That bug was so difficult to deal with most of the time that a lot of my papers wound up being numbered by hand either on the computer or with a pen once I printed it.
Trusted Computing, as theorized in the article, is no more secure than current systems are. It just breeds more blind faith, as a remote machine can only be trusted to verify itself as much as the remote machine was trusted in the first place.
Nearly every example relied on "remote attestation" of the code, which I interpreted as something along the lines of checking the official program's own hash against the one supplied by the remote machine (it's impractical to have the code transmitted to the authenticating machine unless worldwide broadband access becomes pervasive very quickly). While it would be nontrivial to do so, it would not be impossible to maliciously modify a program so that it appears legitimate but does not behave so.
The only possible benefit of TCPA and similar schemes is the ability to offload heavy cryptography to a dedicated chip, and that is rendered moot whenever a flaw is found in its cryptographic algorithm(s) or the chip itself. On the whole, TCPA et al. are just as toothless as the Pentium III serial number issue was.
The point of all this is to get rid of as much arbitrarity as possible, so that the next time something big gets discovered we can look at whatever the rule winds up being and quickly see what to call it.
I still don't believe Microsoft is any more of a monopoly than Intel, but Intel knows how to play the game.
Not really. It's just that Intel got their antitrust suit settled before anyone off Wall St. cared. Why do you think AMD and Intel's chips interoperate so well? The settlement mandates compulsory licensability of each other's x86 (and derivatives like x86-64) -related IP.
Last time I saw a story on this, it was mentioned that the EU can fine up to 10% of MS's worldwide revenue. I wouldn't be surprised if that's significantly less than their profit margin, so even a settlement to the fullest extent of EU law wouldn't dissuade them, unless they were barred completely from the EU member states (and that would wind up screwing the users a lot more than it would screw Redmond).
Well, unless it's going to be funded entirely on grants, donations, and government money, it needs to become economically viable early on. Noone's been able to both come up with a way to do that yet, or if they have, they haven't found any venture capitalists willing to help with it.
The parent's suggestions were to get space economics going. Hence, I based it around the value of things on earth. You are quite right about the value of them in space. However, anything on a body of any significant size, it would be easier to use the indigenous materials wherever possible, and only use offworld materials when absolutely necessary.
Most of your examples can be readily made from either indigenous materials (moon bases, moon telescopes (both optical and radio), mars bases) or be made from small asteroids (Lagrange point bases, large interplanetary ships), so there is little need to mine the asteroids for anything not already ludicrously valuable (such as the aforementioned Helium-3 for clean fusion) both on Earth and in space.
Most of these are chicken-and-egg problems, but throw enough VC money at them and they'll solve themselves. There is however one point that I take issue with:
1. Mining: There are asteroids out there that are nearly entirely composed of precious metals. These would fetch quite a price on the market. The less valuable materials (e.g. water, carbon, hydrogen, iron, etc.) all are very valuable for perpetuating the space economy.
This would only be feasible for incredibly valuable substances (on the order of at least 2-3x the value of platinum). The only thing that would be economically viable right now (or rather in the perpetual '20 years from now' that it would be useful and therefore valuable) is Helium-3.
Whether we have self-sufficient colonies offworld wouldn't matter. If you can nuke anywhere in the world in 15 minutes and the other guys can too, they sure as hell can take out your emergency one-way direct-to-colony shuttle before it gets off the ground or while it's on its way to orbit. And even if it was something more insidious (like dirty bombs or neutron bombs) there's simply not the launch capacity to get very many people off with anything resembling rapidity.
For a few years, mp3.com did this (between when they settled the lawsuits from the major labels and when they got sold to CNet). Granted, a lot of the music on there was pretty shitty, but it was all artist-submitted or label-submitted (in the case of promotional tracks for Madonna and Linkin Park, among others).
Their trance charts were a good indicator of how good an artist was, however, and I got a lot of good, legal trance off of there. I even went so far as to buy a couple CDs from one such artist because I liked their music so well (specifically, Astral Projection's Another World and SFX).
...except edonkey/emule relies on 3rd-party-run servers. Also, (last time i used it at least (a year or so ago)) swarming is quite limited, and nowhere near as positive-feedback-regulated as bittorrent.
I don't even have my own car yet, and the day I got my driver's license, my parents' insurance doubled. And that's with me going to school 700 miles away.
It seems to me that in a car with as many computers as the Prius it shouldn't be that difficult to have a simple driver preferences menu that covers things like the level of detail the driver wants for problems. If it already has driver settings (powered seat positions, radio presets, etc.), it shouldn't be difficult to add this on.
Slashdot Mirror
That bug was so difficult to deal with most of the time that a lot of my papers wound up being numbered by hand either on the computer or with a pen once I printed it.
Trusted Computing, as theorized in the article, is no more secure than current systems are. It just breeds more blind faith, as a remote machine can only be trusted to verify itself as much as the remote machine was trusted in the first place.
Nearly every example relied on "remote attestation" of the code, which I interpreted as something along the lines of checking the official program's own hash against the one supplied by the remote machine (it's impractical to have the code transmitted to the authenticating machine unless worldwide broadband access becomes pervasive very quickly). While it would be nontrivial to do so, it would not be impossible to maliciously modify a program so that it appears legitimate but does not behave so.
The only possible benefit of TCPA and similar schemes is the ability to offload heavy cryptography to a dedicated chip, and that is rendered moot whenever a flaw is found in its cryptographic algorithm(s) or the chip itself. On the whole, TCPA et al. are just as toothless as the Pentium III serial number issue was.
To paraphrase some Congressman, a hundred signs here, a hundred signs there...pretty soon you're talking about real evidence for water on Mars.
I'll bite.
The point of all this is to get rid of as much arbitrarity as possible, so that the next time something big gets discovered we can look at whatever the rule winds up being and quickly see what to call it.
Not really. It's just that Intel got their antitrust suit settled before anyone off Wall St. cared. Why do you think AMD and Intel's chips interoperate so well? The settlement mandates compulsory licensability of each other's x86 (and derivatives like x86-64) -related IP.
Last time I saw a story on this, it was mentioned that the EU can fine up to 10% of MS's worldwide revenue. I wouldn't be surprised if that's significantly less than their profit margin, so even a settlement to the fullest extent of EU law wouldn't dissuade them, unless they were barred completely from the EU member states (and that would wind up screwing the users a lot more than it would screw Redmond).
Well, unless it's going to be funded entirely on grants, donations, and government money, it needs to become economically viable early on. Noone's been able to both come up with a way to do that yet, or if they have, they haven't found any venture capitalists willing to help with it.
The parent's suggestions were to get space economics going. Hence, I based it around the value of things on earth. You are quite right about the value of them in space. However, anything on a body of any significant size, it would be easier to use the indigenous materials wherever possible, and only use offworld materials when absolutely necessary.
Most of your examples can be readily made from either indigenous materials (moon bases, moon telescopes (both optical and radio), mars bases) or be made from small asteroids (Lagrange point bases, large interplanetary ships), so there is little need to mine the asteroids for anything not already ludicrously valuable (such as the aforementioned Helium-3 for clean fusion) both on Earth and in space.
Most of these are chicken-and-egg problems, but throw enough VC money at them and they'll solve themselves. There is however one point that I take issue with:
This would only be feasible for incredibly valuable substances (on the order of at least 2-3x the value of platinum). The only thing that would be economically viable right now (or rather in the perpetual '20 years from now' that it would be useful and therefore valuable) is Helium-3.
Whether we have self-sufficient colonies offworld wouldn't matter. If you can nuke anywhere in the world in 15 minutes and the other guys can too, they sure as hell can take out your emergency one-way direct-to-colony shuttle before it gets off the ground or while it's on its way to orbit. And even if it was something more insidious (like dirty bombs or neutron bombs) there's simply not the launch capacity to get very many people off with anything resembling rapidity.
one connection...one network...one OS...
Ein Volk, ein Reich, ein Fuhrer....
coincidence?
For a few years, mp3.com did this (between when they settled the lawsuits from the major labels and when they got sold to CNet). Granted, a lot of the music on there was pretty shitty, but it was all artist-submitted or label-submitted (in the case of promotional tracks for Madonna and Linkin Park, among others).
Their trance charts were a good indicator of how good an artist was, however, and I got a lot of good, legal trance off of there. I even went so far as to buy a couple CDs from one such artist because I liked their music so well (specifically, Astral Projection's Another World and SFX).
Which, in the world's most common browser, shows up with the same error page.
Actually, if you really want to be pedantic, it'd be Russia or the scandinavian countries.
I'm pretty sure colleges would love that ability too, given how much my dorm uses bittorrent.
...except edonkey/emule relies on 3rd-party-run servers. Also, (last time i used it at least (a year or so ago)) swarming is quite limited, and nowhere near as positive-feedback-regulated as bittorrent.
Gotten McDonald's hotcakes recently? Last I checked, they were still served on styrofoam.
double cupping sounds wasteful, but at least the paper can biodegrade.
Paper can be recycled, too.
I don't even have my own car yet, and the day I got my driver's license, my parents' insurance doubled. And that's with me going to school 700 miles away.
This stuff is already computer controlled, and if a $30 router can have a web administration interface, there's no reason a car couldn't either.
Yes there is, but it's not insurmountable: noone's bothered to port Webmin to TRON yet.
Cars.com claims the Z06 starts at $50k.
First-generation RX-7's were actually carbureted, according to a mention here
It seems to me that in a car with as many computers as the Prius it shouldn't be that difficult to have a simple driver preferences menu that covers things like the level of detail the driver wants for problems. If it already has driver settings (powered seat positions, radio presets, etc.), it shouldn't be difficult to add this on.
...assuming intel figures out where they messed up with Prescotts and don't make the same mistakes with Dothan.
No, the worst (albeit probably unlikely) thing she could do is bury you in minority shareholder lawsuits.