I do believe in space tourism, but unfortunately it is also going to be plagued with some very bad accidents. That alone will hurt its chances of success far more than your financing issues.
Your understanding of failure analysis is flawed. Instead, think about it in terms of probability. You can never get the probability of failure to zero, something can always go wrong. You build in contingencies and over design systems to improve their chances, but they're never zero.
You said: Reliability is of course the second question after price. However, take the Shuttle; it's extremely likely that both crashes are caused by design flaws in the Shuttle; and that the number of flaws that remain undiscovered will decrease over time. Therefore the reliability of the Shuttle should increase, and there's no known limit to how reliable launch vehicles can be.
What you neglect is that the more launches you have, the more often you test that reliability figure (MTBF if you're thinking conservatively, FITs if you're an optimist). The more tourists you send up, the more likely you will kill some of them. It will always be impossible to bring that probability to zero.
So you do your best. You design the best system you can, you analyze every failure and rework the system to keep that failure from happening again. But most of all, you prepare the public for the fact that there WILL be failures.
And when they happen, you move onward. Onward and Upward.
You have all the hardware that's doing the key checking / decrypting. Instead of reverse engineering the key, why not reverse engineer the box?
I seems to me that you have the machine doing the actual calculation on the signed code. Just stop the machine, slow it _way_ down, and watch it do its calculations.
This might necessitate blowing the lid off the chips and actually probing them directly... and this wouldn't be easy by any means... but it would probably be a lot easier than brute-forcing the key.
I heard on the radio this morning that a conservative estimate of the cost of the US attacking Iraq would be on the order of $300B. That makes the space station cost appear kinda paltry.
Imagine what we could do with a space program outfitted with $300 billion dollars (pinky snaps to attention at the corner of my mouth).
A slight modification to your alogirthm for removing commercials:
Instead of searching for black space, just keep a running set of hash values for each of the last 60 seconds or so. Each period of time would have a unique identifier.
Now, keep your TiVo (or freevo or whatever) watching stuff _all_ the time. (They pretty much do this already.) Build up a database of all your hash values (a hash table would be perfect for that.:-) ).
Compare your unique identifiers against your current programming - whenever one of those hash values comes up, you've got "repeated programming." There's only a few kinds of repeated programming - commercials and show intros are two of them. You could use statistical frequency to distinguish between the two. Or you could just give the user the option of seeing the repeated programming. If you save that repeated programming somewhere, you've just creating a sort of run-length encoding (more of a huffman than anything...).
The key to all of this is the hash encoding scheme. You need something that works well even considering how stations tune in differently... perhaps some kind of weighting system, rather than looking for exact matches.
And of course, if you've got one of these things running, and your neighbor on the internet does too... then you should trade them among your friends (with a good automated voting system to throw out kruft and misdirection).
Hope you implement it.:) Consider adding it to the Freevo project, or whichever other open source project you like.
The only problem with him reporting his own news is the underhanded way he did it. He should have just said, "on my site..." or at least not tried to hide his the site by using an IP address.
I suppose, to be fair, he might have changed his URL specifically for this article. But that doesn't seem very likely.
First, I'm currently using a databasing package that I came upon after almost giving up and writing my own. I'd tried scores of different software, but nothing could give me access to a lot of photos, deal with off-line storage easily, and give me powerful databasing/scripting tools useful for finding that important photo. It's called I-Match and can be found at Photools.
The software won't do much for preservation except for its good offline capabilities.
For that, I store photos in a simple system on my hardrive. I break up the directories into 650MB sections with subdirectories that are named with the date as in, "20010927_Wedding". That directory contains subdirectories named "processed", "web", and whatever else makes sense at the time, but the top level contains completely unmodified files in.jpg format.
Now I can copy these directories wholesale to CD-Rs.
Here's the first big point: media sizes always get bigger. So in a few years, I won't duplicate those CD-Rs (there's already two copies of everything right now). Instead, I'll copy them to DVD-R (after they have time to work their bugs/formats out). The DVD-Rs hold a few more CDs worth than the equivalent CD-Rs. A few years from then, my holo-bubble-memory or whatever will hold a few DVD-Rs worth. So in all, I should still be carrying around a relatively small collection of archival data disks despite the growing amount of data.
As for formats... I have a Canon G1 and it has a great RAW format. IMatch even has support for it. There isn't a chance in hell that I would store anything in it without also storing an equivalent high-quality.jpg. JPG is like text. It's going to be around for a long time to come simply because there is so much content in that format already.
From this discussion, however, I will modify my procedure slightly. I'm going to start including a tar'd, bzipped2 version of the xpm tools on each of the CDs.
As for prints, I have no desire for my prints of digital images (so far) to last. The archival copy is the copy on the CD-R, not the print. I _want_ the print to degrade over a few years - it'll keep things fresh. Just so long as I can get at the original data (and the processed version that was used to make that print). Of course... there are artifacts that the printing process introduces that can be considered part of the art....
Supply them all the graphics primitives, teach them how to make a two dimensional array (or fake one), and teach them the loop structure necessary to traverse and iterate the machine.
For extra credit, ask them to find a good solution to the boundary problem. Or ask them to come up with a new life primitive. Or just expose them to all the primitives that are out there.
Finally show them that silly book written by that bright guy on finite automata.
I'm working on an interface to do a VW RioVolt250 interface. When I'm done, the HU controls will run the RioVolt and the RioVolt's time display will be sent up to the HU. Unfortunately, the HU can't handle any kind of text, so ID3 is out... but it does have a couple of toggle switches - one for random and one for scan. I expect I'll set up random to do the normal function on the Riovolt. The "Scan" button will be modified slightly. Under normal situations, I'll have the forward/reverse track move forward and backwards in whatever you're currently playing. With Scan, I think I'll have it move forward and backwards through play lists (the RioVolt is capable of reading WinAmp created play lists). The Disk selector buttons are also an option for this.. but on the VW Head Unit there are only six buttons.
I wouldn't mind modifying the adapter for other radio/MP3 player combinations. It seems like the IPod would be ideal, but I think its only external interface is FireWire. A FireWire driver isn't something that can be put into your average tiny-little microcontroller without some serious work. Are there any other controls on the IPod? IR maybe?
What other headunits would be ideal to support? Is there any particular CD-Changer that works with _A Lot_ of headunits? My system emulates a Panasonic CD changer... but that's just because it's what I had around.
The other difficulty is getting the various automotive connectors needed. Anyone know of a good source for them?
Star Wars has always been about pushing the technological envelope as far as it will go. Sometimes it works. Sometimes they drop the ball. This has not changed since 1977. Look at any of the movies and you will find places where the special effects are truly great, and places where they stink on ice. Why all of a sudden this warrants another "George Lucas sucks" troll of a story is beyond me.
This brings up a good point that I wanted to say in one of the earlier articles on/.
The movie industry, or more specifically, the special effects industry, is a mirror of what technology has been available through the years. Lucas going back and changing the original movies destroys the excellent historical archive of effects techniques by muddying whether one effect is in the original or whether it's only in the re-release version. That information will probably get lost down the road and people will wonder about the technology of a particular period.
While we're on the topic, I'd like to pose a slightly off topic quesiton.
Can you actually give up your ability to sue? The Blockbuster contract that I decided not to sign last night actually said, "You agree not to participate in any class-action suits..." among other crap.
How about the same question but with limited production runs? How could you build say 20 to 100 plastic or metal (whatever is cheap) small electronic cases?
Every GPS on the market with a serial port outputs NMEA format messages (well, almost all, check before you buy). Those messages contain the current time in nice easy-to-read ANSI text.
Here's an example:
$GPGGA,162245.00,4842.6942,N,11342.9476,W,1,05,2 .4,2180.1,M,17.3,M,,*4E^M
The second field there is the time HHMMSS.SS. Here's a link to the NMEA protocol:
http://www.starlinkdgps.com/manuals/invictaproto co l.pdf
If you need more accuracy than that, you could get a GPS with a 1PPS output - the Motorola Oncore series comes to mind. Prices are pretty cheap, though not as cheap as your average consumer model. Tie the 1PPS output to a serial port's DTR line and write a small driver to read the value. Latency of that driver is the prime determinate of your accuracy at that point.
This is pretty neat, but reminds me of something from a few years ago.
I interviewed with Intel and during the interview they said in no uncertain terms that they were actively trying to keep people upgrading their systems, and hence keep the dollars rolling in. At the time, the interviewer said that the technique was largely by helping Micro$oft to keep new OSs coming that required more and more horsepower to run properly.
This is very cool in its own right (or could be, I haven't looked at it completely), but strikes me as another way they can push that curve...
Schneier always mentions that you have to watch out for people motivations. In this case, he should point out that his company makes its living watching for bugs/hacking/vulnerabilities in the systems of the customers that it monitors. He usually does this, but I definitely see it as fodder for Culp to throw back in his face. If the bugs were hidden, Counterpane would have a lot harder time knowing what to look for.
I really liked the point about software companies being liable for the software they produce. The implication from his article was that a firewall manufacturer isn't not liable if a hacker breaks in because of shoddy code in their firewall. Is this true? Anyone know of (or have a subscription to one of those cool legal services) any legal cases that have proved or disproved this?
Skepticism is good. Probably it's better to be occasionally blind to a really good idea than to be inundated with all the crackpot schemes. However, a well spoken person with a good idea should be heard. McMaster's is getting heard. I can't wait to see the outcome of it.
Newton was wrong about a lot of things. It's just that when you simplify reality to the frame of reference that we're living in right now, Newton was mostly right. Still, it's the corner cases that make reality interesting. Look at the junction diode.
I'd like to see Slashdot start a slashbox for each of the following organizations. The data for the slashbox can be taken directly from their home pages - they each have news headlines that a bit of PERL could pull out very easily. In addition, I'd like to see a section before the headlines labeled "top issues" that can be written by the organization themselves. It would be a space for a link to the issues they think most important.
These organizations stand for many of the things talked about on Slashdot. Those that feel strongly about any of the issues supported by one of these organizations should join that organization!
Slashdot Mirror
Try MythTV
You might think it's overkill, but consider it because it can combine lots of separate functions together....
Not just a PVR, but an MP3 jukebox, video server, web/weather/informaiton services and so on...
The Case for Mars is a great read... and now much (all?) of it can be downloaded for free from here:
http://www.nw.net/mars/
Also, check out the Mars society.
I do believe in space tourism, but unfortunately it is also going to be plagued with some very bad accidents. That alone will hurt its chances of success far more than your financing issues.
Your understanding of failure analysis is flawed. Instead, think about it in terms of probability. You can never get the probability of failure to zero, something can always go wrong. You build in contingencies and over design systems to improve their chances, but they're never zero.
You said: Reliability is of course the second question after price. However, take the Shuttle; it's extremely likely that both crashes are caused by design flaws in the Shuttle; and that the number of flaws that remain undiscovered will decrease over time. Therefore the reliability of the Shuttle should increase, and there's no known limit to how reliable launch vehicles can be.
What you neglect is that the more launches you have, the more often you test that reliability figure (MTBF if you're thinking conservatively, FITs if you're an optimist). The more tourists you send up, the more likely you will kill some of them. It will always be impossible to bring that probability to zero.
So you do your best. You design the best system you can, you analyze every failure and rework the system to keep that failure from happening again. But most of all, you prepare the public for the fact that there WILL be failures.
And when they happen, you move onward. Onward and Upward.
You have all the hardware that's doing the key checking / decrypting. Instead of reverse engineering the key, why not reverse engineer the box?
I seems to me that you have the machine doing the actual calculation on the signed code. Just stop the machine, slow it _way_ down, and watch it do its calculations.
This might necessitate blowing the lid off the chips and actually probing them directly... and this wouldn't be easy by any means... but it would probably be a lot easier than brute-forcing the key.
Where's Bruce Schneir when you need him?
As soon as we can simulate life (any life) completely, we have no way of proving we are not a simulation being run by a higher life form.
Basically it doesn't matter if we are a simulation or not. It all comes back to that simple assertion, I think, therefore I am.
If you can think, then you exist. Since you exist, you better damn well make the most of it.
I heard on the radio this morning that a conservative estimate of the cost of the US attacking Iraq would be on the order of $300B. That makes the space station cost appear kinda paltry.
Imagine what we could do with a space program outfitted with $300 billion dollars (pinky snaps to attention at the corner of my mouth).
Mu-hahahahaha....
In general, how do you find the
Should mozilla have a mod to show all these links... or is there one already?
R.I.P. D. Adams. And thanks.
I'd hate to be in a sailboat, relaxing somewhere in the middle of the Pacific when one of these went by.
Or imagine being the poor sucker in the crow's nest looking for land or reattaching a sail.
And I bet they say it's the sailor's fault for being in a shipping lane.
A slight modification to your alogirthm for removing commercials:
Instead of searching for black space, just keep a running set of hash values for each of the last 60 seconds or so. Each period of time would have a unique identifier.
Now, keep your TiVo (or freevo or whatever) watching stuff _all_ the time. (They pretty much do this already.) Build up a database of all your hash values (a hash table would be perfect for that.
Compare your unique identifiers against your current programming - whenever one of those hash values comes up, you've got "repeated programming." There's only a few kinds of repeated programming - commercials and show intros are two of them. You could use statistical frequency to distinguish between the two. Or you could just give the user the option of seeing the repeated programming. If you save that repeated programming somewhere, you've just creating a sort of run-length encoding (more of a huffman than anything...).
The key to all of this is the hash encoding scheme. You need something that works well even considering how stations tune in differently... perhaps some kind of weighting system, rather than looking for exact matches.
And of course, if you've got one of these things running, and your neighbor on the internet does too... then you should trade them among your friends (with a good automated voting system to throw out kruft and misdirection).
Hope you implement it.
(Can a slashdot post be considered prior art?)
Sorry for the (very soon to be moderated down) simple and annoying reply, but...
:-)
Kick ass!
The only problem with him reporting his own news is the underhanded way he did it. He should have just said, "on my site..." or at least not tried to hide his the site by using an IP address.
I suppose, to be fair, he might have changed his URL specifically for this article. But that doesn't seem very likely.
First, I'm currently using a databasing package that I came upon after almost giving up and writing my own. I'd tried scores of different software, but nothing could give me access to a lot of photos, deal with off-line storage easily, and give me powerful databasing/scripting tools useful for finding that important photo. It's called I-Match and can be found at Photools.
.jpg format.
.jpg. JPG is like text. It's going to be around for a long time to come simply because there is so much content in that format already.
The software won't do much for preservation except for its good offline capabilities.
For that, I store photos in a simple system on my hardrive. I break up the directories into 650MB sections with subdirectories that are named with the date as in, "20010927_Wedding". That directory contains subdirectories named "processed", "web", and whatever else makes sense at the time, but the top level contains completely unmodified files in
Now I can copy these directories wholesale to CD-Rs.
Here's the first big point: media sizes always get bigger. So in a few years, I won't duplicate those CD-Rs (there's already two copies of everything right now). Instead, I'll copy them to DVD-R (after they have time to work their bugs/formats out). The DVD-Rs hold a few more CDs worth than the equivalent CD-Rs. A few years from then, my holo-bubble-memory or whatever will hold a few DVD-Rs worth. So in all, I should still be carrying around a relatively small collection of archival data disks despite the growing amount of data.
As for formats... I have a Canon G1 and it has a great RAW format. IMatch even has support for it. There isn't a chance in hell that I would store anything in it without also storing an equivalent high-quality
From this discussion, however, I will modify my procedure slightly. I'm going to start including a tar'd, bzipped2 version of the xpm tools on each of the CDs.
As for prints, I have no desire for my prints of digital images (so far) to last. The archival copy is the copy on the CD-R, not the print. I _want_ the print to degrade over a few years - it'll keep things fresh. Just so long as I can get at the original data (and the processed version that was used to make that print). Of course... there are artifacts that the printing process introduces that can be considered part of the art....
Have your students program the game of life.
Supply them all the graphics primitives, teach them how to make a two dimensional array (or fake one), and teach them the loop structure necessary to traverse and iterate the machine.
For extra credit, ask them to find a good solution to the boundary problem. Or ask them to come up with a new life primitive. Or just expose them to all the primitives that are out there.
Finally show them that silly book written by that bright guy on finite automata.
I'm working on an interface to do a VW RioVolt250 interface. When I'm done, the HU controls will run the RioVolt and the RioVolt's time display will be sent up to the HU. Unfortunately, the HU can't handle any kind of text, so ID3 is out... but it does have a couple of toggle switches - one for random and one for scan. I expect I'll set up random to do the normal function on the Riovolt. The "Scan" button will be modified slightly. Under normal situations, I'll have the forward/reverse track move forward and backwards in whatever you're currently playing. With Scan, I think I'll have it move forward and backwards through play lists (the RioVolt is capable of reading WinAmp created play lists). The Disk selector buttons are also an option for this.. but on the VW Head Unit there are only six buttons.
I wouldn't mind modifying the adapter for other radio/MP3 player combinations. It seems like the IPod would be ideal, but I think its only external interface is FireWire. A FireWire driver isn't something that can be put into your average tiny-little microcontroller without some serious work. Are there any other controls on the IPod? IR maybe?
What other headunits would be ideal to support? Is there any particular CD-Changer that works with _A Lot_ of headunits? My system emulates a Panasonic CD changer... but that's just because it's what I had around.
The other difficulty is getting the various automotive connectors needed. Anyone know of a good source for them?
I don't have any information up yet, but look for updates about this project at:
http://orn.madison-expat.net/tech/hu_interface/
Rudy
Star Wars has always been about pushing the technological envelope as far as it will go. Sometimes it works. Sometimes they drop the ball. This has not changed since 1977. Look at any of the movies and you will find places where the special effects are truly great, and places where they stink on ice. Why all of a sudden this warrants another "George Lucas sucks" troll of a story is beyond me.
This brings up a good point that I wanted to say in one of the earlier articles on /.
The movie industry, or more specifically, the special effects industry, is a mirror of what technology has been available through the years. Lucas going back and changing the original movies destroys the excellent historical archive of effects techniques by muddying whether one effect is in the original or whether it's only in the re-release version. That information will probably get lost down the road and people will wonder about the technology of a particular period.
While we're on the topic, I'd like to pose a slightly off topic quesiton.
Can you actually give up your ability to sue? The Blockbuster contract that I decided not to sign last night actually said, "You agree not to participate in any class-action suits..." among other crap.
Better to read about it from the horse's mouth: JPL. Here's a link to the press release about this.
r el eases/20020528a.html
http://mars.jpl.nasa.gov/odyssey/newsroom/press
Orn
How about the same question but with limited production runs? How could you build say 20 to 100 plastic or metal (whatever is cheap) small electronic cases?
The small time inventor is back.
Rudy
How much accuracy do you need?
Every GPS on the market with a serial port outputs NMEA format messages (well, almost all, check before you buy). Those messages contain the current time in nice easy-to-read ANSI text.
Here's an example:
$GPGGA,162245.00,4842.6942,N,11342.9476,W,1,05,
The second field there is the time HHMMSS.SS. Here's a link to the NMEA protocol:
http://www.starlinkdgps.com/manuals/invictaprot
If you need more accuracy than that, you could get a GPS with a 1PPS output - the Motorola Oncore series comes to mind. Prices are pretty cheap, though not as cheap as your average consumer model. Tie the 1PPS output to a serial port's DTR line and write a small driver to read the value. Latency of that driver is the prime determinate of your accuracy at that point.
A slight update to the post I made a bit ago. Here's an online petition:
http://www.gopetition.com/info.php?petid=536
And here's the nickelodeon "write to nick" website:
http://www.nick.com/blab/write_nick/
But the best would be to write a pen and paper letter to them. Here's a snail mail address:
Nickelodeon / MTV Networks
1515 Broadway
New York, NY 10036
Rudy
This is pretty neat, but reminds me of something from a few years ago.
I interviewed with Intel and during the interview they said in no uncertain terms that they were actively trying to keep people upgrading their systems, and hence keep the dollars rolling in. At the time, the interviewer said that the technique was largely by helping Micro$oft to keep new OSs coming that required more and more horsepower to run properly.
This is very cool in its own right (or could be, I haven't looked at it completely), but strikes me as another way they can push that curve...
Great article, however...
Schneier always mentions that you have to watch out for people motivations. In this case, he should point out that his company makes its living watching for bugs/hacking/vulnerabilities in the systems of the customers that it monitors. He usually does this, but I definitely see it as fodder for Culp to throw back in his face. If the bugs were hidden, Counterpane would have a lot harder time knowing what to look for.
I really liked the point about software companies being liable for the software they produce. The implication from his article was that a firewall manufacturer isn't not liable if a hacker breaks in because of shoddy code in their firewall. Is this true? Anyone know of (or have a subscription to one of those cool legal services) any legal cases that have proved or disproved this?
It seems pretty fundamental.
Rudy
Skepticism is good. Probably it's better to be occasionally blind to a really good idea than to be inundated with all the crackpot schemes. However, a well spoken person with a good idea should be heard. McMaster's is getting heard. I can't wait to see the outcome of it.
Newton was wrong about a lot of things. It's just that when you simplify reality to the frame of reference that we're living in right now, Newton was mostly right. Still, it's the corner cases that make reality interesting. Look at the junction diode.
Rudy
I'd like to see Slashdot start a slashbox for each of the following organizations. The data for the slashbox can be taken directly from their home pages - they each have news headlines that a bit of PERL could pull out very easily. In addition, I'd like to see a section before the headlines labeled "top issues" that can be written by the organization themselves. It would be a space for a link to the issues they think most important.
CDT
ALCU
EFF
These organizations stand for many of the things talked about on Slashdot. Those that feel strongly about any of the issues supported by one of these organizations should join that organization!
Rudy