I'll admit that I've never opened an ODF spreadsheet in Office 2007, and for that matter I've never used the SP2-provided ODF support before, but the Microsoft-funded ODF Convertor plugin (open source, look on Sourceforge) has opened every.odt I've handed it, and the files it produces have worked with OO.o just fine. So... did they break the plugin, is this just a problem with spreadsheets (in which case the summary is extremely misleading), or is the spec really that vague?
It's actually somewhat better than that. I'm not claiming that the POSIX subsystem is *as* complete as an average Linux distro, or as OS X (which actually counts as UNIX, something that NT's POSIX subsystem is far from). However, I've used it to develop software for a Linux device (running on ARM, for that matter) and it was neither difficult nor poorly implemented. The only part I had to use #IFDEF __INTERIX (a macro defined for the version of GCC on the subsystem) for was some non-standard ways of retrieving and setting IP addresses, which the two systems had no common mechanisms for doing.
Most open-source userspace code, unless specifically written with non-standard Linux system calls or similar, will also compile and run on Interix with a minimum of fuss. It's interesting comparing the output of./configure to that of a typical Linux system, but many applications are in fact source-compatible.
The funny thing is, XP isn't that great of an OS. It's a bitch to use as a standard user. It lacks security features that a platform as freqeuntly targeted as Windows really should have, like ASLR. Its scheduler and memory manager algorithms are years out of date, and cause all sorts of problems such as the case where if you don't use your computer for a long time, you come back and find that all your programs have been trimmed to a tiny working set and everything is paged to disk. Its firewall sucks. Its 64-bit support is barely present. At least half of the standard complaints against "Windows" on Slashdot only actually apply to XP and down (I'm not counting ones aimed at a specific version of Windows) and yet people keep treating an 8-year-old OS as The One To Use, even as they complain about all the things that are wrong with it.
Right... because flaws in non-security areas (the pre-release code does actually get security patches for the duration of the pre-release, BTW) are far more likely to get you some malware than a prated copy with $DEITY-only-knows-what slipstreamed into it. I mean, why bother trying to infect people with rootkits when you can just hand out copies that are already infected?
looked at things that went wrong with Vista launch
isn't how I would describe *this* decision, at least. Vista was also freely available in RC form for a year or so. The people who had trouble running the RTM had trouble running the RC (or would have if they'd tried) but the rest of us - myself included - didn't upgrade to RTM for a few months after its release simply because the RC2 did what I needed and was less hassle (cost didn't matter, as a CSE student I get MS software free anyhow).
Acknowledging that you were trying to be funny, Win7 (and to a slightly lesser extent Vista) have pretty much removed the whole "hunt down and install drivers" part of upgrading. On first boot, it asks you for permission to check Windows Update for drivers. If you accept (and I've yet to come across any reason to do so) it will happily pull drivers for every piece of hardware you have, often even including really esoteric stuff (the network-over-usb driver to interface with a Linux-based iMote2 embedded device was in Windows Update. Color me impressed...)
For certain things like video cards you might still want to select your driver version and userspace utilities from the web site, but for most people it is quite seamless. The drivers will all install together too; there's a decent chance (although certainly no guarantee) that you'll need to reboot once - but I've yet to need to reboot twice.
My issue with games like God of War on portable consoles isn't that you physically can't suspend them (that would be an atrocious design failure) but rather that it's too disruptive to the gameplay. Like most action/adventure games, God of War tries to keep the excitement level high and have minimal downtime, which makes sense in a game you want people to spend hours playing.
However, it doesn't make for a game that I find easy to put down at any time, or pick up again later if I had to put it down in a hurry. It's a game where if you have to go somewhere in half an hour, you start looking for a good place to stop - not a game where if you have to get off the bus in 30 seconds you (or rather, I) simply pause and pick up later. You might not have any problem with doing so, but it's not how I like to play games, and I'm not unique in that.
I don't own a PSP or a DS, but I do play games on portable devices. It seems that the most suitable games are really either the ones that take only a few minutes such as you might have on a bus ride and can be abandoned at any time, or ongoing games which can be paused and resumed without the temporal disconnect being too difficult.
The first category might be similar to the kinds of Flash games people play online, or to puzzle challenge games like Audiosurf (I'd love a portable version of this game!). It seems that this type of game should be avaialable in vast quantities - consider the iPhone App Store for an example - but I'm not sure they're commercially viable on a handheld that is expensive to develop for.
The second category suggests RPGs or even point-and-click adventures (Myst would have been fun on a portable), or turn-based strategy. The world should be relatively static while you're idle, and it should be possible to stop from any given point pretty easily. Many of the DS games I've seen seem to be of this category.
So... why is it that so many of the PSP games seem to be straight-up ports of traditional console games, meant to be played for hours (or at least 30 minutes) at a time, and to pack as much excitement as possible into that time? I realize designing a successful game is difficult (and once you've done it, you want to put it on as many platforms as you can), but surely a portable game that naturally takes a break every few minutes (and can be freely abandoned even between breaks), or that can be paused and resumed at any time without needing to remember exactly what you were doing right before, would be best? It seems odd that portable games so rarely give the impression of having been designed for portable gaming.
Fighting games kind of fit the first category, so long as the fights are REALLY short or you can drop out in the middle of a fight and do it later without penalty. Same for racing games. Sports games don't seem to fit well - too much stuff to do over the course of a period too long for one sitting, all of which builds on stuff before it. RPGs consisting of a lot of real-time action don't either. Real-time strategy games certainly don't (although they're rare on consoles anyhow). Things like Pokemon and Legend of Zelda games do make sense, but they don't seem to be the kind of thing you see for PSP.
IE8 has the ability to render using IE7's engine (either with a meta tag in the site, or via the "Comaptibility Mode" button, or - with greater granularity - through the development tools). For testing stuff older than 7, I really can't say. Incidentally, for web dev, the develpment tools are pretty sweet. A couple things in there Firebug could learn from.
You make a good point, except for one thing: ODF is available in MS Office as well. It's a open-source plugin (sponsored by Microsoft, hosted on Sourceforge) and integrates pretty nicely. I've been using it all the way back to when "Office 12" (as it was then called) was in beta, and I've yet to find a file it couldn't open correctly, or one that it saved which opened incorrectly on another office suite.
Office 2007 (and apparently 2003 and XP, although I haven't tried on them) has had an optional plugin for 2+ years that enabled ODF support (for all the od* formats that I'm aware of, at least). It's an open-source project sponsored by Microsoft and listed on Sourceforge: http://odf-converter.sourceforge.net/
I use it to open/convert, edit, and save/export ODF files, and it works fine. Mind, if you prefer OO.o over Office 2007 (I don't, even ignoring the risk of OO.o getting the MSOffice formats wring - something which has never happened using the odf-converter plugin, BTW) then this isn't for you. I'm sure there are other people who would find this useful, though.
For the TL;DR folks: Yes, you can run and compile POSIX apps in a *nix userspace on Windows. For more info, see http://suacommunity.com/
----
Yes, it has since the origin of NT. The NT kernel as we know it was started back when OS/2 was seen as The OS Of The Future, but OS/2 was slow getting started so Microsoft worked on their own thing on the side. They wanted it to be compatible with OS/2 though, so they included an OS/2 subsystem. UNIX was really big at the time also, so they included a POSIX subsystem as well. DOS and Win16 were already doing fairly well, so they included the NTVDM (Virtual DOS machine) with Win16 libraries to support 16-bit programs. Of course, they also included the Win32 API that is what people usually mean when the talk about the "Windows" API.
The OS/2 subsystem was discontinued after Windows 2000, and NTVDM isn't present in x64 versions of NT. However, the POSIX subsystem - including a BSD-derived userspace and GNU build toolchain - are still available in the higher editions of NT. On XP, you need Pro, on Vista you need Ultimate or Enterprise (all server editions should support it). On XP, you first install SFU (Services For UNIX) from Microsoft, which enables the subsystem and installs a basic userland called Interix. On Vista, enabling the subsystem (called SUA, Subsystem for UNIX-based Applications) and installing Interix are separate steps - click on "Turn Windows features on or off" from the Programs and Features control panel, then install "Utilities for Subsystem for UNIX-based Applications" for the Interix libraries, utilities, and build tools.
A package manager providing pre-compiled binaries is available from SUA Community (http://suacommunity.com), or you can use NetBSD's pkgsrc package manager and compile from source, or you can do the standard unpack tarball,./configure && make && sudo make install routine. I use bash as my standard shell in Windows these days, use ssh (including sshd to connect from remote machines) and svn extensively. I've also used Interix to develop and test code for an embedded device that runs Linux; the source compatibility is pretty good.
Wine doesn't solve the problem at all; the reason legacy apps broke on Vista (and I mean really broke, not just need to be run as Administrator) is because Microsoft closed a lot of holes in their APIs and fixed a lot of non-spec behavior that people had relied upon for years (I shit you not, in XP there's memory management code to use an allocation scheme that doesn't explode on double-free for a specific game from over a decade back that relied or the behavior of an old Windows library). Wine doesn't really have those quirks supported; you can tell it to pretend to be a given Windows version for any particular application, but that's about it.
Same apps also tried to install drivers that would do crazy shit in kernel mode (Antivirus, I'm looking at you - but also random other programs like a nautical chart-plotter?!?) which Vista didn't like. These apps wither would work at all or would cause BSODs due to their drivers trying to do kernel-mode stuff that shouldn't have worked in the first place. Since wine supports almost no driver API, it's a pretty safe bet these wouldn't work either.
Home can't connect to Active Directory, so nobody in business would use it anyhow.
I suppose it's vaguely possible that home users will latch onto this feature and buy the higher editions of Win7 to get it, but honestly, I don't think so. The average home user doesn't even know about the application compatibility modes that all versions of Windows since 98 or so have included. I've had people think it was the most amazing thing when I got an XP program to run on Vista simply by running it as Administrator (not that it had a legit reason to, just modifications it made to the install location). People, on average, know very little about running a program beyond click the icon. Installing an optional virtual machine for compatibility is way beyond what a normal home user would do.
I don't know for sure if it can be done while the client is active, but Win7 has built-in capability to mount.VHD files into its filesystem. In theory this makes it quite possible for an AV program on the host to scan the drive of the client.
One of the features of Win7 that was announced early was that it can mount.VHD (Virtual Hard Disk, the format used by Virtual PC) natively (it can even boot off one, so long as the bootloader is on a real partition). So yes, the host AV *should* be able to protect the virtual system.
Firewall is just ridiculous; filter the VPC connection through the host (Win7) network interface, and the host's firewall is the guest's firewall. In fact, on current versions on VPC, if you want to connect the client to a network *without* running it through the host firewall, you need a dedicated NIC (i.e. the host can't connect via that interface).
I'm fairly sure the laws preventing defamation apply at the very least to publicly help organizations and members of goverment organizations, although I'm less sure about government as a complete administration.
In any case, I still disagree that free speech is an acceptable defense in a situation where the only purpose of a communication is to cause unjustified harm to another. I would actually see it as more justified - probably also more harmful, but also more justified - if the captioned pictures had instead been posted publicly; that would at least have indicated that they were intended for humor. As it is, they were sent privately to an individual for whom they would be extremely painful. What you seem to be advocating is not freedom of expression, but freedom to cause emotional anguish.
Consider this: Suppose I developed a device that could determine the image its target would find the most painful to see, and would force the target to see it. Would you defend my right to use this device on unwilling strangers? Would you defend my right to use it on you? On your children? On your parents? I have no idea who you are and I strongly doubt we've ever met, but we're nonetheless probably more acquainted with one another (through this discussion) than whoever sent those photos was with the girl's father.
I hadn't heard of this; it sounds pretty good and I'll check it out. Does anybody know whether the sleep is as "deep" (i.e. can I leave my laptop in sleep for 48 hours without draining the battery) and whether it handles ACPI events (closing or opening the laptop lid, wake-on-LAN, or similar)? I'll take a look anyhow; lack of a suspend state is a real annoyance with a student laptop.
For search, I was referring to the the indexed search tools which ship with Vista as opposed to those installed by default on my Linux boxes. The Vista one is faster by a noticable margin even though the index is larger.
For performance, you're talking about a scheduler design decision here: NT embraced and encouraged kernel-level threads in its core design, and bases scheduling decisions on threads (this makes a certain degree of sense, as context switching between threads is slightly faster than between processes, and creating new threads is less expensive than creating new processes). I'm not sure about Vista and up (NT 6.x) but the 5.x kernel (2000, XP, 2003) will certainly allocate more processor time to applications which use many threads. I wonder, however, whether such a hightly multi-threaded application on NT would perform as well as expected on Linux? Not saying it wouldn't but you didn't specify that it would; you simply picked an example of a case which supported your argument by using the preferred programming paradigm of Linux (POSIX, really), which is not preferred on Windows.
Amen, with one caveat: if the pictures had been completely anonymous (difficult, but a reasonable effort could be made) then no harm would have been done. They family still might have disapproved if it was done without their knowledge, but chances are they never would have known... because nobody would have been able to look up the email address of the girl's father and send him the sickest excuse for a "joke" (giving him the benefit of a doubt that it *might* have been intended as humor; it's possible some people are that tasteless) I've ever heard of.
Let me make that really clear: as I see it, the major fault was not the distribution of the photos, but the identification of the victim in them. THAT is a fairly clear breach of privacy, and should be punished (as should the people who sent the emails, of course - even if it was intended as humor, such things probably count as criminal harassment, and certainly form the basis for a civil suit).
I would not defend their actions, but I would defend their right to take such action.
I find two things terribly wrong with this statement. One of them is that freedom of speech does not, and never has, been intended to allow an attack on private individuals. That's why we have laws covering things from harassment (which this would seem to certainly qualify as) to slander and libel (these don't apply here, but they help illustrate the point where freedom of speech crosses the line). You are free to speak out against a government or other public institution, although you must be able to back up what you say. You are unlikely to wind up in court for privately attacking somebody verbally, although it can happen. In a case like this, though, the sender of those emails ought to face consequences - their actions had no purpose except to harm, you'd be crazy to say that they were not harmful, and there was no form of justification.
Second, regardless of whether the pictures got distributed, the *names* should NEVER have been without express consent. This is more about the cops who released the photos than it is about the assholes who sent the emails, but in any case there is no reason why anybody outside the police department and other organizations directly associated with the event would need to know who this girl was, who her father was, or any other information that could lead somebody to the family's email address. Anonymously distributing the pictures (without consent) as "an example of what a really bad crash can do" still seems a slightly grey area, morally, but there is no reason that they should be distributed without anonymity.
Sorry, but you, personally are missing the point with this line:
it's rather common as I knew a volunteer fireman who kept a photo album of accident pictures and he had shown them to people like it was his pride and joy
Did the pictures personally identify anybody? Was there sufficient information attached that you could find the email addresses of living relatives? If the answer is "no" then there's no harm done, although I personally still wouldn't put the pictures online without blurring out anything that *might* be identifiable. In this case, however, the answer was obviously "yes" which to me says that these people are either horribly insensitive or incredibly negligent (depending on whether it was by intention or accident) and frankly, in either case I think their ought to be consequences.
Or even VirtualPC, which is free and Microsoft (meaning has a known and more-or-less trusted name attched).
I find the complaint about not running those apps "in a secure mode" hilarious, though. Win9x had no secure mode; they were single-user OS's that allowed a few different profiles, and ran on a filesystem which doesn't even support permissions beyond the "I suggest you don't delete this" flag. Yes, security vs. convenience is frequently a trade-off situation; if you want to run programs which weren't designed for a secure system, you have to accept that they aren't secure.
Slashdot Mirror
Interesting, but is it open source? F/OSS OSF plugin for Office
I'll admit that I've never opened an ODF spreadsheet in Office 2007, and for that matter I've never used the SP2-provided ODF support before, but the Microsoft-funded ODF Convertor plugin (open source, look on Sourceforge) has opened every .odt I've handed it, and the files it produces have worked with OO.o just fine. So... did they break the plugin, is this just a problem with spreadsheets (in which case the summary is extremely misleading), or is the spec really that vague?
It's actually somewhat better than that. I'm not claiming that the POSIX subsystem is *as* complete as an average Linux distro, or as OS X (which actually counts as UNIX, something that NT's POSIX subsystem is far from). However, I've used it to develop software for a Linux device (running on ARM, for that matter) and it was neither difficult nor poorly implemented. The only part I had to use #IFDEF __INTERIX (a macro defined for the version of GCC on the subsystem) for was some non-standard ways of retrieving and setting IP addresses, which the two systems had no common mechanisms for doing.
Most open-source userspace code, unless specifically written with non-standard Linux system calls or similar, will also compile and run on Interix with a minimum of fuss. It's interesting comparing the output of ./configure to that of a typical Linux system, but many applications are in fact source-compatible.
sed s/"your own mistake".*\n/"a troll"/g
The whole thing is utterly bogus.
The funny thing is, XP isn't that great of an OS. It's a bitch to use as a standard user. It lacks security features that a platform as freqeuntly targeted as Windows really should have, like ASLR. Its scheduler and memory manager algorithms are years out of date, and cause all sorts of problems such as the case where if you don't use your computer for a long time, you come back and find that all your programs have been trimmed to a tiny working set and everything is paged to disk. Its firewall sucks. Its 64-bit support is barely present. At least half of the standard complaints against "Windows" on Slashdot only actually apply to XP and down (I'm not counting ones aimed at a specific version of Windows) and yet people keep treating an 8-year-old OS as The One To Use, even as they complain about all the things that are wrong with it.
Right... because flaws in non-security areas (the pre-release code does actually get security patches for the duration of the pre-release, BTW) are far more likely to get you some malware than a prated copy with $DEITY-only-knows-what slipstreamed into it. I mean, why bother trying to infect people with rootkits when you can just hand out copies that are already infected?
looked at things that went wrong with Vista launch
isn't how I would describe *this* decision, at least. Vista was also freely available in RC form for a year or so. The people who had trouble running the RTM had trouble running the RC (or would have if they'd tried) but the rest of us - myself included - didn't upgrade to RTM for a few months after its release simply because the RC2 did what I needed and was less hassle (cost didn't matter, as a CSE student I get MS software free anyhow).
Acknowledging that you were trying to be funny, Win7 (and to a slightly lesser extent Vista) have pretty much removed the whole "hunt down and install drivers" part of upgrading. On first boot, it asks you for permission to check Windows Update for drivers. If you accept (and I've yet to come across any reason to do so) it will happily pull drivers for every piece of hardware you have, often even including really esoteric stuff (the network-over-usb driver to interface with a Linux-based iMote2 embedded device was in Windows Update. Color me impressed...)
For certain things like video cards you might still want to select your driver version and userspace utilities from the web site, but for most people it is quite seamless. The drivers will all install together too; there's a decent chance (although certainly no guarantee) that you'll need to reboot once - but I've yet to need to reboot twice.
My issue with games like God of War on portable consoles isn't that you physically can't suspend them (that would be an atrocious design failure) but rather that it's too disruptive to the gameplay. Like most action/adventure games, God of War tries to keep the excitement level high and have minimal downtime, which makes sense in a game you want people to spend hours playing.
However, it doesn't make for a game that I find easy to put down at any time, or pick up again later if I had to put it down in a hurry. It's a game where if you have to go somewhere in half an hour, you start looking for a good place to stop - not a game where if you have to get off the bus in 30 seconds you (or rather, I) simply pause and pick up later. You might not have any problem with doing so, but it's not how I like to play games, and I'm not unique in that.
I don't own a PSP or a DS, but I do play games on portable devices. It seems that the most suitable games are really either the ones that take only a few minutes such as you might have on a bus ride and can be abandoned at any time, or ongoing games which can be paused and resumed without the temporal disconnect being too difficult.
The first category might be similar to the kinds of Flash games people play online, or to puzzle challenge games like Audiosurf (I'd love a portable version of this game!). It seems that this type of game should be avaialable in vast quantities - consider the iPhone App Store for an example - but I'm not sure they're commercially viable on a handheld that is expensive to develop for.
The second category suggests RPGs or even point-and-click adventures (Myst would have been fun on a portable), or turn-based strategy. The world should be relatively static while you're idle, and it should be possible to stop from any given point pretty easily. Many of the DS games I've seen seem to be of this category.
So... why is it that so many of the PSP games seem to be straight-up ports of traditional console games, meant to be played for hours (or at least 30 minutes) at a time, and to pack as much excitement as possible into that time? I realize designing a successful game is difficult (and once you've done it, you want to put it on as many platforms as you can), but surely a portable game that naturally takes a break every few minutes (and can be freely abandoned even between breaks), or that can be paused and resumed at any time without needing to remember exactly what you were doing right before, would be best? It seems odd that portable games so rarely give the impression of having been designed for portable gaming.
Fighting games kind of fit the first category, so long as the fights are REALLY short or you can drop out in the middle of a fight and do it later without penalty. Same for racing games. Sports games don't seem to fit well - too much stuff to do over the course of a period too long for one sitting, all of which builds on stuff before it. RPGs consisting of a lot of real-time action don't either. Real-time strategy games certainly don't (although they're rare on consoles anyhow). Things like Pokemon and Legend of Zelda games do make sense, but they don't seem to be the kind of thing you see for PSP.
IE8 has the ability to render using IE7's engine (either with a meta tag in the site, or via the "Comaptibility Mode" button, or - with greater granularity - through the development tools). For testing stuff older than 7, I really can't say. Incidentally, for web dev, the develpment tools are pretty sweet. A couple things in there Firebug could learn from.
You make a good point, except for one thing: ODF is available in MS Office as well. It's a open-source plugin (sponsored by Microsoft, hosted on Sourceforge) and integrates pretty nicely. I've been using it all the way back to when "Office 12" (as it was then called) was in beta, and I've yet to find a file it couldn't open correctly, or one that it saved which opened incorrectly on another office suite.
http://odf-converter.sourceforge.net/
Office 2007 (and apparently 2003 and XP, although I haven't tried on them) has had an optional plugin for 2+ years that enabled ODF support (for all the od* formats that I'm aware of, at least). It's an open-source project sponsored by Microsoft and listed on Sourceforge: http://odf-converter.sourceforge.net/
I use it to open/convert, edit, and save/export ODF files, and it works fine. Mind, if you prefer OO.o over Office 2007 (I don't, even ignoring the risk of OO.o getting the MSOffice formats wring - something which has never happened using the odf-converter plugin, BTW) then this isn't for you. I'm sure there are other people who would find this useful, though.
For the TL;DR folks:
Yes, you can run and compile POSIX apps in a *nix userspace on Windows. For more info, see http://suacommunity.com/
----
Yes, it has since the origin of NT. The NT kernel as we know it was started back when OS/2 was seen as The OS Of The Future, but OS/2 was slow getting started so Microsoft worked on their own thing on the side. They wanted it to be compatible with OS/2 though, so they included an OS/2 subsystem. UNIX was really big at the time also, so they included a POSIX subsystem as well. DOS and Win16 were already doing fairly well, so they included the NTVDM (Virtual DOS machine) with Win16 libraries to support 16-bit programs. Of course, they also included the Win32 API that is what people usually mean when the talk about the "Windows" API.
The OS/2 subsystem was discontinued after Windows 2000, and NTVDM isn't present in x64 versions of NT. However, the POSIX subsystem - including a BSD-derived userspace and GNU build toolchain - are still available in the higher editions of NT. On XP, you need Pro, on Vista you need Ultimate or Enterprise (all server editions should support it). On XP, you first install SFU (Services For UNIX) from Microsoft, which enables the subsystem and installs a basic userland called Interix. On Vista, enabling the subsystem (called SUA, Subsystem for UNIX-based Applications) and installing Interix are separate steps - click on "Turn Windows features on or off" from the Programs and Features control panel, then install "Utilities for Subsystem for UNIX-based Applications" for the Interix libraries, utilities, and build tools.
A package manager providing pre-compiled binaries is available from SUA Community (http://suacommunity.com), or you can use NetBSD's pkgsrc package manager and compile from source, or you can do the standard unpack tarball, ./configure && make && sudo make install routine. I use bash as my standard shell in Windows these days, use ssh (including sshd to connect from remote machines) and svn extensively. I've also used Interix to develop and test code for an embedded device that runs Linux; the source compatibility is pretty good.
/me laughs
Wine doesn't solve the problem at all; the reason legacy apps broke on Vista (and I mean really broke, not just need to be run as Administrator) is because Microsoft closed a lot of holes in their APIs and fixed a lot of non-spec behavior that people had relied upon for years (I shit you not, in XP there's memory management code to use an allocation scheme that doesn't explode on double-free for a specific game from over a decade back that relied or the behavior of an old Windows library). Wine doesn't really have those quirks supported; you can tell it to pretend to be a given Windows version for any particular application, but that's about it.
Same apps also tried to install drivers that would do crazy shit in kernel mode (Antivirus, I'm looking at you - but also random other programs like a nautical chart-plotter?!?) which Vista didn't like. These apps wither would work at all or would cause BSODs due to their drivers trying to do kernel-mode stuff that shouldn't have worked in the first place. Since wine supports almost no driver API, it's a pretty safe bet these wouldn't work either.
Home can't connect to Active Directory, so nobody in business would use it anyhow.
I suppose it's vaguely possible that home users will latch onto this feature and buy the higher editions of Win7 to get it, but honestly, I don't think so. The average home user doesn't even know about the application compatibility modes that all versions of Windows since 98 or so have included. I've had people think it was the most amazing thing when I got an XP program to run on Vista simply by running it as Administrator (not that it had a legit reason to, just modifications it made to the install location). People, on average, know very little about running a program beyond click the icon. Installing an optional virtual machine for compatibility is way beyond what a normal home user would do.
I don't know for sure if it can be done while the client is active, but Win7 has built-in capability to mount .VHD files into its filesystem. In theory this makes it quite possible for an AV program on the host to scan the drive of the client.
One of the features of Win7 that was announced early was that it can mount .VHD (Virtual Hard Disk, the format used by Virtual PC) natively (it can even boot off one, so long as the bootloader is on a real partition). So yes, the host AV *should* be able to protect the virtual system.
Firewall is just ridiculous; filter the VPC connection through the host (Win7) network interface, and the host's firewall is the guest's firewall. In fact, on current versions on VPC, if you want to connect the client to a network *without* running it through the host firewall, you need a dedicated NIC (i.e. the host can't connect via that interface).
I'm fairly sure the laws preventing defamation apply at the very least to publicly help organizations and members of goverment organizations, although I'm less sure about government as a complete administration.
In any case, I still disagree that free speech is an acceptable defense in a situation where the only purpose of a communication is to cause unjustified harm to another. I would actually see it as more justified - probably also more harmful, but also more justified - if the captioned pictures had instead been posted publicly; that would at least have indicated that they were intended for humor. As it is, they were sent privately to an individual for whom they would be extremely painful. What you seem to be advocating is not freedom of expression, but freedom to cause emotional anguish.
Consider this: Suppose I developed a device that could determine the image its target would find the most painful to see, and would force the target to see it. Would you defend my right to use this device on unwilling strangers? Would you defend my right to use it on you? On your children? On your parents? I have no idea who you are and I strongly doubt we've ever met, but we're nonetheless probably more acquainted with one another (through this discussion) than whoever sent those photos was with the girl's father.
I hadn't heard of this; it sounds pretty good and I'll check it out. Does anybody know whether the sleep is as "deep" (i.e. can I leave my laptop in sleep for 48 hours without draining the battery) and whether it handles ACPI events (closing or opening the laptop lid, wake-on-LAN, or similar)? I'll take a look anyhow; lack of a suspend state is a real annoyance with a student laptop.
For search, I was referring to the the indexed search tools which ship with Vista as opposed to those installed by default on my Linux boxes. The Vista one is faster by a noticable margin even though the index is larger.
For performance, you're talking about a scheduler design decision here: NT embraced and encouraged kernel-level threads in its core design, and bases scheduling decisions on threads (this makes a certain degree of sense, as context switching between threads is slightly faster than between processes, and creating new threads is less expensive than creating new processes). I'm not sure about Vista and up (NT 6.x) but the 5.x kernel (2000, XP, 2003) will certainly allocate more processor time to applications which use many threads. I wonder, however, whether such a hightly multi-threaded application on NT would perform as well as expected on Linux? Not saying it wouldn't but you didn't specify that it would; you simply picked an example of a case which supported your argument by using the preferred programming paradigm of Linux (POSIX, really), which is not preferred on Windows.
Amen, with one caveat: if the pictures had been completely anonymous (difficult, but a reasonable effort could be made) then no harm would have been done. They family still might have disapproved if it was done without their knowledge, but chances are they never would have known... because nobody would have been able to look up the email address of the girl's father and send him the sickest excuse for a "joke" (giving him the benefit of a doubt that it *might* have been intended as humor; it's possible some people are that tasteless) I've ever heard of.
Let me make that really clear: as I see it, the major fault was not the distribution of the photos, but the identification of the victim in them. THAT is a fairly clear breach of privacy, and should be punished (as should the people who sent the emails, of course - even if it was intended as humor, such things probably count as criminal harassment, and certainly form the basis for a civil suit).
I would not defend their actions, but I would defend their right to take such action.
I find two things terribly wrong with this statement. One of them is that freedom of speech does not, and never has, been intended to allow an attack on private individuals. That's why we have laws covering things from harassment (which this would seem to certainly qualify as) to slander and libel (these don't apply here, but they help illustrate the point where freedom of speech crosses the line). You are free to speak out against a government or other public institution, although you must be able to back up what you say. You are unlikely to wind up in court for privately attacking somebody verbally, although it can happen. In a case like this, though, the sender of those emails ought to face consequences - their actions had no purpose except to harm, you'd be crazy to say that they were not harmful, and there was no form of justification.
Second, regardless of whether the pictures got distributed, the *names* should NEVER have been without express consent. This is more about the cops who released the photos than it is about the assholes who sent the emails, but in any case there is no reason why anybody outside the police department and other organizations directly associated with the event would need to know who this girl was, who her father was, or any other information that could lead somebody to the family's email address. Anonymously distributing the pictures (without consent) as "an example of what a really bad crash can do" still seems a slightly grey area, morally, but there is no reason that they should be distributed without anonymity.
Sorry, but you, personally are missing the point with this line:
it's rather common as I knew a volunteer fireman who kept a photo album of accident pictures and he had shown them to people like it was his pride and joy
Did the pictures personally identify anybody? Was there sufficient information attached that you could find the email addresses of living relatives? If the answer is "no" then there's no harm done, although I personally still wouldn't put the pictures online without blurring out anything that *might* be identifiable. In this case, however, the answer was obviously "yes" which to me says that these people are either horribly insensitive or incredibly negligent (depending on whether it was by intention or accident) and frankly, in either case I think their ought to be consequences.
Or even VirtualPC, which is free and Microsoft (meaning has a known and more-or-less trusted name attched).
I find the complaint about not running those apps "in a secure mode" hilarious, though. Win9x had no secure mode; they were single-user OS's that allowed a few different profiles, and ran on a filesystem which doesn't even support permissions beyond the "I suggest you don't delete this" flag. Yes, security vs. convenience is frequently a trade-off situation; if you want to run programs which weren't designed for a secure system, you have to accept that they aren't secure.