Since CowboyNeal insists I have the opportunity to say this, I will.
If you are using an MIT/BSD type license, the license itself allows you to do a lot of things, but if you don't want to kill the goose that laid the golden egg, you have to give back. So you have a moral obligation to open your own work.
If you fail to open up your own work, you can't get help from others, so your work just eventually ends up shriveling up and dying, unless you prop it up with patents and threats and bribes and other immoralities. And even if you prop it up, it does not get the benefit of open peer review, so it ends up inferior.
Also, lurking on the openBSD lists definitely leaves one with the understanding that the projects which use MIT/BSD type license must be careful to keep the licenses in their source trees compatible, just like the GPL. Once you do recognize the need to open it up, BSD/MIT has the same inductive effect as the GPL.
So, while you have a little more time lapse before what comes around goes around, the natural consequences are that anyone using source under BSD/MIT type licenses without opening their own sources basically doesn't get the full benefits of open source.
And while I'm trolling, I'll just note that, the brain being able to emulate a CPU after a manner of speaking, patenting software results in a conundrum: how does one read the patent for understanding without first getting a license? You can't understand software unless you at least try to walk through it in your head.
Or, perhaps, "Why not support one-click maximizing?" And it's a good question. With the OS libraries supporting most of the basic window functionality, the behavior of the maximize/resize button seems like it ought to be customizable. And, for those who like maximized windows, it would be a useful thing, as opposed to having to resize the windows yourself the first time, and then remember to use the resize button instead of trying to resize by hand.
Resizing on Macs may feel awkward if you're used to grabbing any edge to resize. Mac Classic in v. 8/9 allowed you to drag the window by any edge, so I felt sabotaged on MSWindows at first, trying to drag with an edge, and messing up the window size, instead. But Mac OS X is back to dragging by the menu bar only.
But, yeah, I did mean the physical desktop. I guess I'm always going to feel like the interface is constraining until I get a full holodeck. (I don't use a tool belt on the real workbench, either. It gets in the way when I have whatever I'm working on propped up overhead.)
I must admit, I felt the same constraints when working with punched cards, too. But I don't remember that feeling so much with either the Unix command line or the Mac interface. DOS command line and MSWindows before 2000 brought the feeling back a bit.
Hunting covered windows is definitely a frustration. I keep forgetting about expose', so I don't know how well I'll like that. It looks useful, when I remember it (or accidentally hit F9 through F12).
Before expose', I'd just use a non-linear tiling to make the windows I need the most from the most places poke out where they'd be least likely to end up hidden. Somehow, having deliberately put the window there made it easier to remember where it was.
When that didn't work, the last option was to use the dock (or the application menu that classic put in the top-right-hand corner) to get the app, then use the app's Window menu. (That kind of thing is one of the reasons why the menu bar is always at the top of the screent on Macs.) It's a little clumsier than the task bar, but it's easier to tell what you're bringing up before you click.
The problem with MSWindows's task bar is that by the time you need the task bar, you can't tell which is which any more, so you have to remember which it was the last time, or go hunting. Perhaps it's something of a matter of which direction you're used to when you go hunting.
Thinking about using maximized windows, what would you think of being able to put those window tabs all around the edges of your screen, maybe putting the current app's tabs at the top and tabbing the other apps around the other edges, and if it gets too crowded, stack, but never stack the current app's tabs at top? Kind of like the task bar, but separating the curent top-most apps.
Why use an external debugger? Well, for example, when you're using a custom dev tool for some niche tech, you often have to use the debugger that comes with the tool. You may be compiling C with VS and linking the compiled object into some other tool.
Well, for one thing, I don't focus on the window at the top for very long except when I'm surfing the net, and not even then, really. Like I say, the pile of open windows helps me keep track of what I've done, what I'm doing now, and what I need (or want) to do next. When I'm coding, I may even have two or three windows on the same file open, so I can track what I'm working on and the thing it's going to effect with my eyeballs and let my imagination keep track of the larger issues in the background.
Another is that a maximized window, even at 1280x1024, is still significantly less real estate than my physical desktop, so giving up a small amount of screen real estate to be able to grab different windows on my work is, for me, a good trade-off.
It's just different work styles.
How long have those tabbed windows been available? I don't remember them in VS 6.
Are they basically the same thing as tabbed windows in browsers? What happens when you have, oh, say 15 different source files open? If you get multiple rows, or if you have multiple groups, do the rows/groups jump when you bring something from back to front?
Can you mix graphics edit windows and shell windows and external debugger windows and db GUI windows with the VS windows?
Codewarrior on MSWxx, somewhere around v. 5 or 6, developed the ability to break out of the old MDI interface. But trying to use that gives me a feel of how hard it must be to get it to work right in MSWxx. And it doesn't really work the same, although it is an improvement.
BTW, you actually can get the min/max effect in Mac OS X, it just takes a little extra work to set it up. Drag the top-left corner into the top-left corner, drag the resize tab to the bottom-right, and the next time you hit the resize button in does more or less what you want from then until you forget and resized by hand.
MSWindows Multi-document interface window (or whatever that is) interface, like VisualStudio does by default, is actually considered against the HI guidelines in the Mac world. I suppose that's because Mac types assume it's not the role of the computer to be "helping" the user focus on the (single) task at the top of the pile on the screen.
So there is not a single button that you can click to make the current window fill the screen and then click again to shrink the window. There is a button that you click to switch between the current and previous sizes. I rarely use it, but then I rarely (intentionally) maximize windows in MSWindows.
(Once or twice a day I miss the close or minimize button and feel frustrated while I chase the maximize button down where it flew off to so I don't have the app thinking it should open maximized next time I run it.)
I think the negative optimizations of the maximize concept is what induced Apple to invent Expose' and Dashboard.
The minimize button on Mac OS X does work in pretty much the same way as in MSWindows, except that the Dock doesn't have any non-minimized windows in it, and the minimizations don't stack vertically. I find both approaches equally cumbersome, myself, far prefer the windowshades and tabbable folders of Mac OS 8/9.
Perhaps the UI should allow users to choose the behavior of maximize and minimize, and whether running ups and such show up in the dock or maybe could be tabbed on an edge of the screen or something.
Of course, some people don't know about clicking that little triangle to the left of folders.
But the GP want's both the tree view and the list view in one (compound) window, and, yeah that doesn't come stock on a Mac. Some people like it that way, some people don't.
I've gotten used to both, but I still prefer the messy desk with a stack of windows that reminds me where I've been by way of what stacks on what. Mac OS X is almost able to do that again, but not quite yet, I guess.
I hate maximized apps. To me it's like trying to squeeze my mind throw this little tiny window in the ceiling. With lots of windows spread around and overlapping, it's like having my pick from lots of tiny windows in the ceiling.
Yeah, I know, that means that when my boxes boot up, they have open windows lying all over the place. Messy desk. I like it. One thing I miss about the Classic Finder, it kept the pile in the same order I dumped things in it until I moved things around. Made a nice stack for when I had to backtrack trying to solve problems.
I wonder if that grouping by app behavior can be shut off. And it would sure be nice if Finder (and MSWExplorer) would remember the order of the pile when it boots.
I guess maybe this shows one of the bases of disagreement that may be in operation here? Different ways of looking at problems.
Oh, I don't know about you, but when I'm doing serious dev, I have to force-quit MSWExplorer on occasion, too.
Apple was rather careful from the outset to inform everyone that it should be called "Mac OS 9" and not "OS 9".
Microware's point of view was that they didn't need a win, just needed a legal basis to show that they hadn't abandond their trademark.
Essentially, the ruling was exactly that, and, IIRC, left the door open to revisit the issue if Apple were to drop the "Mac" part or attempt to use Mac OS 9 to enter the real-time domain.
I think you're putting too much faith on academia.
Every job one does, one has a responsibility to do one's best. Schooling effects so many people that responsibility is a heavier weight than on, say, someone taking tickets at the local theater.
But the weight of responsibility doesn't somehow magically overcome the fundamental nature of humans to be less than perfect, and even less than ideal.
Now that I have you distracted with a red herring, let me ask you -- How do you know that Sakamura is not right?
For all that you have a fair amount of experience in things Asian, it appears to me that you are operating under the assumption that globalism is not evil. How do you know that is the case? And if it is not the case, can you really argue that Sakamura's position is incorrect, or that the non-optimal current situation is worse than the alternative?
For instance, from my point of view, since I can guess that my children are going to be wanting to record their genealogy on computers, being able to use the ancient kanji is much more important than being able to use the texting smileys.
A quick scan didn't turn up any serious errors, but I didn't have time to dig and compare.
You do have some typos, and one paragraph shows signs of having been eaten by something. And I did notice that you let a desire to debunk get in the way in a couple of places.
In response about Sakamura and the tendency for Japan's society to be too inward facing, it's representative of the entire core issue of internationalization. They may be wrong in many points of fact, but they are right to insist on their right to be wrong.
If internationalization tries to force people and countries to be right, just how far has our so-called modern society advanced from the Crusades?
I'll agree with you that Shift-JIS is a sinister American conspiracy.
I'll agree that much of what has been said in Asia about Unicode is mis-informed.
I'll agree in principle that, if Unicode is not the solution, Mojikyou is definitely not the solution.
But if you look at the history of Unicode, you can understand what caused the doubts and why the doubts, resentment, and resistance linger.
16 bits? Some people still think Unicode is 16 bits. Yes, the Japanese equivalent of an unabridged overflows 16 bits, and trying to fold characters together breaks semantics.
Ten years ago, most of the Unicode Consortium still did not get that. Even now, there are a few holdouts who think that a little clever folding characters together could have solved things.
Folding two Chinese character sets together with Japanese Han and Korean Han? If Cyrillic A and Greek A and English A are all distinct code points in their own language sequences, why did those westerners think Chinese and Japanese should be folded together?
And why did those didn't those westerners understand that fixing the set of Han ideographs was a bit like fixing the set of English root words?
In a very real sense, Unicode is just the old JIS encoding re-arranged and extended to force it to cover things the Japanese printing industry figured were best left until better ways of encoding could be invented.
Sakamura and others may have non-western points of view, and I don't agree with them on everything, but they are not a nut cases. Their points need to be understood by westerners who work with Asian language information encoding.
SPAMbots use their own servers. They can make their own envelopes.
Until we can afford the processor and network time to put a certificate in every item in the envelope, and the processor/network load to look every certificate up, this only pushes the problems back a few weeks.
If you maintain your own black/white/grey lists, then you control what goes in which list. You can fix the lists if they break, as soon as you know about the breakage.
If you share your lists with friends, then you are losing some of the control. But you gain a wider range of coverage, as well. It may or may not be good, but the key point is that you have reason to trust (or not to trust) any particular friend's blacklist, and you are still making the decision.
Particularly, if the sharing mechanism uses some sort of rating system and allows you to tune the shared lists for yourself, you maintain control.
But when you share the lists of people you don't know, you no longer have a valid basis of trust. The only basis of trust that remains is the basis of advertisement. You are essentially giving your freedom into the hands of the group with the best ads.
Centrally administered blacklists are by nature in the latter category.
Therefore, centrally administered blacklists are not a good idea.
Steny H. Hoyer, House of Reps, Maryland, is the guy that proposed this.
I tried to send him some e-mail, but his contact page won't accept you if you admit your zip code is out of his district. Just as well, I can think about what I want to say while I go dig up a valid zip code for his district. Want to be polite when I tell him where he can take this kind of thinking.
BTW, I got IP restricted several months back. I think it was from too much anonymous posting on a single thread. But why four anonymous posts was too many? Maybe it was because I was defending religion in those posts?
Anyway, after a couple of days I could post if I logged in first. Then after about four weeks the ban was dropped.
(Cool down? I needed to cool down? What about the guys who had to swear four times in every sentence to prove that God doesn't exist? Oh well, everywhere has their own kind of censorship, I guess.)
Slashdot Mirror
elsewhere in this thread.
It gives enough information that anyone with the compiler should be able to check it.
... be considered in the first ten of anything.
Which is why I laugh about "innovation".
Since CowboyNeal insists I have the opportunity to say this, I will.
If you are using an MIT/BSD type license, the license itself allows you to do a lot of things, but if you don't want to kill the goose that laid the golden egg, you have to give back. So you have a moral obligation to open your own work.
If you fail to open up your own work, you can't get help from others, so your work just eventually ends up shriveling up and dying, unless you prop it up with patents and threats and bribes and other immoralities. And even if you prop it up, it does not get the benefit of open peer review, so it ends up inferior.
Also, lurking on the openBSD lists definitely leaves one with the understanding that the projects which use MIT/BSD type license must be careful to keep the licenses in their source trees compatible, just like the GPL. Once you do recognize the need to open it up, BSD/MIT has the same inductive effect as the GPL.
So, while you have a little more time lapse before what comes around goes around, the natural consequences are that anyone using source under BSD/MIT type licenses without opening their own sources basically doesn't get the full benefits of open source.
And while I'm trolling, I'll just note that, the brain being able to emulate a CPU after a manner of speaking, patenting software results in a conundrum: how does one read the patent for understanding without first getting a license? You can't understand software unless you at least try to walk through it in your head.
Or, perhaps, "Why not support one-click maximizing?" And it's a good question. With the OS libraries supporting most of the basic window functionality, the behavior of the maximize/resize button seems like it ought to be customizable. And, for those who like maximized windows, it would be a useful thing, as opposed to having to resize the windows yourself the first time, and then remember to use the resize button instead of trying to resize by hand.
Resizing on Macs may feel awkward if you're used to grabbing any edge to resize. Mac Classic in v. 8/9 allowed you to drag the window by any edge, so I felt sabotaged on MSWindows at first, trying to drag with an edge, and messing up the window size, instead. But Mac OS X is back to dragging by the menu bar only.
But, yeah, I did mean the physical desktop. I guess I'm always going to feel like the interface is constraining until I get a full holodeck. (I don't use a tool belt on the real workbench, either. It gets in the way when I have whatever I'm working on propped up overhead.)
I must admit, I felt the same constraints when working with punched cards, too. But I don't remember that feeling so much with either the Unix command line or the Mac interface. DOS command line and MSWindows before 2000 brought the feeling back a bit.
Hunting covered windows is definitely a frustration. I keep forgetting about expose', so I don't know how well I'll like that. It looks useful, when I remember it (or accidentally hit F9 through F12).
Before expose', I'd just use a non-linear tiling to make the windows I need the most from the most places poke out where they'd be least likely to end up hidden. Somehow, having deliberately put the window there made it easier to remember where it was.
When that didn't work, the last option was to use the dock (or the application menu that classic put in the top-right-hand corner) to get the app, then use the app's Window menu. (That kind of thing is one of the reasons why the menu bar is always at the top of the screent on Macs.) It's a little clumsier than the task bar, but it's easier to tell what you're bringing up before you click.
The problem with MSWindows's task bar is that by the time you need the task bar, you can't tell which is which any more, so you have to remember which it was the last time, or go hunting. Perhaps it's something of a matter of which direction you're used to when you go hunting.
Thinking about using maximized windows, what would you think of being able to put those window tabs all around the edges of your screen, maybe putting the current app's tabs at the top and tabbing the other apps around the other edges, and if it gets too crowded, stack, but never stack the current app's tabs at top? Kind of like the task bar, but separating the curent top-most apps.
Why use an external debugger? Well, for example, when you're using a custom dev tool for some niche tech, you often have to use the debugger that comes with the tool. You may be compiling C with VS and linking the compiled object into some other tool.
Why not maximize?
Well, for one thing, I don't focus on the window at the top for very long except when I'm surfing the net, and not even then, really. Like I say, the pile of open windows helps me keep track of what I've done, what I'm doing now, and what I need (or want) to do next. When I'm coding, I may even have two or three windows on the same file open, so I can track what I'm working on and the thing it's going to effect with my eyeballs and let my imagination keep track of the larger issues in the background.
Another is that a maximized window, even at 1280x1024, is still significantly less real estate than my physical desktop, so giving up a small amount of screen real estate to be able to grab different windows on my work is, for me, a good trade-off.
It's just different work styles.
How long have those tabbed windows been available? I don't remember them in VS 6.
Are they basically the same thing as tabbed windows in browsers? What happens when you have, oh, say 15 different source files open? If you get multiple rows, or if you have multiple groups, do the rows/groups jump when you bring something from back to front?
Can you mix graphics edit windows and shell windows and external debugger windows and db GUI windows with the VS windows?
Codewarrior on MSWxx, somewhere around v. 5 or 6, developed the ability to break out of the old MDI interface. But trying to use that gives me a feel of how hard it must be to get it to work right in MSWxx. And it doesn't really work the same, although it is an improvement.
BTW, you actually can get the min/max effect in Mac OS X, it just takes a little extra work to set it up. Drag the top-left corner into the top-left corner, drag the resize tab to the bottom-right, and the next time you hit the resize button in does more or less what you want from then until you forget and resized by hand.
Depends on what you mean by maximize.
MSWindows Multi-document interface window (or whatever that is) interface, like VisualStudio does by default, is actually considered against the HI guidelines in the Mac world. I suppose that's because Mac types assume it's not the role of the computer to be "helping" the user focus on the (single) task at the top of the pile on the screen.
So there is not a single button that you can click to make the current window fill the screen and then click again to shrink the window. There is a button that you click to switch between the current and previous sizes. I rarely use it, but then I rarely (intentionally) maximize windows in MSWindows.
(Once or twice a day I miss the close or minimize button and feel frustrated while I chase the maximize button down where it flew off to so I don't have the app thinking it should open maximized next time I run it.)
I think the negative optimizations of the maximize concept is what induced Apple to invent Expose' and Dashboard.
The minimize button on Mac OS X does work in pretty much the same way as in MSWindows, except that the Dock doesn't have any non-minimized windows in it, and the minimizations don't stack vertically. I find both approaches equally cumbersome, myself, far prefer the windowshades and tabbable folders of Mac OS 8/9.
Perhaps the UI should allow users to choose the behavior of maximize and minimize, and whether running ups and such show up in the dock or maybe could be tabbed on an edge of the screen or something.
Ouch.
Of course, some people don't know about clicking that little triangle to the left of folders.
But the GP want's both the tree view and the list view in one (compound) window, and, yeah that doesn't come stock on a Mac. Some people like it that way, some people don't.
I've gotten used to both, but I still prefer the messy desk with a stack of windows that reminds me where I've been by way of what stacks on what. Mac OS X is almost able to do that again, but not quite yet, I guess.
I hate maximized apps. To me it's like trying to squeeze my mind throw this little tiny window in the ceiling. With lots of windows spread around and overlapping, it's like having my pick from lots of tiny windows in the ceiling.
Yeah, I know, that means that when my boxes boot up, they have open windows lying all over the place. Messy desk. I like it. One thing I miss about the Classic Finder, it kept the pile in the same order I dumped things in it until I moved things around. Made a nice stack for when I had to backtrack trying to solve problems.
I wonder if that grouping by app behavior can be shut off. And it would sure be nice if Finder (and MSWExplorer) would remember the order of the pile when it boots.
I guess maybe this shows one of the bases of disagreement that may be in operation here? Different ways of looking at problems.
Oh, I don't know about you, but when I'm doing serious dev, I have to force-quit MSWExplorer on occasion, too.
Apple was rather careful from the outset to inform everyone that it should be called "Mac OS 9" and not "OS 9".
Microware's point of view was that they didn't need a win, just needed a legal basis to show that they hadn't abandond their trademark.
Essentially, the ruling was exactly that, and, IIRC, left the door open to revisit the issue if Apple were to drop the "Mac" part or attempt to use Mac OS 9 to enter the real-time domain.
I've tried from the links here, I searched from Google. Everything redirects to Google at this point in time.
You know, I can think of better was to fend off a slashdotting.
Yep.
http://www.healthmarketscience.com/company/jobs/de veloper.html
That link redirects to Google now, too.
I wonder why?
A letter to his superiors pointing out the legality issues is capricious?
Hmm. Right now, the URL given for the employer, http://www.healthmarketscience.com/ gets redirected to Google.
So I looked it up on Google and got this URL:http://www.hmsonline.com/.
And it is redirecting to Google as well.Whatever could it mean?
Thankyverymuch. And I have many more where that one came from, unfortunately.
:)
:-/
I think you're putting too much faith on academia.
Every job one does, one has a responsibility to do one's best. Schooling effects so many people that responsibility is a heavier weight than on, say, someone taking tickets at the local theater.
But the weight of responsibility doesn't somehow magically overcome the fundamental nature of humans to be less than perfect, and even less than ideal.
Now that I have you distracted with a red herring, let me ask you -- How do you know that Sakamura is not right?
For all that you have a fair amount of experience in things Asian, it appears to me that you are operating under the assumption that globalism is not evil. How do you know that is the case? And if it is not the case, can you really argue that Sakamura's position is incorrect, or that the non-optimal current situation is worse than the alternative?
For instance, from my point of view, since I can guess that my children are going to be wanting to record their genealogy on computers, being able to use the ancient kanji is much more important than being able to use the texting smileys.
But we need an encoding that can handle both.
With three cpus, couldn't the programmer just decide to dedicate one to physics? Or would that require an API that MS doesn't provide?
That could also be done with one hand, although it would take five fingers.
Of course, morse code could potentially be I/O, but chording could not, unless things of changed there.
A quick scan didn't turn up any serious errors, but I didn't have time to dig and compare.
You do have some typos, and one paragraph shows signs of having been eaten by something. And I did notice that you let a desire to debunk get in the way in a couple of places.
In response about Sakamura and the tendency for Japan's society to be too inward facing, it's representative of the entire core issue of internationalization. They may be wrong in many points of fact, but they are right to insist on their right to be wrong.
If internationalization tries to force people and countries to be right, just how far has our so-called modern society advanced from the Crusades?
I'll agree with you that Shift-JIS is a sinister American conspiracy.
I'll agree that much of what has been said in Asia about Unicode is mis-informed.
I'll agree in principle that, if Unicode is not the solution, Mojikyou is definitely not the solution.
But if you look at the history of Unicode, you can understand what caused the doubts and why the doubts, resentment, and resistance linger.
16 bits? Some people still think Unicode is 16 bits. Yes, the Japanese equivalent of an unabridged overflows 16 bits, and trying to fold characters together breaks semantics.
Ten years ago, most of the Unicode Consortium still did not get that. Even now, there are a few holdouts who think that a little clever folding characters together could have solved things.
Folding two Chinese character sets together with Japanese Han and Korean Han? If Cyrillic A and Greek A and English A are all distinct code points in their own language sequences, why did those westerners think Chinese and Japanese should be folded together?
And why did those didn't those westerners understand that fixing the set of Han ideographs was a bit like fixing the set of English root words?
In a very real sense, Unicode is just the old JIS encoding re-arranged and extended to force it to cover things the Japanese printing industry figured were best left until better ways of encoding could be invented.
Sakamura and others may have non-western points of view, and I don't agree with them on everything, but they are not a nut cases. Their points need to be understood by westerners who work with Asian language information encoding.
SPAMbots use their own servers. They can make their own envelopes.
Until we can afford the processor and network time to put a certificate in every item in the envelope, and the processor/network load to look every certificate up, this only pushes the problems back a few weeks.
What Smallpond says bears repeating --
when reporting spam to anyone but your own ISP (and even them if you are one of certain large ISPs), use a throw-away account.
If you maintain your own black/white/grey lists, then you control what goes in which list. You can fix the lists if they break, as soon as you know about the breakage.
If you share your lists with friends, then you are losing some of the control. But you gain a wider range of coverage, as well. It may or may not be good, but the key point is that you have reason to trust (or not to trust) any particular friend's blacklist, and you are still making the decision.
Particularly, if the sharing mechanism uses some sort of rating system and allows you to tune the shared lists for yourself, you maintain control.
But when you share the lists of people you don't know, you no longer have a valid basis of trust. The only basis of trust that remains is the basis of advertisement. You are essentially giving your freedom into the hands of the group with the best ads.
Centrally administered blacklists are by nature in the latter category.
Therefore, centrally administered blacklists are not a good idea.
Steny H. Hoyer, House of Reps, Maryland, is the guy that proposed this.
I tried to send him some e-mail, but his contact page won't accept you if you admit your zip code is out of his district. Just as well, I can think about what I want to say while I go dig up a valid zip code for his district. Want to be polite when I tell him where he can take this kind of thinking.
BTW, I got IP restricted several months back. I think it was from too much anonymous posting on a single thread. But why four anonymous posts was too many? Maybe it was because I was defending religion in those posts?
Anyway, after a couple of days I could post if I logged in first. Then after about four weeks the ban was dropped.
(Cool down? I needed to cool down? What about the guys who had to swear four times in every sentence to prove that God doesn't exist? Oh well, everywhere has their own kind of censorship, I guess.)
x86 is still x86, even if the libraries are different.