(Disclaimer: IANAL) From the GPL... b) You must cause any work that you distribute or
publish, that in whole or in part contains or is
derived from the Program or any part thereof, to
be licensed as a whole at no charge to all third
parties under the terms of this License. If the terms of another license prohibit you from doing that, it is not GPL compatible.
Of course, the GPL is a copyright license, and does not grant any trademark license. Therefore, an alternate, GPL compatible approach would be to license the software under the exact terms of the GPL, and license the trademark under some other terms(like Mozilla does).
I doubt the vendor would give you a replacement key if they saw the opportunity to make money. As for the legality of using another key on software you own... (Disclaimer IANAL, no warranties, consult your own counsel before doing anything) but... 1) AFAIK, few democratic countries make sites illegal, only specific content. The keys found on these sites are often generated data according to a formula, so the copyright belongs to the site maintainer. It is not even functional, and could hardly even be called a copyright circumvention device. I don't think this would be illegal in many countries at all. In many states(of the world), there is no law against using software which has been lawfully obtained. So entering a key other than as supplied by the vendor into software that you have the right to posses is perfectly lawful.
1) You are under no obligation to let the BSA in. If they knock down your door, complain to the Police that they vandalised your door, broke an entering, and trespassed. 2) They have no right to steal your computer. If they take your computer without a court order(which they won't be able to get if your use is legal), then make a complaint to the police that they stole your computer. Organised crime is taken seriously in most countries, and if they really commit crimes like this, they will soon be declared an Illegal Organisation or the equivalent in your country by the courts.
Except that scheduling the recall is a negative externality resulting from Microsoft's production process, and the cost of that negative externality is not internalised to Microsoft, but rather is met by the customer.
Therefore, it is unethical for Microsoft to not reduce the negative externalities that result from its own poor workmanship in every way it possibly can at no significant cost to itself. The more time people have to prepare for the release of the patch, the better. Therefore, if Microsoft is capable of disclosing an upcoming patch to a certain subset of the population, they have proved that they are capable of disclosing this information publically.
However, having proven that they could give the public more warning, they chose not to. This is unethical.
In practice, Microsoft will probably release the warning to corporate customers at a certain point in time. They will then feel obliged to leave a gap before they release to the general public, otherwise the advantage of the premium service will be diminished. After this, they will feel obliged to give time to ensure everyone has got the warning before they release the patch. Therefore, for the sake of profit, the time until the patch is released is probably also increased. This is also unethical.
The article is not related. Everyone still can call "999"(I assume that is the UK emergency number), but if there is a non-life-threatening situation, they can voluntarily elect(and are encouraged to do so) to call another number. This is very different from selling a good response time, and prioritising another call on the sole basis of how much money you received.
Intel predicts that it will be able to convince the world to abandon the unscalable approach of following standards, including upcoming standards like XForms and IPv6, and open P2P systems, and instead invent its own propietary system.
Wouldn't it be better to spread it out over a long period of time, otherwise they will know it is an organised compaign and will just disregard the calls? Is there any advantage to do this all in one day whatsoever?
From the article... > In this last step, all of the initially queried > nodes percolate the query throughout the network > so that the query is guaranteed to reach a core > sub-network of highly-connected nodes. "Since a > copy of the query is in one of the nodes in the > core network, and since the content list of a > node is cached at one of these high-degree > nodes, one is guaranteed to find the content as > long as at least one node in the network has > it," said Roychowdhury. So in other words, the "major sharers", i.e. nodes which are "high degree", i.e. have a lot of connections, form the "core network", and collectively host the entire index. However, this is starting to lose the advantages of being a peer-to-peer network. Obviously, you can't have it both ways.
for (int i = 0; i < nzones; i++) { Zone* z = zones[i]; z->read_sensors();
if (z->is_public_allowed) continue;
if (!z->detected_movement) { z->state = STATE_VACANT; continue; }
bool allowed = false;
for (int j = 0; j < z->ndetected_rfids; j++) if (is_authorised(z, z->detected_rfids[i], CurrentTime)) { allowed = true; break; } if (allowed) continue;
switch (z->state) { case STATE_VACANT: // They just entered the zone... z->speaker->PlayLoop("big_dog_bark.mp3"); &nbs p; z->state = STATE_DOG_PLAY; z->state_start = CurrentTime; break; case STATE_DOG_PLAY: if (CurrentTime - z->state_start < 20) continue; z->speaker->PlayLoop("death_threat.mp3"); &nbs p; z->state = STATE_THREAT_PLAY; z->state_start = CurrentTime; break; case STATE_THREAT_PLAY: if (CurrentTime - z->state_start < 10) continue; // TODO: Detect co-ordinates of intruder // and point and turn on machine gun. } }
The doomsday threat doesn't work if the attacking party thinks they can take down the entire nuclear capacity of the state before the victim can respond. However, given that North Korea alreay claims to have nukes, demonstrating that they do does not make them any more likely to come under attack.
However, if this is a first weapons test, then they probably don't have very well developed nuclear weapons yet, and some governments might take a "its now or never" approach.
There is no indication in the article that any radioactivity was detected despite the fact this happened on Thursday, which probably means it was a chemical explosive. So it could easily be a larger version of the train explosion earlier this year, either accidental or an act of terrorism or government suppression. If they didn't take safety seriously once, it is likely they routinely move potentially explosive compounds without adequate safety protections.
Ignoring the fact that calculating the time an algorithm takes to run is not the same as running the algorithm, a note about matrix multiplication times...
The most obvious iterative algorithm(i.e. direct from the definition) to multiply two nxn matrices would take theta(n^3). Strassen's divide-and-conquer algorithm, which divides the matrices into 4 submatrices, and does 3 recursive matrix multiplications, takes theta(n^log2(7)) = theta(n^2.81...)
The best known algorithm in terms of asymptotic time is theta(n^2.376), but it does not surpass Strassen's algorithm until n is very huge, so Strassen's is the most practical algorithm known.
It is not Star38's fault, it is the fault of the Telcos who made it possible by switch misconfiguration. However, if their telecoms provider decided to reconfigure the switch properly to block spoofing, and their service suddenly didn't work:), I would have much sympathy for Star38 either. I just hope that this doesn't affect the right of the teleco to fix their service(perhaps "but if we reconfigure our switch, we will get sued"?).
The CallingPartyNumber IE is often recorded and used for this purpose. Q.931 does not include route information, so should this information be spoofed, it would be up to the individual switches to log the previous hop. Law enforcement would then have to visit every switch, and hope they all logged the previous switch. Unfortunately, this isn't usually the case. The solution to this problem is the same as to the IPv4 problem -- at each hop restrict the prefixes allowed to those that make sense.
This is the Display IE in Q.931(see
http://www.telepermit.co.nz/ISDNSpec.html
and look at the TNA134* specs for a copy of Q.931 with slight modifications by Telecom NZ).
However, spoofing the number requires the Calling-Party-Number IE to be spoofed, which at least in NZ, can only be set to a number assigned to your company on the same exchange(or your 0800 number).
The telecos here don't let you "spoof" caller-id even if you have a legitimate reason(for example, the number you are "spoofing" is actually the number of the person really calling, over IP), let alone if you wanted to sell a service to allow customers to deceive people.
I don't see how this is specific to OSS. After all, the same could be true about any software, Free, free, commercial, open, closed, in-house, public domain, etc... There are so many patents, many of them vague, out there, and no company or government(not even those departments dealing with patents) knows them all. Combine this with the fact that every line of source code could violate multiple patents given the simplicity of patents, and the combination of certain lines of code could also violate patents while each line by itself doesn't. So to check a program of size n against a patent database of size m is takes at least O(2^n m) of human resources. No one can supply this much time for any realistic software. Hence, we can never be sure that software we use doesn't violate any patents.
Clearly, whatever platform they are currently using faces the same problem, so unless they have identified a specific problem, this should not affect the migration.
For those who don't want to register, here is my paraphrase(I tried to remove highly technical terms as well)...
1) Ordinary mouse prion protein was cloned by recombinant DNA technology in
E. coli. The recombinant protein was polymerised into amyloid fibrils(the
same stuff that forms in Alzheimer's disease etc...). The fibrils were put
into a mouse brain. The mice used were transgenically modified to express
larger amounts of normal prion protein when compared to wild-type mice. 2) After 380-660 days, the mice got a neurological disorder. 3) Prion protein that was resistant to protease(i.e. a protein which cuts up
another protein) activity was found in the mice brains when the proteins
in the cell were analysed on a Western blot. 4) These protease resistant prion protein extracts caused disease in 150-90
days if put into the brain of either a wild-type or the transgenic mice
in part 1. 5) This suggests that the researchers managed to create a new prion. 6) This helps to prove that prions are infectious proteins.
The abstract for the original Science article is
here.
However, you need to register(free to see abstracts) first. You can also pay to see the fulltext.
The article seems to suggest they don't want a blanket ban on Television. So perhaps they should take several factors into account... a) Which channel b) When c) Will parents have a chance to see whether or not violence is coming up and make a decision beforehand(e.g. displaying symbols in the corner of the screen warning of violence, obscene language, etc..., or warning trailers). Parents could then tell their children, for example, never watch this channel, or only watch this channel between certain hours.
Then it could be up to parents to decide. If parents can stop their children playing with matches, they can stop children watching things they don't want, if they get warnings first.
I voted that they were all frauds because the onMouseOver displays a message different to the actual links. I think that because they didn't explain that they modified the e-mail and the onMouseOver display text wasn't in the original e-mail. That probably explains the low results they are getting
[andrew@hp devel]$ host www.kcckp.net www.kcckp.net has address 194.29.229.125 [andrew@hp devel]$ whois 194.29.229.125 % This is the RIPE Whois server. % The objects are in RPSL format. % % Rights restricted by copyright. % See http://www.ripe.net/ripencc/pub-services/db/copyri ght.html
This site has a relatively limited number of stories, and the people who posted them are a self-selected group. People who got cancer are more likely to post.
Of course, any group of people of a size as large as the group who could be considered an "atomic veterans", and of the same sort of age demographic, would have a reasonable number of people who had cancer.
What would be interesting is a study where individuals were selected randomly from all "atomic veterans", and then a statistical analysis of these, compared to a general group from the population with the same age demographics.
There is a biological expectation that being an "atomic veteran" would increase your risk of cancer, but looking at this site does not provide much evidence for that point due to the lack of statistical validity.
Slashdot Mirror
(Disclaimer: IANAL)
From the GPL...
b) You must cause any work that you distribute or
publish, that in whole or in part contains or is
derived from the Program or any part thereof, to
be licensed as a whole at no charge to all third
parties under the terms of this License.
If the terms of another license prohibit you from doing that, it is not GPL compatible.
Of course, the GPL is a copyright license, and does not grant any trademark license. Therefore, an alternate, GPL compatible approach would be to license the software under the exact terms of the GPL, and license the trademark under some other terms(like Mozilla does).
I doubt the vendor would give you a replacement key if they saw the opportunity to make money. As for the legality of using another key on software you own...
(Disclaimer IANAL, no warranties, consult your own counsel before doing anything) but...
1) AFAIK, few democratic countries make sites illegal, only specific content. The keys found on these sites are often generated data according to a formula, so the copyright belongs to the site maintainer. It is not even functional, and could hardly even be called a copyright circumvention device. I don't think this would be illegal in many countries at all.
In many states(of the world), there is no law against using software which has been lawfully obtained. So entering a key other than as supplied by the vendor into software that you have the right to posses is perfectly lawful.
1) You are under no obligation to let the BSA in. If they knock down your door, complain to the Police that they vandalised your door, broke an entering, and trespassed.
2) They have no right to steal your computer. If they take your computer without a court order(which they won't be able to get if your use is legal), then make a complaint to the police that they stole your computer.
Organised crime is taken seriously in most countries, and if they really commit crimes like this, they will soon be declared an Illegal Organisation or the equivalent in your country by the courts.
Except that scheduling the recall is a negative externality resulting from Microsoft's production process, and the cost of that negative externality is not internalised to Microsoft, but rather is met by the customer.
Therefore, it is unethical for Microsoft to not reduce the negative externalities that result from its own poor workmanship in every way it possibly can at no significant cost to itself. The more time people have to prepare for the release of the patch, the better. Therefore, if Microsoft is capable of disclosing an upcoming patch to a certain subset of the population, they have proved that they are capable of disclosing this information publically.
However, having proven that they could give the public more warning, they chose not to. This is unethical.
In practice, Microsoft will probably release the warning to corporate customers at a certain point in time. They will then feel obliged to leave a gap before they release to the general public, otherwise the advantage of the premium service will be diminished. After this, they will feel obliged to give time to ensure everyone has got the warning before they release the patch. Therefore, for the sake of profit, the time until the patch is released is probably also increased. This is also unethical.
The article is not related. Everyone still can call "999"(I assume that is the UK emergency number), but if there is a non-life-threatening situation, they can voluntarily elect(and are encouraged to do so) to call another number. This is very different from selling a good response time, and prioritising another call on the sole basis of how much money you received.
Intel predicts that it will be able to convince the world to abandon the unscalable approach of following standards, including upcoming standards like XForms and IPv6, and open P2P systems, and instead invent its own propietary system.
It is also a supreme court precedent which defends some quite reasonable fair use rights, such as the right to record and time-shift a broadcast.
Wouldn't it be better to spread it out over a long period of time, otherwise they will know it is an organised compaign and will just disregard the calls? Is there any advantage to do this all in one day whatsoever?
From the article...
> In this last step, all of the initially queried
> nodes percolate the query throughout the network
> so that the query is guaranteed to reach a core
> sub-network of highly-connected nodes. "Since a
> copy of the query is in one of the nodes in the
> core network, and since the content list of a
> node is cached at one of these high-degree
> nodes, one is guaranteed to find the content as
> long as at least one node in the network has
> it," said Roychowdhury.
So in other words, the "major sharers", i.e. nodes which are "high degree", i.e. have a lot of connections, form the "core network", and collectively host the entire index. However, this is starting to lose the advantages of being a peer-to-peer network. Obviously, you can't have it both ways.
for (int i = 0; i < nzones; i++)
// They just entered the zone...
// TODO: Detect co-ordinates of intruder
// and point and turn on machine gun.
{
Zone* z = zones[i];
z->read_sensors();
if (z->is_public_allowed)
continue;
if (!z->detected_movement)
{
z->state = STATE_VACANT;
continue;
}
bool allowed = false;
for (int j = 0; j < z->ndetected_rfids; j++)
if (is_authorised(z, z->detected_rfids[i],
CurrentTime))
{
allowed = true;
break;
}
if (allowed)
continue;
switch (z->state)
{
case STATE_VACANT:
z->speaker->PlayLoop("big_dog_bark.mp3");
&nbs p; z->state = STATE_DOG_PLAY;
z->state_start = CurrentTime;
break;
case STATE_DOG_PLAY:
if (CurrentTime - z->state_start < 20)
continue;
z->speaker->PlayLoop("death_threat.mp3");
&nbs p; z->state = STATE_THREAT_PLAY;
z->state_start = CurrentTime;
break;
case STATE_THREAT_PLAY:
if (CurrentTime - z->state_start < 10)
continue;
}
}
The doomsday threat doesn't work if the attacking party thinks they can take down the entire nuclear capacity of the state before the victim can respond. However, given that North Korea alreay claims to have nukes, demonstrating that they do does not make them any more likely to come under attack.
However, if this is a first weapons test, then they probably don't have very well developed nuclear weapons yet, and some governments might take a "its now or never" approach.
There is no indication in the article that any radioactivity was detected despite the fact this happened on Thursday, which probably means it was a chemical explosive. So it could easily be a larger version of the train explosion earlier this year, either accidental or an act of terrorism or government suppression. If they didn't take safety seriously once, it is likely they routinely move potentially explosive compounds without adequate safety protections.
Ignoring the fact that calculating the time an algorithm takes to run is not the same as running the algorithm, a note about matrix multiplication times...
The most obvious iterative algorithm(i.e. direct from the definition) to multiply two nxn matrices would take theta(n^3). Strassen's divide-and-conquer algorithm, which divides the matrices into 4 submatrices, and does 3 recursive matrix multiplications, takes theta(n^log2(7)) = theta(n^2.81...)
The best known algorithm in terms of asymptotic time is theta(n^2.376), but it does not surpass Strassen's algorithm until n is very huge, so Strassen's is the most practical algorithm known.
It is not Star38's fault, it is the fault of the Telcos who made it possible by switch misconfiguration. However, if their telecoms provider decided to reconfigure the switch properly to block spoofing, and their service suddenly didn't work :), I would have much sympathy for Star38 either. I just hope that this doesn't affect the right of the teleco to fix their service(perhaps "but if we reconfigure our switch, we will get sued"?).
The CallingPartyNumber IE is often recorded and used for this purpose. Q.931 does not include route information, so should this information be spoofed, it would be up to the individual switches to log the previous hop. Law enforcement would then have to visit every switch, and hope they all logged the previous switch. Unfortunately, this isn't usually the case. The solution to this problem is the same as to the IPv4 problem -- at each hop restrict the prefixes allowed to those that make sense.
This is the Display IE in Q.931(see http://www.telepermit.co.nz/ISDNSpec.html and look at the TNA134* specs for a copy of Q.931 with slight modifications by Telecom NZ). However, spoofing the number requires the Calling-Party-Number IE to be spoofed, which at least in NZ, can only be set to a number assigned to your company on the same exchange(or your 0800 number).
The telecos here don't let you "spoof" caller-id even if you have a legitimate reason(for example, the number you are "spoofing" is actually the number of the person really calling, over IP), let alone if you wanted to sell a service to allow customers to deceive people.
I don't see how this is specific to OSS. After all, the same could be true about any software, Free, free, commercial, open, closed, in-house, public domain, etc... There are so many patents, many of them vague, out there, and no company or government(not even those departments dealing with patents) knows them all. Combine this with the fact that every line of source code could violate multiple patents given the simplicity of patents, and the combination of certain lines of code could also violate patents while each line by itself doesn't. So to check a program of size n against a patent database of size m is takes at least O(2^n m) of human resources. No one can supply this much time for any realistic software. Hence, we can never be sure that software we use doesn't violate any patents.
Clearly, whatever platform they are currently using faces the same problem, so unless they have identified a specific problem, this should not affect the migration.
It took 1 to 2 years to *create* the prion. After that, it could be passed on to other normal mice in 90+ days.
For those who don't want to register, here is my paraphrase(I tried to remove highly technical terms as well)...
1) Ordinary mouse prion protein was cloned by recombinant DNA technology in
E. coli. The recombinant protein was polymerised into amyloid fibrils(the
same stuff that forms in Alzheimer's disease etc...). The fibrils were put
into a mouse brain. The mice used were transgenically modified to express
larger amounts of normal prion protein when compared to wild-type mice.
2) After 380-660 days, the mice got a neurological disorder.
3) Prion protein that was resistant to protease(i.e. a protein which cuts up
another protein) activity was found in the mice brains when the proteins
in the cell were analysed on a Western blot.
4) These protease resistant prion protein extracts caused disease in 150-90
days if put into the brain of either a wild-type or the transgenic mice
in part 1.
5) This suggests that the researchers managed to create a new prion.
6) This helps to prove that prions are infectious proteins.
The abstract for the original Science article is here. However, you need to register(free to see abstracts) first. You can also pay to see the fulltext.
The article seems to suggest they don't want a blanket ban on Television. So perhaps they should take several factors into account...
a) Which channel
b) When
c) Will parents have a chance to see whether or not violence is coming up and make a decision beforehand(e.g. displaying symbols in the corner of the screen warning of violence, obscene language, etc..., or warning trailers).
Parents could then tell their children, for example, never watch this channel, or only watch this channel between certain hours.
Then it could be up to parents to decide. If parents can stop their children playing with matches, they can stop children watching things they don't want, if they get warnings first.
I voted that they were all frauds because the onMouseOver displays a message different to the actual links. I think that because they didn't explain that they modified the e-mail and the onMouseOver display text wasn't in the original e-mail. That probably explains the low results they are getting
[andrew@hp devel]$ host www.kcckp.neti ght.html
www.kcckp.net has address 194.29.229.125
[andrew@hp devel]$ whois 194.29.229.125
% This is the RIPE Whois server.
% The objects are in RPSL format.
%
% Rights restricted by copyright.
% See http://www.ripe.net/ripencc/pub-services/db/copyr
inetnum: 194.29.224.0 - 194.29.255.255
netname: IPB-PI2
descr: I/P/B Internet Provider in Berlin
country: DE
admin-c: IH408-RIPE
tech-c: IH408-RIPE
tech-c: JAN-RIPE
status: ASSIGNED PI
mnt-by: RIPE-NCC-HM-PI-MNT
mnt-lower: RIPE-NCC-HM-PI-MNT
mnt-by: LAMBDASOL-MNT
mnt-routes: LAMBDASOL-MNT
mnt-by: IPB-MNT
mnt-routes: IPB-MNT
changed: hostmaster@ripe.net 20000308
changed: jan.czmok@jippiigroup.com 20010929
changed: hostmaster@ripe.net 20011009
changed: hostmaster@ripe.net 20011102
changed: hostmaster@ripe.net 20021111
source: RIPE
route: 194.29.224.0/19
descr: IPB-PI
origin: AS20647
remarks: removed cross-mnt: IPB-MNT
mnt-by: IPB-MNT
changed: czmok@lambda-solutions.de 20010929
source: RIPE
role: IPB Hostmaster
address: I/P/B/ Internet Provider in Berlin GmbH
address: Friedrichstr. 95
address: D-10117 Berlin
e-mail: hostmaster@ipberlin.com
trouble: Bug reports mailto:hostmaster@ipberlin.com
trouble: Urgent bug reports: call +49 30 2096-2951
admin-c: SD192-RIPE
tech-c: SD192-RIPE
nic-hdl: IH408-RIPE
mnt-by: IPB-MNT
changed: hostmaster@ipberlin.com 19980812
changed: hostmaster@ipberlin.com 20010418
source: RIPE
person: Jan-Ahrent Czmok
address: Lambda-Solutions
address: Holunderweg 5
address: 55128 Mainz
address: Germany
phone: +49 160 97 25 6774
e-mail: czmok@lambda-solutions.de
nic-hdl: JAN-RIPE
mnt-by: LAMBDASOL-MNT
changed: czmok@lambda-solutions.de 20030317
source: RIPE
This site has a relatively limited number of stories, and the people who posted them are a self-selected group. People who got cancer are more likely to post.
Of course, any group of people of a size as large as the group who could be considered an "atomic veterans", and of the same sort of age demographic, would have a reasonable number of people who had cancer.
What would be interesting is a study where individuals were selected randomly from all "atomic veterans", and then a statistical analysis of these, compared to a general group from the population with the same age demographics.
There is a biological expectation that being an "atomic veteran" would increase your risk of cancer, but looking at this site does not provide much evidence for that point due to the lack of statistical validity.