it is not a load of bs. My uncle is the leading project manager for longhorn and a few other stuff. It is not a load of bs. How does being able to seamlessly find data seem horrible? What is wrong with the OS being aware of everything going on? Is something wrong with being more efficient? Please explain.
-
We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.
Longhorn in 2006 or later?
Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Nothinman
Elite Member
Office is the better example then, Office 2000, Office XP, Office 2003. It's obvious where two of them go but where does XP fit in? Can you tell just by the name which of those releases it's newer than?
If it's really aware of what all is going on the chance of it being more effecient aren't very good, because it'll take more CPU time for the OS to determine what you're doing. And the WinFS thing replacing NTFS has been reported to have been dropped infavor of a service that runs ontop of NTFS sort of like the Indexing Service (which I hate).
And as for the obfuscation of where my data is,that's the worst idea yet. If I want to search the Internet I do, if I want to search my hard disk I do, I don't want local searches delayed while it searches via MSN (because we know they wouldn't use something like google that they can't control).
And if I were you I'd be carefull about what I said about Longhorn, MS has been fairly quiet about what's really going to be done (probably because things like WinFS were so largely overstated they're starting to realize they can't implement them in the timeline they laid out) and if your uncle has been found to be leaking information, even via you, he'll probably get in trouble.
Nothinman pretty much covered why I think it sounds horrible.
And if your uncle does indeed manage the Longhorn project, accept my appologies, but understand that the net is full of people who claim to have cousins, brothers, uncles, etc etc, who work for X, Z and Y companies.
My uncle has only told me the info that he is allowed to leak. He knows alot more then i do, and alot of it he cannot tell me. Yes he has said it will require more power then any previous OS, but by the time it is released, it is not gonna seem out of the world in requiring so much power. He doesnt even know himself exactly how much power it will need to run. As for the info thing. Im pretty sure that what it will do is search the harddrive, then ask you if you want to search the internet for what you searched for, i dont think it will do it without permission.
But think about it this way. If the OS is fully aware all the time of what is going on within itself, that would make it very hard for viruses to do anything because it would detect it in a snap and inform you about it.
But think about it this way. If the OS is fully aware all the time of what is going on within itself, that would make it very hard for viruses to do anything because it would detect it in a snap and inform you about it.
I'm finding it hard to believe you.
As a project manager, he should know better than to leak anything that is not already common knowledge
to the press.
An historic mistake often made on Microsoft's part. There are still a lot of people on the net (and
in business) making do with older technology. You will only just see the start of a major migration
from NT 4 in 2004, since MS is dropping support for it soon, and the cost of a system to run 2000 on
is now so cheap.
It is good that MS is trying to figure out what people want in a future OS, but they sometimes provide
for those desires at the expense of the power that people want to apply for themselves.
Right... They must be running the alpha versions on something!? And he wouldn't be a project manager
without having gone thru many. many design meetings to establish a target for system requirements for
any programs he is working on. He knows (or has a fair idea of) exactly how much power they
want the average system to have before Longhorn is ready; but he can't tell you because that
would start the rumor mills running, and have people unfairly comparing the alpha against already
established and released versions of the OS.
He should have just said that he can't tell you without compromising the project.
Half the people here can give you a better idea, just based on the directions MS is trying to go
in new PC specifications and technologies they seem excited about.
You mean like the Office 2003 beta did? Its help system would try to connect to microsoft.com first, and
you actually had to force it to install the help files locally. I understand that was fixed in the release version?
(probably due to negative feedback from beta users who did not have an on demand broadband connection
for every machine).
Windows 2000/XP/2003 already have lockdowns on system files that should be preventing virus
exploits. What they need is to finish the work of plugging the obvious security holes that allow the
viruses entry in the first place. Otherwise what you are proposing sounds just like what a decent
virus scanner and firewall should be doing in the first place.
As a project manager, he should know better than to leak anything that is not already common knowledge
to the press.
An historic mistake often made on Microsoft's part. There are still a lot of people on the net (and
in business) making do with older technology. You will only just see the start of a major migration
from NT 4 in 2004, since MS is dropping support for it soon, and the cost of a system to run 2000 on
is now so cheap.
It is good that MS is trying to figure out what people want in a future OS, but they sometimes provide
for those desires at the expense of the power that people want to apply for themselves.
Right... They must be running the alpha versions on something!? And he wouldn't be a project manager
without having gone thru many. many design meetings to establish a target for system requirements for
any programs he is working on. He knows (or has a fair idea of) exactly how much power they
want the average system to have before Longhorn is ready; but he can't tell you because that
would start the rumor mills running, and have people unfairly comparing the alpha against already
established and released versions of the OS.
He should have just said that he can't tell you without compromising the project.
Half the people here can give you a better idea, just based on the directions MS is trying to go
in new PC specifications and technologies they seem excited about.
You mean like the Office 2003 beta did? Its help system would try to connect to microsoft.com first, and
you actually had to force it to install the help files locally. I understand that was fixed in the release version?
(probably due to negative feedback from beta users who did not have an on demand broadband connection
for every machine).
Windows 2000/XP/2003 already have lockdowns on system files that should be preventing virus
exploits. What they need is to finish the work of plugging the obvious security holes that allow the
viruses entry in the first place. Otherwise what you are proposing sounds just like what a decent
virus scanner and firewall should be doing in the first place.
Nothinman
Elite Member
So how do you discern a virus from a normal program? Current virus scanners have a database of signatures of each virus, if MS uses that method they'll be bundling yet another feature and that will make it impossible for Symantec, McAfee, etc to compete in that area. It's not even really possible to detect buffer overflows themselves because x86 hardware doesn't have anything to discern read and execute bits on pages like other CPUs do. On Opteron and Itanic I believe Windows will make use of that security (or will sometime in the future) but for 'normal' people it's not possible yet.
Sure and if you've ever developed any software you know that early versions of big projects look and act almost nothing like the final version and ass things are added memory and speed requirements grow. WinFS may just be a renamed Indexing Service right now, but once they add all the new crap they're saying it'll do it'll probably need atleast twice as much memory to sit idle.
Sure but he also has to take into account the class of machines that will be new and common in the timeframe of the release. They'll probably say it'll run on a 1Ghz machine but it'll probably need a lot more to run nice, just like XP says it'll run on a P233 IIRC.
They do NOT have a target system spec yet. and he knows exactly what he is allowed to tell people. The OS will be smart, not something waiting for info 24/7. It will be familiar with virus like activity and it will be able to tell the difference between games and all that easily.
Falloutboy525a
Member
hmm I remember hearing about a computer becoming aware of it self somewhere.....oh yeah in terminator and we all know that didn't turn out well 😀
techwanabe
Diamond Member
Win 3.11 - 1990
Win NT 4.0 - 1996
Win 2000 Pro - 2000
Win XP Pro - 2002
Let's see, 6 year, then 4 years, then 2 years, to release next version of Windows. Then we gotta get everything to work with it. Hey, I'm not in that much of a hurry! I know MS like selling copies of Window but lets get it right!
Win NT 4.0 - 1996
Win 2000 Pro - 2000
Win XP Pro - 2002
Let's see, 6 year, then 4 years, then 2 years, to release next version of Windows. Then we gotta get everything to work with it. Hey, I'm not in that much of a hurry! I know MS like selling copies of Window but lets get it right!
techwanabe
Diamond Member
I thought of mentioning those as well, but l NT 4.0 was the first decent Office OS... since it went thru several lessor iterations
erikistired
Diamond Member
3.11 was 'windows for workgroups'. according to MS it was released in 1993, not 1990.
however, doing some poking around, there was a windows nt 3.1 released in 1993. then windows nt 3.5 (1994), 3.51 (1995) and finally 4.0 (1996). i don't know that i've ever seen nt 3.1 before. learn something new every day.
windows xp was released on october 25, 2001.
Longhorn may be super uber wonderfull and all that and a "thinking" OS and all that.
But I am sure that once people figure out what MS realy means by all this stuff, then it won't sound nearly as impressive.
But do you remember how much people were freaking out about windows 95? And that was a big turd of a OS. All the big new features existed years and years before Windows 95 ever came out but most people never noticed because they were stuck in DOS land and never before seen a decent OS before.
I've screwed around with w2k3 and that seems nice enough. It's basicly w2k, but actually redesigned to fix lots of the old mistakes. Not anywere the difference between w2k vs NT.
I've been using it for close to six weeks now, (class meets once a week.) and it's only BSOD on me twice and out of a class of 16 computers only one or two has been rendered completely unusable by constant crashes and memory dumps. (seems like a memory issue)
Which is a improvement over w2k and a world of difference from NT.
However the 2006 due date for such a "revolutionary" OS revision seems VERY optimistic.
I mean, this is comming from a company that was off of it's due date for AMD64 port of XP by at least a year.
Anyways the the only real difference I see is that apps will interact with a database, instead of directly with the filesystem. That way you could clear the way for online databases that you could latch into as a subscriber. Like online music databases for instance with the DRM protecting it from Warez people, or businesses could have online internet databases for their nationwide company or something like that.
I can only think of a hundred or so reasons why that's a bad idea, but this is also coming from a company that inflicted the computer world with the windows registry. (one of the worst and most fragile OS design features in recent memory)
But I am sure that once people figure out what MS realy means by all this stuff, then it won't sound nearly as impressive.
But do you remember how much people were freaking out about windows 95? And that was a big turd of a OS. All the big new features existed years and years before Windows 95 ever came out but most people never noticed because they were stuck in DOS land and never before seen a decent OS before.
I've screwed around with w2k3 and that seems nice enough. It's basicly w2k, but actually redesigned to fix lots of the old mistakes. Not anywere the difference between w2k vs NT.
I've been using it for close to six weeks now, (class meets once a week.) and it's only BSOD on me twice and out of a class of 16 computers only one or two has been rendered completely unusable by constant crashes and memory dumps. (seems like a memory issue)
Which is a improvement over w2k and a world of difference from NT.
However the 2006 due date for such a "revolutionary" OS revision seems VERY optimistic.
I mean, this is comming from a company that was off of it's due date for AMD64 port of XP by at least a year.
Anyways the the only real difference I see is that apps will interact with a database, instead of directly with the filesystem. That way you could clear the way for online databases that you could latch into as a subscriber. Like online music databases for instance with the DRM protecting it from Warez people, or businesses could have online internet databases for their nationwide company or something like that.
I can only think of a hundred or so reasons why that's a bad idea, but this is also coming from a company that inflicted the computer world with the windows registry. (one of the worst and most fragile OS design features in recent memory)
QueBert
Lifer
it's a horrible idea becase... it will never work.
to say it will "think" is absurd. I believe what you said, MS is dumb enough to try and reshape the way we interact with our files, for a "newer" and "better" method. But the only way it can "know" what is what. Is files having info in the headers, or some type of id3 of metatags. bottom line Longhorn won't be able to find pictures I have of me on the HD unless I specificly added some info as to "this is a picture of me" to the file in the first place. The idea sounds wonderful to somebody who actually believes PC's will be that advanced anytime soon. But I'm not holding my breath for a "smart PC" especially not one ran by a MS OS 🙂
Finding effective ways to orginize files isn't so bad.
Like for example I use two programs together (I am a linux user obviously): One is called "ripperX" the other is "Zinf". You could use any two programs of this type...
RipperX is a ripping and encoding program for Linux. It pulls sound files off a audio CD using a 3rd program called cdparanioa (that is specificly designed to recover audio files and compisates for things like skips and cdrom jitters to create a 100% quality output). It then takes these .wav files and encodes them into ogg vorbis compressed audio files.
I have them all go into their respective folders based on almum name in a folder called: Music.
Ogg Vorbis is like mp3's in that it is a lossy audio compression format, but more importantly they store information about the album, artist, genre, etc etc in the file's header. RipperX uses cdda server to stick that info into each file.
All this is done in automated steps.
Zinf is a itunes-like program (formally known as freeamp) that has a automatic way of finding songs. I tell it to look in my "Music" directory and then orginizes a "MyMusic" listing by artist and album. It looks much like the explorer mode of windows desktop were you have the + and the - buttons on the tree. (also can have internet feeds and other stuff)
Then from that you can generate new playlists by double clicking on the song/album/artist/or entire database names. Then it lists it by album then artist then track number. then you can randomize, edit it, and create saved playlists. A lot like making mix tapes.
All this is very simple. All I have to do in the whole thing is press go on ripperX and have zinf scan for new songs. Then select the songs or playlists then play the music.
NOW that is a good way to use a database to orginize stuff.
I'd bet that windows longhorn is going to aim for the same way of doing stuff, but with EVERYTHING.
Your pictures may not be automated at first, but I bet if you make any new pictures using the apps in longhorn then it will then have all that info recorded.
However not all file formats like mp3's have the data built in. Like jpeg. And I doubt that MS is stupid enough to try to force everyone to adopt a new jpeg format. 🙂
So they are probably going to do something like apple does (which I don't like) and create a new NTFS-based format like HFS+ (apple's format) does.
Each file becomes 2 parts. The data part and the metadata part. Now in Macs this is totally transparent. Each file is REALY one file, just 2 parts. When transfered over to a format that doesn't support that format it simply strips the metadata away or puts it into special files.
The metadata contains things like file type, creation date, program that made it, what to open the file up with, etc etc. That lots of files don't have extensions like Window's files and the OS simply "knows" what they are. In Apple land it makes this very slick and easy to deal with, but makes networking and interaction with other OSes a bit annoying at times.
So I figure that's what the NTFS extensions stuff people are talking about. NTFS already supports metadata a bit with stuff like ACLs and the like.
That way a person can create a file on one longhorn machine then transfer it to another and not loose the info. But you will lose that info if going to a Linux machine or a Windows XP machine.
Now on top of that I figure Longhorn is going to run a functioning but minimalist database. That way it can keep track of the orgination of the files and the filing system keeps track of the file information.
Probably store the information in a binary form in the database's own form of the registry.
That way when you make a jpeg file (for instance) it will ask for any "extra info" and then encode your user name, file creation date, etc etc. Then it would place the file in a "mypictures" orginization. It will catalog it with any sort of information you will find usefull. Real world location, what album (family, work, taco bell advertising project, etc etc).
Then if your using frontpage you open up the picture orginization and it will have that already orginized and ready to go. Only jpegs and giff become aviable and you press "show this album" and it will rotate around a bit and find what you looking for. You'd have date name and revision numbers to help you locate files.
Then each filetype (or actually information type, written, movie, finacial, image etc etc) will have it's own specific orginizational method. What every MS thinks that will make it's programs seem most usefull.
The apple "finder" program maybe does this already a bit, now that I think about it. (wouldn't be the first time MS ripped off apple's ideas. 😛) But it doesn't orginize stuff exept by real world location.
Now the parts I don't like would be that the OS will attempt to divorce you from the filling system. Very few tools will be aviable to manage the files, and files will probably be stored in a way that would be virtually unmanageable anyways.
Plus with the DRM would tie your files in with each particular machine making a pain in the a** to do anything with your files that the orginal authors didn't think of or want you to do.
One big trouble is that the more rigid and controlling the OS or a program is the usefullness of it decreases rapidly.
Also you could imagine the pain in the butt when a worm can use the database stuff to infect and automaticly spread itself thru the easy-to-use internet.
Like for example I use two programs together (I am a linux user obviously): One is called "ripperX" the other is "Zinf". You could use any two programs of this type...
RipperX is a ripping and encoding program for Linux. It pulls sound files off a audio CD using a 3rd program called cdparanioa (that is specificly designed to recover audio files and compisates for things like skips and cdrom jitters to create a 100% quality output). It then takes these .wav files and encodes them into ogg vorbis compressed audio files.
I have them all go into their respective folders based on almum name in a folder called: Music.
Ogg Vorbis is like mp3's in that it is a lossy audio compression format, but more importantly they store information about the album, artist, genre, etc etc in the file's header. RipperX uses cdda server to stick that info into each file.
All this is done in automated steps.
Zinf is a itunes-like program (formally known as freeamp) that has a automatic way of finding songs. I tell it to look in my "Music" directory and then orginizes a "MyMusic" listing by artist and album. It looks much like the explorer mode of windows desktop were you have the + and the - buttons on the tree. (also can have internet feeds and other stuff)
Then from that you can generate new playlists by double clicking on the song/album/artist/or entire database names. Then it lists it by album then artist then track number. then you can randomize, edit it, and create saved playlists. A lot like making mix tapes.
All this is very simple. All I have to do in the whole thing is press go on ripperX and have zinf scan for new songs. Then select the songs or playlists then play the music.
NOW that is a good way to use a database to orginize stuff.
I'd bet that windows longhorn is going to aim for the same way of doing stuff, but with EVERYTHING.
Your pictures may not be automated at first, but I bet if you make any new pictures using the apps in longhorn then it will then have all that info recorded.
However not all file formats like mp3's have the data built in. Like jpeg. And I doubt that MS is stupid enough to try to force everyone to adopt a new jpeg format. 🙂
So they are probably going to do something like apple does (which I don't like) and create a new NTFS-based format like HFS+ (apple's format) does.
Each file becomes 2 parts. The data part and the metadata part. Now in Macs this is totally transparent. Each file is REALY one file, just 2 parts. When transfered over to a format that doesn't support that format it simply strips the metadata away or puts it into special files.
The metadata contains things like file type, creation date, program that made it, what to open the file up with, etc etc. That lots of files don't have extensions like Window's files and the OS simply "knows" what they are. In Apple land it makes this very slick and easy to deal with, but makes networking and interaction with other OSes a bit annoying at times.
So I figure that's what the NTFS extensions stuff people are talking about. NTFS already supports metadata a bit with stuff like ACLs and the like.
That way a person can create a file on one longhorn machine then transfer it to another and not loose the info. But you will lose that info if going to a Linux machine or a Windows XP machine.
Now on top of that I figure Longhorn is going to run a functioning but minimalist database. That way it can keep track of the orgination of the files and the filing system keeps track of the file information.
Probably store the information in a binary form in the database's own form of the registry.
That way when you make a jpeg file (for instance) it will ask for any "extra info" and then encode your user name, file creation date, etc etc. Then it would place the file in a "mypictures" orginization. It will catalog it with any sort of information you will find usefull. Real world location, what album (family, work, taco bell advertising project, etc etc).
Then if your using frontpage you open up the picture orginization and it will have that already orginized and ready to go. Only jpegs and giff become aviable and you press "show this album" and it will rotate around a bit and find what you looking for. You'd have date name and revision numbers to help you locate files.
Then each filetype (or actually information type, written, movie, finacial, image etc etc) will have it's own specific orginizational method. What every MS thinks that will make it's programs seem most usefull.
The apple "finder" program maybe does this already a bit, now that I think about it. (wouldn't be the first time MS ripped off apple's ideas. 😛) But it doesn't orginize stuff exept by real world location.
Now the parts I don't like would be that the OS will attempt to divorce you from the filling system. Very few tools will be aviable to manage the files, and files will probably be stored in a way that would be virtually unmanageable anyways.
Plus with the DRM would tie your files in with each particular machine making a pain in the a** to do anything with your files that the orginal authors didn't think of or want you to do.
One big trouble is that the more rigid and controlling the OS or a program is the usefullness of it decreases rapidly.
Also you could imagine the pain in the butt when a worm can use the database stuff to infect and automaticly spread itself thru the easy-to-use internet.
erikistired
Diamond Member
were you even using computers when win95 came out? turd of an os? have you ever had to use windows 3.1? stuck in dos land? as opposed to what? linux? in 95? ha. macOS? i wouldn't use macOS now, even tho os x is pretty neat, but not in 95. i looked forward to every new beta cd that arrived because it was one step farther away from windows 3.1. i'm really happy you love linux, but going into every windows thread just to mention linux is kinda getting repetative.
wow, sounds like your pc needs to be looked at. seems like a memory issue? that's not the fault of windows, even tho you're trying to insinuate that. it's amazing that all the time i spent in classes while working on my mcse i never had an nt server crash on me. maybe you just have bad luck?
infected maybe? i'm not sure the registry itself is a problem, as much as crappy software that causes problems with it, or people who don't know what they are doing tooling around in it. if you want to attack windows, try attacking it's security holes, or well, huh. i dunno. most problems i have with it are caused by 3rd party software, so i guess you'll just have to go on about the security holes.
Nothinman
Elite Member
I thought there was a version 3.11 of NT, guess it was just 3.1. Either way it's a minor fact. We still have a Citrix WinFrame server running on top of NT 3.51 where I work, it's a POS but it does it's job and we can't get rid of it because the software it runs doesn't run on anything newer.
Sure it is, it's a binary-only database that's a PITA to script changes to an it's entries are very poorly documented.
erikistired
Diamond Member
'well quit messing around with it! it's not meant to be touched by mere mortals!' ~ bill gates, circa 1996
i had some clients still running old versions of novell (vet software, ick) for the same reason. as far as i know they still are running it, but since i no longer support them i don't seem to mind as much. what happens if you have to upgrade the hardware? is nt 3.51 going to run on an athlon/p4 based system?
Nothinman
Elite Member
But it is meant to be touched by mortals, almost every KB article has you change atleast 1 registry setting to fix or change the behaviour of something.
We don't upgrade the hardware. And hopefully it'll go away soon, the hardware it's used to manage is getting too old and will be replaced eventually.
We don't upgrade the hardware. And hopefully it'll go away soon, the hardware it's used to manage is getting too old and will be replaced eventually.
erikistired
Diamond Member
that's what i meant i guess, what happens if the hardware craps out and you need to use it one. last. time. or one last month. sounds like the problem will be taking care of itself tho.
Settle down there. Window 95 sucked, people wanted it and they bought it. I used DOS and all that jazz, but I was like 12 at the time and didn't have much choice in the matter. (I remember having to turn on extended memory and use a floppy disk to get the extra few K that it needed to run tie fighter.
If I knew then what I know now I would of used MacOS definately or BSD.
Aside from Dos and windows their was OS/2 the BSDs and a few other operating systems that were much better, but most people didn't know anything about them. People used DOS at work. And that was the first and only OS they've ever seen so they used it at home.
I am not saying I was better then anybody, everyone else was in the same boat. But if you objectively looked back at what was aviable at the time, then yes win95 sucked. I mean most people that ended up using it ended up hating it.
Just like most people ended up hating windows 3.11, then wanted win95, then hated it, then got win98, then got winME, then went back to win98 etc etc etc. It wasn't till I made it to windows 98 that I realised that it's stupid to steal OSes when I can find perfectly stable ones that people WANT me to use for free.
Did I say that the hardware problem with the other computer, not a OS. HENCE THE (seems like a memory issue) THING. duh.
And that was a different computer.
My computer crashed twice. The first time I don't remember why, it just happened. Hasn't happened since. The second time was due to a installation of a program that failed and crashed the entire computer. Didn't try installing the program a second time.
That's normal stuff. And I never had NT crash on me either, well in class it is. Although it did take on average 3-4 tries to get a isntall to take on occasion. All that service pack crap.
And so on and so forth. It's (w2k3) better then w2k though. And I am sure that the once MS gets a couple service packs out for it those issues that I and a few of my classmates experiance time to time will go away.
Of course I am still wondering why the hyperterminals in half the machines refuse to have a working number "7".
The trouble is that a OS is designed to be used with "3rd party" programs. That's the whole point of having a OS you use it to run other programs. The registry is a bad idea because it's just so fragile. Uninstalling even the most stupid little program can completely break the thing and then that renders the entire OS unusable.
It's much much better to use plain text files. If you get a power surge and it flips a couple bits in a ascii file, worst thing that happens is that you have to boot up with a floppy disk and manually edit the program to fix the garbled characters. If that happens in the registry you can loose the entire OS, or at least have random BSOD's and crashes, or have programs that all of a sudden stop functioning and need a uninstall/reinstall (which can also cause compounded problems.) There is realy no effective way to edit it if the OS is non-functional.
Then on top of that it's so cryptic and misleading that you need about 6 months of studying to begin to understand what happens when you change a few dwords around, so that even if those "3rd party" programmers wanted to make effective uninstallers and installers it was about near impossible. Most of the time they just left crap laying around in the registry instead of risking it.
I'll try to restrain myself in the future. Sorry.
erikistired
Diamond Member
that's about all i needed to hear. 😛 if you'd had to use win 3.1 you might have a different view on windows 95.
Your digital cam pictures already contain a lot of metadata (in EXIF format) - time taken, size, exposure, camera, camera settings etc.
The file system is as you know it is already just a virtual 'view' of how the files are truely stored - just stream of bits with no hierarchical structure - welcome to the Matrix 🙂 Most people today don't know how to manage their filesystems effectively either, duplicate data, expired data etc... As hard-drive capacity grows this will only get worse. Anything that aids management is a bonus.
DRM doesn't enable the OS to have control - it enables an author to have control. DRM policy can be expressed to constrain by many different facets too e.g. host machine, user, # of copies etc. I wouldn't say that DRM is perfect, but it is necessary in a relatively lawless environment.
Not sure where you were going with the point about binary databases. It's been a long time since I've seen registry corruption being a major problem - probably something to do with the uptake on NTFS. If the point was that it couldn't be edited in notepad say - then a) that's a good thing! b) I guess its relevant to mention that the registry does a lot more than hold segments text. e.g. BLOBs, security keys etc. I'm sure you could give them textual representations - but like the filesystem - that's just a view right? Perhaps you'd prefer longer boot times instead?
LH's WinFS metadata is stored in a subset of SQL Server - a relational database system - just as the registry is a database. Like the registry it's internal format is less relevant - it's robustness, durability, and tools to make it manageable are more relevant.
For a relatively good post - shame Drag that you switched to troller's FUD in your last comments - save it for another forum.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 25K
-
-
-
Question Converting Dell D400P-01 Server Power Supply to Desktop Power supply.- Started by Xenonomous
- Replies: 0