Windows has also descended from the model of its own ancestor, and pretty much giving the user unbridled control of the system. For someone replying to a story about Vista, you don't seem to be familiar with it -- the user privilege model works pretty much identically to Linux with kdesu (elevating to administrator as needed using an admin username & password; except that even if you're running as administrator, everything runs with standard user permissions by default, with the admin explicitely allowing any necessary process elevation). See http://en.wikipedia.org/wiki/User_Account_Control.
This meant going to the Unix model of separating administrator accounts from user accounts by default. This caused problems in many device drivers which had not been properly written to use user level privileges by default. You're confusing two different things here: user account privilege seperation, and kernel / user CPU modes.
Take the latter first. Code running in kernel mode runs on CPU ring 0; User mode, ring 3. What's been causing problems with device drivers is that Vista's new driver framework puts a lot more restrictions on what can run in kernel mode (e.g. kernel mode printer drivers are now banned).
But this is not the same thing as user account privilege seperation, which is a higher level distinction -- for example, different users might only have write privileges to their own home folders, wheras administrators have write privileges to the entire hard drive. This has also cause some compatibility issues, but with programs rather than device drivers -- mostly programs that write to areas they shouldn't (e.g. the root of the C drive), and thus will complain if they aren't run elevated.
So, for example, even if a program is elevated to administrator [user account privileges], the actual code will still almost always be running in User mode [CPU rings].
Moreover, a Feb. 18 report in the London Daily Express showed that there is nearly a third more ice in Antarctica than usual, challenging the global warming crusaders and buttressing arguments of skeptics who deny that the world is undergoing global warming." I'm guessing you're not very familiar with the Daily Express.
(Frankly, I'm surprised they had space to print some bollocks about global warming in between the Diana conspiracy theories and 365-days-a-year Madeline McCann coverage...)
If I hypothesised that the average score produced by throwing a die is 3.5; and you threw a die, and it came up as 1; would that disprove my hypothesis?
Based on the implied point of your post, I'm guessing you think the answer is yes. May I suggest looking up the words "average" and "trend" in a dictionary?
what exactly is the purpose of SmoothText? I don't want to feel that I'm pointing out the obvious, but the very first sentence on the page linked to by the GP gives you the answer...
I think this article is written by some M$ fanboy.
+2, Insightful Oh, come on, Slashdot.
To the parent: Spolsky was the program manager on the Excel team who developed VBA. Would you maybe prefer to read about the MS Office file formats from Erris/twitter, rather than someone who knows about them?
The fact is, Word in its early versions was NOT significantly faster than its competitors and neither was Excel. Word Perfect and Lotus 1-2-3 did everything people needed and they did it within the resource constraints of the day. Yes, because unlike Word; Wordperfect and Lotus 1-2-3 had nice, human-readable XML formats, didn't they?
Newsflash: no, they didn't. The reason Word's.doc design goals didn't produce significantly faster results than its competitors was because its competitors' file formats had exactly the same design goals!
I do believe one of Microsoft's goals here are to assist the process of those binary formats becoming obsolete, to drive Office 2007/2008 adoption. Whilst I agree with your reasoning, your conclusion (that they did it to drive Office 2007/8 adoption) is flawed, since you don't need to upgrade to Office 2007/8 to use the new formats; you just need to install the compatibility pack.
ITs a complete rewrite and the screen shot shows it can not even render text properly at the bottom left of the pic. More information is mentioned here The link you give completely contradicts you. The only part which is being replaced is SystemLayer.dll; the main codebase requiring changes in only "a few places". Hardly a complete rewrite!
and appearently the layers are so proprietary they have to be rewritten. I'm not sure how you came to that conclusion; since the entirety of Paint.NET, source included, is released under the MIT license...!
At some point, you're going have a computer so fast that everything is going to be pretty much instantaneous with everything maxed out. Well, *Windows Explorer* pretty much instantaneously responsive, yeah, but that's not exactly a very high point. Frankly, any computer with >= 1.5GB of RAM will be pretty much as responsive as it can be: there aren't any other limiting factors (Aero doesn't speed up as your GPU gets faster, either it can run it or it can't; and Vista isn't very taxing on the CPU, only RAM). So by that logic, a WEI score of maybe around 3 should be set to "max".
But that would be missing the whole point, since (as I understand it) WEI isn't intended to tell you how well a computer will run Vista (after all, you must have already installed Vista to have run the benchmarker...). It's intended to, in the future, provide an easier alternative to 3rd party programs, especially games, listing detailed system requirements. So in 5 years time, instead of "ID Software game X requires a 1.7GHz CPU (2.5GhZ recommended), 1GB RAM (1.5GB recommended), GPU with 256MB RAM (512GB recommended)... [etc]", you'd in theory just be able to say "Game X requires a computer with a WEI of 4.0, with a WEI of 5 or above recommended".
Not too long ago, I decided to install MacOS 10.4 on a crappy little test machine at work, an old 400mhz G4.... and all the smooth eye-candy was intact. Windows slide and fade in and out of view, transparency works like a charm. Your point? Alpha transparency has been supported in Windows since Windows 2000 and certainly doesn't require hardware acceleration. Ditto sliding and fading -- both things that Windows XP menus did by default way before hardware accelerated window management. But any effects that require hardware acceleration -- e.g. pixel shader 2.0 blurring, as used in Aero Glass's title bars or OS X 10.5's top menu bar -- won't work: if you installed Leopard on that PC, the menubar won't be blurry-translucent, and if you installed Vista, Aero Glass would be switched off since it uses the same effect. So what's your point?
Look at the "Windows Experience Index", and am getting about 60 percent of what I could be. Ummm, 60% of what?
There is no maximum value. To quote Raymond Chen: "Imagine what the world would be like if there were a max value. What happens if the max is 10 and you buy a 10 computer, and then an even faster computer comes out next year - what rating does that computer yet?" (source).
The max you can get on today's absolute best hardware may be around 5.9, but that's not the top end of a scale -- it will certainly increase with time as better hardware comes out and WEI is updated with newer benchmarking tecniques.
Slightly more relevent would be if you said you'd got less than 2, since 2 is what MS claims is the minimum for "Premium Capable".
Firstly, I'm pretty sure the GP was being sarcastic.
More importantly, though, your statement:
the fact is that if a virus runs at a user permission level then it can still affect that user's files Come on, think about that for a minute. Of course something that runs with a user's permissions can affect that user's files, that's what running with that user's permissions means! Having to elevate to above my user level in order to write to my own userspace would rather miss the point of having user permissions! The whole point is, if you run with a user's permissions, you can't affect *anyone else's* files (or system files),
I seem to recall that he stated he was retiring. And back in 2000, didn't he quit then, as well? Prior to 2000, BG was CEO and chairman of the board. In 2000, he quit as CEO, and took up a job as Chief Software Architect. Later this year, he will quit that job, so will no longer be employed by Microsoft; but will still be chairman of its board.
And I agree, totally. UAC is no substitute for kdesu. I don't understand your point. UAC and kdesu work pretty much identically if you run as a standard user, and solve exactly the same problem.
The problem is that, if you're not an Admin in Windows, you might as well throw out your computer. It's just not designed for user-casting in the way of the sudoer. Umm, huh? True, Windoes XP wasn't designed for sudoesque user-casting, but Vista is, using something called UAC -- you must have heard of it, you mentioned it in your previous paragraph...
Security, for example: Sure, Vista has the 'protect the user from himself by continually asking if he really wants to X' features, but I'd be happier with licensing agreements for bundling in AVG and Spybot I'm sorry, but a background antimalware scanner is no substitute for properly enforced privilege seperation -- c.f. Linux and OS X. Which is not to say that it's 'worse' or 'better' in any meaningful sense, since it's not a question of one or the other; they serve different purposes. A knowledgeable user can get by fine just with good security policy (a firewall, privilege seperation, keep everything updated, and don't download crap off the net); but for less knowledgeably users, you want both.
Win9X would also 'launch on top of' DOS, leaving it there all the time It certainly "launched on top of" DOS in the sense of, as I said, using it as a boot loader; but after loading Windows, Windows turned DOS off (source for this and other quotes: Raymond Chen). In other words, interactions with the hardwere were not routed through DOS; e.g. file system operations etc. were "handled by the 32-bit file system manager". So where does the 16-bit code come in? If that manager detected that a program was trying to hook into a DOS the system calls, it would automatically detect that and "jump back into the 16-bit code" to let the hook run.
So, in a sense, "MS-DOS was just an extremely elaborate decoy. Any 16-bit drivers and programs would patch or hook what they thought was the real MS-DOS, but which was in reality just a decoy. If the 32-bit file system manager detected that somebody bought the decoy, it told the decoy to quack".
I can well imagine that a system might spend "20-40%" of its time in 16-bit mode if a lot of the device drivers and programs were 16-bit, but to describe this as "absurd for a 32-bit OS" is silly: it proves nothing about the OS except that it has good backward device driver compatibility. With up to date device drivers and applications, it would spend the entirety of its time once booted up in 32-bit mode.
So actually Win9x is not a shell, by any definition of one:: a DOS shell would translate all system calls to 16-bit DOS calls and pass them along to DOS (which was indeed what Windows 1-3 did).
and in the older versions you could even 'exit' back to DOS, or configure the system to remain in DOS without launching Windows. Nope; you couldn't "exit back to DOS" in Windows 9x, because, again, it wasn't running on top of DOS. You could certainly reboot into DOS, since DOS was used as the bootloader: rebooting into DOS just stopped the boot process before it loaded Windows.
But you just keep trying to rewrite history there buddy. Oh, dear. You were doing so well up till then, too; I was almost fooled into thinking you actually were interested in having an interesting discussion about this. But then, I suppose this is Slashdot: can't have a debate about Windows without randomly throwing around unjustified inflammatory statements. Ah well.
Tim S PS, I liked your prior signature. I think the irrational fear was funny. Thanks! (Can't lay claim to making it up, though; it's quite an old maths joke, as is the new one). As to why I changed it, well... Short answer, mysticgoat gave me phobophobia....
And then you come along and just ruin it. Just. Ruin. It. With all your solid facts and historical accuracies. I hope it made you happy. :'-(
As regards to your sig, I still think that you should thank whatever gods there be that you have no real phobia. Well, I didn't up until I read that. Now I have phobophobia, fear of getting a phobia. I hope you're happy...
The MS-OOXML that Office saves documents in is not the same as the OOXML that MS spec'ed out to ECMA and got submitted to ISO. It's close, but different Lots of people have said that in this thread, but Wikipedia has no such claim and I can't find a source for any differences with a quick Google search. Could you possibly cite a difference between the submitted spec and what MS submitted, or point me to a source that does so? Thanks!
I met a traveller from an antique land,
Who said--"Two vast and trunkless legs of stone
Stand in the desert... Near them, on the sand,
Half sunk a shattered visage lies, whose frown,
And wrinkled lip, and sneer of cold command,
Tell that its sculptor well those passions read
Which yet survive, stamped on these lifeless things,
The hand that mocked them, and the heart that fed;
And on the pedestal, these words appear:
My name is Ozymandias, King of Kings,
Look on my Works, ye Mighty, and despair!
Nothing beside remains. Round the decay
Of that colossal Wreck, boundless and bare
The lone and level sands stretch far away."
Things like autoSpaceLikeWord95 are referenced but not specified. You original poster is right, you need to get some new FUD. AutoSpaceLikeWord95 was specified by the 1st January 2008 Czech ECMA response (AutoSpaceLikeWord95 was CZ-0014, more at CZ-0015 to CZ-0025)
You make some good points, but some rather bad ones.
1. There is already an ISO standard for this same purpose. Since, clearly, different competing standards are bad. Which is why there is only one standard type of screw drive head, Flathead. I once heard someone claim that there were other standards, such as Philips (better for automated assembly) and Pozidriv (allows latge torque without gouging the screw); but I reckon they were lying. I mean, how could competition possibly be better than one standard having a monopoly? Everyone knows how good monopolies are.
4. OOXML is technically very inferior to the existing standard, ISO 26300. For example, OOXML specifies three different implementations of "a table", instead of just one common to different Office applications. This means that you cannot write a "table handling class" as a library, but instead you have to duplicate equivalent functionality several times over. You cite one example where ODF is apparently better than OOXML. And indeed, Wikipedia cites several technical advantages of both ODF over OOXML and, conversely, OOXML over ODF. For example, ODF apparently has only a weakly defined formula syntax, inhibiting ODF spreadsheet implementations based only on the spec (supposedly most implementors just use whatever de facto syntax OO.org decides on). To claim that one format is universally hailed as technically "very inferior" is rather misleading at best.
6. OOXML is controlled by just one corporation... ISO 26300 belongs to ISO. That's a circular argument. It shouldn't be an ISO standard because it currently isn't an ISO standard?! (Granted, aspects of the canonical implementation will probably de fact be decided by what MS Office does, but then the same applies to ODF and OO.o -- see previous item...)
8. ISO 26300 even works with Microsoft Office (up to Office 2003) using a free plugin written by Sun. Microsoft deliberately broke Office 2007 file filters so that this plugin (or any other plugin not written by Microsoft) would not work in Office 2007 That is just plain wrong, and FUD to boot. Not only does a 10 second Google search show that the Sun plugin does support Office 2007, but Microsoft apparently also sponsored their own open-source ODF add-in (hosted on Sourcefourge) for Office, which also supports Office 2007 (& below).
10. It makes no sense to have "choice in standards" How is this different from your first point? Anyway, see my response to that.
Slashdot Mirror
Take the latter first. Code running in kernel mode runs on CPU ring 0; User mode, ring 3. What's been causing problems with device drivers is that Vista's new driver framework puts a lot more restrictions on what can run in kernel mode (e.g. kernel mode printer drivers are now banned).
But this is not the same thing as user account privilege seperation, which is a higher level distinction -- for example, different users might only have write privileges to their own home folders, wheras administrators have write privileges to the entire hard drive. This has also cause some compatibility issues, but with programs rather than device drivers -- mostly programs that write to areas they shouldn't (e.g. the root of the C drive), and thus will complain if they aren't run elevated.
So, for example, even if a program is elevated to administrator [user account privileges], the actual code will still almost always be running in User mode [CPU rings].
(Frankly, I'm surprised they had space to print some bollocks about global warming in between the Diana conspiracy theories and 365-days-a-year Madeline McCann coverage...)
If I hypothesised that the average score produced by throwing a die is 3.5; and you threw a die, and it came up as 1; would that disprove my hypothesis?
Based on the implied point of your post, I'm guessing you think the answer is yes. May I suggest looking up the words "average" and "trend" in a dictionary?
To the parent: Spolsky was the program manager on the Excel team who developed VBA. Would you maybe prefer to read about the MS Office file formats from Erris/twitter, rather than someone who knows about them?
Newsflash: no, they didn't. The reason Word's
But that would be missing the whole point, since (as I understand it) WEI isn't intended to tell you how well a computer will run Vista (after all, you must have already installed Vista to have run the benchmarker...). It's intended to, in the future, provide an easier alternative to 3rd party programs, especially games, listing detailed system requirements. So in 5 years time, instead of "ID Software game X requires a 1.7GHz CPU (2.5GhZ recommended), 1GB RAM (1.5GB recommended), GPU with 256MB RAM (512GB recommended)... [etc]", you'd in theory just be able to say "Game X requires a computer with a WEI of 4.0, with a WEI of 5 or above recommended".
There is no maximum value. To quote Raymond Chen: "Imagine what the world would be like if there were a max value. What happens if the max is 10 and you buy a 10 computer, and then an even faster computer comes out next year - what rating does that computer yet?" (source).
The max you can get on today's absolute best hardware may be around 5.9, but that's not the top end of a scale -- it will certainly increase with time as better hardware comes out and WEI is updated with newer benchmarking tecniques.
Slightly more relevent would be if you said you'd got less than 2, since 2 is what MS claims is the minimum for "Premium Capable".
More importantly, though, your statement: the fact is that if a virus runs at a user permission level then it can still affect that user's files Come on, think about that for a minute. Of course something that runs with a user's permissions can affect that user's files, that's what running with that user's permissions means! Having to elevate to above my user level in order to write to my own userspace would rather miss the point of having user permissions! The whole point is, if you run with a user's permissions, you can't affect *anyone else's* files (or system files),
So, in a sense, "MS-DOS was just an extremely elaborate decoy. Any 16-bit drivers and programs would patch or hook what they thought was the real MS-DOS, but which was in reality just a decoy. If the 32-bit file system manager detected that somebody bought the decoy, it told the decoy to quack".
I can well imagine that a system might spend "20-40%" of its time in 16-bit mode if a lot of the device drivers and programs were 16-bit, but to describe this as "absurd for a 32-bit OS" is silly: it proves nothing about the OS except that it has good backward device driver compatibility. With up to date device drivers and applications, it would spend the entirety of its time once booted up in 32-bit mode.
So actually Win9x is not a shell, by any definition of one:: a DOS shell would translate all system calls to 16-bit DOS calls and pass them along to DOS (which was indeed what Windows 1-3 did). and in the older versions you could even 'exit' back to DOS, or configure the system to remain in DOS without launching Windows. Nope; you couldn't "exit back to DOS" in Windows 9x, because, again, it wasn't running on top of DOS. You could certainly reboot into DOS, since DOS was used as the bootloader: rebooting into DOS just stopped the boot process before it loaded Windows. But you just keep trying to rewrite history there buddy. Oh, dear. You were doing so well up till then, too; I was almost fooled into thinking you actually were interested in having an interesting discussion about this. But then, I suppose this is Slashdot: can't have a debate about Windows without randomly throwing around unjustified inflammatory statements. Ah well.
New sig better?
Ozymandias
by Percy Bysshe Shelley
I met a traveller from an antique land,
Who said--"Two vast and trunkless legs of stone
Stand in the desert... Near them, on the sand,
Half sunk a shattered visage lies, whose frown,
And wrinkled lip, and sneer of cold command,
Tell that its sculptor well those passions read
Which yet survive, stamped on these lifeless things,
The hand that mocked them, and the heart that fed;
And on the pedestal, these words appear:
My name is Ozymandias, King of Kings,
Look on my Works, ye Mighty, and despair!
Nothing beside remains. Round the decay
Of that colossal Wreck, boundless and bare
The lone and level sands stretch far away."