You're vaguely on the right track, but woefully inaccurate in the details.
OS/2 was a joint IBM and Microsoft venture, originally meant to replace DOS on "high end" PCs.
DOS and Windows was meant for "low end" PCs.
Around 1988, Microsoft started working on what was originally meant to be OS/2's replacement, to be run on 386+ machines (OS/2 targeted 286s, at IBM's insistence, and it was a huge mistake). This product eventually became Windows NT, but at the time was known as "OS/2 NT" (which is why those "WNT=VMS+1" comments are so easily demonstrated as wrong).
However, Windows 3.0 was a surprise (to everyone) runaway success and that subsequently caused friction between Microsoft and IBM - IBM still wanted the default API in "OS/2 NT" to be "OS/2", Microsoft now wanted it to be "Windows". Since they couldn't agree, Microsoft took their ball (NT) and left. This was the famous "Microsoft-IBM divorce" and happened in (late, IIRC) 1990.
IBM were then left with the task of taking OS/2 1.x and turning it into OS/2 2.x (and successors). IBM, however, did not "own" OS/2 (at least not all of it) and even as late as OS/2 4.0, were still paying Microsoft royalties for their code.
Also, from about Windows 3.0 onwards, DOS became less and less "the OS" and more and more "the bootloader", as Windows took over the low level functionality. Windows 3.0 was doing memory management, CPU scheduling and some hardware support (video, printers). Windows 3.1 took over more hardware support (disk, network). By Windows 95, DOS was pretty much irrelevant (past bootup), unless you were unlucky enough to still be running old 16 bit code. Microsoft killed DOS in about 1993 - it just took everyone a decade to say goodbye.
As the transition sees developers pushing out binaries for both chips, I don't see a downside of Apple straddling the fence and using both types of chips.
That's probably because you don't have to worry about their shareholders.
Especially once you go into businesses. The difference in price between Apple and Microsoft starts getting huge once you spec out an environment for >50 people. With Apple you know your clients cost you $129, your server $999, Remote Desktop for $499, no limits, everything integrated with Kerberos + LDAP. With Microsoft you got that and then you have to start calculating CAL's for Exchange, CAL's for your Terminal Server, CAL's and server licenses for your SMS and WUS, and each little piece that will make life easier as the admin costs you an extra CAL which is all included in the Server+Client+RemoteDesktop licenses on Mac.
Of course, the Windows environment you've just described (Exchange, Terminal Server, etc) is _vastly_ more functional than the OS X environment you described, so it's not surprising in the least it would cost more.
Next to that, the average Windows machine lasts 3 years before it get's old and slow. I've seen G3's running OS X 10.4 without a hitch, G4's are mainstream in many companies and most haven't even gone to Intel yet.
If someone is happy with a G3 or G4 running OS X, they'll be just as happy with a P2, P3 or P4 running Windows XP. It's *far* kinder to older hardware than OS X is.
at I don't understand is, since Mac software has to be Universal nowadays anyway, why Apple doesn't just permanently keep its lineup as a mix of PPC and x86, picking whichever chip suits the particular machine they're designing at the time? Power6 Xserves along side Core 2 laptops... it sounds good to me!
Because the market for such machines is miniscule and costs in designing, testing and mass-producing computers are non-trivial.
(This is before even getting into the additional costs that would be heaped on developers, discouraging them from writing for the platform.)
In other words, it would cost Apple a lot of money while delivering few, if any, advantages (and a whole bunch of disadvantages).
I've actually heard of kids in middle and high school who use SMS and IM so much that they legitimately don't know how to spell words like "you", "your/you're", and will use internet abbreviations (lol, idk, etc.) in school papers.
Spelling and grammar (especially grammar) were heading downhill long before every kid had a mobile phone. I'll agree the problem has been getting worse in recent years (and that "TXTing" is probably exacerbating the situation), but the stage has been set since the mid-late 80s.
At least in Australia, grammar more complicated than identifying verbs and nouns hasn't been given more than a cursory nod in most schools for about twenty years. Spelling hasn't fared much better - even 10 years ago the typical example of "you're" vs "your" was becoming common.
Personally, I see three main culprits - the first is TV replacing books as "idle entertainment", the second (more the last decade) is the plethora of websites with atrocious spelling and grammar providing a massive feedback loop and the third is the education environment (teachers who have bad English skills and curriculums that basically don't try to teach much more than speaking the language).
Sadly, mathematics is going the same way, with calculators becoming "essential" in primary school years and the ability of people to do problems in their heads more complicated than adding double digit numbers - and especially things like manipulating fractions (eg: for working out percentages) - practically considered black magic to the average 25 year old.
This I do not understand. If it is no longer profitable to sell software, then why would someone keep their source secret?
Because non-trivial software - especially in the commercial sector where all the money is - requires support.
This is before even getting into interesting stuff like Trusted Computing, signed binaries and the like.
If copyright disappeared overnight, why would Microsoft want to spend any resources on improving Windows?
Same reason they do now - keep their customers satisfied.
Assuming that they would no longer be able to make any money from it.
A questionable assumption.
Wouldn't it be better to just make the whole software side public domain and concentrate on making hardware? This way the community would improve on their software, and they wouldn't need to spend money on it anymore.
The "money" is still being spent.
Further, the "public domain" still hasn't demonstrated conclusively it can provide a better product.
With an attitude like that, I can only presume you have little understanding of the GPL's purpose.
Without copyright, the GPL could not work. This is a completely different thing to the GPL not being needed. Should you agree with the philosophy behind the GPL, it is difficult to see how you could consider copyright (or something very similar) unnecessary.
If there was no copyright, then there would be no incentive to keep your source code secret, as you wouldn't be able to sell your software anyways.
Uh, no. Indeed, without copyright the complete opposite would be true - there would be orders of magnitude more incentive to "keep your source secret", since there would no longer be any inherent protection from the legal system.
If you believe abolishing copyright would obviate the need for the GPL, then you haven't thought your cunning plan all the way through. Without copyright the GPL is completely irrelevant, its restrictions unenforcable, and its objectives much less attainable (if not - practically speaking - impossible). Personally this doesn't bother me in the least, since I'm not a fan of either copyright or the GPL - but if you have strong feelings about the latter, you might want to make sure your attitude to the former is consistent with them.
While the GPL is a somewhat... unorthodox... usage of copyright, it's in complete agreement with the fundamental principles underlying it.
Tell me again, how is this part of the intent of copyright law???
Much of the "intent" of copyright has changed over the years, however, one common theme is the ability of the holder of a particular work's copyright to dictate under what conditions it can be used and, particularly, distributed. That is all the GPL does and, hence, code licensed under the GPL is unquestionably being used in line with the "intent" of copyright (ie: you get to say what people do with your code).
The GPL uses copyright law to make sure your work never becomes part of the farse of copyright.
No, it doesn't. Quite the opposite, in fact.
The GPL completely and utterly _relies_ on copyright to have any purpose. Without copyright, the GPL is meaningless and the restrictions it imposes would be impossible.
In other words, for licensing your code under the GPL to mean anything, your work *must* "become part of the farse[sic] of copyright".
Business are used to run Windows on their workstation. Microsoft has never supported additional platforms for long time (Alpha has only had 1 NT version made for it. Itanium had only 1 XP version made for it) because supporting multiple platform is hard for them (the only reason they'll keep support for 64bit x86_64, is that they'll kill 32bits instead and thus they'll still have only 1 main architecture to support). Thus there was a lack of interest from the largest consumers because of this absence (of course there are a lot of shops running Unices. But they aren't profitable on the same level as Dell's consumers).
Windows NT was available on x86, Alpha, MIPS from its first release in 1993. In 1995, NT 3.51 added PPC. Support for MIPS and PPC was dropped around 1997-98 and support for Alpha in about 1999 (Windows 2000 still had an Alpha version up until RC2). In 2001 Itanium support was added and in 2003 x86_64 was added.
There was also an internal port in the mid 90s to SPARC and, IIRC, PA-RISC. This was probably NT 3.51 and/or NT 4.0.
The implication that Windows NT is not, or was not, "portable" and that Microsoft "has difficulty" porting it to new platforms, has no foundation in reality.
This mean that when Windows-the-next (tm) comes out, either there will be a massive switch toward other OS (very likely in university labs) or it will see an even slower reception than Windows Vista is currently experiencing.(very likely on Joe 6-pack's older 32bits machine).
This argument is made frequently (and similar ones have been in the past with earlier releases), but it's difficult to see any sort of logic behind it. Most sales of Windows are through an OEM channel, not individuals buying upgrades. Further, most people who are buying upgrades, are doing so to put on a relatively high-end PC.
The suggestion that the next version of Windows only supporting 64 bit hardware ca. 2010 is going to cause any meaningful migration to other platforms, doesn't even pass the laugh test, given that:
* Most sales of that version of Windows would be for new (ie: x86_64) hardware;
* Of the remainder, most would be for relatively high end (ie: x86_64) hardware;
* The transition period will be measured in years; and
* All mainstream alternatives will almost certainly be in the same position
Computer programs do not all have the same function.
Most computer programs have a common set of identical functionality. Some examples are manipulating windows (resizing, closing, etc), manipulating files (open, save, etc), manipulating text (copy, paste, etc), online help, changing settings.
It is a significant boost to productivity, learnability and ease of use when these common types of functionality are presented in a consistent and predictable fashion.
Further, there are a number of general UI principles - like Fitt's Law - that can also be used (where applicable), regardless of specific implementation.
If you think that Xen is a "royal PITA" to get running, then how do you manage to work with a server?
Most likely, he's trying to do non-trivial things with Xen. Stuff like bridges to multiple vlans, interface bonding, multiple drives, and the like.
Xen *sucks* from an administrative perspective if your environment is anything remotely complicated. VMWare has nothing to fear.
There is a slight learning curve, but after that it's extremely easy to deal with. Much like most Unix things. The massive performance benefit that Xen gives over VMWare Server (the only one I use) is well worth the hour it takes to learn the basics.
It takes a lot more than an hour to get Xen up and running in a non-trivial configuration. The management advantages of VMWare easily make it a better choice than Xen. Stupid amounts of computing power is dirt cheap these days, while people time is only getting more expensive.
Most places in the world, local calls are "free" (which means there is a fixed charge per month), for home lines.
In Australia local calls are charged (but untimed).
In addition you get a phone number that people can call, a phonebook listing, etc. Most places you can't unbundle these.
In Australia, you pay to make calls, not receive them. It follows, therefore, that the "services" you describe revolving around receiving calls, are "free".
The GP is obviously referring to the voice part of the landline, which is not necessary to the operation of the ADSL.
The line rental fee doesn't include any voice services. They cost extra. The line rental fee is exactly that - the cost you pay for a (working) physical connection to the system. Calls (or other things like messagebank) cost extra on top of that.
On both of those days I was modded down for suggesting that the Linux release-when-ready model offered greater security than the Microsoft release-when-we-feel-like-it model.
Say what ? By what definition of 'ready' does Linux have a "release-when-ready" model ?
"Release for the heck of it" would be a better description.
Let me reword that - how come I have to pay for a full suite of voice related services on my line when I just want to use DSL?
What makes you think you do ? The line rental fee doesn't include the cost of any voice services (ie: you pay per call for local, time-based for STD, for message bank, etc).
Yes, you do. You use it for DSL. How else do expect to get ADSL other than over a landline ?
God they're filthy (Telstra) - hopefully we'll have a change of Government soon & get rid of the current spineless Prime Minister John Howard - who can't stand up to Telstra.
Huh ? The Australian Government regulates the hell out of Telstra (and a good thing, too, given the circumstances).
Everyone behind her was following too closely. Plain and simple. They should have time to react, apply their own brakes, change lanes, anything. People need to learn to stop following so closely.
Even if there weren't any crashes, someone stopping suddenly in the fast lane of a motorway is still going to cause chaos at the time and probably for tens of minutes - if not hours - afterwards.
She's not unintelligent (though, being blonde, she did get a certain amount of follicle-related humour directed at her), but she did as she was told, in a pressure-situation.
Anyone who considers driving down one of Britain's Motorways (which are both a) in relatively good condition and b) filled largely with relatively polite, disciplined drivers) a "pressure situation" shouldn't be allowed to hold a license. Neither should anyone who slams on the brakes because the onboard computer said 'stop', either, for that matter.
What's tragic is that she, along with many others, will just laugh this off as some silly "accident", when she should be banned from driving for some time, slapped with a large fine, and given a brutal dressing down by a judge. Incompetent driving like hers is the kind of thing that gets people killed.
Your comments demonstrate an utter lack of comprehension of computer security, as when you say that a home user "has to be" the sys admin.
Who, then, are you proposing will be "administering" their PCs ?
And suggesting that this has anything to do with "scale" is completely irrelevant.
Scale is an inescapable and intrinsic component.
Ironic that you accuse me of having no comprehension of computer security, given your comments.
You've made any number of general statements about NT security design without backing any of them up.
That's because anyone with knowledge of NT knows they are true. Further, anyone without knowledge of NT arguing against them is not worth listening to.
At the same time, you've made general statements about Linux security not being "fine grained" and the like which demonstrate a complete lack of comprehension of both security and Linux security.
(Classical) UNIX (and hence, Linux) security is not at all fine-grained. A user is either root or not root. The root user can do anything because, literally, the security infrastructure for them does not apply. Non-root users can be divided into smaller groups but, ultimately, they are all effectibely equal and the restrictions to each other's resources are almost completely limited to those things which can be represented within the filesystem (with a handful of exceptions, like binding to low ports).
The UNIX security model is both very coarse and based around the assumption that if a user does not have the privileges to do X, then they impersonate another user who does have that privilege (typically - and in many cases, unavoidably - root). This is an inherent security problem, because it both a) negatively impacts auditability and b) means vulnerabilities in the programs running at higher privilege - even temporarily - typically confer ALL the rights of the user (NOT just those necessary to complete the task) to the exploiter.
It is inescapably less capable and less secure than the NT (/VMS, since that's where NT inherited it from) model.
Meanwhile, the reality of the marketplace is that Windows is a disaster as an secure OS.
This statement is meaningless, firstly because it lacks a definition of what you mean by "secure OS" and secondly because no other OS faces the same exposure and requirements as Windows.
You have yet to come up with any ways in which Linux is more secure, although there has been lots of handwaving and examples of configuration problems related to different requirements.
I think that speaks for itself.
Indeed. Just like the people who argue blacks are more prone to criminal behaviour based on the reasoning there's more of them in jail, speak for themselves.
When Linux has eliminated Microsoft as a factor and has the market share of Windows, and when Linux has as many security problems as Windows, email me.
I have a better idea. If any platform manages to do that and do it _without_ the "security problems" Windows has had, and without resorting to something like Trusted Computing, you can email me and have a great time telling me how wrong I was.
Sorry, you do not understand "sudo". It does not give you "System" privilege - it gives you "root" privilege and then only on entry of the root password. On UNIX/Linux, that's almost the same thing, given the power of root on UNIX, but not exactly. The PowerPrompt utility gives you a higher level privilege than the system administrator has.
Firstly, 'root' is more powerful than SYSTEM. Root, literally, does not have any security restrictions imposed on it at all. SYSTEM does and, more importantly, is subject to ACLs just like every other account (whereas root does not).
Secondly, 'sudo' requires the user password, not the root password.
I understand sudo quite well. I have had the frustrating task of trying to use it to allow the secure management of numerous UNIX machines by a team of sysadmins, all with different responsibilities and access levels.
As for privilege escalation, before Windows NT/2000, basically you didn't even need to do it because everybody WAS "root".
DOS-based versions of Windows are not relevant to this discussion.
Even after Windows NT, virtually every exploit in Windows involved "privilege escalation" in the sense that most exploits allow you to do things that are not allowed in UNIX/Linux without root privilege.
For example ?
Spyware couldn't exist on Linux [...]
Of course it could.
- and doesn't - without root privilege - [...]
Much like Windows, then ?
[...] it's trivial on Windows for spyware to set itself up with privileges that a normal user can't handle.
That's because Windows NT has a vastly more fine-grained and capable security model than UNIX's primitive "you're root and can do anything or you're a user and can hardly do anything".
That's WHY the PowerPrompt tool was invented - to give the admin user enough power to deal with spyware that even the sys admin can't get rid of.
"Administrator" and "sys admin" are not synonyms.
The bottom line: Windows has demonstrated since day one that it has a less secure design than UNIX.
The design of Windows is fundamentally more secure than (traditional) UNIX. It lacks the inherent hole of a superuser, it has a vastly more fine-grained and capable security infrastrucutre and it is able to apply that infrastructure throughout the entire OS, rather than (basically) just those things accessible via the filesystem.
You would be hard pressed to find any measurable way that the design of NT is less secure than the "design" of UNIX. And it is trivial to find ways the design of Windows NT is better than the design of UNIX.
The single biggest influence on UNIX's "security record" is its user demographic.
Windows XP continues to install the default user as a sys admin by design [...]
Actually that's a minor configuration issue, not a design problem, and is easily remedied. It exists because most Windows machines are installed in either managed networks (where the default user is not Administrator, anyway) or on unmanaged home desktops (where they essentially must be Administrator). It is a less than ideal, but perfectly valid, engineering tradeoff.
[...] - and that is the single dumbest idea Microsoft ever had (next to the Registry as a single point of failure for the whole system.)
I'm guessing you're as clueless about the Registry as you are about pretty much everything else you've commented on.
That stupidity in itself annihilates any security advantages Microsoft might have ever designed into NT.
In the scenario of an unmanaged end-user desktop, the user *has* to be the "sys admin", basically by definition. If(/when, maybe) Linux had to deal with that problem on any sort of scale, the results will be basically identical. Linux probably won't, however, and it will be OS X that has the pleasure once it starts to hit critical mass.
Following Microsoft's instructions, please explain how to configure the firewall there described to block all TCP traffic except connections on port 1434 from network 192.168.0.0/16. You can't do it. You can block a particular port from operating at all on a particular interface but that's about it. Its... pathetic.
It's also not something that belongs on the type of OS Windows 2000 is (meant for managed corporate networks).
The XP firewall does more, but it still doesn't do the very obvious task: allow connections to port 1434 from these three corporate netblocks and nowhere else.
I haven't actually _tested_ it, but it's certainly possible to input the necessary configuration into XP's firewall.
And that's the whole point, isn't it? Security and the devices which support security do not become requirements in Windows until late in the game... like locking the proverbial barn door after the animals have all fled. In Linux and most of the other Unixes, the security devices tend to make it into the software BEFORE the widespread security events that compel their use.
It's not the "security events" the drive the requirement, it's the expected usage. Home firewalls have only really become a important in the last 5 - 7 years and, ultimately, host-based firewalling is a less than ideal solution with the better one (separate hardware device) being applicable to a significant proportion of installations.
The same requirements do not apply to Linux.
Like perhaps "Internet Connection Sharing" in Windows? And hey, what do you know, Windows can work as a plain router too. It even supports RIP and OSPF natively. What it lacks is any of the tools necessary to make that work securely.
Because it's an incredibly niche configuration situation for Windows. There are relatively few situations where Windows is being used as a router, even fewer where it needs to do anything more complicated than a simple 1:many NAT and practically zero where it needs advanced configuration features.
Again, Linux has a completely different set of requirements because of how it is used and its expected audience.
You are apparently arguing Windows has a major problem because it lacks the configurability out-of-the-box to be used as a high-end (in terms of features) firewalling router device. Your position is, at best, specious. Windows has firewalling capabilities in line with its expected audience and use. That such functionality is less than a platform with an almost completely different set of requirements is not bad programming, but good engineering.
Slashdot Mirror
You're vaguely on the right track, but woefully inaccurate in the details.
OS/2 was a joint IBM and Microsoft venture, originally meant to replace DOS on "high end" PCs.
DOS and Windows was meant for "low end" PCs.
Around 1988, Microsoft started working on what was originally meant to be OS/2's replacement, to be run on 386+ machines (OS/2 targeted 286s, at IBM's insistence, and it was a huge mistake). This product eventually became Windows NT, but at the time was known as "OS/2 NT" (which is why those "WNT=VMS+1" comments are so easily demonstrated as wrong).
However, Windows 3.0 was a surprise (to everyone) runaway success and that subsequently caused friction between Microsoft and IBM - IBM still wanted the default API in "OS/2 NT" to be "OS/2", Microsoft now wanted it to be "Windows". Since they couldn't agree, Microsoft took their ball (NT) and left. This was the famous "Microsoft-IBM divorce" and happened in (late, IIRC) 1990.
IBM were then left with the task of taking OS/2 1.x and turning it into OS/2 2.x (and successors). IBM, however, did not "own" OS/2 (at least not all of it) and even as late as OS/2 4.0, were still paying Microsoft royalties for their code.
Also, from about Windows 3.0 onwards, DOS became less and less "the OS" and more and more "the bootloader", as Windows took over the low level functionality. Windows 3.0 was doing memory management, CPU scheduling and some hardware support (video, printers). Windows 3.1 took over more hardware support (disk, network). By Windows 95, DOS was pretty much irrelevant (past bootup), unless you were unlucky enough to still be running old 16 bit code. Microsoft killed DOS in about 1993 - it just took everyone a decade to say goodbye.
As the transition sees developers pushing out binaries for both chips, I don't see a downside of Apple straddling the fence and using both types of chips.
That's probably because you don't have to worry about their shareholders.
Especially once you go into businesses. The difference in price between Apple and Microsoft starts getting huge once you spec out an environment for >50 people. With Apple you know your clients cost you $129, your server $999, Remote Desktop for $499, no limits, everything integrated with Kerberos + LDAP. With Microsoft you got that and then you have to start calculating CAL's for Exchange, CAL's for your Terminal Server, CAL's and server licenses for your SMS and WUS, and each little piece that will make life easier as the admin costs you an extra CAL which is all included in the Server+Client+RemoteDesktop licenses on Mac.
Of course, the Windows environment you've just described (Exchange, Terminal Server, etc) is _vastly_ more functional than the OS X environment you described, so it's not surprising in the least it would cost more.
Next to that, the average Windows machine lasts 3 years before it get's old and slow. I've seen G3's running OS X 10.4 without a hitch, G4's are mainstream in many companies and most haven't even gone to Intel yet.
If someone is happy with a G3 or G4 running OS X, they'll be just as happy with a P2, P3 or P4 running Windows XP. It's *far* kinder to older hardware than OS X is.
at I don't understand is, since Mac software has to be Universal nowadays anyway, why Apple doesn't just permanently keep its lineup as a mix of PPC and x86, picking whichever chip suits the particular machine they're designing at the time? Power6 Xserves along side Core 2 laptops... it sounds good to me!
Because the market for such machines is miniscule and costs in designing, testing and mass-producing computers are non-trivial.
(This is before even getting into the additional costs that would be heaped on developers, discouraging them from writing for the platform.)
In other words, it would cost Apple a lot of money while delivering few, if any, advantages (and a whole bunch of disadvantages).
I've actually heard of kids in middle and high school who use SMS and IM so much that they legitimately don't know how to spell words like "you", "your/you're", and will use internet abbreviations (lol, idk, etc.) in school papers.
Spelling and grammar (especially grammar) were heading downhill long before every kid had a mobile phone. I'll agree the problem has been getting worse in recent years (and that "TXTing" is probably exacerbating the situation), but the stage has been set since the mid-late 80s.
At least in Australia, grammar more complicated than identifying verbs and nouns hasn't been given more than a cursory nod in most schools for about twenty years. Spelling hasn't fared much better - even 10 years ago the typical example of "you're" vs "your" was becoming common.
Personally, I see three main culprits - the first is TV replacing books as "idle entertainment", the second (more the last decade) is the plethora of websites with atrocious spelling and grammar providing a massive feedback loop and the third is the education environment (teachers who have bad English skills and curriculums that basically don't try to teach much more than speaking the language).
Sadly, mathematics is going the same way, with calculators becoming "essential" in primary school years and the ability of people to do problems in their heads more complicated than adding double digit numbers - and especially things like manipulating fractions (eg: for working out percentages) - practically considered black magic to the average 25 year old.
This I do not understand. If it is no longer profitable to sell software, then why would someone keep their source secret?
Because non-trivial software - especially in the commercial sector where all the money is - requires support.
This is before even getting into interesting stuff like Trusted Computing, signed binaries and the like.
If copyright disappeared overnight, why would Microsoft want to spend any resources on improving Windows?
Same reason they do now - keep their customers satisfied.
Assuming that they would no longer be able to make any money from it.
A questionable assumption.
Wouldn't it be better to just make the whole software side public domain and concentrate on making hardware? This way the community would improve on their software, and they wouldn't need to spend money on it anymore.
The "money" is still being spent.
Further, the "public domain" still hasn't demonstrated conclusively it can provide a better product.
Without copyright, the GPL would not be needed!
With an attitude like that, I can only presume you have little understanding of the GPL's purpose.
Without copyright, the GPL could not work. This is a completely different thing to the GPL not being needed. Should you agree with the philosophy behind the GPL, it is difficult to see how you could consider copyright (or something very similar) unnecessary.
If there was no copyright, then there would be no incentive to keep your source code secret, as you wouldn't be able to sell your software anyways.
Uh, no. Indeed, without copyright the complete opposite would be true - there would be orders of magnitude more incentive to "keep your source secret", since there would no longer be any inherent protection from the legal system.
If you believe abolishing copyright would obviate the need for the GPL, then you haven't thought your cunning plan all the way through. Without copyright the GPL is completely irrelevant, its restrictions unenforcable, and its objectives much less attainable (if not - practically speaking - impossible). Personally this doesn't bother me in the least, since I'm not a fan of either copyright or the GPL - but if you have strong feelings about the latter, you might want to make sure your attitude to the former is consistent with them.
In other words, we beat them at their own game!
How do you figure that ?
While the GPL is a somewhat... unorthodox... usage of copyright, it's in complete agreement with the fundamental principles underlying it.
Tell me again, how is this part of the intent of copyright law???
Much of the "intent" of copyright has changed over the years, however, one common theme is the ability of the holder of a particular work's copyright to dictate under what conditions it can be used and, particularly, distributed. That is all the GPL does and, hence, code licensed under the GPL is unquestionably being used in line with the "intent" of copyright (ie: you get to say what people do with your code).
The GPL uses copyright law to make sure your work never becomes part of the farse of copyright.
No, it doesn't. Quite the opposite, in fact.
The GPL completely and utterly _relies_ on copyright to have any purpose. Without copyright, the GPL is meaningless and the restrictions it imposes would be impossible.
In other words, for licensing your code under the GPL to mean anything, your work *must* "become part of the farse[sic] of copyright".
Business are used to run Windows on their workstation. Microsoft has never supported additional platforms for long time (Alpha has only had 1 NT version made for it. Itanium had only 1 XP version made for it) because supporting multiple platform is hard for them (the only reason they'll keep support for 64bit x86_64, is that they'll kill 32bits instead and thus they'll still have only 1 main architecture to support). Thus there was a lack of interest from the largest consumers because of this absence (of course there are a lot of shops running Unices. But they aren't profitable on the same level as Dell's consumers).
Windows NT was available on x86, Alpha, MIPS from its first release in 1993. In 1995, NT 3.51 added PPC. Support for MIPS and PPC was dropped around 1997-98 and support for Alpha in about 1999 (Windows 2000 still had an Alpha version up until RC2). In 2001 Itanium support was added and in 2003 x86_64 was added.
There was also an internal port in the mid 90s to SPARC and, IIRC, PA-RISC. This was probably NT 3.51 and/or NT 4.0.
The implication that Windows NT is not, or was not, "portable" and that Microsoft "has difficulty" porting it to new platforms, has no foundation in reality.
This mean that when Windows-the-next (tm) comes out, either there will be a massive switch toward other OS (very likely in university labs) or it will see an even slower reception than Windows Vista is currently experiencing.(very likely on Joe 6-pack's older 32bits machine).
This argument is made frequently (and similar ones have been in the past with earlier releases), but it's difficult to see any sort of logic behind it. Most sales of Windows are through an OEM channel, not individuals buying upgrades. Further, most people who are buying upgrades, are doing so to put on a relatively high-end PC.
The suggestion that the next version of Windows only supporting 64 bit hardware ca. 2010 is going to cause any meaningful migration to other platforms, doesn't even pass the laugh test, given that:
* Most sales of that version of Windows would be for new (ie: x86_64) hardware;
* Of the remainder, most would be for relatively high end (ie: x86_64) hardware;
* The transition period will be measured in years; and
* All mainstream alternatives will almost certainly be in the same position
Computer programs do not all have the same function.
Most computer programs have a common set of identical functionality. Some examples are manipulating windows (resizing, closing, etc), manipulating files (open, save, etc), manipulating text (copy, paste, etc), online help, changing settings.
It is a significant boost to productivity, learnability and ease of use when these common types of functionality are presented in a consistent and predictable fashion.
Further, there are a number of general UI principles - like Fitt's Law - that can also be used (where applicable), regardless of specific implementation.
Looks quite easy to me.
That's because you're not doing anything interesting.
Try working with multiple vlans, bonded interfaces, multiple drives in VMs, SANs, etc, then come back.
If you think that Xen is a "royal PITA" to get running, then how do you manage to work with a server?
Most likely, he's trying to do non-trivial things with Xen. Stuff like bridges to multiple vlans, interface bonding, multiple drives, and the like.
Xen *sucks* from an administrative perspective if your environment is anything remotely complicated. VMWare has nothing to fear.
There is a slight learning curve, but after that it's extremely easy to deal with. Much like most Unix things. The massive performance benefit that Xen gives over VMWare Server (the only one I use) is well worth the hour it takes to learn the basics.
It takes a lot more than an hour to get Xen up and running in a non-trivial configuration. The management advantages of VMWare easily make it a better choice than Xen. Stupid amounts of computing power is dirt cheap these days, while people time is only getting more expensive.
Where are you talking about?
Australia. Like the person I replied to said.
Most places in the world, local calls are "free" (which means there is a fixed charge per month), for home lines.
In Australia local calls are charged (but untimed).
In addition you get a phone number that people can call, a phonebook listing, etc. Most places you can't unbundle these.
In Australia, you pay to make calls, not receive them. It follows, therefore, that the "services" you describe revolving around receiving calls, are "free".
The Athlon X2 was superior to the Pentium D. It wasn't until Core 2 Duo that Intel took the lead in desktop CPUs.
I think you mean "regained". Computers did exist before 2003.
The GP is obviously referring to the voice part of the landline, which is not necessary to the operation of the ADSL.
The line rental fee doesn't include any voice services. They cost extra. The line rental fee is exactly that - the cost you pay for a (working) physical connection to the system. Calls (or other things like messagebank) cost extra on top of that.
On both of those days I was modded down for suggesting that the Linux release-when-ready model offered greater security than the Microsoft release-when-we-feel-like-it model.
Say what ? By what definition of 'ready' does Linux have a "release-when-ready" model ?
"Release for the heck of it" would be a better description.
Let me reword that - how come I have to pay for a full suite of voice related services on my line when I just want to use DSL?
What makes you think you do ? The line rental fee doesn't include the cost of any voice services (ie: you pay per call for local, time-based for STD, for message bank, etc).
So I have a landline I never use.
Yes, you do. You use it for DSL. How else do expect to get ADSL other than over a landline ?
God they're filthy (Telstra) - hopefully we'll have a change of Government soon & get rid of the current spineless Prime Minister John Howard - who can't stand up to Telstra.
Huh ? The Australian Government regulates the hell out of Telstra (and a good thing, too, given the circumstances).
Everyone behind her was following too closely. Plain and simple. They should have time to react, apply their own brakes, change lanes, anything. People need to learn to stop following so closely.
Even if there weren't any crashes, someone stopping suddenly in the fast lane of a motorway is still going to cause chaos at the time and probably for tens of minutes - if not hours - afterwards.
She's not unintelligent (though, being blonde, she did get a certain amount of follicle-related humour directed at her), but she did as she was told, in a pressure-situation.
Anyone who considers driving down one of Britain's Motorways (which are both a) in relatively good condition and b) filled largely with relatively polite, disciplined drivers) a "pressure situation" shouldn't be allowed to hold a license. Neither should anyone who slams on the brakes because the onboard computer said 'stop', either, for that matter.
What's tragic is that she, along with many others, will just laugh this off as some silly "accident", when she should be banned from driving for some time, slapped with a large fine, and given a brutal dressing down by a judge. Incompetent driving like hers is the kind of thing that gets people killed.
The woman's car got crunched because the rail crossing was so poorly lit and poorly marked that she didn't know she was on train tracks.
Sounds like a Darwin Awards candidate.
Your comments demonstrate an utter lack of comprehension of computer security, as when you say that a home user "has to be" the sys admin.
Who, then, are you proposing will be "administering" their PCs ?
And suggesting that this has anything to do with "scale" is completely irrelevant.
Scale is an inescapable and intrinsic component.
Ironic that you accuse me of having no comprehension of computer security, given your comments.
You've made any number of general statements about NT security design without backing any of them up.
That's because anyone with knowledge of NT knows they are true. Further, anyone without knowledge of NT arguing against them is not worth listening to.
At the same time, you've made general statements about Linux security not being "fine grained" and the like which demonstrate a complete lack of comprehension of both security and Linux security.
(Classical) UNIX (and hence, Linux) security is not at all fine-grained. A user is either root or not root. The root user can do anything because, literally, the security infrastructure for them does not apply. Non-root users can be divided into smaller groups but, ultimately, they are all effectibely equal and the restrictions to each other's resources are almost completely limited to those things which can be represented within the filesystem (with a handful of exceptions, like binding to low ports).
The UNIX security model is both very coarse and based around the assumption that if a user does not have the privileges to do X, then they impersonate another user who does have that privilege (typically - and in many cases, unavoidably - root). This is an inherent security problem, because it both a) negatively impacts auditability and b) means vulnerabilities in the programs running at higher privilege - even temporarily - typically confer ALL the rights of the user (NOT just those necessary to complete the task) to the exploiter.
It is inescapably less capable and less secure than the NT (/VMS, since that's where NT inherited it from) model.
Meanwhile, the reality of the marketplace is that Windows is a disaster as an secure OS.
This statement is meaningless, firstly because it lacks a definition of what you mean by "secure OS" and secondly because no other OS faces the same exposure and requirements as Windows.
You have yet to come up with any ways in which Linux is more secure, although there has been lots of handwaving and examples of configuration problems related to different requirements.
I think that speaks for itself.
Indeed. Just like the people who argue blacks are more prone to criminal behaviour based on the reasoning there's more of them in jail, speak for themselves.
When Linux has eliminated Microsoft as a factor and has the market share of Windows, and when Linux has as many security problems as Windows, email me.
I have a better idea. If any platform manages to do that and do it _without_ the "security problems" Windows has had, and without resorting to something like Trusted Computing, you can email me and have a great time telling me how wrong I was.
Sorry, you do not understand "sudo". It does not give you "System" privilege - it gives you "root" privilege and then only on entry of the root password. On UNIX/Linux, that's almost the same thing, given the power of root on UNIX, but not exactly. The PowerPrompt utility gives you a higher level privilege than the system administrator has.
Firstly, 'root' is more powerful than SYSTEM. Root, literally, does not have any security restrictions imposed on it at all. SYSTEM does and, more importantly, is subject to ACLs just like every other account (whereas root does not).
Secondly, 'sudo' requires the user password, not the root password.
I understand sudo quite well. I have had the frustrating task of trying to use it to allow the secure management of numerous UNIX machines by a team of sysadmins, all with different responsibilities and access levels.
As for privilege escalation, before Windows NT/2000, basically you didn't even need to do it because everybody WAS "root".
DOS-based versions of Windows are not relevant to this discussion.
Even after Windows NT, virtually every exploit in Windows involved "privilege escalation" in the sense that most exploits allow you to do things that are not allowed in UNIX/Linux without root privilege.
For example ?
Spyware couldn't exist on Linux [...]
Of course it could.
- and doesn't - without root privilege - [...]
Much like Windows, then ?
[...] it's trivial on Windows for spyware to set itself up with privileges that a normal user can't handle.
That's because Windows NT has a vastly more fine-grained and capable security model than UNIX's primitive "you're root and can do anything or you're a user and can hardly do anything".
That's WHY the PowerPrompt tool was invented - to give the admin user enough power to deal with spyware that even the sys admin can't get rid of.
"Administrator" and "sys admin" are not synonyms.
The bottom line: Windows has demonstrated since day one that it has a less secure design than UNIX.
The design of Windows is fundamentally more secure than (traditional) UNIX. It lacks the inherent hole of a superuser, it has a vastly more fine-grained and capable security infrastrucutre and it is able to apply that infrastructure throughout the entire OS, rather than (basically) just those things accessible via the filesystem.
You would be hard pressed to find any measurable way that the design of NT is less secure than the "design" of UNIX. And it is trivial to find ways the design of Windows NT is better than the design of UNIX.
The single biggest influence on UNIX's "security record" is its user demographic.
Windows XP continues to install the default user as a sys admin by design [...]
Actually that's a minor configuration issue, not a design problem, and is easily remedied. It exists because most Windows machines are installed in either managed networks (where the default user is not Administrator, anyway) or on unmanaged home desktops (where they essentially must be Administrator). It is a less than ideal, but perfectly valid, engineering tradeoff.
[...] - and that is the single dumbest idea Microsoft ever had (next to the Registry as a single point of failure for the whole system.)
I'm guessing you're as clueless about the Registry as you are about pretty much everything else you've commented on.
That stupidity in itself annihilates any security advantages Microsoft might have ever designed into NT.
In the scenario of an unmanaged end-user desktop, the user *has* to be the "sys admin", basically by definition. If(/when, maybe) Linux had to deal with that problem on any sort of scale, the results will be basically identical. Linux probably won't, however, and it will be OS X that has the pleasure once it starts to hit critical mass.
Following Microsoft's instructions, please explain how to configure the firewall there described to block all TCP traffic except connections on port 1434 from network 192.168.0.0/16. You can't do it. You can block a particular port from operating at all on a particular interface but that's about it. Its... pathetic.
It's also not something that belongs on the type of OS Windows 2000 is (meant for managed corporate networks).
The XP firewall does more, but it still doesn't do the very obvious task: allow connections to port 1434 from these three corporate netblocks and nowhere else.
I haven't actually _tested_ it, but it's certainly possible to input the necessary configuration into XP's firewall.
And that's the whole point, isn't it? Security and the devices which support security do not become requirements in Windows until late in the game... like locking the proverbial barn door after the animals have all fled. In Linux and most of the other Unixes, the security devices tend to make it into the software BEFORE the widespread security events that compel their use.
It's not the "security events" the drive the requirement, it's the expected usage. Home firewalls have only really become a important in the last 5 - 7 years and, ultimately, host-based firewalling is a less than ideal solution with the better one (separate hardware device) being applicable to a significant proportion of installations.
The same requirements do not apply to Linux.
Like perhaps "Internet Connection Sharing" in Windows? And hey, what do you know, Windows can work as a plain router too. It even supports RIP and OSPF natively. What it lacks is any of the tools necessary to make that work securely.
Because it's an incredibly niche configuration situation for Windows. There are relatively few situations where Windows is being used as a router, even fewer where it needs to do anything more complicated than a simple 1:many NAT and practically zero where it needs advanced configuration features.
Again, Linux has a completely different set of requirements because of how it is used and its expected audience.
You are apparently arguing Windows has a major problem because it lacks the configurability out-of-the-box to be used as a high-end (in terms of features) firewalling router device. Your position is, at best, specious. Windows has firewalling capabilities in line with its expected audience and use. That such functionality is less than a platform with an almost completely different set of requirements is not bad programming, but good engineering.