In addition to the points that other posters have made, you're forgetting one thing - commercial webservers are admined by professionals, while Windows machines at home are most certainly not.
I run Windows XP Pro at home, and despite using P2P, my machine is virus and trojan free. How? I know what I'm doing, and take sensible precautions, including running a software firewall and regularly-updated AV software, and I keep my system patched.
Similarly, commercial webservers are behind firewalls, unnecessary services are shut down, relevant patches are carefully applied, etc.
I seem to remember that each time a Windows/IIS worm is reported here, people start moaning that according to their logs, most of the infected machines they see are on home DSL lines, "admined" by clueless users who shouldn't be on the 'net if they don't know how to keep their machines up to date and secure. If that's the case, comparing them to commercial webhosts running Apache is hardly fair.
As for the amount of time spent patching, well, servers are (or should be!) stripped down to the bare minimum, so the number of required patches will be small. Desktop machines, on the other hand, are generally installed with a much greater variety of things - a server admin wouldn't be patching DirectX, or installing new versions of Media Player or the Journal Viewer, etc.
Indeed - but such failure need not be catastrophic. By the same logic, with the billions upon billions of bits stored on a hard drive, the probability of failure is 1 - but I can still successfully store data on it. With the number of atomic and subatomic particles that make me up, the probability of radioactive decay is almost certainly extremely high - but I'm still here, and I don't have cancer.
Likewise, the failed nanobots may simply not work at all - just floating lifelessly until they're destroyed, by natural means or otherwise. Failing that, perhaps every usage of nanobots will be accompanied by a "dose" of hunter/killer ones, designed specifically to seak out and destroy malfunctioning nanobots.
I agree that care should be taken, but I also agree with the OP that the research should definitely be performed.
which means one more son doesn't come home in a body bag
Well, when it comes to using those battlefield computers in anger, it probably means that someone else's son, on the other side, goes home in a bobybag instead...
Well, under XP I can open a command prompt and type
My Documents\My Music\Artist\Album\track.[ogg|mp3]
and it launches winamp (my default player for those files types) and plays the song. It even does tab-completion.
Even with a complete OO scripting language, I can't see MSH changing that.
On the other hand, I tend to agree with your comments about an OO language being overkill for an interactive shell. But then, shell-scripting is also overkill for an interactive shell, yet you didn't need to use it to play your mp3 in your example. I'm sure the same will be true of MSH - it'll be there when you want/need it, and stay out of the way when you don't.
I've actually switched away from Linux (Mandrake specifically). I've been using various distributions since '97. Back then, if you were using Windows it was either 9x or NT (discounting the few still using 3.x... *shudder*).
Both were unstable, buggy pieces of crap, and Linux, even with it's steeper learning curve and arcane configs was a far preferable alternative.
Time has passed, however, and while Windows has improved almost beyond measure, Linux's improvements, in the areas that I care about, have been rather more incremental. Don't get me wrong, I still think that Linux-based systems are very good, but right now, XP Pro has the edge over them, for me at least (your mileage will of course vary), and to the extent that I spent cold, hard cash buying a copy.
What drove me away from Windows, essentially, was the instability and a number of UI niggles. 2k and XP have fixed stability - perhaps not for servers, but I'm talking about my desktop, not a server. Most of the UI niggles have also gone or can be configured away. I have no reason to spend the time and effort getting Linux setup and configured, and working around things like Office format docs (yes, Star/OpenOffice can handle most, but until it's 100% perfect, I can't use it at work for documents that I have to edit), and the other little things that make using it day to day less easy than using Windows. (Again, YMMV - but I can speak only for myself)
Maybe some Linux company will pull out something special between now and then, and tempt me back, but until that happens, I'm as happy running XP Pro as I ever was running Linux.
I was chatting with a friend on ICQ only this morning about OSX, and how cool it looks, and how tempting buying an Apple machine would be, if only we had the money.
Then I remarked that Longhorn could well also bring some very good stuff - after all, Win2k and XP are pretty-much rock solid as far as desktop expectations and demands go, and Longhorn's 2006 shipping date gives MS plenty of time to learn from OSX's example.
You can be productive in Linux just as easily as in Windows.
Once you've learnt to use it, and adjusted to the new environment and new applications, yes.
For a while after making the switch, though, your productivity is going to take a hit, and that's going to put off a lot of people. Think of it this way - their company pays for the software, so cost is not an issue. What is an issue is getting their job done on time, and if anything is going to jeopardise that, then forget it.
I think it would revert back to them owning the code again
It's not a case of reverting back to the original authors - just because a work has been GPLed (or similar), doesn't mean that it's not copyrighted. Quite the opposite, in fact - it most assuredly is still copyrighted. The author still owns it (unless they transferred their copyright to someone else); releasing it under the GPL doesn't change that.
I'm willing to bet this will mean more money in the bank and more market penetration for Microsoft
Well, the vouchers are going to "past purchasers of Microsoft software". Therefore, in the vast majority of cases, they're going to be used to upgrade or supplement existing MS software owned by these people. They're not going to increase market share at all.
Unfortunately, what is reasonable and what is legal are not always the same thing. Anyone considering embarking on such a project would be very well advised to consult with a lawyer before getting too far into it.
You are _not_ allowed to assault someone who just keyed your car.
More accurately, you're not allowed to key the car of someone who just keyed your car, which is what attacking an attacking computer would be more like.
It seems to me that the main problem with most worms that we've seen so far is the havoc they wreak on the network as a whole, chewing up bandwidth as they propagate, rather than what they do to the machines in question. How would a "good" worm be any different? It has to propagate by the same means, after all. At best, the author may be a little more careful of bandwidth usage, but chances are it'll still cause a problem.
With the release of Windows 2000 and XP, it's also one of the main areas in which Windows can truly be said to be lacking.
Say what you want about servers with uptimes measured in years, XP is stable enough for everyday desktop use. I've not used 2k3 Server, so I can't comment on that, but I assume that (properly maintained) it'll give a Linux-based server a run for its stability money.
Security, on the other hand, is definitely somewhat lacking. Sure, my nearly-always on, nearly-always connected XP Pro box at home is in little or no danger, but that's because I'm careful. I have anti-virus software and a software firewall (as well as XP's own one), and I'm careful about what I do. But I'm the exception, not the rule...
MS needs to concentrate on security, because that's the main area in which their customers find Windows to be lacking, simple as that.
But that's the whole point - that was three revisions of Windows 95, not three versions of Windows. Think 3.0 versus 2.6, as opposed to 2.6 versus 2.4... (Or in the case of Win95, 3.0 vs 2.6 rather than 2.4.3 vs 2.4.2 vs 2.4.1)
(No, I don't know if the differences will be as great as would be expected in that case - but MS don't really play up the version numbers, so there'd be little or no point in incrementing the major number if it wasn't warranted. No-one in control of budget knows about it anyway...)
You remind of CNN for technology. Unbiased "News". feh.
You must be new around here (checks uid - ah yes, 6 figures).
Slashdot has never been about unbiased news in all the time I've been here; it has always had a heavy OSS bias, and especially for Linux. That's not necessarily a bad thing, although I'd argue that it's not necessarily a good thing, either
If you want completely unbiased tech news, you're in the wrong place. On the other hand, your attempt to draw parallels between MS "leaking" betas and Linux test releases is non-sensical. The former is not meant for general release, while the latter most certainly is, at least for those willing to risk using potentially unproven code on their system.
But that's the point - in my experience, the examples in the MSDN (at least those that I've used) are the best that are available, or at least no worse than any other. Now, that may just be because the.net framework and C# are so new that the community haven't had time to catch up yet, but that makes little to no practical difference.
"Anonymous methods are similar to lambda functions in the Lisp programming language."
That's a direct quote from the second bullet point of the first section - it's the fifth sentence of the document. Hardly claiming to have invented them, are they?
Agreed - having used JBuilder Enterprise (7) amongst others, I find myself wishing that VS.NET had refactoring support (I have the Enterprise Architect edition, too). If you have budget for a new IDE, you might want to take a look at Borland's C#Builder. I've only briefly used it, and can't remember what sort of features it has, but I'd expect them to mirror JBuilder's, being from the same company.
Why else would the cops stop the easiest catches, rather than the fastest ones?
Perhaps because the fastest ones are statistically more likely to make a chase of it, thus causing even more risk of an accident? Note that I don't know if the data support this, it's just an idea.
This actually allows for more rapid processing, but opens the door for some pretty stupid mistakes.
Indeed. I dread to think how often, when writing a report or essay, etc, I've read and re-read it while proof-reading it, and every time I've missed the fact that I've missed out a word in a sentence - something non-essential, like "the" or "and".
My brain expects it to be there, as it fits the pattern of the sentence, and so it just fills it in for me as I read through, so I don't notice that it's missing.
So, you have all those stupid fantasies of machines that you "will just talk to in English," and the continuing search for handwriting recognition.
There's a reason for that, and it's tied into something that a lot of people here on slashdot and similar tech-based sites don't understand: most people don't want to have to learn something new just to get their job done.
Executives, as you point out, don't need to type because they "have someone to do that for them". Given that, and that they didn't learn to use a keyboard in school/college (but got plenty of practice using a pen), of course they want a machine that they can write on or talk to.
If nothing else, it'll let them replace their relatively expensive assistants and secretaries with a much cheaper machine, to a greater or lesser extent.
And at the end of the day, why should they learn to use a keyboard? They already have someone to do it for them. Same thing with everyone - there are things that any given person can't do, and they don't learn to, because they can pay someone else. It's easier, if not necessarily cheaper in the long run. For the execs, it's keyboard skills, and IT support, programming, etc. For me, it's plumbing, and car and building maintenance, etc. I have neither the time nor the money to learn these things myself, so I pay other people to do them, as and when I need. If I could buy a cheap machine to do it for me reliably, I probably would.
Slashdot Mirror
In addition to the points that other posters have made, you're forgetting one thing - commercial webservers are admined by professionals, while Windows machines at home are most certainly not.
I run Windows XP Pro at home, and despite using P2P, my machine is virus and trojan free. How? I know what I'm doing, and take sensible precautions, including running a software firewall and regularly-updated AV software, and I keep my system patched.
Similarly, commercial webservers are behind firewalls, unnecessary services are shut down, relevant patches are carefully applied, etc.
I seem to remember that each time a Windows/IIS worm is reported here, people start moaning that according to their logs, most of the infected machines they see are on home DSL lines, "admined" by clueless users who shouldn't be on the 'net if they don't know how to keep their machines up to date and secure. If that's the case, comparing them to commercial webhosts running Apache is hardly fair.
As for the amount of time spent patching, well, servers are (or should be!) stripped down to the bare minimum, so the number of required patches will be small. Desktop machines, on the other hand, are generally installed with a much greater variety of things - a server admin wouldn't be patching DirectX, or installing new versions of Media Player or the Journal Viewer, etc.
Indeed - but such failure need not be catastrophic. By the same logic, with the billions upon billions of bits stored on a hard drive, the probability of failure is 1 - but I can still successfully store data on it. With the number of atomic and subatomic particles that make me up, the probability of radioactive decay is almost certainly extremely high - but I'm still here, and I don't have cancer.
Likewise, the failed nanobots may simply not work at all - just floating lifelessly until they're destroyed, by natural means or otherwise. Failing that, perhaps every usage of nanobots will be accompanied by a "dose" of hunter/killer ones, designed specifically to seak out and destroy malfunctioning nanobots.
I agree that care should be taken, but I also agree with the OP that the research should definitely be performed.
you're not going to believe what you need to do to get oracle to work ;)
:-)
Bah, that's easy - I just email our DBA and ask her to do it
commercial application servers such as Tomcat
.NET framework for Solaris
.NET Framework, not Ximian, although Ximian does have a hand in Mono, the open source implementation of the .NET Framework.
Tomcat is open source; it's one of the Jakarta projects.
compared to Oracle's WebSphere
IBM make WebSphere, not Oracle.
If Ximian would only release the
Microsoft makes the
which means one more son doesn't come home in a body bag
Well, when it comes to using those battlefield computers in anger, it probably means that someone else's son, on the other side, goes home in a bobybag instead...
Well, under XP I can open a command prompt and type
My Documents\My Music\Artist\Album\track.[ogg|mp3]
and it launches winamp (my default player for those files types) and plays the song. It even does tab-completion.
Even with a complete OO scripting language, I can't see MSH changing that.
On the other hand, I tend to agree with your comments about an OO language being overkill for an interactive shell. But then, shell-scripting is also overkill for an interactive shell, yet you didn't need to use it to play your mp3 in your example. I'm sure the same will be true of MSH - it'll be there when you want/need it, and stay out of the way when you don't.
Slash has inserted two spaces into it - one between "url=" and "/library", and another between "command_li" and "ne_utilities"
Copy and paste it into a new browser/tab, strip them out, and it'll work fine.
It is possible to be both pro-Linux and pro-Windows at the same time, you know. Both have their own unique strengths and weakenesses.
The right tool for the right job, and all that...
Hardly.
I've actually switched away from Linux (Mandrake specifically). I've been using various distributions since '97. Back then, if you were using Windows it was either 9x or NT (discounting the few still using 3.x... *shudder*).
Both were unstable, buggy pieces of crap, and Linux, even with it's steeper learning curve and arcane configs was a far preferable alternative.
Time has passed, however, and while Windows has improved almost beyond measure, Linux's improvements, in the areas that I care about, have been rather more incremental. Don't get me wrong, I still think that Linux-based systems are very good, but right now, XP Pro has the edge over them, for me at least (your mileage will of course vary), and to the extent that I spent cold, hard cash buying a copy.
What drove me away from Windows, essentially, was the instability and a number of UI niggles. 2k and XP have fixed stability - perhaps not for servers, but I'm talking about my desktop, not a server. Most of the UI niggles have also gone or can be configured away. I have no reason to spend the time and effort getting Linux setup and configured, and working around things like Office format docs (yes, Star/OpenOffice can handle most, but until it's 100% perfect, I can't use it at work for documents that I have to edit), and the other little things that make using it day to day less easy than using Windows. (Again, YMMV - but I can speak only for myself)
Maybe some Linux company will pull out something special between now and then, and tempt me back, but until that happens, I'm as happy running XP Pro as I ever was running Linux.
I was chatting with a friend on ICQ only this morning about OSX, and how cool it looks, and how tempting buying an Apple machine would be, if only we had the money.
Then I remarked that Longhorn could well also bring some very good stuff - after all, Win2k and XP are pretty-much rock solid as far as desktop expectations and demands go, and Longhorn's 2006 shipping date gives MS plenty of time to learn from OSX's example.
Looks like maybe I was right.
You can be productive in Linux just as easily as in Windows.
Once you've learnt to use it, and adjusted to the new environment and new applications, yes.
For a while after making the switch, though, your productivity is going to take a hit, and that's going to put off a lot of people. Think of it this way - their company pays for the software, so cost is not an issue. What is an issue is getting their job done on time, and if anything is going to jeopardise that, then forget it.
I think it would revert back to them owning the code again
It's not a case of reverting back to the original authors - just because a work has been GPLed (or similar), doesn't mean that it's not copyrighted. Quite the opposite, in fact - it most assuredly is still copyrighted. The author still owns it (unless they transferred their copyright to someone else); releasing it under the GPL doesn't change that.
I'm willing to bet this will mean more money in the bank and more market penetration for Microsoft
Well, the vouchers are going to "past purchasers of Microsoft software". Therefore, in the vast majority of cases, they're going to be used to upgrade or supplement existing MS software owned by these people. They're not going to increase market share at all.
Seems somewhat resonable to me.
Unfortunately, what is reasonable and what is legal are not always the same thing. Anyone considering embarking on such a project would be very well advised to consult with a lawyer before getting too far into it.
You are _not_ allowed to assault someone who just keyed your car.
More accurately, you're not allowed to key the car of someone who just keyed your car, which is what attacking an attacking computer would be more like.
It seems to me that the main problem with most worms that we've seen so far is the havoc they wreak on the network as a whole, chewing up bandwidth as they propagate, rather than what they do to the machines in question. How would a "good" worm be any different? It has to propagate by the same means, after all. At best, the author may be a little more careful of bandwidth usage, but chances are it'll still cause a problem.
With the release of Windows 2000 and XP, it's also one of the main areas in which Windows can truly be said to be lacking.
Say what you want about servers with uptimes measured in years, XP is stable enough for everyday desktop use. I've not used 2k3 Server, so I can't comment on that, but I assume that (properly maintained) it'll give a Linux-based server a run for its stability money.
Security, on the other hand, is definitely somewhat lacking. Sure, my nearly-always on, nearly-always connected XP Pro box at home is in little or no danger, but that's because I'm careful. I have anti-virus software and a software firewall (as well as XP's own one), and I'm careful about what I do. But I'm the exception, not the rule...
MS needs to concentrate on security, because that's the main area in which their customers find Windows to be lacking, simple as that.
Isnt netsh a resource kit binary?
Apparently not - I've not installed the resource kit on this machine (which is running XP Pro), but I definitely have netsh available.
But that's the whole point - that was three revisions of Windows 95, not three versions of Windows. Think 3.0 versus 2.6, as opposed to 2.6 versus 2.4... (Or in the case of Win95, 3.0 vs 2.6 rather than 2.4.3 vs 2.4.2 vs 2.4.1)
(No, I don't know if the differences will be as great as would be expected in that case - but MS don't really play up the version numbers, so there'd be little or no point in incrementing the major number if it wasn't warranted. No-one in control of budget knows about it anyway...)
You remind of CNN for technology. Unbiased "News". feh.
You must be new around here (checks uid - ah yes, 6 figures).
Slashdot has never been about unbiased news in all the time I've been here; it has always had a heavy OSS bias, and especially for Linux. That's not necessarily a bad thing, although I'd argue that it's not necessarily a good thing, either
If you want completely unbiased tech news, you're in the wrong place. On the other hand, your attempt to draw parallels between MS "leaking" betas and Linux test releases is non-sensical. The former is not meant for general release, while the latter most certainly is, at least for those willing to risk using potentially unproven code on their system.
But that's the point - in my experience, the examples in the MSDN (at least those that I've used) are the best that are available, or at least no worse than any other. Now, that may just be because the .net framework and C# are so new that the community haven't had time to catch up yet, but that makes little to no practical difference.
Have you read the specification document?
"Anonymous methods are similar to lambda functions in the Lisp programming language."
That's a direct quote from the second bullet point of the first section - it's the fifth sentence of the document. Hardly claiming to have invented them, are they?
Not the best IDE in my opinon
Agreed - having used JBuilder Enterprise (7) amongst others, I find myself wishing that VS.NET had refactoring support (I have the Enterprise Architect edition, too). If you have budget for a new IDE, you might want to take a look at Borland's C#Builder. I've only briefly used it, and can't remember what sort of features it has, but I'd expect them to mirror JBuilder's, being from the same company.
Why else would the cops stop the easiest catches, rather than the fastest ones?
Perhaps because the fastest ones are statistically more likely to make a chase of it, thus causing even more risk of an accident? Note that I don't know if the data support this, it's just an idea.
This actually allows for more rapid processing, but opens the door for some pretty stupid mistakes.
Indeed. I dread to think how often, when writing a report or essay, etc, I've read and re-read it while proof-reading it, and every time I've missed the fact that I've missed out a word in a sentence - something non-essential, like "the" or "and".
My brain expects it to be there, as it fits the pattern of the sentence, and so it just fills it in for me as I read through, so I don't notice that it's missing.
So, you have all those stupid fantasies of machines that you "will just talk to in English," and the continuing search for handwriting recognition.
There's a reason for that, and it's tied into something that a lot of people here on slashdot and similar tech-based sites don't understand: most people don't want to have to learn something new just to get their job done.
Executives, as you point out, don't need to type because they "have someone to do that for them". Given that, and that they didn't learn to use a keyboard in school/college (but got plenty of practice using a pen), of course they want a machine that they can write on or talk to.
If nothing else, it'll let them replace their relatively expensive assistants and secretaries with a much cheaper machine, to a greater or lesser extent.
And at the end of the day, why should they learn to use a keyboard? They already have someone to do it for them. Same thing with everyone - there are things that any given person can't do, and they don't learn to, because they can pay someone else. It's easier, if not necessarily cheaper in the long run. For the execs, it's keyboard skills, and IT support, programming, etc. For me, it's plumbing, and car and building maintenance, etc. I have neither the time nor the money to learn these things myself, so I pay other people to do them, as and when I need. If I could buy a cheap machine to do it for me reliably, I probably would.