I want to make this abundantly clear: This is not abot Google magically determining your MAC address from you visiting their website, it's about you running software form Google that uses system calls to determine the MAC address and then sends it to Google. Google Earth is not Mozilla Firefox or Internet Explorer, nor does it run inside them. Google Earth is a native application developed and distributed by Google Inc., which you have to install and execute on your machine. And as a native application it's capable of using system calls and of sending data obtained using those system calls to Google.
OK. Let's stipulate that your running Google Earth gives your current MAC address to Google. What does that get Google? How can they use it? Any link between a GoogE client's MAC address and its current geolocation is potentially entirely ephemeral.
And do we actually know that GoogE is snarfing client MAC information?
Even with per-satellite latency, you're nowhere near GEO delay.
You can get close to the flight-time delay of geo (250ms or so) if you include enough instances of per-node transmission time in your back-of-the-envelope calculations. Each node requires a non-zero amount of time to transmit its packet of data to the next hop. For a geostationary, that's a single fixed chunk of latency. (Only one hop). For Iridium et al, that's once per lateral hop. And Inmarsat BGAN has pretty good throughput: 64kbps. Iridium, OTOH, seems to run at best 10 kbps, so each Iridium hop can potentially have more than six times the per-packet per-hop latency.
However, an Air Force Institute of Technology study seems to indicate that simulated Iridium end-to-end latency works out, on average, to 178 ms, so it would seem a bit more responsive than Inmarsat. In fact, this AFIT paper, top of Page 8, indicates that it'd take 14 LEOsat hops to get end-to-end latency comparable to geostationary distance-based delays. So those clever folks at Iridium seem to have found some good optimizations.
All of that said, I agree, Iridium was pure genius. Technically. From the business plan perspective, not so much. But I'm sure the Department of Defense is glad for final outcome. For low-bandwidth data, Iridium modems are simply brilliant. Great way to, for instance, collect data from isolated weather observation systems and pipe them back to the big meteorology centers for analysis and numerical forecasting.
I like your simplistic world. I bet the sky is clear blue and sunny all the time, and the crayon stick-people inhabitants smile a lot.
This last point alone should make any "IT dept" with any common sense at all push to get a new browser in place ASAP.
"IT depts" (to use your scare quotes) are generally not masters of their own fates. They answer to business priorities. Their discretion will vary from organization to organization, but it's a pretty rare IT department that can unilaterally pursue infrastructure improvements without a good business case. And although you would think that getting away from IE6 would make presenting a good business case easy, remember that the bad things you are trying to get away from are pretty much incomprehensible or impossible to those actually making the decisions. So, in that case, the status quo has no downside, whereas upgrading is probably going to be expensive and disruptive.
An illustrative quote from a mid-senior operations manager (comparable to a VP in a business environment): "Unless you can prove that we will have those security problems within the next quarter, I will not approve any disruptive or expensive changes."
Short-sighted? Certainly. I hope you're not surprised.
If your management is too stupid and obstructive to allow this, get a new job - you're working for morons.
The only way to guarantee you're not working for morons is unemployment. (Not even self-employment is necessarily better in that respect.)
install a second browser for use outside of the company network. It's not hard, and there's plenty out there.
That's the best advice I saw in your post. But that's not a guarantee. Not every user is smart enough (or cares enough) to fire up an alternate browser. So you'll have to take technical or training measures. Those aren't free, and irrational resistance can arise around any disruption or expense.
Do your job.
"Just doing my job" is, in most organizations, following orders. If the orders include preserving IE6, the best you can hope for is damage control. If you're allowed that.
As an opening salvo of a war, it's quite feasible. Yes, you'd lose the merchant ship and all its crew. But it's a bargain, in terms of balance of exchange.
Hmm.... you'd have to find people who are willing to die for a chance to kill an opponent of much higher strategic value. I wonder where you'd find that?
BTW, in case you think I'm taking a cheap shot at the "martyr mindset" we see in a lot of the world, I hasten to point out the number of posthumous Medals of Honor awarded to brave American heroes... who were willing to die for a chance to kill an opponent of higher strategic or tactical value. In warfare, regardless of culture, it's not terribly hard to find those willing to lay down their lives to strike a critical blow. And destroying the core of a carrier battle group would qualify, since it's usually the center-of-mass of American power in the region. And even if not the only game in town (like a large fixed land presence with US air power, etc.), bases would make a fine target for a supersonic cruise missile too.
But latency through multi-hop LEO is potentially as bad as geostationary. Absolute distance may be less, but add per-hop packet store-and-forward times.
In my (admittedly limited) first-hand experience, the US military tends to use Iridium for data comm. Stuff which, 20 years ago, would have been landlines with modems. Except you can't really string landline to some mountain in Upickastan, can you?
Google Earth obtains your MAC address an sends it to Google. Now Google has your external IP address and your MAC address.
But they don't have both at the same time. Therefore they can't associate the two. QED.
Do you not actually understand the fundamental limitations of MAC addressing? It's a data-link-layer, non-routeable addressing mode. If you are somehow thinking Google gets your MAC address as you connect to www.google.com, you need to study up on routers and IEEE 802 protocols.
A MAC address is discernable within the local network topology (i.e., cable network segment, FIOS node, WLAN, etc.). It vanishes as soon as your packet passes through a router. And unless you're googling from within a Google datacenter, there are many, many, MANY routers between you and Google.
Nope. Just pointing your browser at www.gmail.com does not expose your MAC address to the Google machine.
Again: The only reason a MAC address is useful is wardriving for geolocation. That's Google geolocation from wifi-enabled devices.
That's because we're still in the low-energy stage of technology. Carrying that much delta-v around is just hard to justify from an engineering perspective. And that is still true if the aliens have to struggle for each kilogram carried in transit. If, OTOH, they have something more exotic (I know that classic Bussard ramjets have been discredited, but let's pretend for argument's sake something like this), a more direct high-energy profile may be desirable, rather than an extended multiple-target gravitational braking profile. And that would imply retrobraking.
Indeed they do. But the public IP of the router has no discernible link to the private IP of the router, which is all they could see if they could (A) sniff WIFI packets, and (B) crack WPA2 to unpack the WIFI packets into their IPv4 goodness.
In other words, with all that, they absolutely cannot connect your public IP address (from the Google web services access logs) to your MAC address (from their Google maps wardriving).
Wardriving, in this case, is just to enhance geolocation of wireless GMaps clients, not to geolocate your router when you connect through it to google for search or whatever.
As far as I can tell, the only danger we pose is that we disagree, civilly, with you.
Or are you falling prey to the "I'm right, and if you disagree you should shut up" fallacy you accuse others?
BTW, "Fundamentalist" is not actually a code-word for "turned his brain off long ago". And it's no more accurate a stereotyping label as "Jew", "Liberal", or "Redneck".
For the record, I and many other fundamentalists assert that separation of church and state is critical. We have no business dictating religious conviction or practice, including the complete absence thereof, and neither does anyone else.
Interesting. The IP address of my WRT54G, at least as visible from the WLAN, is 192.168.0.1.
I bet there are a lot of 192.168.0.1s out there.
Similarly, the IP addresses of the many authorized clients in the household WLAN are various RFC1918 192.168/16 addresses.
Of course Google will never see those addresses, and sniffing MACs will never associate those machines with whatever public IP address the external side of the router winds up with (which is the only one Google will ever see).
Sorry, associating MAC address uncovered by wardriving with IP address uncovered by Google log analysis doesn't work for any NATted network architecture. And for a personal privacy/personal network perspective, that's most of them.
How about, you multiply by 2 and then, at the end, divide by 3.
That leaves the repeating decimal part at the very end of the calculation, where you can "round the final value" like you suggest.
Whereas your suggestion "take 2/3" means you start out with the repeating decimal whose approximation was the source of the problem.
In other words, the root of the problem is the idea of an absolute-precision decimal rendition of "2/3". You haven't escaped it. By doing this division first, you're stuck arbitrarily choosing some approximation for your intermediate value, just like the folks you're criticizing.
Same ep with one of the best renditions of the Xenu story EVAR. If you didn't sense the sarcasm by that point, please RMA your sarcasm detector, because it's clearly non-functional.
It won't be gray goo, then. It'll be jagged rainbow-colored goo that clacks and clicks as it pursues you to disassemble you into your constituent bricks.
You know, I didn't mean to make that rhyme, but why not?
Anyway, I think you have nothing to fear from the impending macrotechnology* apocalypse unless you, or your property, are made of Legos.
*You can't call it nanotechnology; have you seen how big those bricks can get?
cried in the first Mass Effect when one of my team mates had to die (go ahead and get the lols out of the way).
lol
No, actually, not lol. Anyone World of Warcraft player who actually bothered to play out the entire Battle of Darrowshire quest chain, and didn't get misty-eyed at the ending, is a soulless undead thing.
It's a game. But it tells an affecting story. If that isn't, to the slightest degree, art, then nothing is.
It's kinda sad that the evolving thought on Rules of Engagement is migrating back toward "Leave no witnesses" after a couple of decades of "kill the bad guys, don't even scratch the paint".
That just demonstrates that pirates are using their new-found wealth to buy better prostheses. Maybe laser eyes and bionic legs. They'll certainly need it when the zombie apocalypse descends on the measured and honor-bound contest of pirates v. ninjas.
Slashdot Mirror
I want to make this abundantly clear: This is not abot Google magically determining your MAC address from you visiting their website, it's about you running software form Google that uses system calls to determine the MAC address and then sends it to Google. Google Earth is not Mozilla Firefox or Internet Explorer, nor does it run inside them. Google Earth is a native application developed and distributed by Google Inc., which you have to install and execute on your machine. And as a native application it's capable of using system calls and of sending data obtained using those system calls to Google.
OK. Let's stipulate that your running Google Earth gives your current MAC address to Google. What does that get Google? How can they use it? Any link between a GoogE client's MAC address and its current geolocation is potentially entirely ephemeral.
And do we actually know that GoogE is snarfing client MAC information?
Even with per-satellite latency, you're nowhere near GEO delay.
You can get close to the flight-time delay of geo (250ms or so) if you include enough instances of per-node transmission time in your back-of-the-envelope calculations. Each node requires a non-zero amount of time to transmit its packet of data to the next hop. For a geostationary, that's a single fixed chunk of latency. (Only one hop). For Iridium et al, that's once per lateral hop. And Inmarsat BGAN has pretty good throughput: 64kbps. Iridium, OTOH, seems to run at best 10 kbps, so each Iridium hop can potentially have more than six times the per-packet per-hop latency.
However, an Air Force Institute of Technology study seems to indicate that simulated Iridium end-to-end latency works out, on average, to 178 ms, so it would seem a bit more responsive than Inmarsat. In fact, this AFIT paper, top of Page 8, indicates that it'd take 14 LEOsat hops to get end-to-end latency comparable to geostationary distance-based delays. So those clever folks at Iridium seem to have found some good optimizations.
All of that said, I agree, Iridium was pure genius. Technically. From the business plan perspective, not so much. But I'm sure the Department of Defense is glad for final outcome. For low-bandwidth data, Iridium modems are simply brilliant. Great way to, for instance, collect data from isolated weather observation systems and pipe them back to the big meteorology centers for analysis and numerical forecasting.
lol @ anonymous troll.
I like your simplistic world. I bet the sky is clear blue and sunny all the time, and the crayon stick-people inhabitants smile a lot.
This last point alone should make any "IT dept" with any common sense at all push to get a new browser in place ASAP.
"IT depts" (to use your scare quotes) are generally not masters of their own fates. They answer to business priorities. Their discretion will vary from organization to organization, but it's a pretty rare IT department that can unilaterally pursue infrastructure improvements without a good business case. And although you would think that getting away from IE6 would make presenting a good business case easy, remember that the bad things you are trying to get away from are pretty much incomprehensible or impossible to those actually making the decisions. So, in that case, the status quo has no downside, whereas upgrading is probably going to be expensive and disruptive.
An illustrative quote from a mid-senior operations manager (comparable to a VP in a business environment): "Unless you can prove that we will have those security problems within the next quarter, I will not approve any disruptive or expensive changes."
Short-sighted? Certainly. I hope you're not surprised.
If your management is too stupid and obstructive to allow this, get a new job - you're working for morons.
The only way to guarantee you're not working for morons is unemployment. (Not even self-employment is necessarily better in that respect.)
install a second browser for use outside of the company network. It's not hard, and there's plenty out there.
That's the best advice I saw in your post. But that's not a guarantee. Not every user is smart enough (or cares enough) to fire up an alternate browser. So you'll have to take technical or training measures. Those aren't free, and irrational resistance can arise around any disruption or expense.
Do your job.
"Just doing my job" is, in most organizations, following orders. If the orders include preserving IE6, the best you can hope for is damage control. If you're allowed that.
from a couple of young military officers walking down the corridor in a headquarters facility I once worked in:
"I don't mind change, as long as I don't have to do anything different or learn anything new."
That, pretty much, explains why IE6 persists.
Interesting. You've just invented the arsenal ship, albeit in a kind of hillbilly way.
As an opening salvo of a war, it's quite feasible. Yes, you'd lose the merchant ship and all its crew. But it's a bargain, in terms of balance of exchange.
Hmm.... you'd have to find people who are willing to die for a chance to kill an opponent of much higher strategic value. I wonder where you'd find that?
BTW, in case you think I'm taking a cheap shot at the "martyr mindset" we see in a lot of the world, I hasten to point out the number of posthumous Medals of Honor awarded to brave American heroes... who were willing to die for a chance to kill an opponent of higher strategic or tactical value. In warfare, regardless of culture, it's not terribly hard to find those willing to lay down their lives to strike a critical blow. And destroying the core of a carrier battle group would qualify, since it's usually the center-of-mass of American power in the region. And even if not the only game in town (like a large fixed land presence with US air power, etc.), bases would make a fine target for a supersonic cruise missile too.
But latency through multi-hop LEO is potentially as bad as geostationary. Absolute distance may be less, but add per-hop packet store-and-forward times.
In my (admittedly limited) first-hand experience, the US military tends to use Iridium for data comm. Stuff which, 20 years ago, would have been landlines with modems. Except you can't really string landline to some mountain in Upickastan, can you?
lolwhut?
Google Earth obtains your MAC address an sends it to Google. Now Google has your external IP address and your MAC address.
But they don't have both at the same time. Therefore they can't associate the two. QED.
Do you not actually understand the fundamental limitations of MAC addressing? It's a data-link-layer, non-routeable addressing mode. If you are somehow thinking Google gets your MAC address as you connect to www.google.com, you need to study up on routers and IEEE 802 protocols.
A MAC address is discernable within the local network topology (i.e., cable network segment, FIOS node, WLAN, etc.). It vanishes as soon as your packet passes through a router. And unless you're googling from within a Google datacenter, there are many, many, MANY routers between you and Google.
Nope. Just pointing your browser at www.gmail.com does not expose your MAC address to the Google machine.
Again: The only reason a MAC address is useful is wardriving for geolocation. That's Google geolocation from wifi-enabled devices.
you're doing it wrong
That's because we're still in the low-energy stage of technology. Carrying that much delta-v around is just hard to justify from an engineering perspective. And that is still true if the aliens have to struggle for each kilogram carried in transit. If, OTOH, they have something more exotic (I know that classic Bussard ramjets have been discredited, but let's pretend for argument's sake something like this), a more direct high-energy profile may be desirable, rather than an extended multiple-target gravitational braking profile. And that would imply retrobraking.
--Salman Rushdie
Indeed they do. But the public IP of the router has no discernible link to the private IP of the router, which is all they could see if they could (A) sniff WIFI packets, and (B) crack WPA2 to unpack the WIFI packets into their IPv4 goodness.
In other words, with all that, they absolutely cannot connect your public IP address (from the Google web services access logs) to your MAC address (from their Google maps wardriving).
Wardriving, in this case, is just to enhance geolocation of wireless GMaps clients, not to geolocate your router when you connect through it to google for search or whatever.
As far as I can tell, the only danger we pose is that we disagree, civilly, with you.
Or are you falling prey to the "I'm right, and if you disagree you should shut up" fallacy you accuse others?
BTW, "Fundamentalist" is not actually a code-word for "turned his brain off long ago". And it's no more accurate a stereotyping label as "Jew", "Liberal", or "Redneck".
For the record, I and many other fundamentalists assert that separation of church and state is critical. We have no business dictating religious conviction or practice, including the complete absence thereof, and neither does anyone else.
lolwhut? Did this just degenerate into a "Soviet Russia" joke?
Interesting. The IP address of my WRT54G, at least as visible from the WLAN, is 192.168.0.1.
I bet there are a lot of 192.168.0.1s out there.
Similarly, the IP addresses of the many authorized clients in the household WLAN are various RFC1918 192.168/16 addresses.
Of course Google will never see those addresses, and sniffing MACs will never associate those machines with whatever public IP address the external side of the router winds up with (which is the only one Google will ever see).
Sorry, associating MAC address uncovered by wardriving with IP address uncovered by Google log analysis doesn't work for any NATted network architecture. And for a personal privacy/personal network perspective, that's most of them.
How about, you multiply by 2 and then, at the end, divide by 3.
That leaves the repeating decimal part at the very end of the calculation, where you can "round the final value" like you suggest.
Whereas your suggestion "take 2/3" means you start out with the repeating decimal whose approximation was the source of the problem.
In other words, the root of the problem is the idea of an absolute-precision decimal rendition of "2/3". You haven't escaped it. By doing this division first, you're stuck arbitrarily choosing some approximation for your intermediate value, just like the folks you're criticizing.
without the ability to use Flashblock/AdBlock.
No problem, just install Firefox.
Oh, iPhone? never mind.
Same ep with one of the best renditions of the Xenu story EVAR. If you didn't sense the sarcasm by that point, please RMA your sarcasm detector, because it's clearly non-functional.
fnord. fnord fnord fnord, fnord fnord /. fnord fnord.
As far as I can tell, they assert that it is legal, therefore they think it is legal.
That's a good point. They forgot their "IANAL" disclaimer, just so people understand exactly how much their "legal opinion" is worth.
This means that the average Slashdotter is more legally savvy then these two "researchers".
Here you go.
It won't be gray goo, then. It'll be jagged rainbow-colored goo that clacks and clicks as it pursues you to disassemble you into your constituent bricks.
You know, I didn't mean to make that rhyme, but why not?
Anyway, I think you have nothing to fear from the impending macrotechnology* apocalypse unless you, or your property, are made of Legos.
*You can't call it nanotechnology; have you seen how big those bricks can get?
cried in the first Mass Effect when one of my team mates had to die (go ahead and get the lols out of the way).
lol
No, actually, not lol. Anyone World of Warcraft player who actually bothered to play out the entire Battle of Darrowshire quest chain, and didn't get misty-eyed at the ending, is a soulless undead thing.
It's a game. But it tells an affecting story. If that isn't, to the slightest degree, art, then nothing is.
It's kinda sad that the evolving thought on Rules of Engagement is migrating back toward "Leave no witnesses" after a couple of decades of "kill the bad guys, don't even scratch the paint".
That just demonstrates that pirates are using their new-found wealth to buy better prostheses. Maybe laser eyes and bionic legs. They'll certainly need it when the zombie apocalypse descends on the measured and honor-bound contest of pirates v. ninjas.