Doesn't Apple actually have a patent on the idea of a "trash can icon"? It seems Windows and everybody else can get away with a drag&drop target as long as they call it "Wastebasket" or "Recycle bin" or "Dumpster". Even Steve Jobs had to call it a "Recycler" (an image of the recycling logo) when he made the NeXT.
I do think it is strange that Apple has gone the direction everybody else has and made a "pretty" wastebasket icon (most Linux desktops get away with this and can call it "Trash" because the name is a "user option", apparently). It looks like a modern designer office wastebasket. One problem is that such wastebaskets are designed primarily to be non-obtrusive and hide their function. I don't understand why OS/X did not keep the "american suburban aluminum trashcan" as the icon, it was quite recognizable and immediatly identifiable as a trash container, and they seem to have a lock on that, as I have not seen anybody copy it for fear of Apple. They could draw it in 3D and make it all shiny and bright and new, but it would be more of a link to their past and I would think users would like it.
Windows has all pointing devices it finds controlling the same mouse cursor. This works pretty good actually for most programs. With some extra code (which I don't know) you can actually find out about the individual devices and see which one the user moved last, and get other information like pen angle and pressure.
X is completely screwed up. Plugging in a second mouse and one of them will most likely be ignored, but I have also heard of it confusing things so neither works. The official interface is XIE (X Input Extension) which I looked at once but it is a nightmare (though from what I have heard the Windows interface is equivalent, but at least you get movement and clicks without using it). I certainly hope the new X guys address this, it can't be that hard. I would like them to add an interface so all movement produces events that are sets of "device FOO moved to X" (where X is a number), and a fast and simple way of looking up "FOO" and finding out that it is a horizontal or vertical control or buttonN or knobN or another name assigned to it. See Irix GL for hints on how to do this, on it a single and quite simple interface handled all pointing devices, tablets, boxes of buttons and knobs, and the keyboard itself.
I am right-handed, and for the same reason I have also switched to a trackball on the left for the past 6 years. It is an older Kensington with a pool-ball-sized ball. It only has two buttons and I mapped the right one to the left mouse button and the left one to the middle, and pushing both to the right mouse button (this matches the extensive use of middle-mouse button in X and our software). I have gotten quite good at using this as a pointing device, while still easily able to use others computers with normal mice. I can also use a normal mouse with my left hand and no button remapping quite well, this I have on my home machine, but I found that if I swapped the buttons on a normal mouse I got confused.
One annoyance is that newer X systems don't let me do this button remapping. It never did it as a GUI option, but old ones would run a dot-file on login. I can't find any way of getting KDE to run an arbitrary set of commands on login. Instead I am forced to type a command into a terminal after I log in. This is really annoying and a sign that they are turning Linux into Windows and there may be nothing that can be done about it...
Using case to indicate different things has well-established history in mathematics notation back to at least the 1700's, and is often used in phonetics and pronouciation indicators that date back to 1900 at least. It was not invented by K&R.
Don't be an ass. With "enough context" you can replace entire words with meaningless alternatives, or delete them, or insert new ones, or rearrange them, and still "understand" the text. With "enough context" you can understand the text even if whole paragraphs are missing or rearranged. With "enough context" you can understand something written in Chinese.
When Linux appeared there were at least a dozen other equally developed unix clones, including the Minux that Linux was somewhat based on. And there was BSD which already matched what Linux would be perhaps 4 years later (BSD's main holdup was that it was not designed for 386). Yes it is highly likely that any major change in the 1980's would mean the guy named Linus would not have been the developer of the main alternative operating system. But somebody would have.
The only thing that would have stopped a Linux from appearing would be if the main popular system was based enough on Unix to stop the desire for a Unix clone. Unfortunately at the time all possible Unix clones cost money and it was quite obvious that making a CLI and floppy-based filesystem was cheaper than buying anything powerful like Unix. Unix would certainly have worked on even the earliest PC's, it was desinged for PDP-11 that had 64K of memory and disks smaller than the 5.5 inch floppy!
*ALL* graphics formats designed for non-photo images contained the ability to store vectors, so this is no big deal. The first drawing programs in the 1950's stored vectors!
What they are talking about is the use of vectors to store images for all the little icons and buttons and fonts and other pieces of the GUI. I think if you look at your old copy of PowerPoint you will see that all the buttons on the toolbar are bitmaps and they cannot scale.
This is wrong. The main thing Terminal Services does is compress the images when it sends them. Otherwise it works pretty much at the GDI32 level, sending commands like "draw a line" and so on that are equivalent to Xlib.
The main problems with X are:
1. The Xlib graphics library is extremely limited and primitive. It is physically impossible to draw a modern GUI with it, so lots of programs such as GTK revert to drawing an image locally and sending it. Now this is also true of Window's GDI32 but to a lesser extent (for instance you can draw rotated characters with it) so more Windows programs are able to use it to get the display they want.
2. X does not compress the images when sending them, so when things revert to images to get what they want, it is far slower than it needs to be. The attempt to do so (the XImage Extension) turned into a useless bloated over-engineered mess so that nobody used it (something which I'm afraid all Linux and Windows attempts to make a desktop environment are doing). I am unsure if Terminal Services compresses images, but I suspect it does. VNC definately compresses images.
Just to be technical: the current Freedesktop.org standard allows a window manager to implement an arbitrary number of desktops and add or delete a new one at any time.
However I think it lacks any way for a program to create or destroy a desktop. Or perhaps this is possible, but I am pretty certain most or all window managers do not expect this and will either ignore it or crash.
Along with this serious defect is the fact that most window managers do not provide any easy user interface for creating a new desktop, they either fix it at a constant or they require you to change "how many desktops".
So technically, it is about 50% possible, but this is invisible to the average user.
Plan9 has these as well and calls them "union mounts".
Putting the name and a colon at the start is not important but too many people think it is and fail to see how Unix mounts and Plan9 designs are exactly equivalent to the MSDOS filenames, the filenames on many other systems, the URL syntax that is even being pushed by the Linux desktop filesystems, and apparently by the Amiga.
In fact if you assumme colon and slash are not allowed in filenames, it is trivial to swap between colon-prefix and Unix syntax. "name:" turns into "/name/" to generate a Unix-style name. The Unix-style name has several advantages: only one reserved character, support for symbolic links that cross devices, and simpler parsing rules where only the first character needs special handling.
Bah. The "corporate environment" can lock down their machines easily. It's called a *PASSWORD*. You can easily make Windows run a small number of programs and lock everything down with the Administrator password (yes I know it is easier to do this with Linux, but it is possible with Windows).
DRM is an attack on the home user, the independent software developer, and anybody who wants to learn how computers work.
Absolutely 100% of DRM's purpose is to make the computer a method of delivering pay-for-play entertainment. *EVERY* other explanation is a LIE.
This has dire consequences for creativity, one of the drivers for creativity is the need to generate new content to replace already-sold content. A correctly working DRM would allow old entertainment to generate income forever, and in fact new content, by reducing the sale value of old content, is actually undesirable, and will be actively attacked (for instance by requiring a license to "sign" your content so it can be played on consumer machines).
Larry is wrong. As copyright holder you can say anything you want. Linus could have said that programs that start with the letter 'a' are derived from Linux and must be released under the GPL, but all others are not derived.
There seems to be confusion about Microsoft's monopolistic practices.
What Microsoft definately did, and continue to do, is disallow dual-boot machines. This was done to kill off OS2 and BeOS, though it also continues today to prevent anybody from practically trying Linux or any other alternative. Microsoft does not allow a manufacturer to sell a Windows machine that even has the disk partitioned so that you could install Linux without screwing up the Windows installation.
I'm pretty certain Microsoft's actions worked very well for them. If this had not been done, back in 1990 or so all the manufacturers would probably have come up with dual-boot machines, where the other system was BeOS or some other (perhaps manufacturer-proprietary) system, advertised as the "gaming" system. And all the 3D graphics and games would have been made for the gaming system. Big manufacturers would have locked in their own games so they could support their own proprietary systems, and I would expect there would be no Playstation, instead that market would be covered by inexpensive dual-boot machines.
Users would be quite used to and accepting that they have to reboot their machines to switch from work to playing games. But then, to Microsoft's horror, there would be "productivity" applications that would start to appear for the game, advertised clearly as "uses the better graphics, and no reboot necessary!". They predicted this and they did what they could to squash it.
I don't think Microsoft has ever been too concerned about blank machines. Only geeks buy those. Any large corporation that did so and tried to install Windows would either be paying more than if they bought the pre-installed ones, or would be breaking the law.
Before now, these operations could only be done by miniaturizing a small submarine and 5-person crew and injecting them into the bloodstream, so that they could reach the clot and destroy it with a miniaturized industrial laser. This was an extremely expensive operation, and risky due to the fact that the miniaturization only lasted an hour.
Person C definately has to stop distributing the code. Otherwise the original author's copyright is still being violated.
If person C stops distributing the code, then your answer 2) is correct, they are not liable for anything. But, if person C starts complaining and refusing to comply then they are liable. Depending on what happens, person C could lose a lawsuit, or person C (perhaps a huge company with lots of lawyers) might haul the whole case back into court and get the original judgement thrown out.
Suing the project or employee would be exactly a legitimate thing they could do. They sue them for copyright violations. They have to first exactly publish detailed information about what parts of the code are infringing and must be removed. Claiming stuff that is not theirs would probably get their case thrown out, so they must be very specific. This is called "mitigating damages" and not doing it is proof that you are not losing anything due to this copyright violation and in fact may be profiting from it, which can also get your case thrown out.
Suing the users, or threatening to sue the users, is not acceptable and not legal. Very clever of you to put the wrong thing and right thing into the same item so that it is impossible to pick one of them.
As several people had pointed out, the first step is to inform the infringer exactly how they are infringing and try to work out a peaceful means of solving the problem. Who knows, the infringer may agree to pay for the rights to your code, and you make money without going to court at all.
Microsoft was not interested in SCOX either winning or losing the case. Even a complete win would probably be forgotten in a few months. They instead wanted to delay everything as long as possible so that as many scary warnings about how dangerous Linux might be would be published in the PHB press.
People who believe the conspiracy theory think the scientists and government are lying and they actually know that there will be an eruption soon and we will all die but they are lying because they don't want to cause a panic, or for some other more insidious reason, like the fact that the illuminanti have all escaped earth and left the rest of us to die. It is a "conspiracy" because they all got together and organized and agreed to tell this lie.
Colors only need to be matched between two different images displayed on the same screen. This is quite possible even if the screen is adjusted completely wonky.
Anybody using Maya or Shake or any other package designed on SGI machines replaces the mouse with a 3-button one. All 3-button USB mice work. Even the scroll wheel works in lots of Mac programs (though the 3D systems, being designed for non-scroll-wheel mice, usually ignore it).
We all know the closed-source driver is better. But it is not better *because* it is closed source. Some argue that it would be even better open-source because people would fix it, but even if nothing was changed, it would be exactly the same open-source.
The argument is that Intel might demonstrate that releasing the source for something does not cause you to go out of business tomorrow.
The GPL means I OWN MY CODE, and you cannot use it for your own closed-source unless you talk to me and pay me for the right to use it.
It is as far from socalist as you can get and still see the code. The BSD license is "socalist" if you want, though really the people who release that are doing it because they want to, not because the government is telling them to.
But the GPL is applied for strictly selfish and greedy reasons and would may Ayn Rand quite happy.
The real "Socialists" and "Communists" are the people like you who whine "I can't use your code in my closed-source program unless I pay you. Wahh!!!! Unfair!!!! Gimme, gimme, gimme!" Tough luck, jerk. That's called capitalism and selfish greed on my part. Live with it, but don't you dare call me a Communist for wanting to control something I WROTE!
Slashdot Mirror
Doesn't Apple actually have a patent on the idea of a "trash can icon"? It seems Windows and everybody else can get away with a drag&drop target as long as they call it "Wastebasket" or "Recycle bin" or "Dumpster". Even Steve Jobs had to call it a "Recycler" (an image of the recycling logo) when he made the NeXT.
I do think it is strange that Apple has gone the direction everybody else has and made a "pretty" wastebasket icon (most Linux desktops get away with this and can call it "Trash" because the name is a "user option", apparently). It looks like a modern designer office wastebasket. One problem is that such wastebaskets are designed primarily to be non-obtrusive and hide their function. I don't understand why OS/X did not keep the "american suburban aluminum trashcan" as the icon, it was quite recognizable and immediatly identifiable as a trash container, and they seem to have a lock on that, as I have not seen anybody copy it for fear of Apple. They could draw it in 3D and make it all shiny and bright and new, but it would be more of a link to their past and I would think users would like it.
Windows has all pointing devices it finds controlling the same mouse cursor. This works pretty good actually for most programs. With some extra code (which I don't know) you can actually find out about the individual devices and see which one the user moved last, and get other information like pen angle and pressure.
X is completely screwed up. Plugging in a second mouse and one of them will most likely be ignored, but I have also heard of it confusing things so neither works. The official interface is XIE (X Input Extension) which I looked at once but it is a nightmare (though from what I have heard the Windows interface is equivalent, but at least you get movement and clicks without using it). I certainly hope the new X guys address this, it can't be that hard. I would like them to add an interface so all movement produces events that are sets of "device FOO moved to X" (where X is a number), and a fast and simple way of looking up "FOO" and finding out that it is a horizontal or vertical control or buttonN or knobN or another name assigned to it. See Irix GL for hints on how to do this, on it a single and quite simple interface handled all pointing devices, tablets, boxes of buttons and knobs, and the keyboard itself.
I am right-handed, and for the same reason I have also switched to a trackball on the left for the past 6 years. It is an older Kensington with a pool-ball-sized ball. It only has two buttons and I mapped the right one to the left mouse button and the left one to the middle, and pushing both to the right mouse button (this matches the extensive use of middle-mouse button in X and our software). I have gotten quite good at using this as a pointing device, while still easily able to use others computers with normal mice. I can also use a normal mouse with my left hand and no button remapping quite well, this I have on my home machine, but I found that if I swapped the buttons on a normal mouse I got confused.
One annoyance is that newer X systems don't let me do this button remapping. It never did it as a GUI option, but old ones would run a dot-file on login. I can't find any way of getting KDE to run an arbitrary set of commands on login. Instead I am forced to type a command into a terminal after I log in. This is really annoying and a sign that they are turning Linux into Windows and there may be nothing that can be done about it...
Using case to indicate different things has well-established history in mathematics notation back to at least the 1700's, and is often used in phonetics and pronouciation indicators that date back to 1900 at least. It was not invented by K&R.
Don't be an ass. With "enough context" you can replace entire words with meaningless alternatives, or delete them, or insert new ones, or rearrange them, and still "understand" the text. With "enough context" you can understand the text even if whole paragraphs are missing or rearranged. With "enough context" you can understand something written in Chinese.
When Linux appeared there were at least a dozen other equally developed unix clones, including the Minux that Linux was somewhat based on. And there was BSD which already matched what Linux would be perhaps 4 years later (BSD's main holdup was that it was not designed for 386). Yes it is highly likely that any major change in the 1980's would mean the guy named Linus would not have been the developer of the main alternative operating system. But somebody would have.
The only thing that would have stopped a Linux from appearing would be if the main popular system was based enough on Unix to stop the desire for a Unix clone. Unfortunately at the time all possible Unix clones cost money and it was quite obvious that making a CLI and floppy-based filesystem was cheaper than buying anything powerful like Unix. Unix would certainly have worked on even the earliest PC's, it was desinged for PDP-11 that had 64K of memory and disks smaller than the 5.5 inch floppy!
*ALL* graphics formats designed for non-photo images contained the ability to store vectors, so this is no big deal. The first drawing programs in the 1950's stored vectors!
What they are talking about is the use of vectors to store images for all the little icons and buttons and fonts and other pieces of the GUI. I think if you look at your old copy of PowerPoint you will see that all the buttons on the toolbar are bitmaps and they cannot scale.
This is wrong. The main thing Terminal Services does is compress the images when it sends them. Otherwise it works pretty much at the GDI32 level, sending commands like "draw a line" and so on that are equivalent to Xlib.
The main problems with X are:
1. The Xlib graphics library is extremely limited and primitive. It is physically impossible to draw a modern GUI with it, so lots of programs such as GTK revert to drawing an image locally and sending it. Now this is also true of Window's GDI32 but to a lesser extent (for instance you can draw rotated characters with it) so more Windows programs are able to use it to get the display they want.
2. X does not compress the images when sending them, so when things revert to images to get what they want, it is far slower than it needs to be. The attempt to do so (the XImage Extension) turned into a useless bloated over-engineered mess so that nobody used it (something which I'm afraid all Linux and Windows attempts to make a desktop environment are doing). I am unsure if Terminal Services compresses images, but I suspect it does. VNC definately compresses images.
Just to be technical: the current Freedesktop.org standard allows a window manager to implement an arbitrary number of desktops and add or delete a new one at any time.
However I think it lacks any way for a program to create or destroy a desktop. Or perhaps this is possible, but I am pretty certain most or all window managers do not expect this and will either ignore it or crash.
Along with this serious defect is the fact that most window managers do not provide any easy user interface for creating a new desktop, they either fix it at a constant or they require you to change "how many desktops".
So technically, it is about 50% possible, but this is invisible to the average user.
Plan9 has these as well and calls them "union mounts".
Putting the name and a colon at the start is not important but too many people think it is and fail to see how Unix mounts and Plan9 designs are exactly equivalent to the MSDOS filenames, the filenames on many other systems, the URL syntax that is even being pushed by the Linux desktop filesystems, and apparently by the Amiga.
In fact if you assumme colon and slash are not allowed in filenames, it is trivial to swap between colon-prefix and Unix syntax. "name:" turns into "/name/" to generate a Unix-style name. The Unix-style name has several advantages: only one reserved character, support for symbolic links that cross devices, and simpler parsing rules where only the first character needs special handling.
Bah. The "corporate environment" can lock down their machines easily. It's called a *PASSWORD*. You can easily make Windows run a small number of programs and lock everything down with the Administrator password (yes I know it is easier to do this with Linux, but it is possible with Windows).
DRM is an attack on the home user, the independent software developer, and anybody who wants to learn how computers work.
Absolutely 100% of DRM's purpose is to make the computer a method of delivering pay-for-play entertainment. *EVERY* other explanation is a LIE.
This has dire consequences for creativity, one of the drivers for creativity is the need to generate new content to replace already-sold content. A correctly working DRM would allow old entertainment to generate income forever, and in fact new content, by reducing the sale value of old content, is actually undesirable, and will be actively attacked (for instance by requiring a license to "sign" your content so it can be played on consumer machines).
Larry is wrong. As copyright holder you can say anything you want. Linus could have said that programs that start with the letter 'a' are derived from Linux and must be released under the GPL, but all others are not derived.
There seems to be confusion about Microsoft's monopolistic practices.
What Microsoft definately did, and continue to do, is disallow dual-boot machines. This was done to kill off OS2 and BeOS, though it also continues today to prevent anybody from practically trying Linux or any other alternative. Microsoft does not allow a manufacturer to sell a Windows machine that even has the disk partitioned so that you could install Linux without screwing up the Windows installation.
I'm pretty certain Microsoft's actions worked very well for them. If this had not been done, back in 1990 or so all the manufacturers would probably have come up with dual-boot machines, where the other system was BeOS or some other (perhaps manufacturer-proprietary) system, advertised as the "gaming" system. And all the 3D graphics and games would have been made for the gaming system. Big manufacturers would have locked in their own games so they could support their own proprietary systems, and I would expect there would be no Playstation, instead that market would be covered by inexpensive dual-boot machines.
Users would be quite used to and accepting that they have to reboot their machines to switch from work to playing games. But then, to Microsoft's horror, there would be "productivity" applications that would start to appear for the game, advertised clearly as "uses the better graphics, and no reboot necessary!". They predicted this and they did what they could to squash it.
I don't think Microsoft has ever been too concerned about blank machines. Only geeks buy those. Any large corporation that did so and tried to install Windows would either be paying more than if they bought the pre-installed ones, or would be breaking the law.
Before now, these operations could only be done by miniaturizing a small submarine and 5-person crew and injecting them into the bloodstream, so that they could reach the clot and destroy it with a miniaturized industrial laser. This was an extremely expensive operation, and risky due to the fact that the miniaturization only lasted an hour.
Person C definately has to stop distributing the code. Otherwise the original author's copyright is still being violated.
If person C stops distributing the code, then your answer 2) is correct, they are not liable for anything. But, if person C starts complaining and refusing to comply then they are liable. Depending on what happens, person C could lose a lawsuit, or person C (perhaps a huge company with lots of lawyers) might haul the whole case back into court and get the original judgement thrown out.
Part of #2 is the correct answer:
Suing the project or employee would be exactly a legitimate thing they could do. They sue them for copyright violations. They have to first exactly publish detailed information about what parts of the code are infringing and must be removed. Claiming stuff that is not theirs would probably get their case thrown out, so they must be very specific. This is called "mitigating damages" and not doing it is proof that you are not losing anything due to this copyright violation and in fact may be profiting from it, which can also get your case thrown out.
Suing the users, or threatening to sue the users, is not acceptable and not legal. Very clever of you to put the wrong thing and right thing into the same item so that it is impossible to pick one of them.
As several people had pointed out, the first step is to inform the infringer exactly how they are infringing and try to work out a peaceful means of solving the problem. Who knows, the infringer may agree to pay for the rights to your code, and you make money without going to court at all.
Microsoft was not interested in SCOX either winning or losing the case. Even a complete win would probably be forgotten in a few months. They instead wanted to delay everything as long as possible so that as many scary warnings about how dangerous Linux might be would be published in the PHB press.
People who believe the conspiracy theory think the scientists and government are lying and they actually know that there will be an eruption soon and we will all die but they are lying because they don't want to cause a panic, or for some other more insidious reason, like the fact that the illuminanti have all escaped earth and left the rest of us to die. It is a "conspiracy" because they all got together and organized and agreed to tell this lie.
Colors only need to be matched between two different images displayed on the same screen. This is quite possible even if the screen is adjusted completely wonky.
Anybody using Maya or Shake or any other package designed on SGI machines replaces the mouse with a 3-button one. All 3-button USB mice work. Even the scroll wheel works in lots of Mac programs (though the 3D systems, being designed for non-scroll-wheel mice, usually ignore it).
Actually some of the drivers such as Nvidea the *could* include in the download without violating any license with Nvidea or anybody.
However I can't blame them for asking for money for a more complete product.
We all know the closed-source driver is better. But it is not better *because* it is closed source. Some argue that it would be even better open-source because people would fix it, but even if nothing was changed, it would be exactly the same open-source.
The argument is that Intel might demonstrate that releasing the source for something does not cause you to go out of business tomorrow.
If this is true, it means EV1's licenses cost $1000 or less. SCO reported that the total income from selling licenses is $20000.
Are then any real accusations that Apple is doing this?
From what I have heard they have released more code than they are required to.
The GPL means I OWN MY CODE, and you cannot use it for your own closed-source unless you talk to me and pay me for the right to use it.
It is as far from socalist as you can get and still see the code. The BSD license is "socalist" if you want, though really the people who release that are doing it because they want to, not because the government is telling them to.
But the GPL is applied for strictly selfish and greedy reasons and would may Ayn Rand quite happy.
The real "Socialists" and "Communists" are the people like you who whine "I can't use your code in my closed-source program unless I pay you. Wahh!!!! Unfair!!!! Gimme, gimme, gimme!" Tough luck, jerk. That's called capitalism and selfish greed on my part. Live with it, but don't you dare call me a Communist for wanting to control something I WROTE!