Are these new changes happening out of some desire to resemble the old Windows software as little as possible?
Well, yes. It has long been a pastime among Microsoft's Office team to reinvent the wheel rather than using the standard Windows GUI controls. Perhaps the changes introduced in Office get adopted in the next version of Windows (for example the shaded gradient on title bars), perhaps not.
The silly thing is that you end up with a mixture of software using different widget styles since the style of menu to display seems to be burned into the executable. Some apps will have old Windows-style grey menu bars, some will have Office 2003 white menus with dropdown shadows, others the slightly different style used in Office 2002, some draggable and some fixed, but they're all doing the same thing. Even a stock installation of Windows with no third-party apps has different styles for window borders between, say, Control Panel and Command Prompt. Surely the sane way to do things is to have a standard Windows interface for 'please make a menu bar', and then when an innovation like draggable menus or hiding unused menu items comes up, it can apply to all applications consistently. Unfortunately I fear that the Win32 API is too low-level for something like that to work.
(NB I'm not implying that the free software world is any better; historically Unix desktops have been far worse than Windows for lacking a consistent look and feel between applications. It's improving, and distributions like Ubuntu are doing sterling work in trying to harmonize look and feel between programs written with different toolkits. At least a Linux system has only one copy of (say) GTK 2.x installed, so when the GTK appearance changes all the 'g' programs remain consistent.)
Some suggest that for Microsoft, the inconsistency in appearance is deliberate. Once you have the new Office 2000+x installed, applications from year x-1 start to look a bit out of date in comparison. You need to upgrade. Get a new version of Windows and your old Office version isn't quite right any more; you get a slightly dirty feeling using such old software that doesn't quite fit with the rest of the desktop; best to go and buy the latest one just to be on the safe side. You can compare this with the car market where styling changes are made from one year to the next to help make the old model look old-fashioned and encourage buyers to trade up.
The user interface to power steering is just the same as ordinary steering - you turn the wheel. The interface to use antilock brakes isn't any more complicated than old-fashioned breaks. Traction control 'just works'; you don't have to fiddle with settings for it to help.
I guess these are examples of the ideal way to improve things: you don't have to relearn anything to use the improvement, it's just magically better. A shame that so few software improvements follow this path. I guess improved font rendering, faster speed, or better reliability are examples.
That's exactly the analogy. If you plead guilty, you often get a lighter sentence. Those who voluntarily turn themselves in to the police for a murder (which otherwise would have gone undetected) might tend to get shorter prison time than those who get caught. (I don't have the figures - but in principle, it is at least possible to have such a system and it would make some sense.)
The fact that you can't log in as a particular user does not mean the user doesn't exist. sudo runs commands as the root user, for example, so the root user does exist. The article criticizes Windows for running most services as user SYSTEM, although you cannot log in as that.
(Of course I agree that not allowing direct root logins is the right way to do things.)
He seems to argue that Windows is less secure than OS X partly because if your Windows system gets infected, you can't trace the source of the problem, but with OS X you have a better chance of doing so. However I think this is the wrong thing to emphasize. If a piece of malware gets true root access on a system then it can do what it likes, including loading new kernel modules to hide files in the filesystem and so on. It's only lack of skill by some rootkit authors that make them detectable (so in effect, it's security by obscurity; there's a good argument that operating systems should make it as easy as possible to do such nasty things once you get root, so nobody will be tempted to think 'such things are only theoretical').
Now he does mention that most services on OS X don't run with unrestricted privileges, so there is much less chance of malware getting root *in the first place*. This is the important thing to emphasize - not what to hopelessly fiddle with once you are already 0wned.
I guess by root I don't necessarily mean what OS X or BSD or even Linux call root, but the classical Unix notion of the Almighty user who can do anything. Many BSDs have securelevel settings meaning that even root is restricted from doing certain things.
You're forgetting the naming conventions of gaming hardware reviews. Remember that a video card is called a VGA, a hard disk is called a hard drive, memory is called sticks, and most importantly a computer is always called a rig. (A fast computer is a 'sweet rig'.)
I think they could have made more of the 'insensitivity' angle. They should have marketed it with: if you do not go and see Snakes on a Plane, then the terrorists have already won. It would also have helped to bring forward the release date to August 16th, planned date for the liquid explosive attacks on transatlantic jets.
It would be handy if the movie included some suspicious bearded character on the plane who in the end turns out to save it Wesley-Crusher style. I haven't yet seen the film, so for all I know perhaps it does.
I don't get it. Surely a trip to the moon doesn't use more energy than a thousand domestic aeroplane flights, that is, not very much compared to our total energy use each year. What are the numbers here?
Clearly, militaries in law-abiding countries would abide by the terms of the licence, at least as much as any private company would. The army or navy are not above the law and you can sue them just like anyone else for copyright violation. But as you say you couldn't expect Hezbollah or North Korea to have any such qualms. In principle, if you write software that might have military uses, trying to exclude that in the licence is supporting one side against the other.
What if other projects adopt "no military" clauses like we've seen lately?
Then they are not free software. They are shareware 'but you can look at the source code' or something like that. You'd class them together with all the other trial versions, 'evaluation licences' and FREE DOWNLOADS!!! that clog up the net.
A good quality 'professional' CRT (the ones with truly flat screens, as a rough rule) can usually be cranked up to 100Hz refresh, where it's pretty much impossible to see any flicker. I run an Eizo T561 at 1440x1080, 100Hz refresh. If you have a CRT then find out the hsync and vsync range, then play around with gtf(1) to generate a mode you like. Indeed, if you have an LCD you may still want to use gtf(1) but to do the opposite: to find the lowest refresh rate your screen will handle. My fairly ancient IBM flat panel (model 9513) started getting a wobbly display in hot weather when running at 75Hz refresh; lowering the refresh rate fixed that without making the screen flickery.
Do at least try out some of the high-end CRTs from Cornerstone or whoever before you dismiss all CRTs entirely. For brightness, contrast, life of colours and response time they can't be beaten; and while the individual pixels will never be quite as sharp as an LCD a high-quality unit can come close. (And as others mention, it can scale to different resolutions reasonably well.) That said, I don't think I will be buying any more CRTs because they're just too big, hot and heavy.
640x240 is not that small. 'High-resolution' CGA two-colour display was 640x200, and Windows up to 3.0 can run in that. For many years the standard PC resolution was 640x480, and if you choose your fonts carefully it's okay to halve the vertical resolution. The Archimedes had a pleasant and reasonably uncramped GUI in 640x256 (using an 8x8 font designed to work with rectangular pixels), and I think the Amiga and other 'home computers' designed to connect to television sets had a similar resolution.
I think this was in the days before excessive toolbars; the Arc also disposed of menubars and used the middle middle button to pop up a menu, which was a neat way to do things.
It's very easy to write a program that decompresses Wikipedia pages with perfect accuracy from a very small source file.
1. Download the Wikipedia article for, say, Jellied eels. 2. Follow the 'edit' link. 3. Change the page content to a 100 megabyte repeating sequence of 'I am a fish I am a fish'. 4. Generate this as output.
Screenshots are a bit pointless. In 'look and feel', feel is by far the more important. You can keep your transparent xterms and zooming window effects. Give me something ugly, solid and functional any day.
This is about the moral difference between protecting code from modification by signing it versus protecting code from modification by burning it into a PROM.
Perhaps 'moral difference' isn't the best way to judge this. The intent of the GPL is to get more freedom for computer users in practice. Practically speaking, the GPLv2 has been fairly successful here (at least in FSF's view), but as demonstrated by Tivo there are some cases where it's not guaranteeing the freedom of users to change the software running on their computers. If in practice a change to the licence can increase freedom in some cases, it would seem worthwhile.
It wouldn't make much sense to argue that because you can't block all possible ways for unscrupulous manufacturers to work around the GPL, you shouldn't try to block any of them. The existence of one theoretically possible but hardly seen method (burning GPLed code onto a ROM) isn't an argument for not doing something about a more common case (crippling the hardware so that Tivo, Inc. can change what runs on your machine, but you the owner cannot).
He is against DRM, but that's not the point
on
The FSF, GPLv3 and DRM
·
· Score: 4, Insightful
TFA gets it wrong. Richard Stallman is opposed to DRM; look at the 'Defective By Design' real-world protests of earlier this year. But that's not the point here.
Since the beginning the idea of free software (as rms sees it) is that if you use a program, you should have the freedom to modify it, among other freedoms. So if you have a Tivo, you should have the freedom to modify the software that runs on your Tivo. If Linux is GPLed, then it's clearly not allowed for the Tivo manufacturers to ship it with a label saying 'we forbid modifying the software'. It's also not allowed under the GPL for them to try blocking your freedom another way by withholding the source code. But under GPLv2 your freedom to change the program can still be taken away, by the manufacturer making the device only execute signed binaries (for which nobody but the manufacturer has the signing key). GPLv3 as proposed is about making sure your freedom to change the software running on your computer (or Tivo) isn't taken away like this.
Of course anyone can write GPLed software that has DRM restrictions. But if you use it, you should have the right to modify it, and remove the DRM if you don't want DRM on your computer. That is the important point.
Analogously: there is nothing in the GPL against charging a sum of money for the software. You can sell it for as much as you like. But if you do, the person who receives it still gets all the freedoms to use, share and change the program.
The DEC thing, as you say, was 1978, but the Gutenberg spam was in 1971. I think it qualifies as the first spam by the coverage of addresses (what modernday spammer can dream of sending to _every_ address on the Internet?). Maybe DEC's was the first commercial spam.
Slashdot Mirror
The silly thing is that you end up with a mixture of software using different widget styles since the style of menu to display seems to be burned into the executable. Some apps will have old Windows-style grey menu bars, some will have Office 2003 white menus with dropdown shadows, others the slightly different style used in Office 2002, some draggable and some fixed, but they're all doing the same thing. Even a stock installation of Windows with no third-party apps has different styles for window borders between, say, Control Panel and Command Prompt. Surely the sane way to do things is to have a standard Windows interface for 'please make a menu bar', and then when an innovation like draggable menus or hiding unused menu items comes up, it can apply to all applications consistently. Unfortunately I fear that the Win32 API is too low-level for something like that to work.
(NB I'm not implying that the free software world is any better; historically Unix desktops have been far worse than Windows for lacking a consistent look and feel between applications. It's improving, and distributions like Ubuntu are doing sterling work in trying to harmonize look and feel between programs written with different toolkits. At least a Linux system has only one copy of (say) GTK 2.x installed, so when the GTK appearance changes all the 'g' programs remain consistent.)
Some suggest that for Microsoft, the inconsistency in appearance is deliberate. Once you have the new Office 2000+x installed, applications from year x-1 start to look a bit out of date in comparison. You need to upgrade. Get a new version of Windows and your old Office version isn't quite right any more; you get a slightly dirty feeling using such old software that doesn't quite fit with the rest of the desktop; best to go and buy the latest one just to be on the safe side. You can compare this with the car market where styling changes are made from one year to the next to help make the old model look old-fashioned and encourage buyers to trade up.
The user interface to power steering is just the same as ordinary steering - you turn the wheel. The interface to use antilock brakes isn't any more complicated than old-fashioned breaks. Traction control 'just works'; you don't have to fiddle with settings for it to help.
I guess these are examples of the ideal way to improve things: you don't have to relearn anything to use the improvement, it's just magically better. A shame that so few software improvements follow this path. I guess improved font rendering, faster speed, or better reliability are examples.
Don't all mainframes have several power supplies? You should be able to unplug one or two of them without too much consequence.
In any case, we all write our software to be fully transaction-safe and cleanly recoverable after a power failure. Right? Right?
That's exactly the analogy. If you plead guilty, you often get a lighter sentence. Those who voluntarily turn themselves in to the police for a murder (which otherwise would have gone undetected) might tend to get shorter prison time than those who get caught. (I don't have the figures - but in principle, it is at least possible to have such a system and it would make some sense.)
The fact that you can't log in as a particular user does not mean the user doesn't exist. sudo runs commands as the root user, for example, so the root user does exist. The article criticizes Windows for running most services as user SYSTEM, although you cannot log in as that.
(Of course I agree that not allowing direct root logins is the right way to do things.)
He seems to argue that Windows is less secure than OS X partly because if your Windows system gets infected, you can't trace the source of the problem, but with OS X you have a better chance of doing so. However I think this is the wrong thing to emphasize. If a piece of malware gets true root access on a system then it can do what it likes, including loading new kernel modules to hide files in the filesystem and so on. It's only lack of skill by some rootkit authors that make them detectable (so in effect, it's security by obscurity; there's a good argument that operating systems should make it as easy as possible to do such nasty things once you get root, so nobody will be tempted to think 'such things are only theoretical').
Now he does mention that most services on OS X don't run with unrestricted privileges, so there is much less chance of malware getting root *in the first place*. This is the important thing to emphasize - not what to hopelessly fiddle with once you are already 0wned.
I guess by root I don't necessarily mean what OS X or BSD or even Linux call root, but the classical Unix notion of the Almighty user who can do anything. Many BSDs have securelevel settings meaning that even root is restricted from doing certain things.
Hey, this guy has a lower uid than me... I defer to his superior knowledge!
You're forgetting the naming conventions of gaming hardware reviews. Remember that a video card is called a VGA, a hard disk is called a hard drive, memory is called sticks, and most importantly a computer is always called a rig. (A fast computer is a 'sweet rig'.)
I think they could have made more of the 'insensitivity' angle. They should have marketed it with: if you do not go and see Snakes on a Plane, then the terrorists have already won. It would also have helped to bring forward the release date to August 16th, planned date for the liquid explosive attacks on transatlantic jets.
It would be handy if the movie included some suspicious bearded character on the plane who in the end turns out to save it Wesley-Crusher style. I haven't yet seen the film, so for all I know perhaps it does.
I don't get it. Surely a trip to the moon doesn't use more energy than a thousand domestic aeroplane flights, that is, not very much compared to our total energy use each year. What are the numbers here?
Clearly, militaries in law-abiding countries would abide by the terms of the licence, at least as much as any private company would. The army or navy are not above the law and you can sue them just like anyone else for copyright violation. But as you say you couldn't expect Hezbollah or North Korea to have any such qualms. In principle, if you write software that might have military uses, trying to exclude that in the licence is supporting one side against the other.
Then they are not free software. They are shareware 'but you can look at the source code' or something like that. You'd class them together with all the other trial versions, 'evaluation licences' and FREE DOWNLOADS!!! that clog up the net.
What are the odds that the moon turns out to be composed partly of gold, or platinum or palladium? Would moon mining be profitable?
Bart and Lisa are doing some recycling of those plastic ring things that hold six-packs of beer together.
Lisa: And, you have to cut these up first. Otherwise, animals get caught in them.
Bart: Only the stupid ones.
A good quality 'professional' CRT (the ones with truly flat screens, as a rough rule) can usually be cranked up to 100Hz refresh, where it's pretty much impossible to see any flicker. I run an Eizo T561 at 1440x1080, 100Hz refresh. If you have a CRT then find out the hsync and vsync range, then play around with gtf(1) to generate a mode you like. Indeed, if you have an LCD you may still want to use gtf(1) but to do the opposite: to find the lowest refresh rate your screen will handle. My fairly ancient IBM flat panel (model 9513) started getting a wobbly display in hot weather when running at 75Hz refresh; lowering the refresh rate fixed that without making the screen flickery.
Do at least try out some of the high-end CRTs from Cornerstone or whoever before you dismiss all CRTs entirely. For brightness, contrast, life of colours and response time they can't be beaten; and while the individual pixels will never be quite as sharp as an LCD a high-quality unit can come close. (And as others mention, it can scale to different resolutions reasonably well.) That said, I don't think I will be buying any more CRTs because they're just too big, hot and heavy.
I like that review... it starts off reasonable, and gets increasingly Time Cube as it goes on.
640x240 is not that small. 'High-resolution' CGA two-colour display was 640x200, and Windows up to 3.0 can run in that. For many years the standard PC resolution was 640x480, and if you choose your fonts carefully it's okay to halve the vertical resolution. The Archimedes had a pleasant and reasonably uncramped GUI in 640x256 (using an 8x8 font designed to work with rectangular pixels), and I think the Amiga and other 'home computers' designed to connect to television sets had a similar resolution.
I think this was in the days before excessive toolbars; the Arc also disposed of menubars and used the middle middle button to pop up a menu, which was a neat way to do things.
AMD will really miss an opportunity if they call the new chip anything other than Tetrathlon.
uunet was not a website.
It's very easy to write a program that decompresses Wikipedia pages with perfect accuracy from a very small source file.
1. Download the Wikipedia article for, say, Jellied eels.
2. Follow the 'edit' link.
3. Change the page content to a 100 megabyte repeating sequence of 'I am a fish I am a fish'.
4. Generate this as output.
Screenshots are a bit pointless. In 'look and feel', feel is by far the more important. You can keep your transparent xterms and zooming window effects. Give me something ugly, solid and functional any day.
It wouldn't make much sense to argue that because you can't block all possible ways for unscrupulous manufacturers to work around the GPL, you shouldn't try to block any of them. The existence of one theoretically possible but hardly seen method (burning GPLed code onto a ROM) isn't an argument for not doing something about a more common case (crippling the hardware so that Tivo, Inc. can change what runs on your machine, but you the owner cannot).
TFA gets it wrong. Richard Stallman is opposed to DRM; look at the 'Defective By Design' real-world protests of earlier this year. But that's not the point here.
Since the beginning the idea of free software (as rms sees it) is that if you use a program, you should have the freedom to modify it, among other freedoms. So if you have a Tivo, you should have the freedom to modify the software that runs on your Tivo. If Linux is GPLed, then it's clearly not allowed for the Tivo manufacturers to ship it with a label saying 'we forbid modifying the software'. It's also not allowed under the GPL for them to try blocking your freedom another way by withholding the source code. But under GPLv2 your freedom to change the program can still be taken away, by the manufacturer making the device only execute signed binaries (for which nobody but the manufacturer has the signing key). GPLv3 as proposed is about making sure your freedom to change the software running on your computer (or Tivo) isn't taken away like this.
Of course anyone can write GPLed software that has DRM restrictions. But if you use it, you should have the right to modify it, and remove the DRM if you don't want DRM on your computer. That is the important point.
Analogously: there is nothing in the GPL against charging a sum of money for the software. You can sell it for as much as you like. But if you do, the person who receives it still gets all the freedoms to use, share and change the program.
The DEC thing, as you say, was 1978, but the Gutenberg spam was in 1971. I think it qualifies as the first spam by the coverage of addresses (what modernday spammer can dream of sending to _every_ address on the Internet?). Maybe DEC's was the first commercial spam.