if there was SCO code in the linux kernel, there would be a chance to get license fees from a linux user, unless she upgrades to a kernel which doesnt contain SCO code anymore...
If you really would try to build such a crystal, vacanies could very well be the problem. As you said, there is an equilibrium value of defects in any crystal. This equilibrium value is temperature dependant with a exp(-Eform/kT) law, where Eform is the formation enthalpy. High temperature means high rate of defects.
Si single crystals are usually prepared at very high temperatures out of molten Silicon (1414C, Czochralsky method). Essentially, this will lead to a freezing of the defect structure at temperatures close to the melting point, because the lattice reorientation kinetics (point diffusion) also are thermally activated.
You would have to temper the crystal for _very_ long times at temperatures of i.e. 300C to get a thermal equilibrium of defects at this temperature. These times could be >>years !
multicast is not really the answer for traffic spikes as they occured on 9/11. multicasting could be handy if the whole internet paradigm changed from pulling information to a tv-like pushing scheme, losing one of the internet's huge strengths, async distribution of information. remember, one-to-many distribution can only work if the whole audience listens at the same time...
smart caching algorithms could be an answer, assuming the page lifespan prediction of the caches gets a lot better. with an event like 9/11, nobody really wants outdated information, even if it's only 10 minutes old...
>> 10^6 GigaByte = 10^9 Megabyte = 10^15 Byte. > So the hard-disk manufacturers will tell you, > but it's not actually true. > > 1Kb = 2^10 bytes = 1024 bytes
not the hdd manufacturers, but the SI tells me "kilo" is a factor of 1000. there actually had been a SI proposal once, associating the factor 2^10=1024 with the prefix "kibi" - for "kilo binary". so 1 kibibyte would be 1.024 kilobyte.
Just take an encryption tool with a good number of regular users, make it closed source and implement a "government backdoor". Then publish a new version, preferably with a nice GUI and some new encryption methods which are not backwards compatible and let your marketing department do the rest. Sounds familiar ?
Honestly, I can't imagine that AOL and Earthlink won't install any special surveillance equipment in their network. Searching for keywords in >>100 millions of emails takes a little more than a workstation with grep...
It is a well known vulnerability that passwords can be reconstructed out of key timings - but this issue can be avoided easily, i.e. with terminal applications sending the whole password at once.
... And even with chemical extinguishers, it can be difficult. Magnesium cannot only reduce water, but also carbon dioxide. So if you are trying to use a CO2-extinguisher, the Mg fire will crackle and sputter and produce a lot of carbon. A very cheap way to produce coal and fullerenes, though.
The funny thing is: all you're talking about is damaged hifi equipment. Listening to very high amplitudes of sound, such as created by digital code, with headphones or - even worse - with earphones could easily damage your ears. And then ? Sue Sony/whoever for a pair of new ones ???
According to german computer magazine c't (11/2001), about any cordless device can be sniffed - not only logitech. They had an article in the last issue discussing exactly this. Maybe I'll translate it a little later...
Just a summary, now: Cordless devices tend to use a 8 or 16 bit key for identifying (and authenticating) the connection to the base station. So all you need to sniff the keystrokes is another receiver, this code and something actually logging the characters... (i.e. keyghost.)
~$ traceroute pacemaker.george.heart.com
traceroute to pacemaker.george.heart.com (1.2.3.4), 30 hops max, 38 byte packets
1 my-isp.somewhere.de (3.4.5.6) 3.611 ms 2.279 ms 2.033 ms
[...]
8 george-isp.somewhere.com (5.6.7.8) 5.281 ms 4.522 ms 4.147 ms
9 phone.george.heart.com (1.2.3.7) 108.742 ms 105.656 ms 376.795 ms
10 living-room.george.heart.com (1.2.3.6) 180.167 ms 168.235 ms 186.593 ms
11 toilet.george.heart.com (1.2.3.5) 167.976 ms 168.376 ms 168.926 ms
12 heart.george.heart.com (1.2.3.11) 180.141 ms 169.184 ms 168.880 ms
13 pacemaker.george.heart.com (1.2.3.4) 170.063 ms 170.369 ms 169.367 ms
Slashdot Mirror
.. so we'll finally get people watching DivX while doing 90mph on the freeway.
sounds more like blackmail...
if there was SCO code in the linux kernel, there would be a chance to get license fees from a linux user, unless she upgrades to a kernel which doesnt contain SCO code anymore...
..with this is that they finally have to tell WHAT the license would be good for.
I thought Microsoft has bought a license from SCO ?
aren't there 1gb compact flash microdrives for quite a long time ? what's so very new with these things ?
If you really would try to build such a crystal, vacanies could very well be the problem. As you said, there is an equilibrium value of defects in any crystal. This equilibrium value is temperature dependant with a exp(-Eform/kT) law, where Eform is the formation enthalpy. High temperature means high rate of defects.
Si single crystals are usually prepared at very high temperatures out of molten Silicon (1414C, Czochralsky method). Essentially, this will lead to a freezing of the defect structure at temperatures close to the melting point, because the lattice reorientation kinetics (point diffusion) also are thermally activated.
You would have to temper the crystal for _very_ long times at temperatures of i.e. 300C to get a thermal equilibrium of defects at this temperature. These times could be >>years !
multicast is not really the answer for traffic spikes as they occured on 9/11. multicasting could be handy if the whole internet paradigm changed from pulling information to a tv-like pushing scheme, losing one of the internet's huge strengths, async distribution of information. remember, one-to-many distribution can only work if the whole audience listens at the same time...
smart caching algorithms could be an answer, assuming the page lifespan prediction of the caches gets a lot better. with an event like 9/11, nobody really wants outdated information, even if it's only 10 minutes old...
>> 10^6 GigaByte = 10^9 Megabyte = 10^15 Byte.
> So the hard-disk manufacturers will tell you,
> but it's not actually true.
>
> 1Kb = 2^10 bytes = 1024 bytes
not the hdd manufacturers, but the SI tells me "kilo" is a factor of 1000. there actually had been a SI proposal once, associating the factor 2^10=1024 with the prefix "kibi" - for "kilo binary". so 1 kibibyte would be 1.024 kilobyte.
nope. 1 Petabyte = 1000 TeraByte = 10^6 GigaByte = 10^9 Megabyte = 10^15 Byte.
ahem... isn't Boeing the company losing huge market share to the European Airbus group ?
just my $.02
Hey, but there already are Eric Raymond's "Sex Tips for Geeks". Written by the uber-geek Eric S. Raymond himself... :)
... I'd like wearing a kyli_e_ t-shirt even more
scnr.
...does it finally cook coffee or fix my breakfast ?
Just take an encryption tool with a good number of regular users, make it closed source and implement a "government backdoor". Then publish a new version, preferably with a nice GUI and some new encryption methods which are not backwards compatible and let your marketing department do the rest. Sounds familiar ?
Honestly, I can't imagine that AOL and Earthlink won't install any special surveillance equipment in their network. Searching for keywords in >>100 millions of emails takes a little more than a workstation with grep...
I guess 'p' stands for 'personal', so it would be a _p_ersonal _v_ideo _r_ecorder ???
It is a well known vulnerability that passwords can be reconstructed out of key timings - but this issue can be avoided easily, i.e. with terminal applications sending the whole password at once.
... And even with chemical extinguishers, it can be difficult. Magnesium cannot only reduce water, but also carbon dioxide. So if you are trying to use a CO2-extinguisher, the Mg fire will crackle and sputter and produce a lot of carbon. A very cheap way to produce coal and fullerenes, though.
Hey, thats a really great show. I always loved the British humour... Try watching it, hope the servers aren't /.ed soon...
The funny thing is: all you're talking about is damaged hifi equipment. Listening to very high amplitudes of sound, such as created by digital code, with headphones or - even worse - with earphones could easily damage your ears. And then ? Sue Sony/whoever for a pair of new ones ???
fuckedcompany /.ed... pud 1 cmdr 1
According to german computer magazine c't (11/2001), about any cordless device can be sniffed - not only logitech. They had an article in the last issue discussing exactly this. Maybe I'll translate it a little later...
Just a summary, now: Cordless devices tend to use a 8 or 16 bit key for identifying (and authenticating) the connection to the base station. So all you need to sniff the keystrokes is another receiver, this code and something actually logging the characters... (i.e. keyghost.)
...imagine what could be done with traceroute:
~$ traceroute pacemaker.george.heart.com
traceroute to pacemaker.george.heart.com (1.2.3.4), 30 hops max, 38 byte packets
1 my-isp.somewhere.de (3.4.5.6) 3.611 ms 2.279 ms 2.033 ms
[...]
8 george-isp.somewhere.com (5.6.7.8) 5.281 ms 4.522 ms 4.147 ms
9 phone.george.heart.com (1.2.3.7) 108.742 ms 105.656 ms 376.795 ms
10 living-room.george.heart.com (1.2.3.6) 180.167 ms 168.235 ms 186.593 ms
11 toilet.george.heart.com (1.2.3.5) 167.976 ms 168.376 ms 168.926 ms
12 heart.george.heart.com (1.2.3.11) 180.141 ms 169.184 ms 168.880 ms
13 pacemaker.george.heart.com (1.2.3.4) 170.063 ms 170.369 ms 169.367 ms
...too bad noone took an infrared spectrum of the flames. Otherwise, very intense absorption at ~3000-3200cm^-1 would have revealed it all... :)