We're going to have to agree to differ about Windows and NeXTSTEP's stability.
My point is that Windows on decent hardware will give you excellent stability. If it doesn't, you can either a) whine about it on Slashdot or b) remedy the situation by finding out what the problem is and fixing. You should not, however, consider anything that could be deemed "instability" as expected behaviour.
Comparing glibc and IE is comparing apples and oranges. With Linux you'd need to replace glibc with something that implemented the same functionality, but you can change the window manager or browser without difficulty.
IE IS NOT JUST A WEB BROWSER APPLICATION. It's a reusable *module*. It's an OS-level shared library just like glibc, QT, khtml, WebCore/WebKit, Quicktime, or any of a million others.
Your argument that Windows is modular because the modules have dependencies on each other is not only logically unsound, it demonstrates a deep misunderstanding about what "modular" actually means.
Cutting and pasting code isn't modularity - this is what Visual Studio encourages (code snippits). Simply put, nobody but Microsoft would even have thought to add such a feature.
Not being a VS user, I'm afraid I don't know what you're talking about. AfewGoogleresults, however, would suggest your complaint is completely without merit, if not outright lying.
Windows is definitely designed in a very modular fashion. That Microsoft don't have an easy way to custom-build your own Windows "distribution", does not change this.
WMF needed extra patches applied if Office was installed - simply put the code was duplicated from Windows into Office. It should have been in a framework that Office called.
Well, The January patch summary would suggest otherwise, however, the Office team are well-known for going their own way and reimplementing OS functionality. Again, your logic is faulty as this is irrelevant to the modularity of Windows - it's like arguing Linux isn't modular because there's a zillion different widget libraries for X.
Microsoft are not a single, well organised, coherent entity, they're a bunch of different groups that just happen to be grouped under the same umbrella. The Office and Windows teams may as well be different comapanies.
This kind of "design" is totally different to Mac OS X - where there are modules and clear interfaces.
So you're saying there isn't a *single* OS X application that reimplements functionality provided in the OS ? Because that's the logic you're trying to use to argue Windows isn't "modular".
Microsoft have said that Windows Vista is a total rewrite of Windows around these design ideas, clearly XP wasn't!
Windows VIsta isn't a "complete rewrite" in anything outside of staged press interactions. It's Windows NT 6.0 - certainly there have been some major changes and cleanups (particularly to do with internal dependencies), but nothing approaching a "complete rewrite".
Given the alternative those days, OS/2 was certainly superior to everything out there. 2.11 was years ahead of Win95... literally. Maybe a decade ahead of Win2k, literally. They included a web browser before MS included an IP stack.
Windows NT was better technology than OS/2 from the day it was released. Which is hardly surprising, considering it was originally meant to replace it.
That was support for multiple users was added. But when did MS start saying that programs should be developed so they can work in a non-root setting?
Probably long before then, given NT was multiuser since the day it was released.
You are making an extraordinary claim that Microsoft's development guidelines contradicted not only general best practices, but also sheer common sense. I think that demands some proof.
From the development perspective, "LUA-friendly" is just another feature. If you don't need it, you don't use it.
Rubbish. Taking advantage of per-user data stores and avoiding writing or modifying data in system areas unless absolutely necessary is are practices that any remotely competent software should follow without even thinking about it. Shit, I'm only a SysAdmin and even I remember enough from my programming classes to know that.
You just keep doing things the way you usually do. It's not until MS issues best practices and "written for XP"-type certification that development houses are going to change their in-house practices (if then).
Rubbish. If you're a developer with even a shred of pride in writing good software you don't write software that needs Administrator privileges to run.
The "encouragement" I was trying to refer to in the last paragraph was that situation. Microsoft didn't go out there saying "we want everyone to write insecure code". But they created a system that let developers assume they had global read/write to the harddrive. Once the developers fall into that mindset, changing it is going to be a long, slow process.
"Because I can" is not sufficient justification for writing bad software. Developers have had 7+ years to change their habits and code.
That's the encouragement I'm talking about - they set up single-user system, and then tried to change it mid-stride to a multi-user one.
No, the move from DOS-based Windows to NT-based Windows was a punch that had been telegraphed for a good 8 years by the time it actually happened. This included a transitional period of both OSes being - from the developers' perspective - "multi user" of _at least_ 3 years.
It was in no way a "sudden" or "mid-stride" change. It was a planned, structured transition. Developers have had no excuse for releasing apps that needlessly require Administrator level privileges for at _least_ a period of 5 - realistically closer to 7 - years.
It's a deliberate action, but they're not intentionally trying to run a program.
If they're consciously double-clicking an icon, expecting something to happen, it's a deliberate act. I'll grant you the *results* - installing malware instead of seeing the dancing elephants - might not be desired, but that's beside the point, which is that *users will run stuff that will hurt them if they think it will do something they want and this is nearly impossible to "protect" against*.
Except that every user has to know that running code can be dangerous.
Sadly, no. Mainly because most users don't really understand what "running code" (vs, say, "opening a document") really is and why a computer can't understand why something is "bad"[0] when it's so obvious to them.
Every user has to know that, or there's no helping them.
This is kind of the point I'm trying to make.
And most, especially these days, do.
Rubbish. If they did, malware would be an annoyance, not a plague.
If you simply modified the popup box that appears when you double-click a file with an unassigned helper app to say something like "This file appears to be an executable. Executable files may damage your computer and/or files. If you are sure you want to execute this file, right click and tick 'Make File Executable'" it would be a big help.
That the "Open Attachment" dialogs in Outlook have been saying essentially this for ~6 years now, and also defaulting to "Save" rather than "Open", sugges
Given the right PC (back in the day) with supported cards - NeXTSTEP ran just fine (it wasn't unstable).
Sounds just like Windows.
Microsoft stated that they couldn't remove IE because they'd "integrated" it - which seems the opposite of a clean modular implementation.
On the contrary, it's an excellent indicator of a modularity when common functionality is moved into a module and reused, rather than being implemented multiple times.
Would you say Linux wasn't "modular" because ripping glibc out of the typical distribution breaks pretty much everything ?
Further evidence is the WMF flaw that needed several patches applied (clearly that bug was in many places - same code, "cut and pasted" all over Windows and various applications!).
IIRC, the mutliple WMF patches all "fixed" different aspects of the code, not the same thing in different places.
That post should be +5 Informative - CPU utilization was always one of SCSI's biggest benefits over IDE (thank GOD the ATAPI folks have finally almost resolved that issue!) [...]
Uh, that "problem" has been solved for over a decade now.
(Sheesh, I've got a *486* with a DMA-capable IDE controller.)
There are a few reasons why SCSI is (sometimes) [much] faster than IDE. "CPU utilisation" hasn't been one of them since about 1992.
They encouraged it prior to the release of XP. Then they released XP, and changed the way programs are supposed to perform OS operations.
How ? Give some specific examples.
Ok, that was a good step. But you can't expect millions of programs out there to be re-written to do it the new way. Even though they have now changed the way they recommend programs be written for their OS, their previous stance still has repurcussions in the current state of windows software.
Developers have had no excuse not to be writing LUA-friendly applications since about 1998. That's when every shipping version of Windows had support for per-user profiles and registries.
I'd say a whole lot of malware is executed accidentally.
Ignorantly != accidentally. If they double-click an icon, that's deliberate.
Ok, so the people who will jump through a hundred hoops to destroy their system are beyond hope, no matter what architecture they use. But it's the people who encounter a file like [...] and click it that I'm talking about.
Look, I realise it's difficult for a lot of technically-inclined folks to grok, but these sort of people make up the _majority_ of computer users. Heck, the number of people who will even think _at all_ of the possible and expected results before trying to open some arbitrary file would struggle to hit a double-digit percentage.
To most people a PC is like a TV, only less reliable. They don't need to worry about some TV show destroying their PC, why should they worry about some piece of software doing that same thing ?
Particularly when the last ".exe" has been hidden by the OS.
Realistically, this makes no difference. If people will open password protected zipfiles to run software, then they'll follow the one or two instructions in an email it takes to chmod +x a file and run it.
The problem is that double-clicking a file has two meanings - "execute me" and "open me with associated program".
Conceptually, a double click action only has one meaning - "open this" (or "activate this"). Whether this actually entails running an executable or passing the file to a handler program is an implementation detail that the user doesn't - and shouldn't have to - know about (it's like the difference between turning the TV on via the power switch or the remote). I'd be astounded if you could find even a handful of people out of every hundred who knew about the concepts of "running an executable vs opening a file".
If the OS uses an execute bit, then it forces users to manually specify which of these actions are to be performed - which means fewer users accidentally executing what they assumed to be a non-executable file.
It won't work. All it'll do is piss people off who have to go through extra steps to do the things that have been completely transparent up until that day. Then they'll go and set the file executable and run it anyway.
(Also, FWIW, NTFS does have the equivalent of an "execute bit".)
Many of the machines used by inexperienced people I see these days are family machines. There's a Mum account, a Dad account, a little Johnny and little Betsie account.
You'll probably also find that the vast bulk of data on these machines is in a "shared folder". IIRC Windows calls it "Common Files".
Households with multiple computers (at least in Australia, might be different in the US) tend to be those with computer-savvy household members in my experience.
Well, I'm Australian as well and from what I've seen, most "household computers", if they're even used by more than one person, just have a single account that everyone uses.
And if your file system is permissions-based, it means little Johnnies naieve attempt to access the wonderful world of internet porn won't delete all Daddy's financial records.
I'm well aware of what it means, my point is that the vast bulk of non-OS/application data on any given machine i
Well there is an element of this - but NeXTSTEP ran okay on PCs (and at its core Mac OS X is NeXTSTEP).
Clearly you've never tried to get a NeXT up and running on a PC:).
I most of the rest of the key is clearly defined frameworks, and a modular design. I still find it amazing that Microsoft didn't do this - and that they told us they didn't do this.
Linux systems are designed to be run by users, and administered as root. Windows systems, by and large, are impossible to run as anything but root - many programs require root access to work properly, [...]
This is solely an application problem. It has _nothing_ to do with Windows.
[...] and Windows (up until recently) never had the equivelant of a linux sudo to get around that requirement.
It's always had the functionality.
Windows developers have been encouraged for years to write programs dependant on root access.
Encouraged how ? What Microsoft documentation can you provide showing that developers have been told to write applications dependant on Administrator level access ? How do you reconcile this claim with the requirement of the "Made for Windows XP" logo that applications must run in a normal user account ?
Execute permissions prevent accidental execution of malware on Linux, as does not having a stupid system of extensions which are so easily spoofed (especially when default windows behaviour is to hide recognized extensions!).
Very little malware is executed "accidentally". If you seriously think the need to run "chmod a+x" or GUI equivalent is going to stop many people from running their "watching the dancing elephants" program, you're delusional. People are happy to open up password-protected zipfiles to get their malware fix, having to make something executable is barely a speed bump.
The move over to NTFS was good, but it only really hit the public with XP.
Probably because XP was the first release of NT that was aimed at the public. Not that file permissions are particularly important to malware in the typical case.
I still know many people using FAT-based systems. How long has Linux been running a permissions-based filesystem?
Does it matter ? Most unmanaged machines are single user and the most important files on the system are the ones the user has full permissions to anyway.
Windows NT has defaulted to NTFS since the day it was released. If people are running it on FAT, someone has made a conscious decision to change the default.
There's a few architectural security advantages Linux has over windows.
Like what ? Having to manually make files executable is about the only thing I can think of that would come close to this, and at most it's a minor issue.
On the more abstract level, being open source gives Linux the potential to be more secure - it's hard to hide critical vulnerabilities in Linux, whereas MS has a history of doing so for windows.
There's been no shortage of "critical vulnerabilities" in OSS apps that have gone unnoticed for extended lengths of time.
On my new laptop, however, I was browsing around using IE while I waited for firefox to download, and in between the time it took to start the download, and the time it had finished, IE had managed to install a little bugger called Aurora for me . Thanks IE!
there is a HUMONGOUS difference between sex and prostitution.
Not really, just the currency.
quite frankly, women (and men) generally dont become prostitutes as the result of a life long dream they've had. I dont think there are kids who grow up saying "i wanna be a prostitute when i'm older!"
There are lots of careers that could be said about. Garbage collectors, checkout operators, taxi drivers, etc.
The women are generally treated badly and abused and its not what I'd consider a "fun" job. People become prostitutes because they have nothing left and need money, a lot of you are taking a simple black and white view of how it works.
I know of at least two "working girls" who went to school with my girlfriend that are earning well over AU$100k/year, have been since they were ~17 (so nearly ten years), already own one house, are well on the way to owning a second, have completed a Uuniversity degree and are planning to "retire" at ~30 to pursue a more "mundane" career.
Not every prostitute is doing what they do because they're a poor drug addict. Many are doing it because it can be easy, safe work with a high income and guaranteed employment.
Hmm, I seem to remember putting C: and a path into IE being an easy way to access the entire filesystem, including files to which I would not normally have permission.
And you are wrong. Putting a file path into the address bar loads up the filebrowser component (just like in KDE) and restricts you with the same file permissions you would have accessing those same files via regular Explorer or the Desktop (since they're all using the same component).
Sounds like access to a lot more than user space (which is where a browser should be running) to me.
Your logic, other than being based on a false assumption, is flawed.
See the above example. You can install and run firefox as a non admin.
You can run IE as a non admin as well - it runs with the privilege level of the user that starts it.
What's particularly entertaining about your example above is that KDE/Konquerer behaves _exactly_ the same way.
Not really.
Yes, they are. You would struggle to find any major architectural differences.
Sure all three offer HTML parsing services to other apps, but two are clean services to do that with applications that utilize them. IE is not a service, it is an application by itself that leaves a lot more room for the large number of security holes that are the result.
IE is a shared component that any application can (and many do) use to offer that functionality internally.
The IE "application" - iexplore.exe - is just a tiny wrapper around the various IE components.
It's exactly the same as khtml or WebKit/WebCore (or whatever the GNOME one is, I can't remember).
That is because you are erroneously thinking that monopolies are defined by products.
Er, no. Indeed, defining a monopoly by its product would be about the only way you could find Microsoft to be one.
This is not the case. They are defined by markets.
Indeed. And the market definition in this example was "x86 compatible desktop operating systems" (some may like to ruminate on substituting "PPC" for "x86" and how Apple would have fared in that situation). Note that this conveniently excluded Apple and MacOS, considered by most to be not only a viable option, but strong competitor, to Microsoft and Windows.
MS can add any products and features they want so long as they are not combining one product or bundle that combines their monopolized market and another one. You seem to be fundamentally failing to understand the concept of a monopoly or bundling.
No, I'm just confused as to why "bundling" a TCP/IP stack, text editor, calculator, image editor or any of the other myriad "applications" that Windows includes was ok, but a reusable browser component was not. Why is it ok to "integrate" a widget library or C library, but not an HTML renderer (both of which have historically been available seperately) ?
This failing is so common even on Slashdot that it would be a great example of how the public school system fails. It not only failed to teach you this basic concept but it also failed to provide you with the research and critical thinking skills needed to look up and do any basic research on a topic before asserting your uneducated opinion on it.
I did quite a bit of research on the topic back when the Antitrust case was news, and it seemed pretty clear at the time to me that it was a political crusade with little connection to real life. Just the exclusion of Apple/MacOS as a "competitor", alone, marks it as ridiculous. This is before even getting into the silliness of singling out a browser engine from the (relativley) vast amount of functionality that is shipped with Windows (or any other OS) or how Netscape was losing marketshare because their product sucked, not because IE was included in Windows. Microsoft's "integration" of IE was relatively innovative at the time and has since be copied by all the major alternatives (KDE, GNOME, OS X).
I guess this is the kind of reasoning that let other countries in history get away with similar things.
Considering that a) no-one is getting killed or in forced labour (indeed, quite the opposite) and b) they're free to leave whenever they want, I fail to see what's "similar" about them.
Which countries are you thinking of that have zero immigration controls and that don't remove illegal immigrants when they are identified ?
The thing about the Australian government as it stands today is that it's just pure evil.
If you think the Australian Government is "evil", then you're in desperate need of some context, or the word "evil" is completely meaningless your world.
Microsoft were found guilty of illegal use of their monopoly power. They appealed and the verdict stood. Perhaps you're a better lawyer than any of the ones they pay.
It's kind of hard to beat the system when the rules of the game are specifically written to make it impossible (and I'm speaking here of the "market definition" that was used).
One is that it mingles code for file browsing, web browsing, and vital parts of the OS.
No, it doesn't. These things are all modular.
Basically, it mixes code very insecurely in ways that allow interaction with the internet to potentially cause serious changes to the core of the OS. It also allows local users to abuse the Web browser and gain access to escalated privileges.
IE has no more ability to "cause serious changes to the core of the OS" than any other app like Firefox does.
Basically, it is an insecure and basically unfixable architectural mistake.
Architecturally, it's essentially identical to KDE's khtml or OS X's WebCore/WebKit.
The monopoly stuff I'm not really going to comment on, other than saying I disagree with the basic principle that Microsoft is - or ever was - a monopoly, and the idea that they can't add functionality to their product as customers or developers demand is ridiculous.
Unless I'm forgetting something, Apple has never used this "shared video memory" hackery in any of their systems.
Firstly, it's not really "hackery".
Secondly, way back in the 68k days, quite a lot of Macs used system RAM for video RAM.
Can't rule it out as totally impossible that they might in the future, but I'd find it quite surprising.
I wouldn't, particularly on the low end iBook/Mini replacements. Integrated video cards are respectable in terms of performance these days, whereas previously they sucked.
But in Linux breaking KDE just screws up the user's interface, it doesn't really takedown the whole OS.
Which, for people who KDE is the only reason they're using the OS, is the same as "breaking it".
The "problem" you have with Windows isn't that IE is any more integrated into it than khtml is into KDE (because it's not), it's that Microsoft don't sell a version of Windows that runs without a GUI. In the grand scheme of things, this is a tiny, irrelevant detail.
By all accounts, they're going to be around for a very long time yet. However, they're facing serious competition right now on several fronts. They're not the only option any more in operating systems, office suites, server platforms -- in fact, I don't think there's a single area in which they aren't facing some kind of competition.
The situation now is no different to the situation 10, 15, 20 years ago. Certainly, some of the names of the competitors has changed over time, but their presence has not.
Slashdot Mirror
It's hardly an "MO" unique to Microsoft...
Back around 1984, when the first Macintosh came out.
Bollocks.
My point is that Windows on decent hardware will give you excellent stability. If it doesn't, you can either a) whine about it on Slashdot or b) remedy the situation by finding out what the problem is and fixing. You should not, however, consider anything that could be deemed "instability" as expected behaviour.
Comparing glibc and IE is comparing apples and oranges. With Linux you'd need to replace glibc with something that implemented the same functionality, but you can change the window manager or browser without difficulty.
IE IS NOT JUST A WEB BROWSER APPLICATION. It's a reusable *module*. It's an OS-level shared library just like glibc, QT, khtml, WebCore/WebKit, Quicktime, or any of a million others.
Your argument that Windows is modular because the modules have dependencies on each other is not only logically unsound, it demonstrates a deep misunderstanding about what "modular" actually means.
Cutting and pasting code isn't modularity - this is what Visual Studio encourages (code snippits). Simply put, nobody but Microsoft would even have thought to add such a feature.
Not being a VS user, I'm afraid I don't know what you're talking about. A few Google results, however, would suggest your complaint is completely without merit, if not outright lying.
Windows is definitely designed in a very modular fashion. That Microsoft don't have an easy way to custom-build your own Windows "distribution", does not change this.
WMF needed extra patches applied if Office was installed - simply put the code was duplicated from Windows into Office. It should have been in a framework that Office called.
Well, The January patch summary would suggest otherwise, however, the Office team are well-known for going their own way and reimplementing OS functionality. Again, your logic is faulty as this is irrelevant to the modularity of Windows - it's like arguing Linux isn't modular because there's a zillion different widget libraries for X.
Microsoft are not a single, well organised, coherent entity, they're a bunch of different groups that just happen to be grouped under the same umbrella. The Office and Windows teams may as well be different comapanies.
This kind of "design" is totally different to Mac OS X - where there are modules and clear interfaces.
So you're saying there isn't a *single* OS X application that reimplements functionality provided in the OS ? Because that's the logic you're trying to use to argue Windows isn't "modular".
Microsoft have said that Windows Vista is a total rewrite of Windows around these design ideas, clearly XP wasn't!
Windows VIsta isn't a "complete rewrite" in anything outside of staged press interactions. It's Windows NT 6.0 - certainly there have been some major changes and cleanups (particularly to do with internal dependencies), but nothing approaching a "complete rewrite".
No, it doesn't. You'd struggle to find a single way they're similar.
API compatibility != kernel similarities.
Windows NT was better technology than OS/2 from the day it was released. Which is hardly surprising, considering it was originally meant to replace it.
Probably long before then, given NT was multiuser since the day it was released.
You are making an extraordinary claim that Microsoft's development guidelines contradicted not only general best practices, but also sheer common sense. I think that demands some proof.
From the development perspective, "LUA-friendly" is just another feature. If you don't need it, you don't use it.
Rubbish. Taking advantage of per-user data stores and avoiding writing or modifying data in system areas unless absolutely necessary is are practices that any remotely competent software should follow without even thinking about it. Shit, I'm only a SysAdmin and even I remember enough from my programming classes to know that.
You just keep doing things the way you usually do. It's not until MS issues best practices and "written for XP"-type certification that development houses are going to change their in-house practices (if then).
Rubbish. If you're a developer with even a shred of pride in writing good software you don't write software that needs Administrator privileges to run.
The "encouragement" I was trying to refer to in the last paragraph was that situation. Microsoft didn't go out there saying "we want everyone to write insecure code". But they created a system that let developers assume they had global read/write to the harddrive. Once the developers fall into that mindset, changing it is going to be a long, slow process.
"Because I can" is not sufficient justification for writing bad software. Developers have had 7+ years to change their habits and code.
That's the encouragement I'm talking about - they set up single-user system, and then tried to change it mid-stride to a multi-user one.
No, the move from DOS-based Windows to NT-based Windows was a punch that had been telegraphed for a good 8 years by the time it actually happened. This included a transitional period of both OSes being - from the developers' perspective - "multi user" of _at least_ 3 years.
It was in no way a "sudden" or "mid-stride" change. It was a planned, structured transition. Developers have had no excuse for releasing apps that needlessly require Administrator level privileges for at _least_ a period of 5 - realistically closer to 7 - years.
It's a deliberate action, but they're not intentionally trying to run a program.
If they're consciously double-clicking an icon, expecting something to happen, it's a deliberate act. I'll grant you the *results* - installing malware instead of seeing the dancing elephants - might not be desired, but that's beside the point, which is that *users will run stuff that will hurt them if they think it will do something they want and this is nearly impossible to "protect" against*.
Except that every user has to know that running code can be dangerous.
Sadly, no. Mainly because most users don't really understand what "running code" (vs, say, "opening a document") really is and why a computer can't understand why something is "bad"[0] when it's so obvious to them.
Every user has to know that, or there's no helping them.
This is kind of the point I'm trying to make.
And most, especially these days, do.
Rubbish. If they did, malware would be an annoyance, not a plague.
If you simply modified the popup box that appears when you double-click a file with an unassigned helper app to say something like "This file appears to be an executable. Executable files may damage your computer and/or files. If you are sure you want to execute this file, right click and tick 'Make File Executable'" it would be a big help.
That the "Open Attachment" dialogs in Outlook have been saying essentially this for ~6 years now, and also defaulting to "Save" rather than "Open", sugges
Sounds just like Windows.
Microsoft stated that they couldn't remove IE because they'd "integrated" it - which seems the opposite of a clean modular implementation.
On the contrary, it's an excellent indicator of a modularity when common functionality is moved into a module and reused, rather than being implemented multiple times.
Would you say Linux wasn't "modular" because ripping glibc out of the typical distribution breaks pretty much everything ?
Further evidence is the WMF flaw that needed several patches applied (clearly that bug was in many places - same code, "cut and pasted" all over Windows and various applications!).
IIRC, the mutliple WMF patches all "fixed" different aspects of the code, not the same thing in different places.
Uh, that "problem" has been solved for over a decade now.
(Sheesh, I've got a *486* with a DMA-capable IDE controller.)
There are a few reasons why SCSI is (sometimes) [much] faster than IDE. "CPU utilisation" hasn't been one of them since about 1992.
How ? Give some specific examples.
Ok, that was a good step. But you can't expect millions of programs out there to be re-written to do it the new way. Even though they have now changed the way they recommend programs be written for their OS, their previous stance still has repurcussions in the current state of windows software.
Developers have had no excuse not to be writing LUA-friendly applications since about 1998. That's when every shipping version of Windows had support for per-user profiles and registries.
I'd say a whole lot of malware is executed accidentally.
Ignorantly != accidentally. If they double-click an icon, that's deliberate.
Ok, so the people who will jump through a hundred hoops to destroy their system are beyond hope, no matter what architecture they use. But it's the people who encounter a file like [...] and click it that I'm talking about.
Look, I realise it's difficult for a lot of technically-inclined folks to grok, but these sort of people make up the _majority_ of computer users. Heck, the number of people who will even think _at all_ of the possible and expected results before trying to open some arbitrary file would struggle to hit a double-digit percentage.
To most people a PC is like a TV, only less reliable. They don't need to worry about some TV show destroying their PC, why should they worry about some piece of software doing that same thing ?
Particularly when the last ".exe" has been hidden by the OS.
Realistically, this makes no difference. If people will open password protected zipfiles to run software, then they'll follow the one or two instructions in an email it takes to chmod +x a file and run it.
The problem is that double-clicking a file has two meanings - "execute me" and "open me with associated program".
Conceptually, a double click action only has one meaning - "open this" (or "activate this"). Whether this actually entails running an executable or passing the file to a handler program is an implementation detail that the user doesn't - and shouldn't have to - know about (it's like the difference between turning the TV on via the power switch or the remote). I'd be astounded if you could find even a handful of people out of every hundred who knew about the concepts of "running an executable vs opening a file".
If the OS uses an execute bit, then it forces users to manually specify which of these actions are to be performed - which means fewer users accidentally executing what they assumed to be a non-executable file.
It won't work. All it'll do is piss people off who have to go through extra steps to do the things that have been completely transparent up until that day. Then they'll go and set the file executable and run it anyway.
(Also, FWIW, NTFS does have the equivalent of an "execute bit".)
Many of the machines used by inexperienced people I see these days are family machines. There's a Mum account, a Dad account, a little Johnny and little Betsie account.
You'll probably also find that the vast bulk of data on these machines is in a "shared folder". IIRC Windows calls it "Common Files".
Households with multiple computers (at least in Australia, might be different in the US) tend to be those with computer-savvy household members in my experience.
Well, I'm Australian as well and from what I've seen, most "household computers", if they're even used by more than one person, just have a single account that everyone uses.
And if your file system is permissions-based, it means little Johnnies naieve attempt to access the wonderful world of internet porn won't delete all Daddy's financial records.
I'm well aware of what it means, my point is that the vast bulk of non-OS/application data on any given machine i
Clearly you've never tried to get a NeXT up and running on a PC :).
I most of the rest of the key is clearly defined frameworks, and a modular design. I still find it amazing that Microsoft didn't do this - and that they told us they didn't do this.
Huh ?
This is solely an application problem. It has _nothing_ to do with Windows.
[...] and Windows (up until recently) never had the equivelant of a linux sudo to get around that requirement.
It's always had the functionality.
Windows developers have been encouraged for years to write programs dependant on root access.
Encouraged how ? What Microsoft documentation can you provide showing that developers have been told to write applications dependant on Administrator level access ? How do you reconcile this claim with the requirement of the "Made for Windows XP" logo that applications must run in a normal user account ?
Execute permissions prevent accidental execution of malware on Linux, as does not having a stupid system of extensions which are so easily spoofed (especially when default windows behaviour is to hide recognized extensions!).
Very little malware is executed "accidentally". If you seriously think the need to run "chmod a+x" or GUI equivalent is going to stop many people from running their "watching the dancing elephants" program, you're delusional. People are happy to open up password-protected zipfiles to get their malware fix, having to make something executable is barely a speed bump.
The move over to NTFS was good, but it only really hit the public with XP.
Probably because XP was the first release of NT that was aimed at the public. Not that file permissions are particularly important to malware in the typical case.
I still know many people using FAT-based systems. How long has Linux been running a permissions-based filesystem?
Does it matter ? Most unmanaged machines are single user and the most important files on the system are the ones the user has full permissions to anyway.
Windows NT has defaulted to NTFS since the day it was released. If people are running it on FAT, someone has made a conscious decision to change the default.
There's a few architectural security advantages Linux has over windows.
Like what ? Having to manually make files executable is about the only thing I can think of that would come close to this, and at most it's a minor issue.
On the more abstract level, being open source gives Linux the potential to be more secure - it's hard to hide critical vulnerabilities in Linux, whereas MS has a history of doing so for windows.
There's been no shortage of "critical vulnerabilities" in OSS apps that have gone unnoticed for extended lengths of time.
On my new laptop, however, I was browsing around using IE while I waited for firefox to download, and in between the time it took to start the download, and the time it had finished, IE had managed to install a little bugger called Aurora for me . Thanks IE!
Maybe not.
(Although I bet you weren't running as a non-Admin user, as well.)
Not really, just the currency.
quite frankly, women (and men) generally dont become prostitutes as the result of a life long dream they've had. I dont think there are kids who grow up saying "i wanna be a prostitute when i'm older!"
There are lots of careers that could be said about. Garbage collectors, checkout operators, taxi drivers, etc.
The women are generally treated badly and abused and its not what I'd consider a "fun" job. People become prostitutes because they have nothing left and need money, a lot of you are taking a simple black and white view of how it works.
I know of at least two "working girls" who went to school with my girlfriend that are earning well over AU$100k/year, have been since they were ~17 (so nearly ten years), already own one house, are well on the way to owning a second, have completed a Uuniversity degree and are planning to "retire" at ~30 to pursue a more "mundane" career.
Not every prostitute is doing what they do because they're a poor drug addict. Many are doing it because it can be easy, safe work with a high income and guaranteed employment.
And you are wrong. Putting a file path into the address bar loads up the filebrowser component (just like in KDE) and restricts you with the same file permissions you would have accessing those same files via regular Explorer or the Desktop (since they're all using the same component).
Sounds like access to a lot more than user space (which is where a browser should be running) to me.
Your logic, other than being based on a false assumption, is flawed.
See the above example. You can install and run firefox as a non admin.
You can run IE as a non admin as well - it runs with the privilege level of the user that starts it.
What's particularly entertaining about your example above is that KDE/Konquerer behaves _exactly_ the same way.
Not really.
Yes, they are. You would struggle to find any major architectural differences.
Sure all three offer HTML parsing services to other apps, but two are clean services to do that with applications that utilize them. IE is not a service, it is an application by itself that leaves a lot more room for the large number of security holes that are the result.
IE is a shared component that any application can (and many do) use to offer that functionality internally.
The IE "application" - iexplore.exe - is just a tiny wrapper around the various IE components.
It's exactly the same as khtml or WebKit/WebCore (or whatever the GNOME one is, I can't remember).
That is because you are erroneously thinking that monopolies are defined by products.
Er, no. Indeed, defining a monopoly by its product would be about the only way you could find Microsoft to be one.
This is not the case. They are defined by markets.
Indeed. And the market definition in this example was "x86 compatible desktop operating systems" (some may like to ruminate on substituting "PPC" for "x86" and how Apple would have fared in that situation). Note that this conveniently excluded Apple and MacOS, considered by most to be not only a viable option, but strong competitor, to Microsoft and Windows.
MS can add any products and features they want so long as they are not combining one product or bundle that combines their monopolized market and another one. You seem to be fundamentally failing to understand the concept of a monopoly or bundling.
No, I'm just confused as to why "bundling" a TCP/IP stack, text editor, calculator, image editor or any of the other myriad "applications" that Windows includes was ok, but a reusable browser component was not. Why is it ok to "integrate" a widget library or C library, but not an HTML renderer (both of which have historically been available seperately) ?
This failing is so common even on Slashdot that it would be a great example of how the public school system fails. It not only failed to teach you this basic concept but it also failed to provide you with the research and critical thinking skills needed to look up and do any basic research on a topic before asserting your uneducated opinion on it.
I did quite a bit of research on the topic back when the Antitrust case was news, and it seemed pretty clear at the time to me that it was a political crusade with little connection to real life. Just the exclusion of Apple/MacOS as a "competitor", alone, marks it as ridiculous. This is before even getting into the silliness of singling out a browser engine from the (relativley) vast amount of functionality that is shipped with Windows (or any other OS) or how Netscape was losing marketshare because their product sucked, not because IE was included in Windows. Microsoft's "integration" of IE was relatively innovative at the time and has since be copied by all the major alternatives (KDE, GNOME, OS X).
Microsoft is not a
Considering that a) no-one is getting killed or in forced labour (indeed, quite the opposite) and b) they're free to leave whenever they want, I fail to see what's "similar" about them.
Which countries are you thinking of that have zero immigration controls and that don't remove illegal immigrants when they are identified ?
No. "Non residents" who are in the country legally are free to move around the country as they please.
Barbed wire ringed camps out in the desert for concentrating illegals in the same place?
Actually it's for keeping illegal immigrants away from everyone else and making escape difficult and inadvisable.
If you think the Australian Government is "evil", then you're in desperate need of some context, or the word "evil" is completely meaningless your world.
How are these "features" going to cause problems when they never get used ?
You also insist on comparing KDE to Windows...which is FAR from the case.
No, I compare a Linux distro with KDE to Windows.
Windows is an OS, KDE is a nice pretty addon thing. You can use a GUI in Linux without Gnome or KDE...just plain ol X server and a WM.
Yes, as I said, your complaint is that Microsoft don't sell a version of Windows without the GUI. Which is, as I said, irrelevant in the big picture.
khtml integrated into KDE isn't anywhere near the same thing as IE being integrated into the OS, except in a superficial manner.
In fact, the architectures of khtml and IE are basically identical.
kthml doesn't affect the inner workings of the system, IE does.
No, it doesn't. It's just a shared component that runs in the context of the current user.
It's kind of hard to beat the system when the rules of the game are specifically written to make it impossible (and I'm speaking here of the "market definition" that was used).
No, it doesn't. These things are all modular.
Basically, it mixes code very insecurely in ways that allow interaction with the internet to potentially cause serious changes to the core of the OS. It also allows local users to abuse the Web browser and gain access to escalated privileges.
IE has no more ability to "cause serious changes to the core of the OS" than any other app like Firefox does.
Basically, it is an insecure and basically unfixable architectural mistake.
Architecturally, it's essentially identical to KDE's khtml or OS X's WebCore/WebKit.
The monopoly stuff I'm not really going to comment on, other than saying I disagree with the basic principle that Microsoft is - or ever was - a monopoly, and the idea that they can't add functionality to their product as customers or developers demand is ridiculous.
Firstly, it's not really "hackery".
Secondly, way back in the 68k days, quite a lot of Macs used system RAM for video RAM.
Can't rule it out as totally impossible that they might in the future, but I'd find it quite surprising.
I wouldn't, particularly on the low end iBook/Mini replacements. Integrated video cards are respectable in terms of performance these days, whereas previously they sucked.
In fact, it's "tied in" in an almost identical fashion to IE in Windows.
[...] and you're free to use Firefox, Camino, Opera or any other browser.
Wow. Just like Windows.
Which, for people who KDE is the only reason they're using the OS, is the same as "breaking it".
The "problem" you have with Windows isn't that IE is any more integrated into it than khtml is into KDE (because it's not), it's that Microsoft don't sell a version of Windows that runs without a GUI. In the grand scheme of things, this is a tiny, irrelevant detail.
Uh, no. If you delete everything "browser related" out of a KDE-based distro, KDE will break.
The situation now is no different to the situation 10, 15, 20 years ago. Certainly, some of the names of the competitors has changed over time, but their presence has not.