Large ball, only 2 buttons. The interface is a *rectangular* 9-pin, currently it has an adapter to a circular serial port on the end. I have used this on many machines since the initial SGI I bought it for. Only way I can avoid getting RSI. I am worried that they are going to stop putting the serial port on new machines, but the brand new one I have now still has it (and one for the keyboard, where I have an old Microsoft Natural keyboard with proper-sized cursor keys).
I use it on the left, and I have swapped the buttons around so the left one is the middle button (paste), the right one is the primary button, and hitting both is the third button (that pops up menus normally).
I also keep a regular usb mouse on the right so I can use the scroll wheel and also to test user interfaces using the mouse.
Unfortunaltly this worked great in my previous RedHat Linux machine, in that apparently there was a bug and xmodmap only altered the trackball and not the mouse, allowing me to swap the buttons around only on the trackball. Unfortunatley the bug appears to be "fixed" on the new Ubuntu system and xmodmap changes both the trackball and mouse. I did waste some time trying to change those "CorePointer" and "SendCoreEvents" and other keywords on xorg.conf, but could not get any different behavior except for making xmodmap not work for *either* of them.
I don't want a penguin either. How about a (gasp) word. Like "meta". Then people might be able to instruct other people how to operate the computer over the phone or email or in print without going through weird contortions trying to identify this key. If they have that other key Microsoft added, try printing "menu" on it.
Have you ever been able to try a system where clicking in the window does not raise it? I believe this is the main reason "overlapping windows are not usable" as you say. Older X window managers did not raise the windows unless you clicked the title bar (see fvwm for example), and many programs were created that used many overlapping windows (FrameMaker and Gimp for instance). These were much more usable then than they are today, due to modern Linux systems insisting on copying Windows "click raises window" behavior.
I agree that some new ideas need to be done, but they may not be as dramatic as you think. Primary "innovations" I would like to see are:
1. Get rid of click-raises-window and let the application decide how to stack it's windows. This is easily done on X now, but as long as it is optional it is useless as no program can take advantage of it. (notice that on a click THE APPLICATION CAN RAISE ITSELF. It is NOT a change in user interface, only a change in implementation! Any arguments about whether this is user friendly or not are IRRELEVANT!)
2. Get rid of titlebars and all other decoration on the windows. The application should be responsible for drawing every pixel in it's window and responding to all mouse clicks, including raising, moving, and resizing itself. This will stop the lock-in to current ideas about window manipulation and allow much more freedom in use of screen space.
Go after the users that are *UPLOADING*. Especially if they are trying to make money doing so (which also means there is information such that you can track them down).
They can flood the free services with garbage so the chances are reduced that somebody downloading will get something useful and thus reduce the value of the downloading service. However this is likely to be a losing battle that will just waste everybody's time. However I think it is a legitimate method.
And (the part that is forgotten) also supply a service that is *better* (not worse) than what users can get for free, and charge for it. Note that DRM makes it worse than downloading free stuff, so it must not be there or this is not going to work. Use watermarks, or even the threat of watermarks, and it will be far more effective as a deterrent for copying. As long as you don't have DRM, then your service could well be better: because the user knows they will quickly get a high-quality copy of exactly the thing they expect, because they will get things immediatly when they are available, that they can get non-downloadable tie-ins (tickets, trinkets, posters, branded clothing, etc), and also they will know they are getting it legally (which does have some positive value).
1. I came to the conclusion that it is essentially redundant and suffering from the "me too!" problem.
2. I also got tired of looking for equivalents and "alternatives" for my Windows apps.
You have written that Linux, simultaneously, acts too much like Windows, and does not act enough like Windows! How about that? Apparently the reason Linux is a failure is that it does not matter what it does, everything is wrong!
Hey, the truth is, the reason Linux has not succeeded is that it is not installed by default on a machine that the average person buys if they go and buy a "computer". It is totally irrelevant whether Linux is crap or the greatest invention of mankind, or what it does or does not do or whether anything is a copy of anything else. Any argument about this is totally pointless, as the explanation for Linux's failure is bloody obvious. It is insane that so many people here continue to rant for so long.
Really shameful is that the only interesting thing about this article, the fact that somebody got pissed off enough from arguing with kernel developers that they quit, is buried in a mess of stupid rants about why Linux is better or worse than Windows.
Yes, Microsoft can probably make the absolute best ODF document processor in the world.
That is absolutely irrelevant, and has nothing to do with Microsoft's fears. The best document processor will only give them some large fraction of the market. Right now they have ONE HUNDRED PERCENT of the market, due to file format lock in. See the difference?
If you cannot see why Microsoft fears ODF, you are as ignorant as they come. You might want to make sure you really own all those bridges you think you have!
ODF provides namespaces for exactly this purpose. You can add tags that readers are allowed to skip, to convey information that is outside the standard. These tags have rules so they are recognizable.
Obviously this can be abused, I would expect Word to do this extensively if they are forced to read/write ODF. I also expect Open Office to do so. Expect lots of arguments about whether some use of this is abusive.
HOWEVER:
The first difference with OOXML is that unlike these tags, the document is not required to simulate the results using the "standard" part, thus any document with these tags in it will not be readable unless the reader understands them.
The second difference is that there is no "namespaces". Instead there is a set of tags specifically designed for undocumented parts of Word. There is no rule for recognizing the optional tags, or rules for creating new ones. Other programs that want to insert non-standard information are not able to do so (unless they abuse these existing tags, which they probably will).
I think the parent was complaining about XML itself. In the "uses XML" criteria, both ODF and OOXML are equally hideous.
It would have been nice if ODF had avoided the XML bandwagon. I feel that a document format should have the requirement that a UTF-8 text file with newline only seperating the paragraphs and a very small set of ASCII characters (perhaps only 0-31) not occurring should be a legal document. XML fails this test.
Fighting XML is even harder than fighting the billion-dollar Microsoft FUD machine. It just isn't going to happen.
It should be pretty obvious. Microsoft is afraid of losing lock-in.
Yes it is quite certain that if ODF was required, Microsoft word would read/write ODF. And Microsoft word would almost certainly still be the number 1 word processor, and just like.doc format today it is likely that 95% of the ODF documents would never be read or written by anything other than Microsoft word. And Microsoft would (initially) make exactly the same amount of money as they do now. They may even make a big windfall, if the ODF read/write is only a feature of the new version that people need to buy.
The difference is that a number-2 word processor could then at least exist.
Microsoft is not worried about Open Office, that is just another bit of FUD they throw out (they act like there is some physical impossibility of any program other than an open-source Open Office working with ODF, which is a blatently false, but unstated, premise, of all their arguments).
What they are worried about is a *commercial* number-2 word processor. Say Google-word. Or maybe a company we never heard of. But suddenly no "something is wrong with open source" arguments will work (whether these are FUD or not), and any other argument against it will sound like Microsoft is claiming that they are the only company legally allowed to write software.
Such software would cut far more into Word sales than Open Office (I think the result would be 50% Word, 40% this competitor, and 10% divided amoung Open Office, a dozen other free open-source products, and 5 or 6 other commercial attempts). Retaining their market share would also require them to compete on functionality by developing the software, further cutting profits.
More serious is that it removes a possible lock-in for server products for the office. Even if a place uses 100% Word, the pointy-haired boss may actually have a hint and question why the "microsoft document server" they are thinking of buying will not work with this possible competitor, and for the same price they can buy the IBM unit that works with both. Microsoft will be forced to make such products that work with both or they will lose all the sales. But they will then lose that lock in, and then lose the lock in of things that run on or talk to these servers, etc, etc.
No. Michael Sweet would still own the copyright, even if he had released the code GPL3. Apple would get *exactly* the same thing as they do now (including the ability to release it under any license they want, even the GPL3), therefore "fear of GPL3" is not a reason for this.
Since Michael Sweet owned the copyright, Apple gets exactly the same rights to the code. It is totally irrelevant whether the code had been previously released under the GPL, the GPL3, or not at all, or BSD, or as closed-source, or anything.
Therefore, whether you like or hate the GPL3, it has NOTHING to do with this decision.
This is due to that goddamm "click to top" behavior that was copied from Windows. This basically means that when you click in a window, it raises before telling the program.
If you read old release notes for X11 you will see they dropped this behavior because it was obviously bad.
Unfortunatly it reappeared in about 1995 when window managers started copying Windows in order to be "more user friendly" which is code for "act as much like Windows as possible".
In case you cannot understand why this behavior is bad, here is a hint: A program can raise ITSELF in response to a click and thus achieve exactly the same user interface without this stupid behavior. There is NO reason for the system to do this at all. It is not a question of whether it is good behavior from the user's point of view. It is simply that it is TRIVIAL for this decision to be left to the programs.
Click-to-top has completely ruined multiple-window apis (including Gimp) and caused endless frustration by requiring tiled or "mdi" or other abonimations or require displays to be crammed into increasing small areas to avoid overlap.
Hey, and "child windows" or "transient for" or layers or "keep on top" or all that other crap is NOT a solution. They are all bandaid attempts to patch it, by making a complex system of rules for disabling the "click to top", when in fact they could disable it ALL the time and everything would be trivial and easy, and Linux would have programs with GUI's that were far better than anything on Windows due to the ability to make overlapping windows useful.
You realize that what you are saying is that you think there should be no copyright on any material that is given away for free. Are you really sure you think this is a good idea? At least the advertising industry would be very hurt by this, and I'm sure it would have enormous other implications.
I don't know where you are getting your information but it is absolutely bogus.
The goal of the GPL is so that whoever uses the GPLed code is forced to release their modifications to that code. It actually is pretty weak on that, too, since anybody who just modifies the code for their own use and does not distribute it does not have to do anything (several other licenses that are not considered open source try to do this, and are disliked by GPL proponents).
At no point does anything in the GPL say anything about the end users own code.
Re:If your looking for logic in coding conventions
on
Are 80 Columns Enough?
·
· Score: 1
The MDA also had the attribute byte. Some of the bits controlled intensity. Other bits made it reverve video, blink, and underscore.
In any case the same logic would mean 80x25x2 is just under 4K which would also be a useful cutoff when such chips were available.
I suspect a much more important reason for the 24/25 line convention was that there are only 480 or so useful scan lines on NTSC video and crt screens were based on similar technology.
If you reprogram your Tivo, you *will* violate the warranty. That is not what the GPL3 is trying to do. Trying to confuse this with warranties is just more FUD.
Gnome / GTK certainly already has a "canvas" object (ie a drawing surface that collects drawing objects and can redraw itself in response to exposure events). Not sure what you are talking about here. A canvas has nothing to do with embedding widgets into other programs. Possibly this is another Microsoft basterdization of a common computer science term to mean something completely different?
"Commercial Customers who need a different license, please contact Matthias L. Jugel for more information and pricing. Include details why you need a different license and what you need it for in your email."
Slashdot Mirror
Large ball, only 2 buttons. The interface is a *rectangular* 9-pin, currently it has an adapter to a circular serial port on the end. I have used this on many machines since the initial SGI I bought it for. Only way I can avoid getting RSI. I am worried that they are going to stop putting the serial port on new machines, but the brand new one I have now still has it (and one for the keyboard, where I have an old Microsoft Natural keyboard with proper-sized cursor keys).
I use it on the left, and I have swapped the buttons around so the left one is the middle button (paste), the right one is the primary button, and hitting both is the third button (that pops up menus normally).
I also keep a regular usb mouse on the right so I can use the scroll wheel and also to test user interfaces using the mouse.
Unfortunaltly this worked great in my previous RedHat Linux machine, in that apparently there was a bug and xmodmap only altered the trackball and not the mouse, allowing me to swap the buttons around only on the trackball. Unfortunatley the bug appears to be "fixed" on the new Ubuntu system and xmodmap changes both the trackball and mouse. I did waste some time trying to change those "CorePointer" and "SendCoreEvents" and other keywords on xorg.conf, but could not get any different behavior except for making xmodmap not work for *either* of them.
I don't want a penguin either. How about a (gasp) word. Like "meta". Then people might be able to instruct other people how to operate the computer over the phone or email or in print without going through weird contortions trying to identify this key. If they have that other key Microsoft added, try printing "menu" on it.
Have you ever been able to try a system where clicking in the window does not raise it? I believe this is the main reason "overlapping windows are not usable" as you say. Older X window managers did not raise the windows unless you clicked the title bar (see fvwm for example), and many programs were created that used many overlapping windows (FrameMaker and Gimp for instance). These were much more usable then than they are today, due to modern Linux systems insisting on copying Windows "click raises window" behavior.
I agree that some new ideas need to be done, but they may not be as dramatic as you think. Primary "innovations" I would like to see are:
1. Get rid of click-raises-window and let the application decide how to stack it's windows. This is easily done on X now, but as long as it is optional it is useless as no program can take advantage of it. (notice that on a click THE APPLICATION CAN RAISE ITSELF. It is NOT a change in user interface, only a change in implementation! Any arguments about whether this is user friendly or not are IRRELEVANT!)
2. Get rid of titlebars and all other decoration on the windows. The application should be responsible for drawing every pixel in it's window and responding to all mouse clicks, including raising, moving, and resizing itself. This will stop the lock-in to current ideas about window manipulation and allow much more freedom in use of screen space.
I believe he was implying that Microsoft would not have made C# except to compete with Sun's Java. By that reasoning C#'s existence is due to Sun.
I didn't know Intel was out of business. Thanks for pointing that out.
With the "runtime exception", the GPL version *is* the commercial product (assuming by "commercial" you mean "can be used in closed source").
Go after the users that are *UPLOADING*. Especially if they are trying to make money doing so (which also means there is information such that you can track them down).
They can flood the free services with garbage so the chances are reduced that somebody downloading will get something useful and thus reduce the value of the downloading service. However this is likely to be a losing battle that will just waste everybody's time. However I think it is a legitimate method.
And (the part that is forgotten) also supply a service that is *better* (not worse) than what users can get for free, and charge for it. Note that DRM makes it worse than downloading free stuff, so it must not be there or this is not going to work. Use watermarks, or even the threat of watermarks, and it will be far more effective as a deterrent for copying. As long as you don't have DRM, then your service could well be better: because the user knows they will quickly get a high-quality copy of exactly the thing they expect, because they will get things immediatly when they are available, that they can get non-downloadable tie-ins (tickets, trinkets, posters, branded clothing, etc), and also they will know they are getting it legally (which does have some positive value).
Hmm, let's try reading what you wrote carefully:
1. I came to the conclusion that it is essentially redundant and suffering from the "me too!" problem.
2. I also got tired of looking for equivalents and "alternatives" for my Windows apps.
You have written that Linux, simultaneously, acts too much like Windows, and does not act enough like Windows! How about that? Apparently the reason Linux is a failure is that it does not matter what it does, everything is wrong!
Hey, the truth is, the reason Linux has not succeeded is that it is not installed by default on a machine that the average person buys if they go and buy a "computer". It is totally irrelevant whether Linux is crap or the greatest invention of mankind, or what it does or does not do or whether anything is a copy of anything else. Any argument about this is totally pointless, as the explanation for Linux's failure is bloody obvious. It is insane that so many people here continue to rant for so long.
Really shameful is that the only interesting thing about this article, the fact that somebody got pissed off enough from arguing with kernel developers that they quit, is buried in a mess of stupid rants about why Linux is better or worse than Windows.
You and the parent are talking about the trivial-to-detect case.
// as recommended above, big deal; // unfortunately b is still a dangling pointer!
Setting the pointer to null does not really help in the real bugs, which are more like this:
Foo* a = new Foo;
Foo* b = a;
delete a;
a = 0;
Nice strawman there.
Yes, Microsoft can probably make the absolute best ODF document processor in the world.
That is absolutely irrelevant, and has nothing to do with Microsoft's fears. The best document processor will only give them some large fraction of the market. Right now they have ONE HUNDRED PERCENT of the market, due to file format lock in. See the difference?
If you cannot see why Microsoft fears ODF, you are as ignorant as they come. You might want to make sure you really own all those bridges you think you have!
ODF provides namespaces for exactly this purpose. You can add tags that readers are allowed to skip, to convey information that is outside the standard. These tags have rules so they are recognizable.
Obviously this can be abused, I would expect Word to do this extensively if they are forced to read/write ODF. I also expect Open Office to do so. Expect lots of arguments about whether some use of this is abusive.
HOWEVER:
The first difference with OOXML is that unlike these tags, the document is not required to simulate the results using the "standard" part, thus any document with these tags in it will not be readable unless the reader understands them.
The second difference is that there is no "namespaces". Instead there is a set of tags specifically designed for undocumented parts of Word. There is no rule for recognizing the optional tags, or rules for creating new ones. Other programs that want to insert non-standard information are not able to do so (unless they abuse these existing tags, which they probably will).
I think the parent was complaining about XML itself. In the "uses XML" criteria, both ODF and OOXML are equally hideous.
It would have been nice if ODF had avoided the XML bandwagon. I feel that a document format should have the requirement that a UTF-8 text file with newline only seperating the paragraphs and a very small set of ASCII characters (perhaps only 0-31) not occurring should be a legal document. XML fails this test.
Fighting XML is even harder than fighting the billion-dollar Microsoft FUD machine. It just isn't going to happen.
It should be pretty obvious. Microsoft is afraid of losing lock-in.
.doc format today it is likely that 95% of the ODF documents would never be read or written by anything other than Microsoft word. And Microsoft would (initially) make exactly the same amount of money as they do now. They may even make a big windfall, if the ODF read/write is only a feature of the new version that people need to buy.
Yes it is quite certain that if ODF was required, Microsoft word would read/write ODF. And Microsoft word would almost certainly still be the number 1 word processor, and just like
The difference is that a number-2 word processor could then at least exist.
Microsoft is not worried about Open Office, that is just another bit of FUD they throw out (they act like there is some physical impossibility of any program other than an open-source Open Office working with ODF, which is a blatently false, but unstated, premise, of all their arguments).
What they are worried about is a *commercial* number-2 word processor. Say Google-word. Or maybe a company we never heard of. But suddenly no "something is wrong with open source" arguments will work (whether these are FUD or not), and any other argument against it will sound like Microsoft is claiming that they are the only company legally allowed to write software.
Such software would cut far more into Word sales than Open Office (I think the result would be 50% Word, 40% this competitor, and 10% divided amoung Open Office, a dozen other free open-source products, and 5 or 6 other commercial attempts). Retaining their market share would also require them to compete on functionality by developing the software, further cutting profits.
More serious is that it removes a possible lock-in for server products for the office. Even if a place uses 100% Word, the pointy-haired boss may actually have a hint and question why the "microsoft document server" they are thinking of buying will not work with this possible competitor, and for the same price they can buy the IBM unit that works with both. Microsoft will be forced to make such products that work with both or they will lose all the sales. But they will then lose that lock in, and then lose the lock in of things that run on or talk to these servers, etc, etc.
No. Michael Sweet would still own the copyright, even if he had released the code GPL3. Apple would get *exactly* the same thing as they do now (including the ability to release it under any license they want, even the GPL3), therefore "fear of GPL3" is not a reason for this.
The Titanic ship was a 1/20th scale miniature, not CG.
However there was heavy amounts of CG added. The water, sky, smoke, the wire rigging, and of course all the people, were added CG.
Since Michael Sweet owned the copyright, Apple gets exactly the same rights to the code. It is totally irrelevant whether the code had been previously released under the GPL, the GPL3, or not at all, or BSD, or as closed-source, or anything.
Therefore, whether you like or hate the GPL3, it has NOTHING to do with this decision.
This is due to that goddamm "click to top" behavior that was copied from Windows. This basically means that when you click in a window, it raises before telling the program.
If you read old release notes for X11 you will see they dropped this behavior because it was obviously bad.
Unfortunatly it reappeared in about 1995 when window managers started copying Windows in order to be "more user friendly" which is code for "act as much like Windows as possible".
In case you cannot understand why this behavior is bad, here is a hint: A program can raise ITSELF in response to a click and thus achieve exactly the same user interface without this stupid behavior. There is NO reason for the system to do this at all. It is not a question of whether it is good behavior from the user's point of view. It is simply that it is TRIVIAL for this decision to be left to the programs.
Click-to-top has completely ruined multiple-window apis (including Gimp) and caused endless frustration by requiring tiled or "mdi" or other abonimations or require displays to be crammed into increasing small areas to avoid overlap.
Hey, and "child windows" or "transient for" or layers or "keep on top" or all that other crap is NOT a solution. They are all bandaid attempts to patch it, by making a complex system of rules for disabling the "click to top", when in fact they could disable it ALL the time and everything would be trivial and easy, and Linux would have programs with GUI's that were far better than anything on Windows due to the ability to make overlapping windows useful.
By "end users own code" I mean "code written by that end user". Not sure what you are defining it as.
Anyway according to my definition, the GPL says absolutely nothing about the end users own code.
You realize that what you are saying is that you think there should be no copyright on any material that is given away for free. Are you really sure you think this is a good idea? At least the advertising industry would be very hurt by this, and I'm sure it would have enormous other implications.
I don't know where you are getting your information but it is absolutely bogus.
The goal of the GPL is so that whoever uses the GPLed code is forced to release their modifications to that code. It actually is pretty weak on that, too, since anybody who just modifies the code for their own use and does not distribute it does not have to do anything (several other licenses that are not considered open source try to do this, and are disliked by GPL proponents).
At no point does anything in the GPL say anything about the end users own code.
The MDA also had the attribute byte. Some of the bits controlled intensity. Other bits made it reverve video, blink, and underscore.
In any case the same logic would mean 80x25x2 is just under 4K which would also be a useful cutoff when such chips were available.
I suspect a much more important reason for the 24/25 line convention was that there are only 480 or so useful scan lines on NTSC video and crt screens were based on similar technology.
If you reprogram your Tivo, you *will* violate the warranty. That is not what the GPL3 is trying to do. Trying to confuse this with warranties is just more FUD.
Gnome / GTK certainly already has a "canvas" object (ie a drawing surface that collects drawing objects and can redraw itself in response to exposure events). Not sure what you are talking about here. A canvas has nothing to do with embedding widgets into other programs. Possibly this is another Microsoft basterdization of a common computer science term to mean something completely different?
"Commercial Customers who need a different license, please contact Matthias L. Jugel for more information and pricing. Include details why you need a different license and what you need it for in your email."
Please name the library to prove you are not a troll.