Burn the contents of the tar archive onto a CD. Mount the CD over the original directory structure. Use find(1)'s -fstype option to locate all the files that aren't on the CD, copy them to an empty disk image, then eject the CD. Remount the disk image over the original directory, delete all the files in the directory, then unmount the disk image. The files identical in name to those that were on the disk image (which are those that weren't on the CD) won't be deleted thanks to the peculiarities of mount(2).
Do us a favor next time you want to quote a three-paragraph post in order to respond to three words, and look up "figures of speech." Especially, "synecdoche."
The similarity between these things and NeXT's Property Lists (now called "Old-School Property Lists" that Apple/NeXT has standardized on XML) is incredible. Some things are changed, like having a specification instead of just assuming that the recipient will parse it and figure it out, but the likeness is there. I wonder if any of the proto people at google had experience with plists, or if it's just a case of convergent design.
Everything old-school is new-school again, I guess.
It doesn't help that CSS2 is ridiculously complex, and CSS3 even more so. I think one of the things that kept IE back was that all the great open-source and power-user browsers that jumped on board to implement it, screwed up. Remember Acid2? Remember how it took until 2005 before ANY browser passed the test? And remember how the test designers realized there was a a bug in the test?
Now think about this: Acid2 only tested a few narrow aspects of CSS2 compliance. Who's to say there aren't more bugs that no one understands in the various gearhead browsers on the market?
I think the W3C is part of the problem, no question. I haven't done any HTML for a while, but I remember the time I was doing it quite wellâ"there were so many standards to choose from, all put out by the same organization. Do I use HTML 4 Strict, or Transitional? Some browsers didn't deal well with Strict, but the W3C said Transitional was bad and you were an idiot for using it (even though they still had a fully-operational checker for it, the address of which I still have saved as a shortcut). If you wanted to use frames (they were bad back then, then they were good for awhile, then bad again, and now I don't care) there was another specification you had to worry about.
And then there's XHTML. I couldn't for the life of figure out what it was for. I asked someone, who told me that it was like "like HTML, only it was XML." "What's XML?" "A mark-up language." "You mean like SGML?" "XML is SGML." "Oh, so is HTML, they must be the same thing." "No, with XML you must close IMG tags." "Oh, that's... useful."
And don't get me started on the travesty that is cascading style sheets. It was years before anyone had even read through the documents so they could implement, and some things still aren't done. All the while, of course, we've been getting hammered with every-more-annoying floating CSS ads, such that I'd almost rather still be dealing with MARQUEE text instead.
In summary: CSS and Web Standards were a good idea ruined by a bunch of incompetent jerks.
I think you care about this, too. The fiber you care about is not necessarily the last mile to your house; it's whatever the weakest link in the chain is. When connecting to a remote host in Europe, North America, or Japan, then it's probably the last mile to your house (incidentally, it's probably also the least-useful mile, since it can only send packets from your to the ISP, and vice-versa).
But if you want to have a live peer-to-peer connection with a business contact in Dubai, or if you want to watch Thabo Mbeki's home videos of him wiggling out of international responsibility on YouTube.sa, then you're going to wish you had the deep-sea fiber. Those cables will bring broadband to a lot more people than a fatter pipe to your house will, and some of those people make more money in a day than you do in a year.
You might be wrong. We've already laid a ton of fiber down to serve Asia, North America, Europe. A lot of that is still unused.
There's not so much fiber serving the Carribbean, the Middle East, and Africa now, but the capacity for demand is growing. Saudi Arabia, UAE, and Bahrain are using their oil wealth to build whole new cities that will compete, not with industrial cities like Delhi or Beijing, but with New York, London, Silicon Valleyâ"the places where money is made on ideas, not extractive resources or physical products. They are trying to build first-class universities too. They are going to demand top-notch informatics and telecom capabilities, and thanks to your boss's car, they have the cash to get it.
Africa and the Caribbean are a bit different, but Egypt and South Africa are in good positions to make use of it. Some of the more stable countries (Morocco, Tanzania) could grow into it within a decade, and that's assuming that Lagos and Zimbabwe don't fix themselves up (I wouldn't hold your breath, but if it happened, they'd need fiber to sell oil/food/minerals). I suspect the Caribbean line is intended for Cuba, one the US ends its brain-dead embargo.
They were books on murder investigation, not committal. He could easily play that up to say that he killed her in a fit of passion and afterwards tried to cover it up. That sort of explains the arrogance: he'd read the books, and figured he would be able to outsmart the police and the DA.
Oddly enough, that seems to fit the evidence better than premeditation. If he'd killed her with malice aforethought, I'd imagine he'd have been able to dispose of the body without so much mess.
No, there are people who divide some things but not others into two groups, and people who divide things into two groups sporadically.
It's like the Epimenides paradox ("All Cretans are liars", says Epimenides, who is a Cretan). A liar doesn't lie all the time, only when it suits him. If a liar had to lie all the time, his deception wouldn't be very effective because everyone would know he was lying.
One telling of the story of Hercules has him match wits with Nereus, the Old Man of the Sea, who has a reputation for lying but cleverly tells two true statements to Hercules, who then falls for the third one when it turns out to be a lie.
In the first place, Russian has tenses. it's not Chinese.
In the second place, this is the sort of thing jurors are more qualified to judge. If she only uses the present tense, then your scenario is more likely. If she uses the present tense to refer to Nina but the past tense to refer to something else, that's a different kettle of fish. Not proof that the mother would be lying, mind you, but something for jurors to keep in mind.
Anyway, at this point it's a bit irrelevant, enah?
The Reiser koan: Three crazy people set up a bizarre love triangle/business. One murders his wife, the second kills eight others and maybe a ninth, and the third is dead. Who is the more guilty?
Or maybe it should be a joke line: A hacker, a Russian mail-order bride, and a bisexual serial killer into bondage walk into a bar...
A lot of people probably wanted Hans to be innocent because he's part of the "tribe", but if you look back, the evidence against him was a bit shakyâ"mostly circumstantial, plus the testimony of a nutcase who said he murdered eight other people. And let's not forget the procession of other men Nina had been in contact with, any one of whom could have been an internet stalker.
He was found guilty anyway, and now he's come clean, so I guess it was the right verdict. But that doesn't mean everyone who thought he was innocent in the past was experiencing cognitive dissonance, only that they weren't on the jury.
Okay, if a Cocoa app announces a service to the operating system in Mac OS X, that service is available from the application's menu on every other Cocoa app.
Practically every commercial graphics application, even those that compete with Adobe products, supports Photoshop.psd files. Most RAW digital camera formats also work well with pro graphics software, whereas GIMP requires (at the very least) the intercession of another program, dcraw.
Most personal finance/tax software plays well together; support for H&R's.txf format. AFAIK there aren't even any competitors in the open source arena, so maybe this is just a cheap shot.
If the photo is depicting something newsworthy, or everyone gives permission in advance, maybe.
According to the courts, if you're out in public you have not reasonable expectation of privacy. That may be, but I certainly think you have a reasonable expectation of not having your face plastered all over some guy's website just for walking around outside.
And... So would you rather have someone who doesn't care about how something works/knows what works write something or would you rather have someone who uses it all the time write something? It is like saying, would you rather a graphics program be written by an artist or a songwriter? The songwriter may make a graphics program that is nice for him, but doesn't satisfy the needs of an artist.
It's not really a question of artist vs. songwriter, but rather would you want two different kinds of graphics programs to play nicely together? Do you want your sound file editor to be able to interact well with your MIDI composer? Do you want your music ripper, your jukebox program, and your CD burner to all work together? And so on.
Again, most people using Linux are not artists nor do they use GUIs much, so their needs are different then the ones of other people. So they write programs to fit the needs they have.
It may have been true years ago that most Linux users may have stuck with the console, but these days any given command line is likely to be running in x/k/gterm. Nearly everyone using a web-browser is doing it through a GUI, music players all use GUIs, digital photo management is done through a visual interface, chatting (even on IRC) is done with a graphical program, email, etc etc etc. Even so, GUI is still treated as a second-class citizen or sorts, since if mouse-and-clicking something might be too hairy, or if adding another control would take up too much space, the developer can just resort to the command line. The end user can't always do that, but they do always make a note of the shortcoming. In other words, if you can't do it simply with a mouse, for a lot of users you can't do it at all.
Yep, and as you have seen with all the Visual Basic crap that floats around for Windows.
Yes, I sure have. But I've also seen plenty of good, useful apps. Quicken, Photoshop, most games, inventory management systems, library back-ends, and so on. And hey, you never know when you might decide you want a GUI for SCP or something like that. It may be easier to write and compile command-line programs for Unix than for Windows, but the ease of building graphical apps on Windows or even Mac is way ahead of Linux, and as almost no work is being done in this area on Linux, the gap is going to get wider.
A quick anecdote: the first version of Quake, for Windows, was developed on the original Next OS because the tools for development, even on other platforms, were so far ahead of everything else. No matter how good Wine gets, as long as developing for Windows is easier than for Linux, FOSS will be trailing, not leading, on the desktop.
I think you've misunderstood a lot of what I had to say. Those opinions are not my own, they are what I've imputed to authors of many open-source software communities, having observed and even participated in some of their discussions. I don't agree with the perspective I listed above, but I enumerated and explained them so that the OP (and whoever else) could gain some insight into the problem.
And for the record, it is possible to remove the GUI component of Mac OS X, just edit/etc/ttys, same as any Unix that auto-boots to GUI.
The problem with using command line tools is first that it causes a performance hit, and second that it can cause serious problems with argument flags. rm(1) causes problems if the first file happens to be named '-Rf ~', but that won't happen if unlink(2) is called directly.
Graphical tools, even if they do the same thing as a command-line tool, should be treated by developers as first-class priorities rather than something to hack together to satisfy GUI people.
This is a perfect example of my point 5. I'm not saying that Open Source doesn't have advantages, that they don't fix bugs, or that the developers involved aren't (mostly) wonderful people. In fact, I'd even say that the quality of open-source command-line tools is one the reason why such tools are even still viable. Be Linux, GNU got its start as a non-brain-dead version of the standard Unix tools, plus Emacs.
But the nuance of my point is lost on you, because I have something critical to say about Open Source. Therefore everything I say about Open Source is negative, therefore I disagree with you, therefore I'm wrong. Therefore I should shut up and work for EA or something.
If you read what I wrote, and your reply, carefully, you'll see that I said absolutely nothing with respect to friendliness or the user-to-developer connection. I didn't even say that developers don't take pride in their work; I said they don't care much for projects they don't use or work on, which affects how well applications interact with each other. I could write a whole book on this topic, but if you read about some of the problems other OS's have had with getting interoperability from glibc, or anyone getting copy/paste to work in X, you'll see what I mean.
Slashdot Mirror
You're just jealous you couldn't think of something so simultaneously clever and over-involved.
Mod parent -1 Horseshit.
yacc is a compiler, what do you think the two c's stand for?
Burn the contents of the tar archive onto a CD. Mount the CD over the original directory structure. Use find(1)'s -fstype option to locate all the files that aren't on the CD, copy them to an empty disk image, then eject the CD. Remount the disk image over the original directory, delete all the files in the directory, then unmount the disk image. The files identical in name to those that were on the disk image (which are those that weren't on the CD) won't be deleted thanks to the peculiarities of mount(2).
You're welcome.
The next bug will be in Boolean logic. After that, OpenBSD devs will start fixing structural engineering errors the Tower of Pisa.
Do us a favor next time you want to quote a three-paragraph post in order to respond to three words, and look up "figures of speech." Especially, "synecdoche."
Thanks.
The similarity between these things and NeXT's Property Lists (now called "Old-School Property Lists" that Apple/NeXT has standardized on XML) is incredible. Some things are changed, like having a specification instead of just assuming that the recipient will parse it and figure it out, but the likeness is there. I wonder if any of the proto people at google had experience with plists, or if it's just a case of convergent design.
Everything old-school is new-school again, I guess.
It doesn't help that CSS2 is ridiculously complex, and CSS3 even more so. I think one of the things that kept IE back was that all the great open-source and power-user browsers that jumped on board to implement it, screwed up. Remember Acid2? Remember how it took until 2005 before ANY browser passed the test? And remember how the test designers realized there was a a bug in the test?
Now think about this: Acid2 only tested a few narrow aspects of CSS2 compliance. Who's to say there aren't more bugs that no one understands in the various gearhead browsers on the market?
You can't blame MS for that, only the W3C.
I think the W3C is part of the problem, no question. I haven't done any HTML for a while, but I remember the time I was doing it quite wellâ"there were so many standards to choose from, all put out by the same organization. Do I use HTML 4 Strict, or Transitional? Some browsers didn't deal well with Strict, but the W3C said Transitional was bad and you were an idiot for using it (even though they still had a fully-operational checker for it, the address of which I still have saved as a shortcut). If you wanted to use frames (they were bad back then, then they were good for awhile, then bad again, and now I don't care) there was another specification you had to worry about.
And then there's XHTML. I couldn't for the life of figure out what it was for. I asked someone, who told me that it was like "like HTML, only it was XML." "What's XML?" "A mark-up language." "You mean like SGML?" "XML is SGML." "Oh, so is HTML, they must be the same thing." "No, with XML you must close IMG tags." "Oh, that's... useful."
And don't get me started on the travesty that is cascading style sheets. It was years before anyone had even read through the documents so they could implement, and some things still aren't done. All the while, of course, we've been getting hammered with every-more-annoying floating CSS ads, such that I'd almost rather still be dealing with MARQUEE text instead.
In summary: CSS and Web Standards were a good idea ruined by a bunch of incompetent jerks.
For the same reason that being too dumb to get the joke doesn't prevent you from posting to slashdot.
I think you care about this, too. The fiber you care about is not necessarily the last mile to your house; it's whatever the weakest link in the chain is. When connecting to a remote host in Europe, North America, or Japan, then it's probably the last mile to your house (incidentally, it's probably also the least-useful mile, since it can only send packets from your to the ISP, and vice-versa).
But if you want to have a live peer-to-peer connection with a business contact in Dubai, or if you want to watch Thabo Mbeki's home videos of him wiggling out of international responsibility on YouTube.sa, then you're going to wish you had the deep-sea fiber. Those cables will bring broadband to a lot more people than a fatter pipe to your house will, and some of those people make more money in a day than you do in a year.
You might be wrong. We've already laid a ton of fiber down to serve Asia, North America, Europe. A lot of that is still unused.
There's not so much fiber serving the Carribbean, the Middle East, and Africa now, but the capacity for demand is growing. Saudi Arabia, UAE, and Bahrain are using their oil wealth to build whole new cities that will compete, not with industrial cities like Delhi or Beijing, but with New York, London, Silicon Valleyâ"the places where money is made on ideas, not extractive resources or physical products. They are trying to build first-class universities too. They are going to demand top-notch informatics and telecom capabilities, and thanks to your boss's car, they have the cash to get it.
Africa and the Caribbean are a bit different, but Egypt and South Africa are in good positions to make use of it. Some of the more stable countries (Morocco, Tanzania) could grow into it within a decade, and that's assuming that Lagos and Zimbabwe don't fix themselves up (I wouldn't hold your breath, but if it happened, they'd need fiber to sell oil/food/minerals). I suspect the Caribbean line is intended for Cuba, one the US ends its brain-dead embargo.
Some guy on kernel.org: Will this bug in reiserfs ever get fixed, Hans?
Reiser: I'm incarerated, Lloyd!
We should set up a scholarship fund (or Namesys should set it up and we can send in money) that will pay for the kids to get CS degrees.
Yeah, I'm sick.
So you're saying that OpenBSD stopped supporting ipfilter because...
They were books on murder investigation, not committal. He could easily play that up to say that he killed her in a fit of passion and afterwards tried to cover it up. That sort of explains the arrogance: he'd read the books, and figured he would be able to outsmart the police and the DA.
Oddly enough, that seems to fit the evidence better than premeditation. If he'd killed her with malice aforethought, I'd imagine he'd have been able to dispose of the body without so much mess.
No, there are people who divide some things but not others into two groups, and people who divide things into two groups sporadically.
It's like the Epimenides paradox ("All Cretans are liars", says Epimenides, who is a Cretan). A liar doesn't lie all the time, only when it suits him. If a liar had to lie all the time, his deception wouldn't be very effective because everyone would know he was lying.
One telling of the story of Hercules has him match wits with Nereus, the Old Man of the Sea, who has a reputation for lying but cleverly tells two true statements to Hercules, who then falls for the third one when it turns out to be a lie.
In the first place, Russian has tenses. it's not Chinese.
In the second place, this is the sort of thing jurors are more qualified to judge. If she only uses the present tense, then your scenario is more likely. If she uses the present tense to refer to Nina but the past tense to refer to something else, that's a different kettle of fish. Not proof that the mother would be lying, mind you, but something for jurors to keep in mind.
Anyway, at this point it's a bit irrelevant, enah?
The Reiser koan:
Three crazy people set up a bizarre love triangle/business. One murders his wife, the second kills eight others and maybe a ninth, and the third is dead. Who is the more guilty?
Or maybe it should be a joke line:
A hacker, a Russian mail-order bride, and a bisexual serial killer into bondage walk into a bar...
A lot of people probably wanted Hans to be innocent because he's part of the "tribe", but if you look back, the evidence against him was a bit shakyâ"mostly circumstantial, plus the testimony of a nutcase who said he murdered eight other people. And let's not forget the procession of other men Nina had been in contact with, any one of whom could have been an internet stalker.
He was found guilty anyway, and now he's come clean, so I guess it was the right verdict. But that doesn't mean everyone who thought he was innocent in the past was experiencing cognitive dissonance, only that they weren't on the jury.
Okay, if a Cocoa app announces a service to the operating system in Mac OS X, that service is available from the application's menu on every other Cocoa app.
Practically every commercial graphics application, even those that compete with Adobe products, supports Photoshop .psd files. Most RAW digital camera formats also work well with pro graphics software, whereas GIMP requires (at the very least) the intercession of another program, dcraw.
Most personal finance/tax software plays well together; support for H&R's .txf format. AFAIK there aren't even any competitors in the open source arena, so maybe this is just a cheap shot.
Anyway, three examples good enough for you?
If the photo is depicting something newsworthy, or everyone gives permission in advance, maybe.
According to the courts, if you're out in public you have not reasonable expectation of privacy. That may be, but I certainly think you have a reasonable expectation of not having your face plastered all over some guy's website just for walking around outside.
A person's face should be his trademark.
It's not really a question of artist vs. songwriter, but rather would you want two different kinds of graphics programs to play nicely together? Do you want your sound file editor to be able to interact well with your MIDI composer? Do you want your music ripper, your jukebox program, and your CD burner to all work together? And so on.
It may have been true years ago that most Linux users may have stuck with the console, but these days any given command line is likely to be running in x/k/gterm. Nearly everyone using a web-browser is doing it through a GUI, music players all use GUIs, digital photo management is done through a visual interface, chatting (even on IRC) is done with a graphical program, email, etc etc etc. Even so, GUI is still treated as a second-class citizen or sorts, since if mouse-and-clicking something might be too hairy, or if adding another control would take up too much space, the developer can just resort to the command line. The end user can't always do that, but they do always make a note of the shortcoming. In other words, if you can't do it simply with a mouse, for a lot of users you can't do it at all.
Yes, I sure have. But I've also seen plenty of good, useful apps. Quicken, Photoshop, most games, inventory management systems, library back-ends, and so on. And hey, you never know when you might decide you want a GUI for SCP or something like that. It may be easier to write and compile command-line programs for Unix than for Windows, but the ease of building graphical apps on Windows or even Mac is way ahead of Linux, and as almost no work is being done in this area on Linux, the gap is going to get wider.
A quick anecdote: the first version of Quake, for Windows, was developed on the original Next OS because the tools for development, even on other platforms, were so far ahead of everything else. No matter how good Wine gets, as long as developing for Windows is easier than for Linux, FOSS will be trailing, not leading, on the desktop.
I think you've misunderstood a lot of what I had to say. Those opinions are not my own, they are what I've imputed to authors of many open-source software communities, having observed and even participated in some of their discussions. I don't agree with the perspective I listed above, but I enumerated and explained them so that the OP (and whoever else) could gain some insight into the problem.
And for the record, it is possible to remove the GUI component of Mac OS X, just edit /etc/ttys, same as any Unix that auto-boots to GUI.
The problem with using command line tools is first that it causes a performance hit, and second that it can cause serious problems with argument flags. rm(1) causes problems if the first file happens to be named '-Rf ~', but that won't happen if unlink(2) is called directly.
Graphical tools, even if they do the same thing as a command-line tool, should be treated by developers as first-class priorities rather than something to hack together to satisfy GUI people.
This is a perfect example of my point 5. I'm not saying that Open Source doesn't have advantages, that they don't fix bugs, or that the developers involved aren't (mostly) wonderful people. In fact, I'd even say that the quality of open-source command-line tools is one the reason why such tools are even still viable. Be Linux, GNU got its start as a non-brain-dead version of the standard Unix tools, plus Emacs.
But the nuance of my point is lost on you, because I have something critical to say about Open Source. Therefore everything I say about Open Source is negative, therefore I disagree with you, therefore I'm wrong. Therefore I should shut up and work for EA or something.
If you read what I wrote, and your reply, carefully, you'll see that I said absolutely nothing with respect to friendliness or the user-to-developer connection. I didn't even say that developers don't take pride in their work; I said they don't care much for projects they don't use or work on, which affects how well applications interact with each other. I could write a whole book on this topic, but if you read about some of the problems other OS's have had with getting interoperability from glibc, or anyone getting copy/paste to work in X, you'll see what I mean.