, without running afowl of and IP.
The one tough thing WINFS aims to do that would be simple in user space is it hopes to be able to look in files and gleen some atributes form them. This is great if you can hook into some of the libraries form office or adobe et al, it saves you from having to implement parseing for all that stuff. I am not quite sure how you solve that one at the FS level. I just fear a user space system will get real crufty real fast and break when major changes occur to the files and their real attribes on disk that the DB can't know about. Like if a mount point gets moved or everything is resotored form a tarball and the dates get changed/permissions change a little because someone was careless. I think overall getting the neccecary info form the user when new files are created would be a fair compromise, the only issues is rule one of DATA "crap in crap out".
Then there are all the problems that you mostly have to deal with wether you do it in the FS or as some user space hack/bloatware thing:
Note that file creation would constitute just that you would want/need for efficency archives to contain all that info for the file in them, so the user does not have to enter it. Makefiles and the like would have to be update to do magic and fill in that data for the output files. Then you naturally have to fix all the gui tool kits so their fileIO dialogs support that info, any apps with custom dialogs will need to be patched as will console apps. Some sort of default values would be need for apps that just can't resonably support collecting that info as well. I don't want to have to fill in values everytime I "cat" somethig, I mean to unlink moments later.
I think its clear there are lots of differcult usability problems to solve. Some could probably extend and of the major OSS filesystems to include some extra attributes and add a crude query system, its all a question of what do you really do with it once you have it. I am sure R&D at Microsoft is just as perplexed on that point as I am. I feel sory for them since the marketing dept has been pushing this as the next big thing for almost a decade now, the pressure must be intense.
Repeal the 17th Amendment TODAY! Also Please Read http://www.gnu.org/philosophy/right-to-read.html
Dude... You messed up the TTR quote :)
:)
Chode: What the hell you been fixing these past few days?!?
Gus: The transdigital freon converter.
Chode: And what does that do?
Gus: It makes Ice Cubes.
Chode: Wuh... You mean to tell me, that with all the CRAP that's broken on this ship, you start with the fucking ICE MACHINE?
Gus: Now listen to ME, you fat purple dung pile... As the ship's engineer, I decide what gets fixed first. So if you don't like it, go screw yourself.
Chode: That does it! Come here... You've had this coming for along time [pulls out baseball bat] now I said get over here!
Gus: You wanna piece of me? [Middle finger becomes propane torch] Bring it ON!
Chode: [Smacks Gus with baseball bat]
Gus: Oh, this is fair... [Is beaten down by Chode]
[Sex android Six quietly enters bridge]
Chode: Oh, hello Six! [drops baseball bat]
Six: Captain, you know how fighting with Gus aggrivates your irritable bowel syndrome... Oooh, would ya like a handjob?
FWIW, download here. I love that short film... Especially Six >:P. The neat thing is they made a series of it on SciFi... Don't know when season 2's coming out though.
Go ahead... mod me OT. I still think it's hilarious
True but Reiser4 is available now. Someone just needs to build a front-end into Gnome/KDE.
Other examples of vaporware in Linux:
- integrated NVidia or ATI drivers
This doesn't fit the definition of Vaporware because no one ever claimed it was going to happen. Besides, you have to download the drivers for Windows too.
- working USB 2 of Firewire support
Works for me, I don't know what problem you are having.
- fast boot-up times
25 seconds including init on a 700Mhz machine is fast enough for me.
What alternate reality are you living in?
Time makes more converts than reason
Is there any project for a similiar file system in linux? If it ever comes out I'm sure our Linux systems will be able to read it. And everyone else should begin complaining that Windows won't read EXT3/XFS/Reiser out of the box.
It's a perfect time for being wasted.
A perfect time to watch the stars.
- Burden Brothers, "Beautiful Night"
When I was at college one of the girls I went out with had a step mother who had no ability to organise her own information.
In her rolodex type phone number finder she had several of her friends listed under "H" for "Home number" with a sublist of name and numbers. She had a similar setup for "W" for "work numbers" and "M" for "mobile numbers" with a list of peoples numbers.
Obviously the cards for "H", "W", "M" where quite full as most people where listed there. Other cards where almost empty.
I asked her why she didn't organise people by first names or last names. She looked stunned that at the suggestion.
I would hate to see how this lady organises her computer files, but a search facility no mater how bad would help her alot.
Elivs
--
Sorry about any typoos in my post, Im having a busy day.
BobFS. There I knew you could.
The real intent is to simulate conceptual processing - or at least conceptual organization of data.
Unfortunately it doesn't even come close.
None of this can be done (done well, anyway) without some decent simulation of human conceptual processing.
And since Microsoft chose to piss away their R&D money on a one-time stock prop scheme, I guess they won't be the ones who succeed in doing it. If they even care, which I doubt.
Richard Steven Hack - This sig is TOO GODDAMN SHORT TO DO ANYTHING USEFUL WITH! MORONS!
If I'm reading your post correctly, you're (slightly) mistaken. HKCU can be located anywhere (including on a UNC share if you're clever about hating your users), along with the entire "user" directory tree... none of the "per-user" stuff needs to be tied to a specific box or partition, exactly as you wish.
In a perfect world, the user-side of what you're suggesting can be done without too much headache at all - and in fact is mostly done in larger shops via combinations of roaming profiles, policies and "home directories" in the user props.
Isolating apps is another issue altogether, sadly - most of them are so ill behaved (Flash and Real, for example, sticking shit in the $sysroot tree) that it takes more time to create this "isolation" than it'd be worth - compare the time of your ideal strategy (and the hair pulling that's required due to shithead vendors) versus something as trival as Ghost. From a time-required standpoint, when \Winnt gets hosed... App Isolation loses hands down since I can Ghost in about four minutes.
Cheers,
help me i've cloned myself and can't remember which one I am
As soon as Apple or major Linux distros start running a WinFS'ish-like file system, Microsoft won't bother to foist their code on the world. Why should they? They drag their feet on every project they can until competition forces them to react.
[Microsoft | Bill Gates] != innovator
Ruby on Rails Screencast
If you are asking me, it would be "better" if files just had metadata properties associated with them and their metadata and contents stored in an easily searchable manner. So, I would say it's probably "worse" to have that functionality implemented in your desktop environment, or any higher layer that does not interface directly with a filesystem or is implemented into a filesystem itself.
You can only access a desired file by its directory path (whether one or more), unless you get involved into a lower level filesystem-specific calls. The directory structure is an unnecessary "static" interface to files since files can be classified more efficiently by indexing their own extended metadata and contents.
Well, WinXP is deadly slow already. Every machine I've run that's had XP on it has run like crap basically.
The response time to almost anything OS related in XP is like treacle compared to my home machine with runs (fanfare of trumpets) Win ME. Who would have thunk it ?
Note that this is on decent > 1 GHz machines, whereas my home machine is a lowly Pentium III 550.
If I ever get round to getting a new machine, I guess I may have to keep a windows partition around, but my main OS will be one of the Linux distribs.
-- "It's not stalking if you're married!" My Wife.
Yes. But storage is not a layer on top of a filesystem but rather a service or a feature of a desktop environment. For a true filesystem implementation, there should be a new OS filesystem API.
Wasn't BeOS's BeFS something similar to this?
It was a next generation file system, that afaik, is still superior to many modern filesystems. It even had methods for storing meta data from custom file types (ie- mp3), so you could search for an "artist" field with "Cibo Matto" in it, or whatever.
Also, it used a set block size (1, 2, or 4K) rather than a set # of blocks.
i miss BeOS...... *sniff*
...spike
Ewwwwww, coconut...
maybe you would be interested in This.
Jisho - A Japanese English German Russian French Dictionary for the rest of us.
WinFS, and any "filesystem" structured around the data, rather than the form of the data (eg. files), is more than just "content searching". An index of the content of the data is one metadata type. File event dates (create/update/last-read), access directories, archivablilty, MIME type, compression/encryption, application defaults, and data-specific pointers (packages, components, multidimensional scales, etc) are all even more useful data about the data than just some data contained in the dataset. Especially as human senses operate by association of related data, modelled as database schema relations.
The old "filesystem" leverages human experience with filing cabinets, fast becoming a lost art, into working with computers. It's a 1960s era hierarchical schema, long surpassed by the relational model for expressing human operations on data. Microsoft is so tied to the file metaphor that it can't produce anything but vaporware like "WinFS" (or OLEDB, or all the other pure marketsprach) to replace their legacy data tier. Linux isn't tied to such an albatross. We can get content searching, and all kinds of other human-sensible data operations, when we've moved to a modern data tier, and make Microsoft computing look as archaic as VAX/VMS. Let the good times roll!
--
make install -not war
You make it sound like these are features that Microsoft's cutomers need...
In reaity, these are features that Microsoft needs!
"[Regarding the 'cloud,'] ownership was what made America different than Russia." -- Woz
Although it makes a nice tagline and dig in the ribs for Microsoft -- same delayed technology, different century, yuck, yuck -- the Cairo Object File System (OFS) and WinFS bear no resemblance to one another. Having worked in the Cairo/NT group at the tail end of the former and suffered through uncountable meetings about the goals/architecture/benefits of the latter prior to leaving MS, I can say this with some certainty. Saying they're the same internally or architecturally because both strive(d) to provide the ability to find any document by any properties or content (aka "information at your fingertips"... remember that?) is just vacuous -- you might as well talk about similarities between file-systems that support shell wildcard expansions * and ?.
OFS was about a lot of things, probably too many things. It was designed during the "object wars" and things like copeland and pink and opendoc were in the headlines. Document-centered work was the proposed user paradigm, where structured documents contained nested opaque data from many different applications, and so applications wouldn't need or want to know the difference between a top-level document or a sub-part of a document. This user paradigm did not entirely come to pass, and so an entire file and object-system architecture and shell user-experience premised on it was canned.
That said, a few features from "OFS" did survive into NT/XP, including:
From what I saw to date, WinFS seemed to be about the data/XML paradigm of data format transparency, not about opaque nested/contained data like OFS. It seems to be pursuing a different usage paradigm. At least I think so.
It's a confusing thing, and it shouldn't be. The basic idea of fusing a DB and a FS is dead simple, and if every OS offered structured and unstructured data, a set of simple core schemas, federated query across the two forms of data, and transactional/ACID cross-references between them, you could build many applications more easily. Why WinFS keeps taking so many more bits to describe itself than this is beyond me.
n@
You don't need an SQL database hiding inside your file system if you want to provide unified searching across disparate data sources (email, office, websites, SQL, etc). People have been doing it for years. Bill's just chosen the wrong means to the right end.
That's not Webster's Dictionary. That's just another cheapass website which tries to make money by taking Wikipedia's content and jamming some ads on it. And webster-dictionary has the added quality of trying to rip off the good name of the real Webster's dictionary
(I'm pretty sure Webster's Dictionary's trademark has long since passed into a more nebulous place.)
"Never attribute to malice that which can be adequately explained by stupidity." -- Hanlon's Razor
Commercial attempts at object-oriented filesystems predate Microsoft's concepts. They include not just Pink and Copland, but also Sun's Spring a research project always just short of product. All of these efforts suffered from being dead slow. So again we see attributions of an original idea to Microsoft that if one has a long enough memory has precursors.
It is also interesting to note that Jim Allchin was in charge of Cairo at Microsoft way back when. The curiousity is whether there have been real optimizations in speed for use in WinFS beyond those supplied by Moore's Law. Perhaps not enough to justify its immediate deployment.
What, you mean like the desktop database from MacOS circa 1990?
microschlong is now certainly going to lose out by postponing the ill-conceived version of a
database file system that continues to float around in its money pit.
the real problem as i see it isn't not having search capabilities, but just allowing users and gently forcing them to keep better order in their directory structure. the real problem with current file systems is that they were designed with one parameter in mind, speed/stability... in other words, to simply avoid losing data.
to achieve a file system that goes beyond this and actually allows easy (quick) retrieval of files is the goal, not clueing in to google's
success and saying, people just want to put in key words. they don't. they just need to be able
to stop worrying about where to put a file that has multiple metadata associated with it. currently they will more or less randomly have to decide that a file on model airplanes should go in with their directory called airplanes or models, and whether the airplanes directory should be inside the models directory or models inside airplanes is an even more problematic question the current user must ask himself every time he stores something new.
inevitably which ever way the directory structure is formed now, the user soon realizes he needs another folder for airplanes or models that are airplanes but not models or models but not airplanes. this user's attempted solution to this problem then compounds the his subsequent attempt to access a file that is simply about airplanes or models. the point is that we need to enhance the concept of directory rather than assume users want key word searching. even if the current search feature were faster and boolean based, it just isn't precise enough to allow faster retrieval in my opinion.
we need a file system that upgrades directories to 'zones'. zones could operate like venn diagrams. a zone name would be made up of a list of all the categories in it, e.g. airplanes, models. and that way, the user has a smaller shift in the way he thinks about storing things.
it really just allows him to do what he soon realizes he should have done with his models and airplanes directories before, to combine them into a directory called model airplanes. but it adds the additional intelligence of keeping the separation of metadata terms by replacing the flat directory name with a list of descriptive terms. if im not mistaken, the natural way to implement this is to have the hierarchy of zones automatically regenerated by the computer using the zone name terms list in each zone. this i think would be less overhead for the computer than a full-blown approach that tries to allow the user to relate apples and oranges.
I don't see it like this. File metadata andcontent are already tracked by journaling filesystems like XFS and Reiser and they do not result in "one blob of unmaintainable code." If implemented properly (and I am not proposing that WinFS would or will be), neither should this.
Again, I don't see how this is "inseparable" and from what. You should be able to specify when you create a filesystem what type of optional data you want indexed, if any. One good change that should result from this type of filesystem is making static directory trees obsolete; and defining a better, more intuitive interface to a filesystem.
These are valid arguments for today but may not hold up for future. There are a lot of features implemented in most layers of software today that would have been a big waste of resources just a decade ago.
I think you make good points and it would be interesting to see someone try to make it really work cross-platform and IP clean, maybe tie it in to some of the work being done to redo system config files with XML so that you end up with an object interface for the system configuration.
LSB would then be kind of an LDAP standard for locating those configuration objects, you could even layer in the Kerberos hooks to provide a real distributed management facility for a cluster or subnet.
As to today and future, that's really the problem. Industry has issues that need to be addressed now, not at some indeterminate future point. It's like Itanium -- great theoretical future benefits, but how do you solve current needs? Where's the short-term payback to justify the vision?
I do not fail; I succeed at finding out what does not work.
I loved that short too. I remember seeing it on SciFI with a lan buddy of mine and our jaws just about dropped. That short has been floating around for ages it seems. glad the writer/animator/whatever got a show
I disagree with the statement that "Every filesystem is a database at heart."
Most filesystems store files in block aligned files, meaning that they are poorly suited for storing phone number sized objects. The phone number sized objects that they do store (owner, permissions, etc.), are usually statically allocated with one off data structures, and very rigid in their implementation details.
Reiser4 tries to live up to your description, but we are the oddballs in the business.
Arg, is boot time the biggest issue for servers?
It is when you have to reboot them as often as Windows.
(bah-dum-dum)
Thank you, I'll be here all week...
As far as i can see, there are two different concepts in that thing:
- The real FS part: ReiserFS-like storing of a file/dir architecture, which is nice, disk-space-savey and all, but has no consequences on the way people work. Furthermore it already exists: i'm using it right now.
- The self-organized document hierarchy and search capabilities, which might change the way people work for the best, as far as it's restrained to *very specific parts* of your data. Who would trade a well crafted UNIX dirs architecture for a key indexed FS? What about dirs related documents, like a hierarchy of Java packages? What about URL accessible documents? What about implicit (not already keyword-based) relations between documents? And so on... In most cases, this stuff would have to emulate a standard file hierarchy anyway, which would probably result in system resource overhead only, or would require that you specify explicit keywords (not really knowing how they would impact the search algorythm), which would result in user resource overhead only.
You get my point: this stuff must be an option, and it belongs to the user interface, as in DBFS or Google, with a standard lib/API for easy re-usability by tiers software. It would be of no use with MOST of the files, in my system anyway.
WinFS is not even a solution looking for a problem, it's a problem seeking naive clients for its solution, IMHO.
"Take away our PlayStations
And we're a third-world nation"
A.D.
Where is it... It's on my hard drive somewhere... I have no idea.
I mod down pyramid schemes in sigs.
WinFS is not an ordinairy filesystem with an index on steroids. It's a system which allows you to store files as regular objects together with dataobjects you don't see as files, like an email contact, a customer entity from your accounting program, etc. etc.
Because of this, you can query, from the shell, on not only your files, but also your objects from other applications. THIS is the real thing about WinFS. WinFS offers the developer a neat, clean, OO api for accessing / querying the objects inside WinFS as well.
Files with a great search, like spotlight, it's very very great, but is a developer of a desktop application able to store its data directly into the filesystem so the data of the app can be queried from the shell as well?
Never underestimate the relief of true separation of Religion and State.
here. It's independEnt, BTW.
Again, I don't see how this is "inseparable" and from what.
If M$ says it is, it damn well is - to you, the user.
You should be able to specify when you create a filesystem what type of optional data you want indexed, if any.
Hello! This is M$ we are talking about!
You should be able to do allot of things with Windows, that you can't...
Atleast, these are just my 2c...
Ethics is what you say you do. Morals is what you actually do.
Actually in commercial boxed distros like Mandrake 10, the Nvidia and ATI drivers come with the distro. You will promptly upgrade these most of the time, but you are not forced to download them.
What do you expect from a 3 year old OS like XP anyway? You still have to patch the hell out of it if you ever reinstall it and during the patching time you're praying to the internet gods that your box doesn't get owned.
(I'm pretty sure Webster's Dictionary's trademark has long since passed into a more nebulous place.)
It has. Why, just look it up at Webster's Dictionary!
Now, that's all up to the implementation. It doesn't matter if it's a native filesystem or a bunch of shell scripts using thousands of text files as the database, on top of N other hacks served over a Python based userspace VFS module implementation, as long as it works. In other words, we can always improve and rework the implementation later, when more scalability, speed or beauty is required.
The big problem is in the shifting of the desktop paradigm. Most people are so used to the way computers work now, and they're reluctant enough to learn that it'd take another decade to re-teach them. Conversely, there are also the nerds who absolutely want to know where their data is actually stored, instead of accepting a mere result set of a database. They want to run Unix scripts on the files, so the abstraction must be backwards compatible with the notion of traditional path names. (E.g. "grep frobnicated /sysdb/files/of/type/tex/*")
Even the current desktop paradigm is unintentionally abused: many still have problems with creating file system hierarchies, putting all files to the same flat directory -- or desktop directory. A "more intuitive" file system might help these people, or just let them not take advantage of a more sophisticated system instead.
Roughly, most people who can organize things can do it already and those who don't, will always fail (overstatement intended). Since file access is such a fundamental part of any computer user interface, this thing must be really thought of and weaved into the whole operating system, rather than just writing a new FS and going for the shortest route to enable the new features in traditional user interfaces.
Everybody talks about Tigers search capability, i.e. find word within map in PDF format.
/ 1004717.htm l
Windows XP has had this capability since 2001. Try downloading the free PDF plug-in for MS Indexing Service from Adobe's web-site.
http://www.adobe.com/support/salesdocs
MS Indexing Service already has the ability to search text, html and MS Office documents built-in.
The advantage of XP's Indexing Service (available since NT4!!!) is that it runs when the machine is idle.
OK, WinFS will offer a lot more and MS Indexing Service isn't perfect but it is there now and it is free (not open source def.)........
...so that as many applications can provide as much meta-data as possible. So that your e-mail program, browser, digicam software and whatnot can provide. Not to mention there's finally a use for templates. You could even have specific fields that provide metadata beyond the type of the template itself.
Imagine an "invoice" template where you could have a "name" field. Now you can search for "invoices to Bob", which won't bother with every other form of communication (meetings, contracts, whatnot) and not with other people's invoices. There's a lot of opportunities there, if the system supports it.
Kjella
Live today, because you never know what tomorrow brings
Also you can optimize the hell out of a linux bootup process, all those nice friendly init scripts can be replaced with a hardcoded multithreaded C program, or even removed (resulting in the box booting to whatever login daemons you specified in /etc/inittab instantly after mounting the fs)
Infact, sometime i plan to rewrite a hardcoded init process that just loads up my minimal services without any fancy shit..
http://spamdecoy.net - free throwaway anonymous email - avoid spam!
No, windows server 2003 presents the login screen faster but the network services aren't all loaded yet, and if your running a server surely you want the services running and available?
Linux won't display the login box until after everything is initialized, windows will half load itself, show the login and continue loading in the background.
http://spamdecoy.net - free throwaway anonymous email - avoid spam!
yeah, it's called shitty technology on a shitty platform.
I'm glad it's behind schedule.
They can't seem to get virus immunity or Browsing right either.
I would rather use a Mac, Solaris, SysV, AIX, OS2, NeXT, Palm, HPUX, Linux, Osborne (yes..I had one), or abacus any day. Hope the project melts on the back burner. I have a feeling it's going to be sitting there for some time.
It's not too much difficult to get a 20 seconds boottime (from grub to GDM) in any machine using linux.
And you don't even need to rebuild custom kernels or hardcode anything, just launch GDM as soon as posible, then you load the other services in parallel just like windows do. Some other tricks explained in also help ;-).
8 out of the 20 seconds are used for the linux kenrel to load, and the rest are mainly Xfree loading and GDM & GTK loading....
Now imagine if you use a lightweight Xserver like kdrive and port kdrive to a fast-booting kernel such as FreeBSD: I bet you could get under 10 seconds boottimes. And there are still much more tricks:
I'm thinking about building a new distro with some of these ideas and more, if you're interested, edulix@jabber.org ;-).
Cheers,
Edulix.
--- [1] http://forums.gentoo.org/viewtopic.php?t=131142&po stdays=0&postorder=asc
[2] http://www.intel.com/technology/efi/
So MicroSoft is going to index files for us? Let's look at their record: * A Start menu thingy "find" that is slow, mostly broken, and doesnt have the smarts to not search .DLL and .EXE files or their own DllPrefetch or Cookie or system32 directories when I'm looking for plain text.
* A background file indexing thingie that seems to take forever.
From these brain-dead search folks you're expecting something GOOD?
*
This sounds a lot like Reiser4. Metadata, transaction support, lots of other stuff. All there. link. Hopefully, this will make it into the kernel someday, although it might be 2.7 material.
When MS Access 1.0 was launched, MS's Access team said MS's ultimate vision was to have everything in the system relationally stored - which makes sense, see stuff like Gnome Storage.
...I forget) are poor implementations of SQL, and SQL isn't relational at all. SQL is a misimplementation of a few of the relational ideas carrying severe arbitrary limitations.
Problem is, MS Access (and MS SQL Server, and their engines Jet and...
Most probably MS will never come to push this until they get the relational theory right. But with the MS Access and MS SQL Server pushing the party line of 'SQL is relational, but objects are better', they most probably will never get there.
Perhaps Gnome Storage has a better chance, because PostgreSQL is such a nimble system. But it still is SQL. Rel looks like being a potentially conceptually better solution as far as the data language side goes, but it still needs a huge amount of work on the storage engine side.
Leandro Guimarães Faria Corcete DUTRA
DA, DBA, SysAdmin, Data Modeller
GNU Project, Debian GNU/Lin
Keep in mind the past of MS. See a thing, make a minor improvement, make it your own. The storage FS issues has been a good concept, but implimation has been lacking until now with Reiser4. MS has not had a Xerox to emmulate until now.
Comment removed based on user account deletion
In an abstract, theoretical way, that makes sense, but I'm not so sure it applies readily to the real world.
Consider: there are many different filing systems in use, and people frequently transfer files from one to another. This loses metadata that can't be stored on most FSs.
Many file types (pictures, audio, even HTML) already include metadata embedded in the file.
And many types of metadata apply only to particular types of file.
So, if you put this in the FS itself, you risk losing metadata when transferring between filesystems, and you have issues syncing the FS metadata with any embedded metadata.
The more I hear about this, the more I think that maybe Apple's solution (Spotlight) is more realistic: you leave metadata embedded in the file, and have a generic way of accessing it that allows searching. That way, you make best use of the information that's already there, you don't have to create anything new that'll get lost in file transfers, it can work with any FS, and no changes are needed to applications.
Ceterum censeo subscriptionem esse delendam.
Microsoft started shipping SQL Server for OS/2 in 1989. In '92 it shipped MS SQL Server 4.2 (a 16-bit version) and in '93 shipped MS SQL Server for Windows NT (32-bit version). All of this well before Cairo.
Small nitpick, the last line spoken by Six, was revised for the TV release. Dunno why. I do have teh original trailer when it came out, Six said something to the effect of Hyperdrive not being useable when entering (or near) the Edgeworth Hypersmiley.
So rise up, all ye lost ones, as one, we'll claw the clouds.
Good fucking grief, osViews.com. You can't spell when it's right there in from of your dumbass face and all your stupid ass has to do is block and copy. Retarded.
Fact-index.com not only puts up Wikipedia content with Google ads, it's actually started making substantial financial donations to Wikipedia!
http://rocknerd.co.uk
yeahhh
It's a lot easier to take a straightforward idea like the GUI from Xerox and recode and market it than to do something like this, another example that MS has been talking about for years is voice recognition. These things lie somewhere outside their basic business model, more a dream or ambition than an actual ability.
NTFS is outperformed by ext2, but most sensible users prefer the lower performance and higher data integrity of NTFS and ReiserFS. Overhead is a lame argument against improving OS services.
System's??
That's it. Get out of my English.
Such an indexed WinFS could be the big stick of Microsoft against both Oracle and Google.
.. in anticipation. And hedge your stock bets.
You should shiver
I'm still trying to figure out what people mean by 'social skills' here.
None appear to know it .. right?
Just because it's legal and moral doesn't mean it's not lame. But it's nice to hear that Fact-index sends money back to Wikipedia.
"Never attribute to malice that which can be adequately explained by stupidity." -- Hanlon's Razor
The other problem is that it's not uncommon in the database world to spend far more disk indexing complex data for access than it actually takes to store the raw information itself.
So what?
Do you really want the possibility that your inseperable all-in-one file system is using more space for the equivalent of directory entries than for data itself?
I don't really care.
Let me explain.
Back in Feb 1984, when the Macintosh first began shipping, there were lots of arguments very similar to some of the ones you are making (in the parent post).
Example argument: But think how inefficient it is just to draw one character onto the display! The processor has to read the glyph for the font, and then individually set the right pixels in the display bitmap. Lots of other complex graphics processing as well. The font's character may be partially clipped by a window stacked on top of it. The pixels to set for the font have to be loacted within the display frame buffer bitmap -- offset by the window location, and possibly other nested panels within the window where the text is drawn. On an IBM PC, you just change one byte in the textmode frame buffer and the video card's character generator displays the character on your green-screen monitor.
Now fast forward to the 21st century. Does anybody care? Even modern Linux now runs pure "textmode" console screens in graphics mode and the kernel manipulates the individual pixels in a graphics framebuffer (and displays a nice penguin graphic in the same framebuffer). Nobody even uses hardware text mode anymore.
My Point...
Stop caring so much about machine efficiency and start caring about usability.
I don't really care if the search indexing system takes ten times, or one hundred times, or one thousand times the disk space as the raw data. As long as today's hardware can handle it, I don't really care. If today's hardware can't handle it, then tomorrow's hardware will, and we'll see these innovations appear as soon as common everyday hardware can handle it. All I really care about is that I can ask the computer to: "Show me all OpenOffice.org Writer documents, where I mention the name of any of our employees who live in New York, are over 9 feet tall, and have green hair.".
Remember this isn't about special cases like a user too lazy to sort their home directory or documents folder
Like lazy Mac (or Windows) users too lazy to learn convenient DOS commands.
Make the machine fit the man, not the man fit the machine.
In fifty years, there will be people complaining because our artificially intelligent speech and vision enabled applications, that can instantly access all the world's data and intelligently filter it, and even think about problems for us, require "bloated" computers using 500 Petabytes of disk storage, and 256 Terabytes of RAM. That is just too bloated. I prefer the old applications that only required 800 Terabytes of disk, and merely 16 Terabytes of RAM. Ah, those were the days of machine efficiency, back when a computer and its applications fit into a nice "slim" configuration of only 16 Terabytes of RAM. Ahhhhh.
Those who would give up liberty in exchange for security and DRM should switch to Microsoft Palladium!
Explain to me how System Restore doesn't solve the problem you describe here?
Hans,
Please investigate whether your work is patentable. If it is patentable and you do not patent it, Microsoft -- or some other party -- will surely patent something close enough that they can appropriate your intellectual capital with no compensation whatsoever. In fact, they may even be able to charge you for the privilege of having your own work forever denied to you.
If you truly have invented something new and worthwhile, taking the extra step to protect it is an important act of self-defense.
"Genius may have its limitations, but stupidity is not thus handicapped." --Elbert Hubbard (1856-1915)
(jargon) vaporware -
Damn. That is pretty harsh.
If it's with acknowledgement and the GFDL, that's just fine - because Wikipedia is a site that Slashdots itself, and the less users the better ;-)
http://rocknerd.co.uk
How many M$ developers are needed to develop a big stick ?
120 = 6 teams of 20 to assess the competitors sticks
12 to design a revolutionary new stick
24 to browse old source code for old sticks that can be adapted into a big stick.
48 to query customers for stick features
12 to redesign the big stick
1 in India to carve the prototype
20 in quality assurance
100 to test the big sticks on
1 to go to Walmart and buy a baseball bat to show the patent office in the patent application for a big stick
60 key account managers to force all companies into OEM contracts to sell exclusivly the M$ stick.
12 to spread FUD about competitors sticks.
total so far: 410 M$ developers
Disclaimer: I buy M$ sticks, too.
I'm still trying to figure out what people mean by 'social skills' here.
Like the 1450XLD. Built-in double-sided disk drive (before the XF551 was released), built-in speech synthesizer, 520ST keyboard layout. I still drool over that (and the 1MB ICD MIO, but those days are long gone for me.)
Here's the thing, instead of reinventing a pretty darn round wheel, Apple has focused on what it is the user wants. Imagine that. They started with use cases in mind. Steve Jobs' demo mentions iTunes live searching of tracks and their metadata and showed how extending this functionality to the entire OS made logical sense. Did they have to rewrite a file system from scratch? No, they extended their existing file system with richer metadata and created an API for developers to plug their own file types into it.
WinFS is already obsolete, old news, borderline vapor-ware.
Spotlight is in my hands.
Not to mention there are a number of 3rd-party solutions out there for the windows world that accomplish what WinFS aims to do.
Extraordinary Vacations. Exceptional Prices