Well, obviously they have already dealt with multi-platformness. You think that's a maintenance nightmare as well? Of course not. They probably did their best to isolate the platform-specific stuff and put the rest in separate libs.
Maintaining separate code bases for multiple platforms is a nightmare. I expect IBM have come to that realisation too for Lotus Notes and weighed up their options for consolidating the code base. I expect they looked at cross-platform widget sets like QT or wxWidgets, AJAX,.NET and Java / Eclipse. AJAX is probably too light,.NET is a waste of time for cross-platform, and widgets still mean you rely on C++ for all its faults. The attractiveness of Eclipse is that you already get a very mature workspace framework for nothing, virtually 99% reusable code and SWT ensures very good native platform performance.
I'm actually amazed that Java isn't used much more for big projects. Client UIs have always been a problem for Java. Swing is okay, but SWT is already very mature and native. I expect IBM saw the robust workspace platform they played no small part in creating and thought it a good way to reboot the Lotus Notes experience.
It would be interesting to quantify how much wasted time and energy were caused by slow bootups. To me, it makes absolutely no sense that people leave their machines on overnight and perhaps the reason for this is because it then takes 5 mins in the morning for it to start again. I reckon that flash ram in hard drives + versions of Windows / Linux that enable sleep mode (not standby, sleep) by default would save hundreds of millions if not billions of dollars in wasted energy each year.
WINE is a handy tool, but apps that run under it suck. They suck because use Windows conventions for file names, registry settings, paths, menu structure and look and feel. Even if you recompiled the app using winelib they would still suck. After a lot of effort you might be able to hack some of the Windows-ness out of it, but it would be a nightmare to maintain.
But in yet another defeat for personal freedom (and another win for the moneyed interests), the courts have found that this is a violation of copyright law.
What about the personal freedoms of the copyright holder who doesn't want people selling bastardized copies of their intellectual property?
Perhaps you are being contrary, perhaps you are being stupid. If you read what I said, you would realise there are far easier ways to accomplish the same thing than individually walking through a movie, one screenshot at a time.
This is kind of a stupid exploit all right. Yes it's possible to capture screen, just as it's possible to do the same with every e-book. The next question is who is going to bother to do it. It seems far easier to just hack the software player, or wait for the inevitable dongle which streams HDCP to any device of your choosing. That's even assuming your average pirate would even be bothered to go to those lengths. I'm sure even a downsampled image on a non-copy protected device is more than adequate for viewing and ripping purposes.
I'm not American, but my understanding of the term "redneck" is that it refers to a certain underclass of uneducated, criminally inclined, trailer park dwelling white folks. Doesn't sound very racist to me.
Reverse engineering the Windows API is hard without changes. WINE has been at it for years, and they still haven't gotten to the point that most Win32 apps will run seamlessly over Linux. Part of the problem is that "Windows API" sounds like some neat little term when it fact you're talking tens of thousands of function calls spanning hundreds of DLLs, plus thousands more registry settings on top. And that's just to get you to Windows 95/98 functionality or so. On top of that you've got COM, DCOM, OLE2, DirectX, humoungous components like the IE,.NET, ODBC, the list goes on and on. You will never get to reverse engineer it all. Never.
You might get pretty good emulation, but unless Apple licenced the entire source code of Windows, they'd never get it to work seamlessly. It's such a big deal that it would make more sense if OS X supported some kind of virtualisation so that Microsoft could engineer a version of XP or Vista that sat on top of it and ran at near native speeds.
A more relevant question - what damned difference does it matter what it runs, just as long as it works?
Linux PDAs have traditionally been cursed with buggy software, awful handwriting recognition, crashes and high prices. But hey, it runs Linux right? Certainly Linux is not the cause of these issues, but it seems symptomatic of certain manufacturers that they think they can release some junk and get people to buy it simply because it runs Linux.
Fortunately we're getting to the stage where Linux is reliable and mostly behind the scenes. What OS is running underneath is an irrelevance to most people. They'd rather that their PDA / phone did what it was meant to do, namely make calls, take notes, make appointments, store addresses etc. If it runs Linux then all well and good, but a piece of crap running Linux is still a piece of crap.
So you'd rather they ship an empty OS? or include 5 dvds of every single competition that they can legally redistribute? I don't see where theres a better alternative.
A better question is why a full blown media player, music ripper, web browser, or other application should even be considered part of the OS. But if Microsoft are going to ship such things and freeze out the competition, then yes they should be required to ship alternatives on their DVD. Since they have 1.2Gb of space left even on the largest Vista dist, this does not seem unreasonable. Alternatively they could put a very straightforward post-installation step which offers to download their own, or competitor's products from the web or from a supplemental disk.
Neither option is insurmountable or technically challenging.
I'm all in favor of Microsoft opening some of their interfaces, but let's face it -- there is NOTHING that Microsoft can do with their apps that a third party can't. It just requires the third party to write more software that looks like native stuff. Big deal.
Yes there is. Microsoft can install their software by default on people's machines. This in itself enough that the vast majority of users will never go looking for anything else to use. Competition doesn't merely suffer - it dies on the vine.
True of GTA 1 & 2, possibly true of GTA 3+ on the PS2. Certainly not true on the PC. One of the most annoying aspects of playing GTA 3 / VC / SA on the PC is that you can turn 360 degrees and everything behind you - vehicles, pedestrians etc. has respawned as something else. The only things that pass the rotate test are stolen / wrecked cars, enemies & bodies, but they too have some low timeout. If your typical MMPORG (e.g. WoW) can have literally hundreds of player chars, all individual and replete in their own kit on screen at once, then I don't see a technical constraint with GTA being able to track a couple of hundred peds and cars in your immediate vicinity.
The reason that it exists I feel has more to do with GTA being a port. But if they can revamp the graphics and controller, then why not the game engine? Here's hoping the next gen game is attempts to resolve these issues. The driving aspect is actually very good, but the environment is too static and predictable. It needs smarter peds, less obvious spawning, proper indoor areas, destructable landscapes and long term memory of the mayhem you've caused.
Hmm, it's hard to see how anyone can claim Firefox is ugly in any sense. It's minimalist, uncluttered, intuitive and attractive. It even works fairly well on OS X since it picks up the native look & feel through the theme engine.
That might work but the next question is why is the button so wide if the only place it pushes properly is in the middle? Doesn't sound like good design to me.
It's hard to see MS Vista being out by year end. Beta 2 of Vista is pretty a rough piece of software. The UAC is a disaster zone (and infuriating), the thing is a memory hog, the UI is a dog's dinner of conflicting look & feels and major distributables such as MSDE,.NET 1.1 don't even work properly. Aside from a few extra apps for calendaring and DVD authoring (which could have release as an add-on for XP), there is little to see that is compelling about it.
I was actually quite looking forward to Vista until I tried it.
Return the thing for a replacement or refund. Or a class action lawsuit if you feel that strongly. These things are only a month old and people are resorting to potentially warranty invalidating hacks to fix overheating, CPU whine, case yellowing and now squidgy button issues when they should be kicking Apple for releasing a lemon.
The core developers don't have to be involved. Every major dist maintains their own kernel with patches. The Red Hats and Novells of this world could esily ensure that Linus et al never have to deal with a binary driver. They would write and maintain the ABI over the kernel and ensure that any differences were hidden from both sides. I doubt the kernel changes significantly in a major release cycle (e.g. 2.6) that they couldn't maintain and provide an ABI that did this.
There are philosophical objecttions to binary drivers but the reality is that many OEMs won't open source their drivers for very sensible business reasons. Even if they were open source, very few users are capable or inclined to build a driver module, especially if their dist doesn't even contain a toolchain. Binary drivers are essential if Linux is ever to conquer the home / consumer market.
The problem with Linux comes when it doesn't support your hardware. With windows you pop in a disk and the driver installs. On Linux you go without and hope someone will put you out of your misery by writing a driver. Mostly my experience has been good but I've had two occasions where I spent hours and hours of searching and googling, desperately trying to get a sound chipset to work and a wireless card. I managed to get neither of them to work although both were allegedly supported.
Linux desperately needs a decent binary driver layer, one which is common to all dists with a common packaging system so that manufacturers can ship a driver which works on all of them. I'm actually surprised that NVidia, ATI, VMWare and the dists don't band together to produce such a thing. It must be a pain for all concerned in the current model.
"the device could conceivably serve as an everyday tool as common as the mobile phone or the palmtop."
It could but it ain't going to happen. No one but the most extreme, purist form of nerd is going to be seen dead wearing one of these things. Might have applications in military or warehousing, but those kind of people would probably as happy to have something that sits in a holster until required.
If there was an XP2004 and an XP2006 released, you wouldn't see the bitching. XP's biggest problem at this point is just that it's old and clunky.
Old and clunky is sometimes a good thing, assuming it works to begin with. Where I work, they're still using Java 1.3 based JRun for a lot of server code because it works. Upgrading or moving to Websphere implies a lot risk, usually in getting the config to behave properly. Perversely they're embracing.NET when it's already gone through 1.0, 1.1 and now 2.0 - I don't think they quite realise that they're creating a big headache for themselves.
But anyway the same is true for operating systems, particularly for enterprise servers. Companies already despise the merry-go-round of upgrades that MS put them through for operating systems and Office. My company actually upgraded me to Office 2003 recently which shows how far behind they are. This is probably why Microsoft go for longer periods without an upgrade and then make companies deal with the shitstorm when it hits in their own time. Vista is going to be a category 10 shitstorm.
Apple has a different strategy - release lots of times, which in my view has good and bad points. It's good since it means more incremental and probably more stable updates. It's bad because you're basically paying for a point upgrade and "features" like Dashboard or rejigging Finder (again) which Apple tack on to make the upgrade look more compelling. It also means that apps constantly expect and only run on the last few releases and you're forced onto an upgrade treadmill if you want the new stuff. Apple get away with it because they're dealing with consumers and toy server installs. They have the lee-way to do things that neither Microsoft or Linux / Unix vendors can afford to do.
Anyway, Microsoft still release lots of interim technology things, and still do. New versions of DirectX, IE,.NET and so on don't need to be tied to the OS, and indeed I don't believe they should be. Generally speaking Windows upgrades thus far have not been too bad, but judging from my experiences with Vista beta 2, I think MS may be in a spot of trouble.
I agree that encryption usability is terrible. I think X509 is terrible and that's *integrated* into the UI. I think PGP / GPG encryption would actually far easier to work into a UI since there is no requirement to get a cert from anywhere.
I don't think crypto is going to be ubiquitous while it is the mess it is now. I think PGP has a very good UI, but it costs money (though 8.0 is free from pgpi.com). I think GPG has terrible UI (none at all), and what tools that exist to put a UI on it such as WinPT are pretty sucky too. I don't think GPG did itself any favours by only shipping as an executable-only format using stdin & stdout to report everything. GPG *needs* a reentrant DLL version of itself which allows front-ends to talk in terms of structures and error messages rather than decipher what comes out of stdin. I've seen wrapper libs around gpg but it's not enough, it needs a reentrant version. Not only would this make the thing work better from the UI, but key servers etc. would be much faster too.
Still, once you get Enigmail working with GPG, it's an excellent crypto system. I have a customer who insists on talking to me with OpenPGP and Enigmail makes it very easy. IIRC crypto speed and file sizes are far better in OpenPGP over X509 too.
I think it would be very difficult to enforce a ban on P2P for a number of reasons. First, most P2P networks run off any port. If certain ports (e.g. the BT ones) are monitored, you can bet your boots that the clients will soon use random ones. Secondly, it would be straightforward enough to add a bit of crypto to the mix, i.e. prevent packet sniffers seeing *what* two hosts are passing between them. Thirdly, if necessary P2P software could even mask out IP addresses known to reside inside Spain from other Spanish up/downloaders. So all the ISP would know is that you're moving a lot of encrypted traffic in and out of Spain from another country but it would be exceedinly difficult to prevent it, unless by misfortune you were talking to some kind of honeypot set up by the RIAA or whoever.
That in my view is the biggest problem with P2P - honeypots. It would be trivial to monitor IP addresses who request a certain file off your machine. Once they've downloaded a few chunks, you start gathering stats. When the person in question tops your list, perhaps on multiple requests you email your counterparts in those countries for prosecution. Therefore it seems to me that the future of P2P relies on a Freenet-lite kind of model, being able to cache encrypted chunks of popular data, random data exchange between nodes, introducing a couple of hops for requests and so on. Not something so anal about anonymity that it runs like a sloth, but something that makes the burden of proof far, far harder for people who are monitoring these networks.
I know myself as a Windows developer that WINE can be an invaluable resource to understanding how the Win32 API works. For the most part the APIs in Windows are fairly well documented, but occasionally you'll come across one which takes a LPVOID which could be one of many structures, all of which have flags, and size members and not enough documentation, or it doesn't make clear who allocates or releases what, or has some flag with a meaningless name and description. WINE can help a lot in those situations.
I guess MS Engineers are no different. Sometimes looking at someone else's implementations of their own APIs can be less confusing that attempting to make sense of the garbled 10 year old documentation that was written in-house or not even locatable. Still, I would have thought MS would be slapping their staff hard for doing that, for fear of somehow "infecting" their own code by mistake.
As for MS and this process. I can well imagine that if you are faced with providing documentation that you make sure it's written by a remedial English class, translated into Mandarin, then to Urdu and finally back into English. Then ensure to sort in a useful way such as by how many vowels there are on each page and copiously hyperlink every single instance of "and", "then", "if" but nothing else.
X.509 is a heap a crap for a number of reasons, but top of my list is that obtaining a cert for it is such a monumental bother that it renders the whole exercise pointless. You can either pay for a cert or go a-hunting for a free cert, but either way, the CAs want everything but a stool sample before they issue you with a cert that expires in anywhwere from 60 days to a year away. If you're lucky you might find a cert that lasts longer, but all that pain and suffering gets you a cert which the CAs do not validate or vouch for in any way.
Conversely, all it takes to to use PGP is a few seconds with the wizard to generate your key and you're all set. It doesn't ask you for your passport number or your ssn, or your birthday or anything else. Integration with mail readers is slightly harder, but solutions exist for PGP and GPG for most popular readers. Once you have the key you're all set to use it for as long as you like. If you're paranoid about impersonation you can even get a few friends to sign your key.
I have 2.5 harddrive in my laptop, 40gb in size. I set the defragger off on Friday night and several hours later it hadn't even gotten 1/3 the way through. My laptop doesn't have a very stressful life - it has some Mozilla source on there alright, but mostly it boots up, connects to the internet and shuts down.
That means to me that NTFS has serious real world shortcomings, not only for forcing me to defrag the disk but allowing it to degrade so much that performance takes a dive and it takes a protracted session to set it right. I don't understand why any FS in this day and age requires manual defragmentation. I'm not singling out ntfs, except because it has the least excuse of any FS. To me, it makes no sense at all since it *must* be possible for the driver or the OS to reasonably detect inactivity and be able to do some defragging in the background. I wonder how many man years have been lost to sluggish performance or defrag sessions thanks to MS. I'm sure that some Unix systems are as bad, but I suspect that they don't have nearly the issues of ntfs.
Slashdot Mirror
Maintaining separate code bases for multiple platforms is a nightmare. I expect IBM have come to that realisation too for Lotus Notes and weighed up their options for consolidating the code base. I expect they looked at cross-platform widget sets like QT or wxWidgets, AJAX, .NET and Java / Eclipse. AJAX is probably too light, .NET is a waste of time for cross-platform, and widgets still mean you rely on C++ for all its faults. The attractiveness of Eclipse is that you already get a very mature workspace framework for nothing, virtually 99% reusable code and SWT ensures very good native platform performance.
I'm actually amazed that Java isn't used much more for big projects. Client UIs have always been a problem for Java. Swing is okay, but SWT is already very mature and native. I expect IBM saw the robust workspace platform they played no small part in creating and thought it a good way to reboot the Lotus Notes experience.
It would be interesting to quantify how much wasted time and energy were caused by slow bootups. To me, it makes absolutely no sense that people leave their machines on overnight and perhaps the reason for this is because it then takes 5 mins in the morning for it to start again. I reckon that flash ram in hard drives + versions of Windows / Linux that enable sleep mode (not standby, sleep) by default would save hundreds of millions if not billions of dollars in wasted energy each year.
WINE is a handy tool, but apps that run under it suck. They suck because use Windows conventions for file names, registry settings, paths, menu structure and look and feel. Even if you recompiled the app using winelib they would still suck. After a lot of effort you might be able to hack some of the Windows-ness out of it, but it would be a nightmare to maintain.
What about the personal freedoms of the copyright holder who doesn't want people selling bastardized copies of their intellectual property?
Perhaps you are being contrary, perhaps you are being stupid. If you read what I said, you would realise there are far easier ways to accomplish the same thing than individually walking through a movie, one screenshot at a time.
This is kind of a stupid exploit all right. Yes it's possible to capture screen, just as it's possible to do the same with every e-book. The next question is who is going to bother to do it. It seems far easier to just hack the software player, or wait for the inevitable dongle which streams HDCP to any device of your choosing. That's even assuming your average pirate would even be bothered to go to those lengths. I'm sure even a downsampled image on a non-copy protected device is more than adequate for viewing and ripping purposes.
I'm not American, but my understanding of the term "redneck" is that it refers to a certain underclass of uneducated, criminally inclined, trailer park dwelling white folks. Doesn't sound very racist to me.
You might get pretty good emulation, but unless Apple licenced the entire source code of Windows, they'd never get it to work seamlessly. It's such a big deal that it would make more sense if OS X supported some kind of virtualisation so that Microsoft could engineer a version of XP or Vista that sat on top of it and ran at near native speeds.
Linux PDAs have traditionally been cursed with buggy software, awful handwriting recognition, crashes and high prices. But hey, it runs Linux right? Certainly Linux is not the cause of these issues, but it seems symptomatic of certain manufacturers that they think they can release some junk and get people to buy it simply because it runs Linux.
Fortunately we're getting to the stage where Linux is reliable and mostly behind the scenes. What OS is running underneath is an irrelevance to most people. They'd rather that their PDA / phone did what it was meant to do, namely make calls, take notes, make appointments, store addresses etc. If it runs Linux then all well and good, but a piece of crap running Linux is still a piece of crap.
A better question is why a full blown media player, music ripper, web browser, or other application should even be considered part of the OS. But if Microsoft are going to ship such things and freeze out the competition, then yes they should be required to ship alternatives on their DVD. Since they have 1.2Gb of space left even on the largest Vista dist, this does not seem unreasonable. Alternatively they could put a very straightforward post-installation step which offers to download their own, or competitor's products from the web or from a supplemental disk.
Neither option is insurmountable or technically challenging.
Yes there is. Microsoft can install their software by default on people's machines. This in itself enough that the vast majority of users will never go looking for anything else to use. Competition doesn't merely suffer - it dies on the vine.
True of GTA 1 & 2, possibly true of GTA 3+ on the PS2. Certainly not true on the PC. One of the most annoying aspects of playing GTA 3 / VC / SA on the PC is that you can turn 360 degrees and everything behind you - vehicles, pedestrians etc. has respawned as something else. The only things that pass the rotate test are stolen / wrecked cars, enemies & bodies, but they too have some low timeout. If your typical MMPORG (e.g. WoW) can have literally hundreds of player chars, all individual and replete in their own kit on screen at once, then I don't see a technical constraint with GTA being able to track a couple of hundred peds and cars in your immediate vicinity.
The reason that it exists I feel has more to do with GTA being a port. But if they can revamp the graphics and controller, then why not the game engine? Here's hoping the next gen game is attempts to resolve these issues. The driving aspect is actually very good, but the environment is too static and predictable. It needs smarter peds, less obvious spawning, proper indoor areas, destructable landscapes and long term memory of the mayhem you've caused.
Hmm, it's hard to see how anyone can claim Firefox is ugly in any sense. It's minimalist, uncluttered, intuitive and attractive. It even works fairly well on OS X since it picks up the native look & feel through the theme engine.
That might work but the next question is why is the button so wide if the only place it pushes properly is in the middle? Doesn't sound like good design to me.
I was actually quite looking forward to Vista until I tried it.
Return the thing for a replacement or refund. Or a class action lawsuit if you feel that strongly. These things are only a month old and people are resorting to potentially warranty invalidating hacks to fix overheating, CPU whine, case yellowing and now squidgy button issues when they should be kicking Apple for releasing a lemon.
There are philosophical objecttions to binary drivers but the reality is that many OEMs won't open source their drivers for very sensible business reasons. Even if they were open source, very few users are capable or inclined to build a driver module, especially if their dist doesn't even contain a toolchain. Binary drivers are essential if Linux is ever to conquer the home / consumer market.
Linux desperately needs a decent binary driver layer, one which is common to all dists with a common packaging system so that manufacturers can ship a driver which works on all of them. I'm actually surprised that NVidia, ATI, VMWare and the dists don't band together to produce such a thing. It must be a pain for all concerned in the current model.
It could but it ain't going to happen. No one but the most extreme, purist form of nerd is going to be seen dead wearing one of these things. Might have applications in military or warehousing, but those kind of people would probably as happy to have something that sits in a holster until required.
Old and clunky is sometimes a good thing, assuming it works to begin with. Where I work, they're still using Java 1.3 based JRun for a lot of server code because it works. Upgrading or moving to Websphere implies a lot risk, usually in getting the config to behave properly. Perversely they're embracing .NET when it's already gone through 1.0, 1.1 and now 2.0 - I don't think they quite realise that they're creating a big headache for themselves.
But anyway the same is true for operating systems, particularly for enterprise servers. Companies already despise the merry-go-round of upgrades that MS put them through for operating systems and Office. My company actually upgraded me to Office 2003 recently which shows how far behind they are. This is probably why Microsoft go for longer periods without an upgrade and then make companies deal with the shitstorm when it hits in their own time. Vista is going to be a category 10 shitstorm.
Apple has a different strategy - release lots of times, which in my view has good and bad points. It's good since it means more incremental and probably more stable updates. It's bad because you're basically paying for a point upgrade and "features" like Dashboard or rejigging Finder (again) which Apple tack on to make the upgrade look more compelling. It also means that apps constantly expect and only run on the last few releases and you're forced onto an upgrade treadmill if you want the new stuff. Apple get away with it because they're dealing with consumers and toy server installs. They have the lee-way to do things that neither Microsoft or Linux / Unix vendors can afford to do.
Anyway, Microsoft still release lots of interim technology things, and still do. New versions of DirectX, IE, .NET and so on don't need to be tied to the OS, and indeed I don't believe they should be. Generally speaking Windows upgrades thus far have not been too bad, but judging from my experiences with Vista beta 2, I think MS may be in a spot of trouble.
I don't think crypto is going to be ubiquitous while it is the mess it is now. I think PGP has a very good UI, but it costs money (though 8.0 is free from pgpi.com). I think GPG has terrible UI (none at all), and what tools that exist to put a UI on it such as WinPT are pretty sucky too. I don't think GPG did itself any favours by only shipping as an executable-only format using stdin & stdout to report everything. GPG *needs* a reentrant DLL version of itself which allows front-ends to talk in terms of structures and error messages rather than decipher what comes out of stdin. I've seen wrapper libs around gpg but it's not enough, it needs a reentrant version. Not only would this make the thing work better from the UI, but key servers etc. would be much faster too.
Still, once you get Enigmail working with GPG, it's an excellent crypto system. I have a customer who insists on talking to me with OpenPGP and Enigmail makes it very easy. IIRC crypto speed and file sizes are far better in OpenPGP over X509 too.
That in my view is the biggest problem with P2P - honeypots. It would be trivial to monitor IP addresses who request a certain file off your machine. Once they've downloaded a few chunks, you start gathering stats. When the person in question tops your list, perhaps on multiple requests you email your counterparts in those countries for prosecution. Therefore it seems to me that the future of P2P relies on a Freenet-lite kind of model, being able to cache encrypted chunks of popular data, random data exchange between nodes, introducing a couple of hops for requests and so on. Not something so anal about anonymity that it runs like a sloth, but something that makes the burden of proof far, far harder for people who are monitoring these networks.
I guess MS Engineers are no different. Sometimes looking at someone else's implementations of their own APIs can be less confusing that attempting to make sense of the garbled 10 year old documentation that was written in-house or not even locatable. Still, I would have thought MS would be slapping their staff hard for doing that, for fear of somehow "infecting" their own code by mistake.
As for MS and this process. I can well imagine that if you are faced with providing documentation that you make sure it's written by a remedial English class, translated into Mandarin, then to Urdu and finally back into English. Then ensure to sort in a useful way such as by how many vowels there are on each page and copiously hyperlink every single instance of "and", "then", "if" but nothing else.
Conversely, all it takes to to use PGP is a few seconds with the wizard to generate your key and you're all set. It doesn't ask you for your passport number or your ssn, or your birthday or anything else. Integration with mail readers is slightly harder, but solutions exist for PGP and GPG for most popular readers. Once you have the key you're all set to use it for as long as you like. If you're paranoid about impersonation you can even get a few friends to sign your key.
That means to me that NTFS has serious real world shortcomings, not only for forcing me to defrag the disk but allowing it to degrade so much that performance takes a dive and it takes a protracted session to set it right. I don't understand why any FS in this day and age requires manual defragmentation. I'm not singling out ntfs, except because it has the least excuse of any FS. To me, it makes no sense at all since it *must* be possible for the driver or the OS to reasonably detect inactivity and be able to do some defragging in the background. I wonder how many man years have been lost to sluggish performance or defrag sessions thanks to MS. I'm sure that some Unix systems are as bad, but I suspect that they don't have nearly the issues of ntfs.