Quick comment from a rover driver, since I'm being a media whore today anyway....
Last I heard they'd found bound water, and the surface was a lot hotter than they expected it to be.
This is correct -- in Spirit's vicinity, the water content is something like a few percent of the soil. This is exciting not because it's news that there's water in the Martian soil (we knew that already, from Odyssey measurements), but because there's water where we are -- it means Spirit has water right under her feet. Also because it's "ground truth" for the orbital measurements.
The higher temperatures are probably due to the (clearing) dust storm. Spirit is almost too warm, which is about the last problem we ever expected to have (but I'd rather have this problem than most others I can think of!).
Incidentally, there probably is liquid water on Mars -- or, more precisely, under Mars; it's all in the range of 100m to 2km below the soil. Surface water would sublime.
Maybe I'm just being pessimistic, but I worry about the trend toward cheaper long distance, especially cheap international calling.
Why? Well, if you think telemarketing calls are bad now, wait until every business on the planet can afford to call you. Just like spam, but with your damn phone ringing off the hook 24 hours a day.
You can bet there's somebody in Japan who can afford to bug you for 3c/min, if it helps them sell a few more useless widgets.
``Every improvement in communication makes the bore more terrible,'' as Frank Moore Colby wrote.
This will be very useful in my plan to create a giant killer robot with the brain of a rat.
Does this remind anyone else of the Simpsons episode where they go to Itchy and Scratchy Land?
Re:I've always enjoyed these Google toys
on
Google's new toys
·
· Score: 2
My favorite is Google Sets. I use it to look for new musical artists. For example, if I type in a few band names in a similar genre that I like it returns a list containing other similar bands.
I'd never thought of that, so I tried it after reading this post. I put in my three favorite bands/artists: Green Day, Rage Against the Machine, and Eminem.
The first item (other than the ones I put in) on the resulting list was "Backstreet Boys."
I think there's a reason Google Sets hasn't made it out of the lab yet.:-)
I'm building a little Linux-based media box myself -- not as some kind of homebrew TiVo killer, but as something to show slideshows on the TV (mmm... Hawaii), play MP-3s while I'm reading, etc. But mainly it's just supposed to be a fun project.
I went with the
VIA EPIA 800,
which features an 800 MHz VIA C3 CPU with on-board TV-out (and much more) for a mere US$120. And it's tiny for a full-featured computer, just 170mm x 170mm.
Plus, it has built-in audio, on-board Ethernet (though I've plugged in a wireless PCI card), USB, etc. -- great for a project like this. And it runs GNU/Linux like a champ.
I put it in a US$90
black mini-ITX case
so it doesn't look like a computer -- it looks like an A/V component, fitting right in with the VCR and the cable box.
Finally, it runs very cool and very quiet (or it will once I remove the noisy hard disk and make it boot off the network instead) -- just one small CPU fan required. The CPU isn't that powerful -- mostly around the speed of a 600MHz Celeron, and the FPU is a little weaker than that -- but it's plenty of power for me.
There's a lot more info about this and similar hardware at
mini-itx.com. I think a big market is developing around these little boxes.
Did you hear what I wrote about the neighborhoods? They won't put them in neighborhoods.
And keep in mind that governing bodies have an official policy of neutrality; they're not going to build AIs that aren't neutral.
I believe that you believe all of this is true of the Orlando project. You know more about it than I do. I hope it remains true, and I hope it will be true for all other similar projects. I just don't think it will (in the general case, I mean); once the tech is in place, the smart money says that it will be abused eventually.
By the way, there's another way that the algorithms are a "colorblind": skin detection algorithms detect everyone as almost exactly the same skin color (but with different intensity). (One notable exception is asiatic skin tones, which are slightly different - but only slightly - almost not even statistically different).
OK, that's interesting, and I didn't know it. I don't think it changes much, though. If my eyes can tell the difference, a computer's eyes will be able to tell the difference -- if not now, then some day.
Here in the US we have a constitutional right to a fair trial which includes due process ( more below) one of the biggest aspects of this is teh ability to challenge your accuser in a court of law. If a machine enforves teh law who is your accuser?
Yeah, haven't these people seen the Star Trek episode "Court Martial"?
Kirk's Lawyer: Rights, sir, human rights -- the Bible, the Code of Hammurabi
and of Justinian, Magna Carta, the Constitution of the United States, fundamental declarations of the Martian colonies, the statutes of Alpha 3 -- gentlemen, these documents all speak of rights.
Rights of the accused to a trial by his peers, to be represented by counsel, the rights of cross-examination, but most importantly,
the right to be confronted by the witnesses against him -- a right to which my client has been denied.
Your Honor, that is ridiculous. We produced the witnesses in court. My learned opponent had the opportunity to see them, cross-examine them -- all but one! The most devastating witness against my client is not a human being. It's a machine,
an information system. The computer log of the Enterprise.
Some foolish part of me can't wait for one of these systems to give me a ticket. "Your honor, the most devastating witness against me...."
One of the things I like best about this is that unlike policemen, cameras are colorblind.
The software behind the cameras will do whatever you tell it to do. If someone decides the cameras should racially profile for some reason, they can be made to do it. (E.g., in principle, they could be told something like this: "If there are too many people in view to spy on them all at once, process the data for dark-skinned people first.") Cameras may not be racist, but they don't have a conscience, either.
Plus there's the issue of where the cameras are installed. I expect we'll just happen to see a lot more of them installed in black neighborhoods than in white neighborhoods. Note that I'm not saying that this is ipso facto the wrong choice, if that's where your city's street crime happens to be. But the fact that City X's cameras don't preferentially spy on black people instead of white people matters less if they're installed only in 99%-black neighborhoods.
Poor example: you shouldn't move into the middle of the intersection until you can leave it safely too. The middle of an intersection is a really dangerous place to stop, which is why you shouldn't stop there. If you do, then you are (in most states) in violation of the traffic regulations and you should get a ticket, whether its from a camera or a cop.
This does indeed violate the traffic laws in many places -- for example, in California. The problem is that you often cannot hope to make a left turn at all if you follow this rule, as when there's no left-turn signal (that is, no protected left). So you can wait five hours for your chance to make a legal left turn, with other cars stacking up behind you the whole time, or you can do the "wrong" thing because it's the only practical thing to do. (Or you can make three right turns instead -- uh, yeah.) Guess what everyone does.
This is one of the problems with delegating law enforcement to a machine: a cop knows better than to ticket people for this, a camera doesn't.
My proposal for the next version's expansion of RMGPT:
RMS Mandated GNU/POSIX Threads.
Aikido and Perl
on
Ask Larry Wall
·
· Score: 3, Interesting
First, many thanks for Perl, which has saved me much anguish.
I know you're an aikidoka, and after studying aikido for a year, I've come to see several similarities between aikido and Perl. For example, Perl gives you a nice feeling of blending with the problem instead of struggling directly against it, just as you blend with and redirect your attacker's energy rather than directly confronting it in aikido. Similarly, TMTOWTDI ("there's more than one way to do it") in aikido as well as in Perl (at least in my dojo, where understanding and reaching the goal is more important than slavishly copying the sensei).
My question is, did you consciously approach Perl with aikido in mind (or vice versa:-)? Or is it just that they both appeal to your personality in the same way?
Also, incidentally, what style of aikido do you practice?
To be honest, I don't see the importance of this. Let's say that Linux installs faster 100% of the time. So?
The importance of this isn't just the installation time, it's the installation experience, including the hassle and confusion involved with installing each OS. Lots of people have complained that Linux was difficult to install, often using this as a reason not to try it or a reason why "Linux will never be ready for the desktop." One of the author's points is that, compared to Windows, Linux is no longer so hard to install after all (indeed, in many ways, it's clearly easier and smoother). Whatever the difficulty of installing Linux was supposed to prove in the past, this is an argument to the contrary.
So people who used to bash (no pun intended) Linux for being difficult to install will have to find something else to complain about. And then we'll fix that.... [Insert "Step 3: Profit!" joke here.]
The patents will never be granted because they don't represent original work. We can prove this by the fact that open source development consists entirely of chasing tail lights, and never involves doing anything original. We know that's true because Microsoft says so. Q.E.D.
The examples he gives for usage of null pointers are both wrong. When a null pointer (whether written as 0 or NULL) is passed to a varargs function, it should be cast to a pointer of the appropriate type.
Indeed. In the particular case in question, passing a pointer to printf(), this should be (void *) 0 or (void *) NULL.
At least he's right when he says "The following is coded wrong.":-)
Bar is also mistaken on at least one other ANSI/ISO C-related point. He writes:
values of type size_t should use the Z size modifier [to printf], like so:
In fact, the Z modifier in the %Zu construction is non-standard. There was no portable way to print a size_t in the original ANSI/ISO C (C89). C99 (the 1999 revision of the ISO C standard) uses a lower-case z instead, so portable code should use %zu instead. Of course, the kernel is intended for compilation with gcc, not just any compiler, so Bar's example is correct for the kernel but is not (as he claims) standard.
Microsoft, or any other non-GPL developer, would be blocked from taking that code and linking to it. They'd be blocked from adding it to their code in order to efficiently use it. [...]
In your example, Microsoft would find themselves unable to benefit from the good GPL'd code. They'd develop their own "ActiveSwoop" compression which is guaranteed to be incompatible with the GPL'd code. Their huge userbase and marketshare gives "ActiveSwoop" considerable market viability even if it's not as good as the GPL code.
That's still no reason to ban releasing government-funded code under the LGPL. Microsoft (or anyone else) would have to release any modifications to the LGPLd code itself, but they could maintain proprietary control over the code they linked to it. And compatibility is assured.
To be fair, you don't seem to be arguing against the LGPL, only the GPL. But it's telling that Microsoft doesn't seem to make this distinction -- they lump the GPL and the LGPL together, implicitly or explicitly, and contrast them both equally with the BSD license. They want to pretend that their (already dubious) argument against the GPL translates to the LGPL, and it doesn't.
Finally, this incompatibility "guarantee" you speak of isn't a guarantee at all. If the file formats and protocols are documented and freely implementable by anyone, interoperability is entirely possible. XML and TCP/IP are two of many successful examples.
Under that logic, we should've been worried about the manufacturers of carriages and whips when the automobile came on the scene. How about all the people who ground away at lengthy calculations to produce mathematical tables and such whose jobs were eliminated by computers?
Not to mention that the TV, movie, and (music) recording industries themselves displaced earlier, well-established forms of entertainment: live theater (stage plays, vaudeville, circuses, you name it) and symphony orchestras, for instance. Imagine if the affected businesses had successfully sued Edison et al. for destroying their business models.
I'd recommend
Linux Core Kernel Commentary,
whose first edition was
favorably reviewed
right here on Slashdot. It's unlike other kernel books in that it examines much of the core of the Linux kernel code line by line; it's a good way to pick up a lot of the code's idioms and to learn to think like a kernel programmer, but this approach necessarily narrows its coverage. Depending on your specific interests, this might be the perfect book for you, or it might serve well as a companion volume to a book with broader/different focus.
The publisher has a
sample chapter
online (though their HTML looks weird to me; I hope it looks better in your browser). Also, you can read a little more about the book, find links to online reviews, get errata listings, and so on, at its
support site.
The pressure the boss was putting on us was awful, and he didn't really even understand what we were doing, even though he was the one demanding it. I think she read it in a trade mag somewhere. God I'd do a lot more work if she didn't read that shit.
Maybe he/she was just distracted by his/her sex change operation.
Microsoft will donate new servers to slashdot... of course, the slash code will have to be ported over to visual basic... slashdot's
url will change to slashdot.msn.com...
Slashdot Mirror
This is correct -- in Spirit's vicinity, the water content is something like a few percent of the soil. This is exciting not because it's news that there's water in the Martian soil (we knew that already, from Odyssey measurements), but because there's water where we are -- it means Spirit has water right under her feet. Also because it's "ground truth" for the orbital measurements.
The higher temperatures are probably due to the (clearing) dust storm. Spirit is almost too warm, which is about the last problem we ever expected to have (but I'd rather have this problem than most others I can think of!).
Incidentally, there probably is liquid water on Mars -- or, more precisely, under Mars; it's all in the range of 100m to 2km below the soil. Surface water would sublime.
Still waiting to drive ....
Woo-hoo! We're number one! We're number one! We're number one!
In your face, Japan!
Maybe I'm just being pessimistic, but I worry about the trend toward cheaper long distance, especially cheap international calling.
Why? Well, if you think telemarketing calls are bad now, wait until every business on the planet can afford to call you. Just like spam, but with your damn phone ringing off the hook 24 hours a day.
You can bet there's somebody in Japan who can afford to bug you for 3c/min, if it helps them sell a few more useless widgets.
``Every improvement in communication makes the bore more terrible,'' as Frank Moore Colby wrote.
Does this remind anyone else of the Simpsons episode where they go to Itchy and Scratchy Land?
I'd never thought of that, so I tried it after reading this post. I put in my three favorite bands/artists: Green Day, Rage Against the Machine, and Eminem.
The first item (other than the ones I put in) on the resulting list was "Backstreet Boys."
I think there's a reason Google Sets hasn't made it out of the lab yet. :-)
I went with the VIA EPIA 800, which features an 800 MHz VIA C3 CPU with on-board TV-out (and much more) for a mere US$120. And it's tiny for a full-featured computer, just 170mm x 170mm.
Plus, it has built-in audio, on-board Ethernet (though I've plugged in a wireless PCI card), USB, etc. -- great for a project like this. And it runs GNU/Linux like a champ.
I put it in a US$90 black mini-ITX case so it doesn't look like a computer -- it looks like an A/V component, fitting right in with the VCR and the cable box.
Finally, it runs very cool and very quiet (or it will once I remove the noisy hard disk and make it boot off the network instead) -- just one small CPU fan required. The CPU isn't that powerful -- mostly around the speed of a 600MHz Celeron, and the FPU is a little weaker than that -- but it's plenty of power for me.
There's a lot more info about this and similar hardware at mini-itx.com. I think a big market is developing around these little boxes.
Where can I read more about this newfangled "air carriage" you speak of? Is it anything like Doctor Flavinbottom's ocean-going mechanical horse?
Truly, this is an age of marvels.
I believe that you believe all of this is true of the Orlando project. You know more about it than I do. I hope it remains true, and I hope it will be true for all other similar projects. I just don't think it will (in the general case, I mean); once the tech is in place, the smart money says that it will be abused eventually.
OK, that's interesting, and I didn't know it. I don't think it changes much, though. If my eyes can tell the difference, a computer's eyes will be able to tell the difference -- if not now, then some day.
Some foolish part of me can't wait for one of these systems to give me a ticket. "Your honor, the most devastating witness against me ...."
The software behind the cameras will do whatever you tell it to do. If someone decides the cameras should racially profile for some reason, they can be made to do it. (E.g., in principle, they could be told something like this: "If there are too many people in view to spy on them all at once, process the data for dark-skinned people first.") Cameras may not be racist, but they don't have a conscience, either.
Plus there's the issue of where the cameras are installed. I expect we'll just happen to see a lot more of them installed in black neighborhoods than in white neighborhoods. Note that I'm not saying that this is ipso facto the wrong choice, if that's where your city's street crime happens to be. But the fact that City X's cameras don't preferentially spy on black people instead of white people matters less if they're installed only in 99%-black neighborhoods.
This does indeed violate the traffic laws in many places -- for example, in California. The problem is that you often cannot hope to make a left turn at all if you follow this rule, as when there's no left-turn signal (that is, no protected left). So you can wait five hours for your chance to make a legal left turn, with other cars stacking up behind you the whole time, or you can do the "wrong" thing because it's the only practical thing to do. (Or you can make three right turns instead -- uh, yeah.) Guess what everyone does.
This is one of the problems with delegating law enforcement to a machine: a cop knows better than to ticket people for this, a camera doesn't.
RMS Mandated GNU/POSIX Threads.
First, many thanks for Perl, which has saved me much anguish.
I know you're an aikidoka, and after studying aikido for a year, I've come to see several similarities between aikido and Perl. For example, Perl gives you a nice feeling of blending with the problem instead of struggling directly against it, just as you blend with and redirect your attacker's energy rather than directly confronting it in aikido. Similarly, TMTOWTDI ("there's more than one way to do it") in aikido as well as in Perl (at least in my dojo, where understanding and reaching the goal is more important than slavishly copying the sensei).
My question is, did you consciously approach Perl with aikido in mind (or vice versa :-)? Or is it just that they both appeal to your personality in the same way?
Also, incidentally, what style of aikido do you practice?
The importance of this isn't just the installation time, it's the installation experience, including the hassle and confusion involved with installing each OS. Lots of people have complained that Linux was difficult to install, often using this as a reason not to try it or a reason why "Linux will never be ready for the desktop." One of the author's points is that, compared to Windows, Linux is no longer so hard to install after all (indeed, in many ways, it's clearly easier and smoother). Whatever the difficulty of installing Linux was supposed to prove in the past, this is an argument to the contrary.
So people who used to bash (no pun intended) Linux for being difficult to install will have to find something else to complain about. And then we'll fix that .... [Insert "Step 3: Profit!" joke here.]
I thought of a solution to this. Reformulate the movie with Dick van Dyke as a crimefighting doctor, and call it Diagnosis: Mordor.
Thanks, I'll be here all week. You've been a great audience!
The patents will never be granted because they don't represent original work. We can prove this by the fact that open source development consists entirely of chasing tail lights, and never involves doing anything original. We know that's true because Microsoft says so. Q.E.D.
Indeed. In the particular case in question, passing a pointer to printf(), this should be (void *) 0 or (void *) NULL.
At least he's right when he says "The following is coded wrong." :-)
Bar is also mistaken on at least one other ANSI/ISO C-related point. He writes:
In fact, the Z modifier in the %Zu construction is non-standard. There was no portable way to print a size_t in the original ANSI/ISO C (C89). C99 (the 1999 revision of the ISO C standard) uses a lower-case z instead, so portable code should use %zu instead. Of course, the kernel is intended for compilation with gcc, not just any compiler, so Bar's example is correct for the kernel but is not (as he claims) standard.
Fair enough. After all, 640 MB of RAM should be enough for anybody. ;-)
That's still no reason to ban releasing government-funded code under the LGPL. Microsoft (or anyone else) would have to release any modifications to the LGPLd code itself, but they could maintain proprietary control over the code they linked to it. And compatibility is assured.
To be fair, you don't seem to be arguing against the LGPL, only the GPL. But it's telling that Microsoft doesn't seem to make this distinction -- they lump the GPL and the LGPL together, implicitly or explicitly, and contrast them both equally with the BSD license. They want to pretend that their (already dubious) argument against the GPL translates to the LGPL, and it doesn't.
Finally, this incompatibility "guarantee" you speak of isn't a guarantee at all. If the file formats and protocols are documented and freely implementable by anyone, interoperability is entirely possible. XML and TCP/IP are two of many successful examples.
Somehow I see it as deeply ironic that this bill's number is a power of 2 (2**11).
Not to mention that the TV, movie, and (music) recording industries themselves displaced earlier, well-established forms of entertainment: live theater (stage plays, vaudeville, circuses, you name it) and symphony orchestras, for instance. Imagine if the affected businesses had successfully sued Edison et al. for destroying their business models.
The publisher has a sample chapter online (though their HTML looks weird to me; I hope it looks better in your browser). Also, you can read a little more about the book, find links to online reviews, get errata listings, and so on, at its support site.
Oh, and I, er, happen to know the author. :-)
Dude, have you seen their product?
Maybe he/she was just distracted by his/her sex change operation.
I'm not touching that. (So to speak.)
And they'll have to rename it "backslashdot."