That's scarey. Some of the devices you are mentioning get used in large networks (Lantronix term servers, for example). And a lot of them are on home Internets.
It will only work if the destination has a broken IP implementation. IFF the implementation is correct, it's supposed to either forward or drop any IP packets with a destination that isn't configured on the interface. It's never supposed to process them locally. Of course on a POS that you can't change the IP address on, one can't be sure it's actually handling the IP protocol correctly. Which would mean that putting more than one of them on a switch is probably a bad idea.
ICMP uses IP.
You are correct that ARP doesn't use IP.
The LAT protocol doesn't use IP, but nobody uses LAT any more.
There used to be audio distribution protocols that ran in the MAC layer on 100Mb Ethernet, but I believe they switched to running under IP when switches and routers learned how to do vLans and QOS.
Appletalk used to avoid IP, but I believe they switched, too.
Protocols the use IP are routeable; so, they are limited in utility.
See http://en.wikipedia.org/wiki/EtherType for a more complete list.
Mucking with the ARP tables won't work, because the IP addresses will be wrong in the packets (unless the IP implementation in the pumps is sorely lacking).
OTOH, using vLans on a cheap managed switch is better than the 16 USB Ethernet adapters the poster suggested.
This won't work if the pumps have a real IP implementation, because the destination IP address in the packets will be wrong and they will get dropped.
Don't fill up pages and pages of Slashdot arguing about which methodology fad is best...
You can get to mid-town from JFK that fast if you pay the cabbie a flat fee that's about 150% of the usual fair in advance. You might want to keep your eyes closed during the ride
Most places running lotteries forbid employees from playing (of at least from accepting the payout).
In fact, the public service ethics laws in most states would automatically forbid accepting the payout as a conflict of interest... Massachusetts' laws do and so do the Federal ethics laws. Many states base their own laws on the Federal laws; so, given the number of states in that multiple state lottery, there have to be a number of ethics laws that could be used to prosecute just on the basis of accepting the payout.
Wait until the first big snow storm when all the driverless cars stop in the middle of the road because it's too slippery and they aren't programmed for anything else.
why did charge him with escape from custody? I.e., They must have been able to prove that he sent the e-mail.
Marvin Gaye might disagree about what his kids deserve. It could even be that he left the works to his kids with the express instructions to do exactly this: Remember that the music industry has had a long history of ripping off black artists from Marvin Gaye's generation. There is definitely a lot of residual resentment.
And before you think this is a tragedy: The flip side of that is that the music industry might get more interested in supporting artists who actually are capable of creating original material. Given that it's easier to sell copies of something popular than something new, I won't be holding my breath for anything to change, but that is one possible outcome.
The news played a clip where they cross-faded between the two songs and I couldn't hear where the cross-fade happened. I also couldn't hear any difference at all between the two clips in that isolated context (which is all the matters for copyright infringement). One could certainly argue that the ability to edit songs with a computer makes it easier to demonstrate things like this. I've also heard this done where completely different songs are fit together in very clever ways (sorry, I don't remember where); so, part of this might be that the courts haven't caught up with computers.
But I think the over-riding issue here is that the admitted they copied from it, by calling what they did a tribute. Most of the time when someone does a tribute to another artists work, they cover the songs with different arrangements and pay royalties to the other artist. The "tribute" thing has happened before with ZZ Top and John Lee Hooker (La Grange vs Boogie Chillin', I think). It was much more overt in that case - I believe it never went to court. But that sets an industry precedent for using the term "tribute" in this manner as a face saving maneuver when somebody knows they got caught.
Right on. I have used the Ancestry Y chromosome database and had a hit where ALL the alleles matched. I emailed the guy and we can't figure out how were are related: It's definitely way more than the number of generations Ancestry suggested was likely. So if one allele doesn't match you could be talking a common ancestor many generations back. That's not exactly a close relative.
Ancestry's predictions for how close a match were not very accurate for the Y chromosome database. In fact, they aren't using the Y database any more. They are using something that uses data from all chromosomes and in every case where I could check it from the database, it has been accurate as to how far back the common ancestor was.
Does anyone out there really think Leonard Nimoy would be honored by this?
By and large goto is OK when used as an exit path from a nested context (as others here have shown in examples). Some languages provide explicit constructs for this (like variants or exit and continue that allow one to specify which nested context is being exited). C doesn't have these; so, one has to make do.
BTW, if you want to know what's bad about gotos, track down some Fortran programs that were written in the 60s and you will see where Dijkstra was coming from