4. The box remains off the net during the entire install: no registering, no setting up an ISP, no activation, no network configuration, no nothing. (BTW, the only networking component that I install is tcp/ip. All the other MS stuff never gets on the machine.)
Using TCP/IP may have been a mistake. It was, after all, the vector by which the malware installed itself to begin with.
A better approach may be to do this with two computers, where one is the machine onto which you need to install XP and the other is already up & running with whatever operating system you like.
This second computer will act as a bridge to the internet, speaking TCP/IP only on its WAN interface, and speaking a non-routable protocol like NetBEUI to the XP machine on the LAN interface.
This way, the XP machine can only speak to other local machines.
With a setup like this, you can download the necessary service packs and other updates to the gateway machine -- people have already explained this in some detail elsewhere in this discussion -- and then the XP box can access the updates by regular old fashioned Windows file sharing.
Once you have the minimal updates, then and only then does it make sense to turn on TCP/IP support on the XP machine.
In otherwords, remember the gag from one of the later Hitchhiker's books where it is said that the trick to learning how to fly is to fling yourself at the ground with great vigor, but forget to actually land? Douglas Adams was joking around, of course, but this isn't such a bad description of how orbital dynamics work.
Bodies in orbit are flinging themselves to the side at exactly the same speed that the earth is pulling them inwards, such that they just happen to keep missing the ground.
I'm going to mangle the math here, for which I apologize in advance, but please bear with me, because this stuff is neat. For a body the size of the earth, the speed you need to keep a body in orbit -- that is, the speed at which you have to fling yourself perpindicular to the planet's downward pull in order to sustain a perpetual balance above the planet's surface -- is something like 17,500 miles per hour. At lower orbits, this translates to an orbital time of around 90 minutes, which is what the space shuttle sees. If you push out from the earth though, there's a height at which the rate at which you fall towards/around the earth just happens to be identical to the speed at which the planet rotates on its access. This height works out to something like 22,000 miles up, and if you put a body in orbit there, it will just happen to give the illusion of hovering over a single point on the planet's surface, perpetually, more or less forever.
Hey presto, geosynchronous orbit.
Douglas Adams was talking about something else -- something that doesn't seem to be possible -- but he happened to convey a very good sense of how orbits work in the process.
Maybe instead of adding such features to Firebird, you should pick up copies of Strunk & White's The Elements of Style and Lynne Truss' Eats, Shoots & Leaves. And read them. And get rid of that crutch!
When software can't do a good job -- and in this case, it can't -- then the best tool you have available is your brain. Use it! And the best tool for enriching your brain is a good book. Read them!
I'll admit I was being sarcastic, but I was also trying to be constructive. "Elements of Style" is a wonderful book, and a superb substitute for any software crutch you will find. "Eats, Shoots, and Leaves" is also wonderful, and while it is a bit Anglo-centric (I'm not sure all the rules given are valid for generally accepted American usage), she does a much better job at getting the reader to think about and internalize proper punctuation style than any software could ever do.
Also, speaking of books, throw away your thesaurus. It's one thing to use an obscure word that packs a lot of meaning into a succinct phrase; it's quite another to say a simple thing with flowery, obfuscating language that adds little. Evey grade school English teacher knows the difference, and any good teacher will penalize students that writes to "impress" when they should really be writing to "express" -- a subtle, but crucial, distinction.
A software thesaurus is the worst kind of software writing tool, because it lets stumbling writers puff up flimsy writing in the most transparent way imaginable. Don't get caught!:-)
Is it a 12" model? I thought those were basically identical to the 12" iBook but with a metal shell instead of plastic.
In any case, I don't have information on how Apple chooses to designate machines as far as the initial letter-codes go. The part I'm more sure about -- because an employee at an Apple store pointed it out to me -- is that the next three digits represent year & week that the device was manufactured.
And just to add another datapoint, Powermac G4s also seem so to start with "XB". If the CPU class is encoded in the serial number, that must come up after the datestamp. The same may also be the case of their laptops -- maybe they all start with "UV", but I don't have access to any Powerbook G4s, old clamshell iBooks, or older Powerbook G3s to compare.
I have their "Paul's Boutique" album from 1989 (A pretty good CD, IMO). I've seen a blurb somewhere that said that an album like this couldn't be made today, given that each track contains at least 2 or 3 recognizable samples. In fact, I'd say that the samples are so prominent, they are the main "musical instrument" on the record.
Well, yeah. It's not such a stretch to assume that that was the whole point back then. If you listen to an album like "License to Ill" , it's like a scavenger hunt to track down where all the samples came from. Hey, it's the drums from Led Zeppelin's "When The Levee Breaks"; hey, it's a Beatles riff; hey, it's that guy from "Good Times", etc.
And the same thing was true for other rappers in the eighties. If you listen to an old Public Enemy or Run-DMC album, they also were chock full of recognizeable riffs.
When I was a kid, this whole sampling business bugged the hell out of me -- how can they call it music when all they're doing is re-arranging the rock music I liked? It took me a while to come around and realize that they were making something new out of older things, just as the rock bands I liked were doing in more indirect ways (Led Zeppelin ripped off Robert Johnson & made heavy metal, while Robert Johnson ripped old black folk music and made the blues); it took me longer still to realize that all this sampling business is really very postmodern and therefore enjoyable in the same way that I enjoy blazingly referential Tarantino movies, retro-futuristic product design like the new Mini Cooper, the scattershot eclecticism of a computer language like Perl, or yes the rap of a Run-DMC or Public Enemy. All of these borrow heavily from that which came before them, but they make something somehow original out of the pieces, so the at-first obvious familiarity is actually distracting you from how brilliant & new this really is. You *think* you've seen this before, but once you get your head around it, you realize that it really is a whole new thing. Clever trick, that.
But I digress.
Back in the early 90s, Rolling Stone magazine had a constant series of articles about sampling, which really came to a head with Vanilla Ice's butchery of David Bowie & Queen's "Under Pressure" with his "Ice, Ice, Baby", which preserved not only a several second long sample, but the whole song had the same basic melodical feel as the original. Vanilla Ice and his record label got in a lot of trouble for that (I forget if they lost the lawsuit per se, but they were forced to change their tactics). Meanwhile, you had people like Dr. Dre and Snoop Dogg raiding everything they could find in James Brown's back catalog, turning all his old funk & soul hits into, basically, the exact same songs, but with more drugs & prostitution. James Brown was somewhat less than flattered.
As a result, as you say, the labels were forced to start doing more bookkeeping for the sampling done by their artists. That wasn't to say that sampling ended -- bands like Massive Attack & Portishead put out fantastic late-90s albums with lots of recognizable samples -- but somewhere in the fine print had to be some kind of legal acknowledgement that old material was being used, and if recognizeable snippets of more than a couple of seconds were being used (that is, if they were going beyond simple fair use citations of the original material), they had to get permission from the original artists.
In recent years, the only time bands have really run into trouble are ones that made the mistake that Vanilla Ice made, where not only the sample but the whole song seems like a simple remake of the original song. So for example, on "When The Levee Breaks", Led Zeppelin started the song with these massive, booming drums for a few bars before bringing in the guitars & other instruments, while the Beastie Boys took that drum loop and used it as the foundation for "Rhymin' And Stealin'" -- which sounds almost nothing like
FYI, the opening characters in the serial codes Apple uses indicate the product class & manufacture date of the product.
The initial letters of the serial string indicate the class of product. The letters "UV", for example, seem to be the code for iBooks, while iPods may start with "JQ" and a Powermac G5 may start with "XB". (I haven't yet figured out how often the letters are changed -- if, for example, the original iBook or the later G4 iBook had codes other than "UV" -- but whatever.)
More interestingly, the three digits following those first two letters indicate the year and the week during which that device was manufactured. So, for example, by setting the recall range to iBooks with codes from UV117XXXXXX to UV342XXXXXX, what they're admitting is that they had a problem from the 17th week of 2001 -- that is, the week starting Sunday, 22 Apr 2001 -- through the 42nd week of 2003 -- that is, Sunday, 12 Oct 2003. Approximately -- I'm not sure what day of the week they start counting on, or if Wednesday 1 Jan counts as being part of the 53rd week of the previous year, but again, whatever.
So, what they're saying is that all iBooks manufactured over a 30 month range had at least the possibility of a severe manufacturing defect.
As silly as these two might sound, IMHO, they are the single biggest barriers to adoption of open source productivity tools.
Nah, deploring the state of math education because kids today all use calculators, that's silly. Human minds aren't good at any kind of math beyond basic arithmatic, and using an electronic tool to increase speed & accuracy can be a very valuable thing for anyone. Complaining about something so obviously beneficial is silly.
But depending on Microsoft Word to enforce good grammar? Now that's just absurd. Just as humans are bad at math while computers can do it with ease, we're all naturally talented with language while computers are crude at best. While computers can help people be better at arithmatic, I can't see how they can do anything to improve one's language abilities.
Where did you pick up this habit? How do you function this way?
Maybe instead of adding such features to Firebird, you should pick up copies of Strunk & White's The Elements of Style and Lynne Truss' Eats, Shoots & Leaves. And read them. And get rid of that crutch!
Those who fail to learn from history, are condemned to repeat it.
And those who did not fail to learn history will also be condemned to repeat it, but they'll at least recognize the shit as being, in fact, shit, when the uneducated monkeys continue to fling it at the fan as they have always done.
At least the monkeys seem to be enjoying themselves, which is more than I can say for the miserable bastards who went & educated themselves...
Real audio was a mature technology when they (NPR) first signed on for their service. It has served them well. If you want them to consider free alternatives, make a presentation to them.
Exactly. Look at what happened with Car Talk (story one, story two). Click & Clack didn't like the way that Real tries to abuse their users either, so they reluctantly tried switching to Windows Media instead, knowing that this solution wasn't much better. After getting lots of complaints, they switched back to Real, but not before getting Real to relax on some of their policies.
And you may have other options, as well. In Boston, WBUR provides streams in each of RealAudio, Windows Media, and Quicktime, and even goes so far as to tell users how to listen to the station through iTunes (or, but they don't quite spell this out, any other player that can take an MPEG-4 URL as a stream source).
But if these formats aren't enough for you, and you've just got to have these shows in your format of choice (and you're using a Mac, but I think we can take that as given if you're an NPR junkie:-), then maybe you should take a look at Audio Hijack, which is a neat little program for, either on-demand or on a schedule, starting up an audio stream in your player of choice (or the site's player of choice, as the case may be) and capturing the output as AIFF files. (Actually, it does far more, and can record any audio on your Mac, but I'm trying to stay focused on internet audio here.)
These files can in turn be converted by a program like Amadeus II or LAME into MP3 or OGG or what have you; Audio Hijack can even fire off the converter program for you automatically if you want it to. The audio quality of a Real->AIFF->MP3 recording may not be anything spectacular, but for talk radio this isn't such a bad compromise (hint: an episode of This American Life averages around 30mb this way). For a year or two now, I've been idly wondering how to do a decent TiVo for radio on my computer, and now with this I think I've found a pretty good solution...
I think a fun game would be to find the longest way to print "Hello World", without unnecessary filler functions or comments.
Welcome to the world of the JAPH, or "Just Another Perl Hacker" game, in which frighteningly talented peoplee try to come up with the most convoluted possible ways they can think of to output the string 'just another perl hacker'.
Part of me is impressed by how expressive this language is, that it allows people all these inventive ways to do the same simple task.
Part of me has no trouble seeing way games like this scare a lot of people away from what really is a very nice language.
In any case, I don't think Perl is the only language where this meme has taken hold, but it certainly seems to have taken it farther than any other language I know of. If you really want to play this game of yours, jump right in -- you're hardly alone...
I think there's a simple explanation: this is the diagram for how one CPU is cooled.
Both of the CPUs in the top end machine will have a complete copy of this subsystem, with their own fan & coolant loops.
The icons labeled "G5" in this diagram are not the CPUs themselves (why aim the fan to the left of the hot bit?) but rather are pumps or some other kind of controller apparatus.
Actually, from some of what I've read, the new 90nm fabrication process is producing a CPU so small that traditional air cooling methods weren't efficient enough to dissipate excess heat away from the processor, so the liquid cooling was necessary to keep heat under control.
That's not to say that the chip runs all that hot (as you note, it compares very well to other CPU families here) or that sound dampening wasn't a factor (it was, obviously). These things are true, of course. But it's also true that these chips are both smaller & hotter than the original G5s, and this forced the engineers to come up with a more efficient heat transfer method.
That the solution they came up with is both quieter & cooler than air cooled systems is just one of those little elegant flourishes that these guys are getting so annoyingly predictable about:-)
My guess is you're looking at the cooling system for a single processor. Notice the single fan, compared to the two fans for the two processors in the side view a bit lower down on the page you referenced.
This guy has it exactly right: the diagram in the image/video only shows how a single CPU is cooled. With a second CPU, everything shown is duplicated. The two blocks labeled "G5" to the right of the fan & heat sink seem to be cooling pumps or something.
But of course, "magical" is a funnier description. It just doesn't happen to be what's going on here.
The most interesting explanation I'm aware of for Atlantis -- and all the other western flood myths (Noah's ark, Gilgamesh, etc) goes back even further, to the end of the last ice age, when sea levels were lower and the Mediterranean basin may have been a relatively small, dry basin.
Now, this may seem all the wildest speculation, but it is not entirely so, for if we examine a submarine contour map of the Straits of Gibraltar, we find there is an enormous valley running up from the Mediterranean deep, right through the Straits, and trenching some distance out on to the Atlantic shelf.... This refilling of the Mediterranean, which by the rough chronology we are employing in this book may have happened somewhere between 30,000 and 10,000 B.C., must have been one of the greatest single events in the pre-history of our race.... Suddenly the ocean waters began to break through over the westward hills and to pour in upon these primitive peoples--the lake that had been their home and friend became their enemy; its waters rose and never abated; their settlements were submerged; the waters pursued them in their flight. Day by day and year by year the waters spread up the valleys and drove mankind before them. Many must have been surrounded and caught by the continually rising salt flood. It knew no check; it came faster and faster; it rose over the tree-tops, over the hills, until it had filled the whole basin of the present Mediterranean and until it lapped the mountain cliffs of Arabia and Africa. Far away, long before the dawn of history, this catastrophe occurred.
So, we have a huge cataclysmic event that would have been common to all the people living in the Mediterranean basin, possibly going up past the Bosporous to the Black Sea.
And because nearly all ancient communities seem to have sprung up along sea coasts and river banks, it seems reasonable to assume that the ancient coastline of the Mediterranean (and Black Sea) would have been thickly populated, while the "inland" areas that form the current coastline would have been populated sparsely if at all.
With that in mind, it seems obvious that whatever remains of any civilizations that preceded ones like Greece & Egypt would have been in areas that are now submerged. The survivors of this cataclysm would have been dispersed across the region, where their stories may well have evolved into the various flood myths that have been handed down to us today. This would help explain why nearly all of these civilizations have flood myths, while also explaining why these stories vary so much.
Of course, anyone that keeps up with their Cryptogram newsletters would recognize immediately that, like most things that are described as "one time pads", these are very probably not one time pads. Quoting from Bruce Schneier's essay from the above linked newsletter:
So, let me summarize. One-time pads are useless for all but very specialized applications, primarily historical and non-computer. And almost any system that uses a one-time pad is insecure. It will claim to use a one-time pad, but actually use a two-time pad (oops). Or it will claims to use a one-time pad, but actually use a steam cipher. Or it will use a one-time pad, but won't deal with message re-synchronization and re-transmission attacks. Or it will ignore message authentication, and be susceptible to bit-flipping attacks and the like. Or it will fall prey to keystream reuse attacks. Etc., etc., etc.
One-time pads may be theoretically secure, but they are not secure in a practical sense. They replace a cryptographic problem that we know a lot about solving -- how to design secure algorithms -- with an implementation problem we have very little hope of solving. They're not the future. And you should look at anyone who says otherwise with deep and profound suspicion.
In the original poster's defence, I don't actually see him using the term "one time pad" anywhere other than the headline, which may have been put in by the Slashdot staff. In any case, the term is almost certainly being misused here.
Apple's implementation of Linux PAM may not be complete, but that doesn't change the fact that that's what they've been using since Panther came out. This isn't really a debatable point: all of Apple's documentation refers to Linux-PAM, the string 'linux' shows up 15 times in the pam manpage, etc. They got it from Linux.
If, as you say, they aren't using it pervasively, that's a different matter. Maybe by the time 10.4 comes out, the left & right hands at Apple will have had a nice little chat, and you'll finally be able to do a graphical login with PAM. In any case, the Linux version of PAM is available in OSX today, and (at least in some contexts) it can be used the same way it can be used on Linux.
OSX just uses Linux-PAM for authentication, so if you can get these cards working on Linux, the exact same procedure should work on your Macs. Further, any documentation describing how to get these cards working on Linux should also apply to OSX.
I posted this late to the earlier discussion about this hole, but chances are not many people saw it. Looks like I'm not the only one thinking this now though, so why not post it twice...
It occurs to me that the help:// URI protocol problem could be broader than is generally being portrayed so far.
Consider: the fundamental issue here is that an OSX web browser -- Safari in the original reports, but apparently also Mozilla etc -- is acting as a broker for any URI that the user may come across, delegating the request out to external handler programs. Whether those external programs handle their URIs safely may be an open question.
The problem isn't really that Safari or Help is broken, but that the interaction between them, arising from the URI handling mechanism on OSX, is leading to Unintended Consequences.
OSX can handle many different URI namespaces, some of which seem to be used nowhere other than OSX. I'm having a hard time finding an exhaustive list of the URI protocols that OSX supports, but a partial list includes, in no particular order:
So far, I can think of published vulnerabilities in the telnet:// and now help:// protocols [and this article expands on those points], but is that the end of it, or is the whole framework vulnerable to these sorts of attacks?
I have a hunch that we're just seeing the thin edge of the wedge. If that's the case, then a real fix is going to hit a lot of the system, which in turn will mean that Apple is going to have a hell of a time doing proper bug testing on whatever fix they come up with. And if that is the case, then the patches we see for this through Software Update seem likely to be partial, possibly unreliable or ineffective, and may lead to system instability. (I'd hope that isn't the case, but if the fix has to be at some fundamental level of the UI then it can't be ruled out.)
It may be that we don't see a true correction to this situation until at least the next 10.3 point release, if not 10.4...
...and to think I'd almost forgotten how hard it is to have a level-headed conversation with you...:-)
The office I work in has dozens of computers, of which around 70% run Debian, 15% are Macs, and 15% are Windows. NIS may be old and flaky, yes, but it works fine on the Debian side, and hooking the Macs into it seemed easier than starting over from scratch. Yes, something like LDAP &/or Kerberos would be nicer in the long run, but for now we're stuck with NIS, so please accept integration with NIS as a necessary constraint at this point.
And yes, I realize how the upgrade options differ, and based on what we're trying to do, upgrade installs seem to make the most sense. I don't get why the groupthink is so set against them -- I've never had any problems in using them, and the time saved in not reconfiguring machines seems worthwhile. I realize I'm out of step in thinking that upgrade installs aren't a problem, but I'm willing to be convinced that there are real problems in doing it this way -- it's just that you're not going to win me over by berating me here...:-). For lack of a calm, clear explanation of how clean installs really are better, I think I'll stick with this for now...
Sorry, I'm not trying to be argumentative, I'm just surprised that the clean installs are so much faster -- I had no idea that it made that much difference.
But please, humor me for a minute: will a clean install preserve things like user accounts, network settings, install applications, application settings under/Library, etc?
If a clean install wipes all this stuff out, then the 20 minutes saved isn't worth it if I have to go around and reconfigure every machine manually. Of the machines I was setting up, for example, each has one or two local user accounts with local home directories, and gets dozens of network user accounts with network home directories via NIS and NFS. That all has to be reconstructed after a clean install, right? Even if that can be scripted, it can still be a pain in the ass to do that many times.
In any case, I think net-booting or net-installing systems is going to be easier in the long run. We're doing that now with all the Debian machines we've got, and while setting up the infrastructure to support it is a pain in the ass, it does make management of individual machines much easier in the long run. We were using FAI to install Debian over the network in 3-5 minutes, but even that doesn't compare to the benefits of network booting. Now we (I) just have to set up the same thing on the Mac side...
Okay, so reading over this subthread, it seems that everyone but me is getting 10 minute installs, where I'm getting a third of that at best. My only explanation is that I'm doing an upgrade install, partly because that seems the easiest way to preserve existing NFS and NIS settings that we have for everything on our network, and apparently that Just Takes Longer.
That said, someone mentioned being able to do network installs in three minutes. Is this playing off OSX Server capabilities, or are people actually putting the standard OSX Client software on the network and installing that somehow? I'd be happy to give that a try if someone could offer a URL or two explaining what's involved...
Slashdot Mirror
Using TCP/IP may have been a mistake. It was, after all, the vector by which the malware installed itself to begin with.
A better approach may be to do this with two computers, where one is the machine onto which you need to install XP and the other is already up & running with whatever operating system you like.
This second computer will act as a bridge to the internet, speaking TCP/IP only on its WAN interface, and speaking a non-routable protocol like NetBEUI to the XP machine on the LAN interface.
This way, the XP machine can only speak to other local machines.
With a setup like this, you can download the necessary service packs and other updates to the gateway machine -- people have already explained this in some detail elsewhere in this discussion -- and then the XP box can access the updates by regular old fashioned Windows file sharing.
Once you have the minimal updates, then and only then does it make sense to turn on TCP/IP support on the XP machine.
In otherwords, remember the gag from one of the later Hitchhiker's books where it is said that the trick to learning how to fly is to fling yourself at the ground with great vigor, but forget to actually land? Douglas Adams was joking around, of course, but this isn't such a bad description of how orbital dynamics work.
Bodies in orbit are flinging themselves to the side at exactly the same speed that the earth is pulling them inwards, such that they just happen to keep missing the ground.
I'm going to mangle the math here, for which I apologize in advance, but please bear with me, because this stuff is neat. For a body the size of the earth, the speed you need to keep a body in orbit -- that is, the speed at which you have to fling yourself perpindicular to the planet's downward pull in order to sustain a perpetual balance above the planet's surface -- is something like 17,500 miles per hour. At lower orbits, this translates to an orbital time of around 90 minutes, which is what the space shuttle sees. If you push out from the earth though, there's a height at which the rate at which you fall towards/around the earth just happens to be identical to the speed at which the planet rotates on its access. This height works out to something like 22,000 miles up, and if you put a body in orbit there, it will just happen to give the illusion of hovering over a single point on the planet's surface, perpetually, more or less forever.
Hey presto, geosynchronous orbit.
Douglas Adams was talking about something else -- something that doesn't seem to be possible -- but he happened to convey a very good sense of how orbits work in the process.
When software can't do a good job -- and in this case, it can't -- then the best tool you have available is your brain. Use it! And the best tool for enriching your brain is a good book. Read them!
I'll admit I was being sarcastic, but I was also trying to be constructive. "Elements of Style" is a wonderful book, and a superb substitute for any software crutch you will find. "Eats, Shoots, and Leaves" is also wonderful, and while it is a bit Anglo-centric (I'm not sure all the rules given are valid for generally accepted American usage), she does a much better job at getting the reader to think about and internalize proper punctuation style than any software could ever do.
Also, speaking of books, throw away your thesaurus. It's one thing to use an obscure word that packs a lot of meaning into a succinct phrase; it's quite another to say a simple thing with flowery, obfuscating language that adds little. Evey grade school English teacher knows the difference, and any good teacher will penalize students that writes to "impress" when they should really be writing to "express" -- a subtle, but crucial, distinction.
A software thesaurus is the worst kind of software writing tool, because it lets stumbling writers puff up flimsy writing in the most transparent way imaginable. Don't get caught! :-)
Is it a 12" model? I thought those were basically identical to the 12" iBook but with a metal shell instead of plastic.
In any case, I don't have information on how Apple chooses to designate machines as far as the initial letter-codes go. The part I'm more sure about -- because an employee at an Apple store pointed it out to me -- is that the next three digits represent year & week that the device was manufactured.
And just to add another datapoint, Powermac G4s also seem so to start with "XB". If the CPU class is encoded in the serial number, that must come up after the datestamp. The same may also be the case of their laptops -- maybe they all start with "UV", but I don't have access to any Powerbook G4s, old clamshell iBooks, or older Powerbook G3s to compare.
Well, yeah. It's not such a stretch to assume that that was the whole point back then. If you listen to an album like "License to Ill" , it's like a scavenger hunt to track down where all the samples came from. Hey, it's the drums from Led Zeppelin's "When The Levee Breaks"; hey, it's a Beatles riff; hey, it's that guy from "Good Times", etc.
And the same thing was true for other rappers in the eighties. If you listen to an old Public Enemy or Run-DMC album, they also were chock full of recognizeable riffs.
When I was a kid, this whole sampling business bugged the hell out of me -- how can they call it music when all they're doing is re-arranging the rock music I liked? It took me a while to come around and realize that they were making something new out of older things, just as the rock bands I liked were doing in more indirect ways (Led Zeppelin ripped off Robert Johnson & made heavy metal, while Robert Johnson ripped old black folk music and made the blues); it took me longer still to realize that all this sampling business is really very postmodern and therefore enjoyable in the same way that I enjoy blazingly referential Tarantino movies, retro-futuristic product design like the new Mini Cooper, the scattershot eclecticism of a computer language like Perl, or yes the rap of a Run-DMC or Public Enemy. All of these borrow heavily from that which came before them, but they make something somehow original out of the pieces, so the at-first obvious familiarity is actually distracting you from how brilliant & new this really is. You *think* you've seen this before, but once you get your head around it, you realize that it really is a whole new thing. Clever trick, that.
But I digress.
Back in the early 90s, Rolling Stone magazine had a constant series of articles about sampling, which really came to a head with Vanilla Ice's butchery of David Bowie & Queen's "Under Pressure" with his "Ice, Ice, Baby", which preserved not only a several second long sample, but the whole song had the same basic melodical feel as the original. Vanilla Ice and his record label got in a lot of trouble for that (I forget if they lost the lawsuit per se, but they were forced to change their tactics). Meanwhile, you had people like Dr. Dre and Snoop Dogg raiding everything they could find in James Brown's back catalog, turning all his old funk & soul hits into, basically, the exact same songs, but with more drugs & prostitution. James Brown was somewhat less than flattered.
As a result, as you say, the labels were forced to start doing more bookkeeping for the sampling done by their artists. That wasn't to say that sampling ended -- bands like Massive Attack & Portishead put out fantastic late-90s albums with lots of recognizable samples -- but somewhere in the fine print had to be some kind of legal acknowledgement that old material was being used, and if recognizeable snippets of more than a couple of seconds were being used (that is, if they were going beyond simple fair use citations of the original material), they had to get permission from the original artists.
In recent years, the only time bands have really run into trouble are ones that made the mistake that Vanilla Ice made, where not only the sample but the whole song seems like a simple remake of the original song. So for example, on "When The Levee Breaks", Led Zeppelin started the song with these massive, booming drums for a few bars before bringing in the guitars & other instruments, while the Beastie Boys took that drum loop and used it as the foundation for "Rhymin' And Stealin'" -- which sounds almost nothing like
FYI, the opening characters in the serial codes Apple uses indicate the product class & manufacture date of the product.
The initial letters of the serial string indicate the class of product. The letters "UV", for example, seem to be the code for iBooks, while iPods may start with "JQ" and a Powermac G5 may start with "XB". (I haven't yet figured out how often the letters are changed -- if, for example, the original iBook or the later G4 iBook had codes other than "UV" -- but whatever.)
More interestingly, the three digits following those first two letters indicate the year and the week during which that device was manufactured. So, for example, by setting the recall range to iBooks with codes from UV117XXXXXX to UV342XXXXXX, what they're admitting is that they had a problem from the 17th week of 2001 -- that is, the week starting Sunday, 22 Apr 2001 -- through the 42nd week of 2003 -- that is, Sunday, 12 Oct 2003. Approximately -- I'm not sure what day of the week they start counting on, or if Wednesday 1 Jan counts as being part of the 53rd week of the previous year, but again, whatever.
So, what they're saying is that all iBooks manufactured over a 30 month range had at least the possibility of a severe manufacturing defect.
Ouch.
Nah, deploring the state of math education because kids today all use calculators, that's silly. Human minds aren't good at any kind of math beyond basic arithmatic, and using an electronic tool to increase speed & accuracy can be a very valuable thing for anyone. Complaining about something so obviously beneficial is silly.
But depending on Microsoft Word to enforce good grammar? Now that's just absurd. Just as humans are bad at math while computers can do it with ease, we're all naturally talented with language while computers are crude at best. While computers can help people be better at arithmatic, I can't see how they can do anything to improve one's language abilities.
Where did you pick up this habit? How do you function this way?
Maybe instead of adding such features to Firebird, you should pick up copies of Strunk & White's The Elements of Style and Lynne Truss' Eats, Shoots & Leaves. And read them. And get rid of that crutch!
Just trying to be helpful... :-)
And those who did not fail to learn history will also be condemned to repeat it, but they'll at least recognize the shit as being, in fact, shit, when the uneducated monkeys continue to fling it at the fan as they have always done.
At least the monkeys seem to be enjoying themselves, which is more than I can say for the miserable bastards who went & educated themselves...
Exactly. Look at what happened with Car Talk (story one, story two). Click & Clack didn't like the way that Real tries to abuse their users either, so they reluctantly tried switching to Windows Media instead, knowing that this solution wasn't much better. After getting lots of complaints, they switched back to Real, but not before getting Real to relax on some of their policies.
And you may have other options, as well. In Boston, WBUR provides streams in each of RealAudio, Windows Media, and Quicktime, and even goes so far as to tell users how to listen to the station through iTunes (or, but they don't quite spell this out, any other player that can take an MPEG-4 URL as a stream source).
But if these formats aren't enough for you, and you've just got to have these shows in your format of choice (and you're using a Mac, but I think we can take that as given if you're an NPR junkie :-), then maybe you should take a look at Audio Hijack, which is a neat little program for, either on-demand or on a schedule, starting up an audio stream in your player of choice (or the site's player of choice, as the case may be) and capturing the output as AIFF files. (Actually, it does far more, and can record any audio on your Mac, but I'm trying to stay focused on internet audio here.)
These files can in turn be converted by a program like Amadeus II or LAME into MP3 or OGG or what have you; Audio Hijack can even fire off the converter program for you automatically if you want it to. The audio quality of a Real->AIFF->MP3 recording may not be anything spectacular, but for talk radio this isn't such a bad compromise (hint: an episode of This American Life averages around 30mb this way). For a year or two now, I've been idly wondering how to do a decent TiVo for radio on my computer, and now with this I think I've found a pretty good solution...
Welcome to the world of the JAPH, or "Just Another Perl Hacker" game, in which frighteningly talented peoplee try to come up with the most convoluted possible ways they can think of to output the string 'just another perl hacker'.
Part of me is impressed by how expressive this language is, that it allows people all these inventive ways to do the same simple task.
Part of me has no trouble seeing way games like this scare a lot of people away from what really is a very nice language.
In any case, I don't think Perl is the only language where this meme has taken hold, but it certainly seems to have taken it farther than any other language I know of. If you really want to play this game of yours, jump right in -- you're hardly alone...
I think there's a simple explanation: this is the diagram for how one CPU is cooled.
Both of the CPUs in the top end machine will have a complete copy of this subsystem, with their own fan & coolant loops.
The icons labeled "G5" in this diagram are not the CPUs themselves (why aim the fan to the left of the hot bit?) but rather are pumps or some other kind of controller apparatus.
Make more sense? I think so...
Actually, from some of what I've read, the new 90nm fabrication process is producing a CPU so small that traditional air cooling methods weren't efficient enough to dissipate excess heat away from the processor, so the liquid cooling was necessary to keep heat under control.
That's not to say that the chip runs all that hot (as you note, it compares very well to other CPU families here) or that sound dampening wasn't a factor (it was, obviously). These things are true, of course. But it's also true that these chips are both smaller & hotter than the original G5s, and this forced the engineers to come up with a more efficient heat transfer method.
That the solution they came up with is both quieter & cooler than air cooled systems is just one of those little elegant flourishes that these guys are getting so annoyingly predictable about :-)
This guy has it exactly right: the diagram in the image/video only shows how a single CPU is cooled. With a second CPU, everything shown is duplicated. The two blocks labeled "G5" to the right of the fan & heat sink seem to be cooling pumps or something.
But of course, "magical" is a funnier description. It just doesn't happen to be what's going on here.
<elwood> These can't be stopped. They're on a mission from Gahd. </elwood>
The most interesting explanation I'm aware of for Atlantis -- and all the other western flood myths (Noah's ark, Gilgamesh, etc) goes back even further, to the end of the last ice age, when sea levels were lower and the Mediterranean basin may have been a relatively small, dry basin.
In H. G. Wells' Outline of History, there is this interesting passage:
So, we have a huge cataclysmic event that would have been common to all the people living in the Mediterranean basin, possibly going up past the Bosporous to the Black Sea.
And because nearly all ancient communities seem to have sprung up along sea coasts and river banks, it seems reasonable to assume that the ancient coastline of the Mediterranean (and Black Sea) would have been thickly populated, while the "inland" areas that form the current coastline would have been populated sparsely if at all.
With that in mind, it seems obvious that whatever remains of any civilizations that preceded ones like Greece & Egypt would have been in areas that are now submerged. The survivors of this cataclysm would have been dispersed across the region, where their stories may well have evolved into the various flood myths that have been handed down to us today. This would help explain why nearly all of these civilizations have flood myths, while also explaining why these stories vary so much.
It seems reasonable to me...
Oh, you can get anything you want at Apple's Restaraunt
You can get anything you want at Apple's Restaraunt
Just log right in, it's just to the left
Just a half a inch from the RDF
You can get anything you want at Apple's Restaraunt
Fair enough, but that's not what you said the first time around :-)
In the original poster's defence, I don't actually see him using the term "one time pad" anywhere other than the headline, which may have been put in by the Slashdot staff. In any case, the term is almost certainly being misused here.
Apple's implementation of Linux PAM may not be complete, but that doesn't change the fact that that's what they've been using since Panther came out. This isn't really a debatable point: all of Apple's documentation refers to Linux-PAM, the string 'linux' shows up 15 times in the pam manpage, etc. They got it from Linux.
If, as you say, they aren't using it pervasively, that's a different matter. Maybe by the time 10.4 comes out, the left & right hands at Apple will have had a nice little chat, and you'll finally be able to do a graphical login with PAM. In any case, the Linux version of PAM is available in OSX today, and (at least in some contexts) it can be used the same way it can be used on Linux.
OSX just uses Linux-PAM for authentication, so if you can get these cards working on Linux, the exact same procedure should work on your Macs. Further, any documentation describing how to get these cards working on Linux should also apply to OSX.
I posted this late to the earlier discussion about this hole, but chances are not many people saw it. Looks like I'm not the only one thinking this now though, so why not post it twice...
It occurs to me that the help:// URI protocol problem could be broader than is generally being portrayed so far.
Consider: the fundamental issue here is that an OSX web browser -- Safari in the original reports, but apparently also Mozilla etc -- is acting as a broker for any URI that the user may come across, delegating the request out to external handler programs. Whether those external programs handle their URIs safely may be an open question.
The problem isn't really that Safari or Help is broken, but that the interaction between them, arising from the URI handling mechanism on OSX, is leading to Unintended Consequences.
OSX can handle many different URI namespaces, some of which seem to be used nowhere other than OSX. I'm having a hard time finding an exhaustive list of the URI protocols that OSX supports, but a partial list includes, in no particular order:
So far, I can think of published vulnerabilities in the telnet:// and now help:// protocols [and this article expands on those points], but is that the end of it, or is the whole framework vulnerable to these sorts of attacks?
I have a hunch that we're just seeing the thin edge of the wedge. If that's the case, then a real fix is going to hit a lot of the system, which in turn will mean that Apple is going to have a hell of a time doing proper bug testing on whatever fix they come up with. And if that is the case, then the patches we see for this through Software Update seem likely to be partial, possibly unreliable or ineffective, and may lead to system instability. (I'd hope that isn't the case, but if the fix has to be at some fundamental level of the UI then it can't be ruled out.)
It may be that we don't see a true correction to this situation until at least the next 10.3 point release, if not 10.4...
Oh nevermind. Thanks for the "help" and "suggestions"...
...and to think I'd almost forgotten how hard it is to have a level-headed conversation with you... :-)
The office I work in has dozens of computers, of which around 70% run Debian, 15% are Macs, and 15% are Windows. NIS may be old and flaky, yes, but it works fine on the Debian side, and hooking the Macs into it seemed easier than starting over from scratch. Yes, something like LDAP &/or Kerberos would be nicer in the long run, but for now we're stuck with NIS, so please accept integration with NIS as a necessary constraint at this point.
And yes, I realize how the upgrade options differ, and based on what we're trying to do, upgrade installs seem to make the most sense. I don't get why the groupthink is so set against them -- I've never had any problems in using them, and the time saved in not reconfiguring machines seems worthwhile. I realize I'm out of step in thinking that upgrade installs aren't a problem, but I'm willing to be convinced that there are real problems in doing it this way -- it's just that you're not going to win me over by berating me here... :-). For lack of a calm, clear explanation of how clean installs really are better, I think I'll stick with this for now...
Sorry, I'm not trying to be argumentative, I'm just surprised that the clean installs are so much faster -- I had no idea that it made that much difference.
But please, humor me for a minute: will a clean install preserve things like user accounts, network settings, install applications, application settings under /Library, etc?
If a clean install wipes all this stuff out, then the 20 minutes saved isn't worth it if I have to go around and reconfigure every machine manually. Of the machines I was setting up, for example, each has one or two local user accounts with local home directories, and gets dozens of network user accounts with network home directories via NIS and NFS. That all has to be reconstructed after a clean install, right? Even if that can be scripted, it can still be a pain in the ass to do that many times.
In any case, I think net-booting or net-installing systems is going to be easier in the long run. We're doing that now with all the Debian machines we've got, and while setting up the infrastructure to support it is a pain in the ass, it does make management of individual machines much easier in the long run. We were using FAI to install Debian over the network in 3-5 minutes, but even that doesn't compare to the benefits of network booting. Now we (I) just have to set up the same thing on the Mac side...
Okay, so reading over this subthread, it seems that everyone but me is getting 10 minute installs, where I'm getting a third of that at best. My only explanation is that I'm doing an upgrade install, partly because that seems the easiest way to preserve existing NFS and NIS settings that we have for everything on our network, and apparently that Just Takes Longer.
That said, someone mentioned being able to do network installs in three minutes. Is this playing off OSX Server capabilities, or are people actually putting the standard OSX Client software on the network and installing that somehow? I'd be happy to give that a try if someone could offer a URL or two explaining what's involved...