I'm really surprised that they actually took windowsupdate offline.
Just to clarify (though I think you understand this), they unregistered the windowsupdate.com domain, but did *not* take the Windows Update service offline.
The windowsupdate.com domain was only a "redirector" domain anyway. The service itself is hosted at windowsupdate.microsoft.com, but apparently the worm-burners weren't very clever on this point, and they pointed at the redirector domain instead of the real thing. This let Microsoft retire the redirector as part of a defense against this particular worm. I'd say MS is just being opportunistic, because basically very few people ever used windowsupdate.com to get to the Windows Update service anyway. For example, if you click Tools | Windows Update in IE you are sent to windowsupdate.microsoft.com, not to windowsupdate.com.
This article discusses how astronauts get lots of useful pictures that we would not get from satellites.
And I've seen lots of articles like this one explaining how an astronaut discovered something unexpected and that would have been missed without the astronaut there.
I've also seen articles (sorry, no links handy) where on-the-ground scientists talk about how they can execute much richer experiments in space because there are people up there.
You might not think these are great examples, and it's true that given enough time, technology could do most of the things that astronauts are doing. Plus the claims that most experiments are autonomous seems true enough. But even with the autonomous experiments, there have been reports back about the people on board being able to see something unexpected, to make calibrations in ways that could not have been anticipated, to make unexpected (and otherwise impossible) repairs to important and pricey equipment.
The common thread here seems to be that having real people on the ISS has generated lots of ideas, with respect to both science and experimentation, that might not have been thought up for some time by ground-based researchers, and certainly not by robots. In other words, they seem to speed up the efficiency of our learning and research up there. And it's possible that having real people on board something like the ISS will help guide researchers in this way for a long time, no matter how far out that research and learning curve goes. What we might look for is the point of diminishing returns on that curve - the time when having astronauts on board, while still adding value, doesn't add enough value to justify their cost or the risk to their lives. From what I've read, we aren't very close to that situation yet.
I appreciate the reply, and would like to ask a follow up question. Remember, IAJADBG (I am just a database guy).
That's not an issue if you're only dealing with messages, rather than people (e.g. the "twins paradox"). To send a message between two reference frames, all you have to do is send a pulse of light between them.
But this would be communicating with gravity permutations, not light pulses. What would the equivalent of color or spectrum shift be for gravity?
This assumes that you can send faster-than-light messages in two different reference frames that are moving with a high relative velocity - you bounce the message back and forth between the reference frames, and the net result is that it arrives at its point of origin before it was sent.
But doesn't this argument assume that the special theory of relativity applies? Because if you look at the general theory of relativity, you have to account for (speaking from the point of view of one reference frame only) the massively negative acceleration as the message is stopped and bounced back. I think that if you work it out that this negative acceleration will cancel out the temporal paradox that is being suggested.
I've seen a persuasive presentation of this on the Web someplace, but it's too late to look for it now. Maybe a real relativity physicist could comment?
Yes, space weather forecasting is a good thing
on
SOHO's Antenna Jammed
·
· Score: 1
What do you do, move the Earth out of the way? Put your tinfoil hat on? Why bother forecasting the Sun's weather, not much you can do.
Strong storms can be deadly to spacewalking astronauts. The crew of the International Space Station, while inside, is generally not endangered, but they do have a special protective area they can go to in a severe storm.
Advance warning of impending storms allows satellite operators to reduce the risk of damage to some satellites by shutting down electronics. Engineers anticipate problems in an effort to recover damaged satellites before they are lost. Extra staff can be brought in. Agreements are made to shift signals to other satellites.
Re:The coolest thing on that site
on
Platform Evangelism
·
· Score: 4, Funny
> The coolest thing on that site... is the Windows Media file of the seattle earthquake.
But did you notice how none of the computers rebooted? That ought to silence people who say Windows isn't fault-tolerant!
> I think one of Lou's contributions to the IBM culture is act of being more tight > lipped in public. He openly referred to the IT industry as a media "circus"...
What you're saying is correct as far as it goes, but seems to me overly restrictive about what XML derivitives and other markup languages are intended to be used for. The difference between "emphasis" and "emphasize this" is a difference in grammar. The result is still that a rendering device emphasizes something.
Similarly, the difference between "location" and "locate this" is a difference in grammar, but the result would be that a rendering device could locate something.
XML is supposed to create a document. That's a pretty broad category of things, and (to the amazement of some/.ers) includes a lot of document types that would *not* be interpreted by a Web browser. So suppose there is an XML document someplace in the world that looks like this:
How is this *not* a valid XML document (I don't mean the syntax, I mean the concept)? It is a perfectly good use of the XML concept, and it is perfectly useful for remotely controlling a telescope, whether the RTML interpreter is in the scope's hardware or in a piece of planetarium software. This is true even though the document contains "instructions" - i.e., it says that I want the scope to photograph M13 at 20:05 UT for 30 seconds and M92 at 20:06 for 30 seconds (and it also gives the location of M13 and M92, although the interpreter is likely to already know this).
Please understand that I just made this format up, I haven't tried to look up the actual RTML format anyplace. But I just don't see why in the world anybody would object to somebody doing something like this with XML. It looks like it's a spot-on application of what XML is for.
Note that my example is *not* a programming language, but (from the astronomer's point of view) a format for conveying instructional *data*.
How do you use a markup language to control a telescope?
The same way you use other types of XML to control, say, a Web browser.
Documents have content, and content can include procedural instructions. Lots of telescopes have programmable interfaces, and RTML could be used to either:
Tell a program running on a computer hooked up to a telescope what the telescope should do; or
Tell the telescope what to do directly.
There are lots of programs available that can interpret instructions and transmit control commands to scopes (in the spirit of the first bullet point above). I don't know if there are (m)any telescopes available that accept RTML directly (in the spirit of the second bullet point), but there might be. My scopes are not computerized, so when I use telescope-control software, it is always on scopes that belong to somebody else... so I never have to worry about implementation details like that very much.
> It sure is good to know that our children will be safe from being accidentally redirected to dangerous, kid-unfriendly sites like www.cnn.com when they're hanging around nickelodeon.kids.us. Way to go, GWB.
I know you were being cynical, but...
To look at this another way, how hard is it going to be for me to set up filters that allow all URLs in the kids.us domain plus any individual sites I want my kids to go to (such as CNN or, heaven forbid, Fox News and ESPN)? It sounds very easy to me, a big practical improvement for parents, especially compared to the current state of affairs.
There is a little town in Iowa called Riverside which bills itself as "the future birthplace of James T. Kirk." Every year, they have a geeky trekkie party thing - you know, the kind of thing where lots of middle-aged, overweight guys dress themselves up in tin-foil outfits supposed to make them look like Enterprise crew members. Or worse yet, they dress like tribbles. But I digress.
At least once, they tried to get JTK himself to come and grace their little party. I don't know if they offered to pay him or not, so I'm not saying he should have gone, but his response was basically something like "there's no way in @#^$% that I'm going to Riverside, Iowa." Whether or not he was going to accept the invitation, you'd think the guy could extend basic courtesy.
So, it doesn't surprise me that Kirk seems a little curt. I just think he really might not care much about things trekkie, and he might not care much what other people think about him.
True, but my point was just that you couldn't start with computers from 54 years (or whatever) in the future.:o)
I just wanted to start a thread where people would talk about the problem intelligently and mathematically, so I started with the simplistic math just to generate some excitement (it appears I succeeded!). Kind of like bad journalism. Maybe I should be censored by the/. community for this - nobody every does stuff like that on these pages, right?
Seriously, I've done lots of advanced math, but I quit a decade or so ago, and so now I'm at a point where I was interested in seeing the good math, but was feeling far too lazy (read: efficient?) to do the math myself.
Several posters made some interesting comments, but the post that finally came up with what I thought was the most useful information can be found here.
> Keep in mind this is two days before the full moon, so you're going to miss a lot of low magnitude meteors.
This is true, but Leonidforecasts predict rates of 100 to 350 meteors per 15 minutes, even after adjusting for the brightness of the moon!
You're right that we'll be missing lots of low-mag meteors, and it would be great to get this shower near a New Moon. In fact, I think last year's Leonid shower was near New Moon, though here in Iowa it was just a Clouded Moon. But 350 meteors per 15 minutes is still a big deal - I can't wait!
Catch-22: Unfortunately, if you started cracking it today, you'd be using today's computers.:o)
But I agree with you that 54 years is probably the more practical guess. The interesting thing to me is that 54 years isn't very long really. Which means somebody might actually be working on this with some justifiable optimism.
The main problem I would like to hear a real crypto-head comment on, however, is what the appropriate order of complexity is for "weak elliptic curve" crypto. We are assuming it is exponential - add a bit to the key, and the complexity doubles. But is that right? It certainly might not be. Anybody know the answer?
> Obviously something is wrong with your thinking. Hint: take a computational complexity class.
You're obviously right, and (having done quite well, thanks, in all my PhD linear and nonlinear programming classes) I knew it when I posted. That's a little embarrassing to admit, but I still thought it would be a fun post that might make people think a little bit.
Problem is, I don't use any of those nifty advanced math skills anymore, so I just posted the "easy lie." You know: "statistics, damned statistics, and bad math posted in a pop-science forum."
I hope it still communicates to people the degree to which the size difference in the keys might matter. I actually thought another poster did a better job when s/he said something like "add another bit and you've just pushed Moore's law back 18 months."
Oh, and I should add (with respect to the impossibility claim) that 36,028,797,018,963,968 years is approximately 2.5 million times the estimated age of the universe.
So, while we might not want to use the word "impossible," I suppose we could say something like "unimaginably unlikely given current technologies."
Since a 163-bit key is 2^54 times more complex than a 109-bit key, and it took 4 years for the 109-bit key, aren't we looking at at least 4 * 2^53 years, not even figuring in the elliptical complexity (which I admit I would need to read up on)?
According to calc.exe, 4 * 2^53 years is 36,028,797,018,963,968 years. Anybody want to start working on that one?
...but stupid too. Did you see this part? Apparently they are not only possible to copy, but (sometimes) impossible to play! Quick, get me a Sharpie(R)!
Unfortunately, some audio CD players and in-car players use PC CD drives, and will not legitimately play a protected CD you have paid for. Nor can people play music CDs on their PCs.
> Microsoft probably won't put in any significant look and feel changes until much later in the testing program,
Just as a sidebar, this may not be true. As an MS partner, I got to see an early, build-of-the-day, pre-beta version of Windows XP, and all the look-and-feel changes were pretty much there already. When the RTM finally arrived, nothing in the look-and-feel area really surprised me, because it hadn't changed much from the early version I had seen.
What *had* changed, however, was the speed. In the early version, some of the menu rendering was so slow that the demonstrator would click on the start button and then talk to us a bit while the menu built. He said that a lot of optimizations are turned on fairly late in the development cycle. So maybe new GUI features are already built into Longhorn, but not running at optimized speeds yet.
For what it's worth, these screenshots look like doctored-up XP screenshots to me, but I'm no expert on this...
The Czech Republic recently became the first country in the world to pass a dark-sky law. From what I've heard, it works. The idea is simple enough: shine the light on the things you're trying to light up, and quit shining so much of it into the sky. Details are here (as well as lots of other places).
For those who think this doesn't matter, wouldn't it be swell if light polution became so pervasive that we couldn't see that next mass-extinction event meteor heading our way? The headlines read: Doomsday Meteor Arrives Unannounced. Subtitle reads: At least the few survivors had a well-lit view of the damage.:o)
Slashdot Mirror
I'm really surprised that they actually took windowsupdate offline.
Just to clarify (though I think you understand this), they unregistered the windowsupdate.com domain, but did *not* take the Windows Update service offline.
The windowsupdate.com domain was only a "redirector" domain anyway. The service itself is hosted at windowsupdate.microsoft.com, but apparently the worm-burners weren't very clever on this point, and they pointed at the redirector domain instead of the real thing. This let Microsoft retire the redirector as part of a defense against this particular worm. I'd say MS is just being opportunistic, because basically very few people ever used windowsupdate.com to get to the Windows Update service anyway. For example, if you click Tools | Windows Update in IE you are sent to windowsupdate.microsoft.com, not to windowsupdate.com.
This article discusses how astronauts get lots of useful pictures that we would not get from satellites.
And I've seen lots of articles like this one explaining how an astronaut discovered something unexpected and that would have been missed without the astronaut there.
I've also seen articles (sorry, no links handy) where on-the-ground scientists talk about how they can execute much richer experiments in space because there are people up there.
You might not think these are great examples, and it's true that given enough time, technology could do most of the things that astronauts are doing. Plus the claims that most experiments are autonomous seems true enough. But even with the autonomous experiments, there have been reports back about the people on board being able to see something unexpected, to make calibrations in ways that could not have been anticipated, to make unexpected (and otherwise impossible) repairs to important and pricey equipment.
The common thread here seems to be that having real people on the ISS has generated lots of ideas, with respect to both science and experimentation, that might not have been thought up for some time by ground-based researchers, and certainly not by robots. In other words, they seem to speed up the efficiency of our learning and research up there. And it's possible that having real people on board something like the ISS will help guide researchers in this way for a long time, no matter how far out that research and learning curve goes. What we might look for is the point of diminishing returns on that curve - the time when having astronauts on board, while still adding value, doesn't add enough value to justify their cost or the risk to their lives. From what I've read, we aren't very close to that situation yet.
Now those 1,000 monkeys can actually do something useful.
I appreciate the reply, and would like to ask a follow up question. Remember, IAJADBG (I am just a database guy).
That's not an issue if you're only dealing with messages, rather than people (e.g. the "twins paradox"). To send a message between two reference frames, all you have to do is send a pulse of light between them.
But this would be communicating with gravity permutations, not light pulses. What would the equivalent of color or spectrum shift be for gravity?
This assumes that you can send faster-than-light messages in two different reference frames that are moving with a high relative velocity - you bounce the message back and forth between the reference frames, and the net result is that it arrives at its point of origin before it was sent.
But doesn't this argument assume that the special theory of relativity applies? Because if you look at the general theory of relativity, you have to account for (speaking from the point of view of one reference frame only) the massively negative acceleration as the message is stopped and bounced back. I think that if you work it out that this negative acceleration will cancel out the temporal paradox that is being suggested.
I've seen a persuasive presentation of this on the Web someplace, but it's too late to look for it now. Maybe a real relativity physicist could comment?
What do you do, move the Earth out of the way? Put your tinfoil hat on? Why bother forecasting the Sun's weather, not much you can do.
From space.com:
And, more importantly for most of us, from msnbc:
But did you notice how none of the computers rebooted? That ought to silence people who say Windows isn't fault-tolerant!
Rimshot!
Similarly, the difference between "location" and "locate this" is a difference in grammar, but the result would be that a rendering device could locate something.
XML is supposed to create a document. That's a pretty broad category of things, and (to the amazement of some
Please understand that I just made this format up, I haven't tried to look up the actual RTML format anyplace. But I just don't see why in the world anybody would object to somebody doing something like this with XML. It looks like it's a spot-on application of what XML is for.
Note that my example is *not* a programming language, but (from the astronomer's point of view) a format for conveying instructional *data*.
Documents have content, and content can include procedural instructions. Lots of telescopes have programmable interfaces, and RTML could be used to either:
- Tell a program running on a computer hooked up to a telescope what the telescope should do; or
- Tell the telescope what to do directly.
There are lots of programs available that can interpret instructions and transmit control commands to scopes (in the spirit of the first bullet point above). I don't know if there are (m)any telescopes available that accept RTML directly (in the spirit of the second bullet point), but there might be. My scopes are not computerized, so when I use telescope-control software, it is always on scopes that belong to somebody elseTo look at this another way, how hard is it going to be for me to set up filters that allow all URLs in the kids.us domain plus any individual sites I want my kids to go to (such as CNN or, heaven forbid, Fox News and ESPN)? It sounds very easy to me, a big practical improvement for parents, especially compared to the current state of affairs.
Just my 0.02.
There is a little town in Iowa called Riverside which bills itself as "the future birthplace of James T. Kirk." Every year, they have a geeky trekkie party thing - you know, the kind of thing where lots of middle-aged, overweight guys dress themselves up in tin-foil outfits supposed to make them look like Enterprise crew members. Or worse yet, they dress like tribbles. But I digress.
At least once, they tried to get JTK himself to come and grace their little party. I don't know if they offered to pay him or not, so I'm not saying he should have gone, but his response was basically something like "there's no way in @#^$% that I'm going to Riverside, Iowa." Whether or not he was going to accept the invitation, you'd think the guy could extend basic courtesy.
So, it doesn't surprise me that Kirk seems a little curt. I just think he really might not care much about things trekkie, and he might not care much what other people think about him.
Just my 0.02.
Didn't we already cover this a couple of hundred years ago? Well, here in the US anyway...
No taxation on net presentation! (rimshot!)
I am hman hrakunapunt.
True, but my point was just that you couldn't start with computers from 54 years (or whatever) in the future. :o)
/. community for this - nobody every does stuff like that on these pages, right?
I just wanted to start a thread where people would talk about the problem intelligently and mathematically, so I started with the simplistic math just to generate some excitement (it appears I succeeded!). Kind of like bad journalism. Maybe I should be censored by the
Seriously, I've done lots of advanced math, but I quit a decade or so ago, and so now I'm at a point where I was interested in seeing the good math, but was feeling far too lazy (read: efficient?) to do the math myself.
Several posters made some interesting comments, but the post that finally came up with what I thought was the most useful information can be found here.
Aargh ... how will it take until we start the "but is MySQL really an RDBMS" thread here?
... there, I've just done it!
(don't click here!)
(or here!)
Oops, wait
> Keep in mind this is two days before the full moon, so you're going to miss a lot of low magnitude meteors.
This is true, but Leonid forecasts predict rates of 100 to 350 meteors per 15 minutes, even after adjusting for the brightness of the moon!
You're right that we'll be missing lots of low-mag meteors, and it would be great to get this shower near a New Moon. In fact, I think last year's Leonid shower was near New Moon, though here in Iowa it was just a Clouded Moon. But 350 meteors per 15 minutes is still a big deal - I can't wait!
Thank you. I knew somebody would have the real goods on this question.
Catch-22: Unfortunately, if you started cracking it today, you'd be using today's computers. :o)
But I agree with you that 54 years is probably the more practical guess. The interesting thing to me is that 54 years isn't very long really. Which means somebody might actually be working on this with some justifiable optimism.
The main problem I would like to hear a real crypto-head comment on, however, is what the appropriate order of complexity is for "weak elliptic curve" crypto. We are assuming it is exponential - add a bit to the key, and the complexity doubles. But is that right? It certainly might not be. Anybody know the answer?
> Obviously something is wrong with your thinking. Hint: take a computational complexity class.
You're obviously right, and (having done quite well, thanks, in all my PhD linear and nonlinear programming classes) I knew it when I posted. That's a little embarrassing to admit, but I still thought it would be a fun post that might make people think a little bit.
Problem is, I don't use any of those nifty advanced math skills anymore, so I just posted the "easy lie." You know: "statistics, damned statistics, and bad math posted in a pop-science forum."
I hope it still communicates to people the degree to which the size difference in the keys might matter. I actually thought another poster did a better job when s/he said something like "add another bit and you've just pushed Moore's law back 18 months."
Oh, and I should add (with respect to the impossibility claim) that 36,028,797,018,963,968 years is approximately 2.5 million times the estimated age of the universe.
So, while we might not want to use the word "impossible," I suppose we could say something like "unimaginably unlikely given current technologies."
Since a 163-bit key is 2^54 times more complex than a 109-bit key, and it took 4 years for the 109-bit key, aren't we looking at at least 4 * 2^53 years, not even figuring in the elliptical complexity (which I admit I would need to read up on)?
According to calc.exe, 4 * 2^53 years is 36,028,797,018,963,968 years. Anybody want to start working on that one?
Who cares if they change the shell? As long as they publish API's for the middleware pieces, how could we possibly complain?
> Microsoft probably won't put in any significant look and feel changes until much later in the testing program,
Just as a sidebar, this may not be true. As an MS partner, I got to see an early, build-of-the-day, pre-beta version of Windows XP, and all the look-and-feel changes were pretty much there already. When the RTM finally arrived, nothing in the look-and-feel area really surprised me, because it hadn't changed much from the early version I had seen.
What *had* changed, however, was the speed. In the early version, some of the menu rendering was so slow that the demonstrator would click on the start button and then talk to us a bit while the menu built. He said that a lot of optimizations are turned on fairly late in the development cycle. So maybe new GUI features are already built into Longhorn, but not running at optimized speeds yet.
For what it's worth, these screenshots look like doctored-up XP screenshots to me, but I'm no expert on this...
The Czech Republic recently became the first country in the world to pass a dark-sky law. From what I've heard, it works. The idea is simple enough: shine the light on the things you're trying to light up, and quit shining so much of it into the sky. Details are here (as well as lots of other places).
:o)
For those who think this doesn't matter, wouldn't it be swell if light polution became so pervasive that we couldn't see that next mass-extinction event meteor heading our way? The headlines read: Doomsday Meteor Arrives Unannounced. Subtitle reads: At least the few survivors had a well-lit view of the damage.