Or they could throttle back the traffic on behalf of the ISPs in question to just above the maximum point that they think the congested peer will support and/or start sending ICMP Source Quench packets to the ISP's end users. The ISPs would, in all likelihood, be none the wiser, and it would release Level 3's transit bandwidth for their customers that actually care about providing decent services to their end users. It's still extra load on L3's edge routers, but at least it alleviates any problems in their core and on other customer's edges.
It's just a metaphor, and that's the way I used it - the whole point is the premise, not the outcome, just like Schroedinger's Cat. That the oft proposed outcome of boiled frog is actually false (and, yes, I was aware of that) is actually a positive thing in my view since the alternative would be that the analogy implies that we're fucked and might as well happily continue doing all we can to screw up the environment since we're doomed anyway. Other than that, it's also a useful analogy to explain why global warming doesn't equate to warmer and sunnier weather but just means more turbulence in the heated medium (e.g. more variable weather) for people that don't understand how increased heat works at the atomic level and how that then translates into the macroscopic level, so don't expect it to go away any time soon.
Global warming was always a terrible name because the imagery was all wrong.
I don't know what image you had, but the one I got was of the frog swimming in a pan full of water while the heat was slowly turned up. Replace "frog" with "Life on Earth", "pan" with "Earth" and "water" for "seas and atmosphere" and I think it's a pretty clear image of what the proponents of global warming, climate change, or whatever other nomenclature you want to assign the process, are trying to get across. While I can see that changing the name might break with some of the misconceptions people have, it also makes me think of this sketch by George Carlin and rearranging the deckchairs on the Titanic.
Or the amount of tape in the cartridge could be smaller. Maybe a return to DAT sized cartridges?
It's probably a horses for courses thing though. There are plenty of applications, particularly in the scientific research arena, that generate vast amounts of data that can be spooled off to tape for archiving/distribution pretty much at leisure before the hot storage is wiped for re-use. Even with Internet2, I could easily imagine somewhere like Arecibo, CERN or the SKA being all over this kind of storage capacity given the old bandwidth of physical media in a truck adage.
Blanket approvals and template documents that you can cut and paste notifications into are the way to go, especially when it's on a schedule like MS, Adobe & Oracle. If they push back, suggest a documented process (this is ITIL, right? You can avoid the need for a CAB if it's an approved and documented procedure) where you push the patches to a few test systems on Tuesday (in the case of MS) then deploy to the rest later in the week - whatever they are happy with - if there are no issues. Depending on your timezone Tuesday PM or Wednesday AM are good slots for weekly CABs to pick up this; push to the test servers on the day, then the rest at the end of the week. For *nix, i've done updates this way for anything that didn't require a reboot so only stuff like Kernel updates and major low-level libraries needed to get approval via a CAB.
For everything else, it's your call. Either the patch waits for the next regular CAB or you play the game and keep calling emergency CABs when there are justifiably critical updates, such as Heartbleed, or for the inevitable critical updates from MS every second Tuesday that impact your systems. The best tactic is to embrace ITIL and make it work for you, not allow them to make you jump through hoops and spend your time crafting unique documents for every patch. It also serves as a useful procedure check to make sure you don't mess up and have a contingency plan for when you do, and ultimately, if you get it right, you still get to dictate the schedule and make them do things in ways that you are happy to work with.
Heartbleed is a score for closed source. Those trying to spin it like this is open source working are delusional.
So, if this were to have happened in a closed source library, another company would have been looking at the code in order to discover the bug *how*, exactly? Even if the bug had been found by a white hat, the only recourse would have been to raise a bug report with the vendor and hope they actually did something about it. The failure for open source here isn't the development model, it's the fact it took two years for the vaunted "many eyes" to get around to looking at new code in a critical piece of the tool chain. As I noted, that's something that can easily be addressed by forcing commits be vetted before acceptance, and potentially other ways too, but again, you could also apply that approach in a closed source shop.
This, and I suspect a lot of shilling by proprietary software vendors playing up the "many eyes make bugs shallow" thing. This wasn't so much a failure of the open source model as it was a failure to properly vet commits to the code of a project before accepting them into the main tree, and that could happen just as easily on a closed source development model as an open source one. That might be OK for small hobby projects, and perhaps even major projects that don't have quite so major ramifications in the event of a major flaw, but hopefully this will serve as a wake up call for projects that aim to form some kind of critical software infrastructure. For such projects requiring that commits be reviewed and "signed off" by one or more other developers would perhaps have caught this bug, and others like it, and could perhaps work very well in conjuction with some of the bug-bounty programmes out there. Of course, "Find a flaw in our pending commits, and get paid!" only works if the code is open for inspection...
Some countries don't even have personal income tax, and apart from the U.S. I don't know of any others that require their citizens pay income taxes on wages earned overseas. Admittedly several of the countries on the list are not the best places to live, but for non-USians it's perfectly possible to avoid paying income tax altogether.
A better solution might be to move the original sender's "From" to another header ("Return-Path", "Reply-To", - whatever works best for the list software/admin) and set a new "From" to an address that would feed any replies to the list's submission/moderation queue. If the address of the person replying is on the mailing list or the list accepts any submission address, it goes into the normal queue for remailing, if not it either gets discarded as a bogus reply that is probably spam or goes into a moderation queue, depending on the list.
This is still an implementation flaw in the way DMARC and SPF work with mailing lists rather than a problem with mailing lists though, so the onus really belongs with DMARC and SPF to better provide a way to support mailing lists. Including a way to specify in the DMARC/SPF configuration for the that the sender is a mailing list and that they need to validate the original sender against a different header instead - "X-Originally-From", rather than the mailing list's domain in the current "From", perhaps?
Hopefully they will name it something beginning with "P" so mnemonics like "My Very Easy Method, Just Set Up Nine Planets" and so on all work again, or has anyone come up with a good one for the current 8 planet setup?
It's a view point in the UK that has been around for years - at least a few decades - although where it originated from and when I have no idea. Quite possibly it was the Daily Mail or a similar rag going off on one of the their diatribes about "declining standards" or whatever they had a bee in their bonnet over that month. From comparing notes though it does seem C of E schools in the UK generally force much less dogma and indoctrination upon impressionable infants compared with church schools of other faiths, and even other christian denominations. I guess you really do reap what you sow...:)
The graphs certainly back up the idea that the best way to raise an atheist is to send the child to a Church of England school (in my case I was an atheist by the age of nine), but I suspect that the increasingly secularisation in UK education has something to do with that as well. When the only primary school in a small rural town is a church school (usually that would be C of E, but sometimes Catholic) and you have a typical rural UK demographic representing both major christian denominations plus a scattering of other faiths that school tends to get coerced into providing a more agnostic education if it wants financial support from the local government.
I think there's more to it than just being exposed to skepticism from existing atheists/agnostics too. You get much more exposure to people who are from different cultures and religions that you might in your own little neighbourhood, both knowingly and unknowingly, and when that penny drops, that's when the thinking part kicks in. Generally you are going to you realise that, hey, they are not that unlike us and we actually share many of the same views on life - most religions teach the same core principles wrapped up in some slightly different stories, after all. It's fairly well understood that major cities with cosmopolitan populations tend to be more open minded and their populations tend to have a less religious view than those from more rural communities, so I suspect this is just the same principle manifesting itself on a much grander scale.
It might actually be more than that. Worst case, the screen in in 80x25 text mode (assuming a PC), which gives 2,000 binary bits, but if you start playing around with extended ASCII graphics characters you could probably encode a KB of data quite easily. Hardly a crash dump, but easily enough to get across the essentials.
That would be excellent if this happened, although unlikely given how much the local population that supports the tourist trade is likely to rely on that same mobile coverage. I go on vacation to *get away* from the daily grind, yet of late it has got to the point that you can't go anywhere without someone yakking on a mobile phone, and I go to some pretty out of the way places to try and make that happen. The absolute last thing you want to hear when you reach Everest Base Camp, slightly out of breath from the lack of oxygen and effort, and are just starting to take in the amazing view is:
*Latest naff ringtone*
"Hello...?"
*pause*
"Yes, I'm climbing Mount Everest!"
Inmarsat managed to eliminate the northen arc based on differences in expected doppler differences of the signal pings, when the last ping was received, and assuming a conservative fuel consumption to that point, there would have been insufficient fuel left for the plane to make land, hence it went down in the ocean. It's important to note that Inmarsat is unable to say where exactly, only that it is within a given range of the location where last known ping was now known to have been sent from, which is where the search for wreckage is now centred. I gather this is the result of a highly unorthodox set of data analysis that is well outside normal procedures for determining location, hence the reason it's taken so long - some of the techniques they used probably haven't ever been done before.
Still vastly better than what it was only a day ago, and there seems to be a lot more possible debris sightings in the search area which I take as a sign they might be in the right area and will hopefully pin it down some more. The race now is to find it before the black box transmitters go silent, a task for which the US is dispatching some specialist search gear apparently, because that's probably the only hope of giving the bereaved a chance at some closure left now.
I don't think it's going to be quite the VR nirvana that some people are expecting, at least not for some of the more involved games that would benefit the most from VR - simpler console based stuff will be fine, although I'd expect there to be a similar level of bandwagon jumping crap that we had when the first "Multimedia PCs" were all the rage. Having a device like the Oculus Rift strapped in front of your eyes is a double edged sword; yes, you are totally immersed in the virtual environment, but you are also much more limited in your interactions with the real one. You are going to need to have situational awareness of both worlds, and do everything in the real one pretty much by touch alone, and that's likely a more limiting factor than some people might be expecting.
There's a lot of people planning on using the Oculus to play Star Citizen when it comes out, yet this is a traditional old school style PC flight sim at heart which, as many old timers will attest, even with all the buttons and other controls on a HOTAS setup, you often still needed some controls on the keyboard. The game also has an FPS mode that many of those same players are planning on using with a mouse rather than a stick, so that most likely means that the left hand will be moving between throttle and keyboard and the right between stick and mouse. Sure, most PC gamers can touch type, but with the Oculus we won't even have the benefit of our peripheral vision to find the home keys and get our bearings to find the key(s) we want, and I think that might be harder to do quickly than some people expect, particularly when they are in the middle of a dogfight or attempted boarding. That's not to say it's an unsolvable problem, some extra thought on control to key mappings might be enough to avoid most mis-steps, and I expect to see a lot of work going into making input devices much more tactile to help with this over the next few years - braille keyboards for hardcore VR gamers anyone?
Perhaps. By not applying Full Disclosure to the identity of the "insider" that has resulting in this you could accuse John Cartright of breaching his and the list's principles, but without knowing the details of the threat (and the list has resistant many such threats in the past) it's difficult to know what the consequence of that might be. Or maybe there is no really significant threat other than some inconvenience, but this is just the straw that broke the camel's back. If not taking down this list would result in the breach of a court order, then this is almost certainly the right tack to take, regardless of how painful it might seem, unless we are expecting John to potentially become another fugitive from justice, like Edward Snowden?
Sure,it's a sad day for freedom of information, and will no doubt have negative consequences due to more information being known only those with malicious intentions and companies sweeping issues under the rug due to lack of exposure, but even so I don't think it's ont that is worth compromising your life over, let alone expecting someone else to do so.
Ahh, yes! That low, low, price doesn't include the price of building renewable-power plants to replace those coal fired ones that are to be shutdown, nor does it even include any budget for sending out lots of blankets to prevent millions from freezing to death during the next cold spell. Nope, that part is entirely left down to others, and specifically to the government and thus taxpayers money. So, yeah, $50b to buy and shutdown the plants, and then what, maybe $500b of pork on a good day to replace them with renewables? Sure, sounds like a good deal to me...
I think I'll just chalk this up as another ill thought out scheme that'll never work.
It's really either this or tape. Just be sure to verify your backups, make sure that you can actually restore, and keep a copy off the data off site at a suitably remote family member/friend's place. Personally, I use external USB3 disks as they are cheap and can be left copying data overnight, especially so if you structure the data so that it's easy to segment data on the RAID across multiple external drives, and replace them long before any MTBF kicks in. Also unlike some tape systems, don't require particular software to backup/restore - any directory copy/sync tool of choice will do, and for restoring specific files you can even use a file manager or command line copy.
It also reduces the time it takes to write the file out to disk or memory card. That could have a small knock on effect in a number of areas like the burst length on cameras and battery life on mobile devices (assuming that the new codec isn't much more CPU intensive). If the extra 10% compression improvement mentioned in the summary is from photographic images rather than illustrations then that could be quite a significant difference.
I'm guessing FB sees it more about data mining and the ad revenue (the product) than the actual users (the raw material). If they tack an brief ad on the end of each message and charge for a premium ad-free service, then it becomes more a case of how many messages do WhatsApp users send to each each other that we can make money off. Even charging the advertiser something ridiculous like 0.01c/message, given the rate typical teens messages each other that's going to add up pretty quickly, although I doubt it's going to ever add up to $16b though - especially if they really are shedding users at the rates implied in the tech press.
I think they mean it's something that would need to be pushed out by each of the hardware vendors as a 4.2 OS update, not something that Google could patch via the Play Store update mechanism as would be the case if the issue was with one of their apps built on top of GMS. Kind of like expecting Microsoft to fix a bug in a PC's BIOS. Given how badly vendors are doing at upgrading to new versions of the OS, I suspect that getting them to go back and patch a version that is already out of date is going to be an even harder mountain to climb.
Slashdot Mirror
Or they could throttle back the traffic on behalf of the ISPs in question to just above the maximum point that they think the congested peer will support and/or start sending ICMP Source Quench packets to the ISP's end users. The ISPs would, in all likelihood, be none the wiser, and it would release Level 3's transit bandwidth for their customers that actually care about providing decent services to their end users. It's still extra load on L3's edge routers, but at least it alleviates any problems in their core and on other customer's edges.
It's just a metaphor, and that's the way I used it - the whole point is the premise, not the outcome, just like Schroedinger's Cat. That the oft proposed outcome of boiled frog is actually false (and, yes, I was aware of that) is actually a positive thing in my view since the alternative would be that the analogy implies that we're fucked and might as well happily continue doing all we can to screw up the environment since we're doomed anyway. Other than that, it's also a useful analogy to explain why global warming doesn't equate to warmer and sunnier weather but just means more turbulence in the heated medium (e.g. more variable weather) for people that don't understand how increased heat works at the atomic level and how that then translates into the macroscopic level, so don't expect it to go away any time soon.
I don't know what image you had, but the one I got was of the frog swimming in a pan full of water while the heat was slowly turned up. Replace "frog" with "Life on Earth", "pan" with "Earth" and "water" for "seas and atmosphere" and I think it's a pretty clear image of what the proponents of global warming, climate change, or whatever other nomenclature you want to assign the process, are trying to get across. While I can see that changing the name might break with some of the misconceptions people have, it also makes me think of this sketch by George Carlin and rearranging the deckchairs on the Titanic.
Or the amount of tape in the cartridge could be smaller. Maybe a return to DAT sized cartridges?
It's probably a horses for courses thing though. There are plenty of applications, particularly in the scientific research arena, that generate vast amounts of data that can be spooled off to tape for archiving/distribution pretty much at leisure before the hot storage is wiped for re-use. Even with Internet2, I could easily imagine somewhere like Arecibo, CERN or the SKA being all over this kind of storage capacity given the old bandwidth of physical media in a truck adage.
Blanket approvals and template documents that you can cut and paste notifications into are the way to go, especially when it's on a schedule like MS, Adobe & Oracle. If they push back, suggest a documented process (this is ITIL, right? You can avoid the need for a CAB if it's an approved and documented procedure) where you push the patches to a few test systems on Tuesday (in the case of MS) then deploy to the rest later in the week - whatever they are happy with - if there are no issues. Depending on your timezone Tuesday PM or Wednesday AM are good slots for weekly CABs to pick up this; push to the test servers on the day, then the rest at the end of the week. For *nix, i've done updates this way for anything that didn't require a reboot so only stuff like Kernel updates and major low-level libraries needed to get approval via a CAB.
For everything else, it's your call. Either the patch waits for the next regular CAB or you play the game and keep calling emergency CABs when there are justifiably critical updates, such as Heartbleed, or for the inevitable critical updates from MS every second Tuesday that impact your systems. The best tactic is to embrace ITIL and make it work for you, not allow them to make you jump through hoops and spend your time crafting unique documents for every patch. It also serves as a useful procedure check to make sure you don't mess up and have a contingency plan for when you do, and ultimately, if you get it right, you still get to dictate the schedule and make them do things in ways that you are happy to work with.
So, if this were to have happened in a closed source library, another company would have been looking at the code in order to discover the bug *how*, exactly? Even if the bug had been found by a white hat, the only recourse would have been to raise a bug report with the vendor and hope they actually did something about it. The failure for open source here isn't the development model, it's the fact it took two years for the vaunted "many eyes" to get around to looking at new code in a critical piece of the tool chain. As I noted, that's something that can easily be addressed by forcing commits be vetted before acceptance, and potentially other ways too, but again, you could also apply that approach in a closed source shop.
This, and I suspect a lot of shilling by proprietary software vendors playing up the "many eyes make bugs shallow" thing. This wasn't so much a failure of the open source model as it was a failure to properly vet commits to the code of a project before accepting them into the main tree, and that could happen just as easily on a closed source development model as an open source one. That might be OK for small hobby projects, and perhaps even major projects that don't have quite so major ramifications in the event of a major flaw, but hopefully this will serve as a wake up call for projects that aim to form some kind of critical software infrastructure. For such projects requiring that commits be reviewed and "signed off" by one or more other developers would perhaps have caught this bug, and others like it, and could perhaps work very well in conjuction with some of the bug-bounty programmes out there. Of course, "Find a flaw in our pending commits, and get paid!" only works if the code is open for inspection...
Some countries don't even have personal income tax, and apart from the U.S. I don't know of any others that require their citizens pay income taxes on wages earned overseas. Admittedly several of the countries on the list are not the best places to live, but for non-USians it's perfectly possible to avoid paying income tax altogether.
A better solution might be to move the original sender's "From" to another header ("Return-Path", "Reply-To", - whatever works best for the list software/admin) and set a new "From" to an address that would feed any replies to the list's submission/moderation queue. If the address of the person replying is on the mailing list or the list accepts any submission address, it goes into the normal queue for remailing, if not it either gets discarded as a bogus reply that is probably spam or goes into a moderation queue, depending on the list.
This is still an implementation flaw in the way DMARC and SPF work with mailing lists rather than a problem with mailing lists though, so the onus really belongs with DMARC and SPF to better provide a way to support mailing lists. Including a way to specify in the DMARC/SPF configuration for the that the sender is a mailing list and that they need to validate the original sender against a different header instead - "X-Originally-From", rather than the mailing list's domain in the current "From", perhaps?
Hopefully they will name it something beginning with "P" so mnemonics like "My Very Easy Method, Just Set Up Nine Planets" and so on all work again, or has anyone come up with a good one for the current 8 planet setup?
It's a view point in the UK that has been around for years - at least a few decades - although where it originated from and when I have no idea. Quite possibly it was the Daily Mail or a similar rag going off on one of the their diatribes about "declining standards" or whatever they had a bee in their bonnet over that month. From comparing notes though it does seem C of E schools in the UK generally force much less dogma and indoctrination upon impressionable infants compared with church schools of other faiths, and even other christian denominations. I guess you really do reap what you sow... :)
The graphs certainly back up the idea that the best way to raise an atheist is to send the child to a Church of England school (in my case I was an atheist by the age of nine), but I suspect that the increasingly secularisation in UK education has something to do with that as well. When the only primary school in a small rural town is a church school (usually that would be C of E, but sometimes Catholic) and you have a typical rural UK demographic representing both major christian denominations plus a scattering of other faiths that school tends to get coerced into providing a more agnostic education if it wants financial support from the local government.
I think there's more to it than just being exposed to skepticism from existing atheists/agnostics too. You get much more exposure to people who are from different cultures and religions that you might in your own little neighbourhood, both knowingly and unknowingly, and when that penny drops, that's when the thinking part kicks in. Generally you are going to you realise that, hey, they are not that unlike us and we actually share many of the same views on life - most religions teach the same core principles wrapped up in some slightly different stories, after all. It's fairly well understood that major cities with cosmopolitan populations tend to be more open minded and their populations tend to have a less religious view than those from more rural communities, so I suspect this is just the same principle manifesting itself on a much grander scale.
It might actually be more than that. Worst case, the screen in in 80x25 text mode (assuming a PC), which gives 2,000 binary bits, but if you start playing around with extended ASCII graphics characters you could probably encode a KB of data quite easily. Hardly a crash dump, but easily enough to get across the essentials.
That would be excellent if this happened, although unlikely given how much the local population that supports the tourist trade is likely to rely on that same mobile coverage. I go on vacation to *get away* from the daily grind, yet of late it has got to the point that you can't go anywhere without someone yakking on a mobile phone, and I go to some pretty out of the way places to try and make that happen. The absolute last thing you want to hear when you reach Everest Base Camp, slightly out of breath from the lack of oxygen and effort, and are just starting to take in the amazing view is:
*Latest naff ringtone*
"Hello...?"
*pause*
"Yes, I'm climbing Mount Everest!"
It kind of ruins the moment, you know?
Maybe we should #AskJenny about that?
Inmarsat managed to eliminate the northen arc based on differences in expected doppler differences of the signal pings, when the last ping was received, and assuming a conservative fuel consumption to that point, there would have been insufficient fuel left for the plane to make land, hence it went down in the ocean. It's important to note that Inmarsat is unable to say where exactly, only that it is within a given range of the location where last known ping was now known to have been sent from, which is where the search for wreckage is now centred. I gather this is the result of a highly unorthodox set of data analysis that is well outside normal procedures for determining location, hence the reason it's taken so long - some of the techniques they used probably haven't ever been done before.
Still vastly better than what it was only a day ago, and there seems to be a lot more possible debris sightings in the search area which I take as a sign they might be in the right area and will hopefully pin it down some more. The race now is to find it before the black box transmitters go silent, a task for which the US is dispatching some specialist search gear apparently, because that's probably the only hope of giving the bereaved a chance at some closure left now.
I don't think it's going to be quite the VR nirvana that some people are expecting, at least not for some of the more involved games that would benefit the most from VR - simpler console based stuff will be fine, although I'd expect there to be a similar level of bandwagon jumping crap that we had when the first "Multimedia PCs" were all the rage. Having a device like the Oculus Rift strapped in front of your eyes is a double edged sword; yes, you are totally immersed in the virtual environment, but you are also much more limited in your interactions with the real one. You are going to need to have situational awareness of both worlds, and do everything in the real one pretty much by touch alone, and that's likely a more limiting factor than some people might be expecting.
There's a lot of people planning on using the Oculus to play Star Citizen when it comes out, yet this is a traditional old school style PC flight sim at heart which, as many old timers will attest, even with all the buttons and other controls on a HOTAS setup, you often still needed some controls on the keyboard. The game also has an FPS mode that many of those same players are planning on using with a mouse rather than a stick, so that most likely means that the left hand will be moving between throttle and keyboard and the right between stick and mouse. Sure, most PC gamers can touch type, but with the Oculus we won't even have the benefit of our peripheral vision to find the home keys and get our bearings to find the key(s) we want, and I think that might be harder to do quickly than some people expect, particularly when they are in the middle of a dogfight or attempted boarding. That's not to say it's an unsolvable problem, some extra thought on control to key mappings might be enough to avoid most mis-steps, and I expect to see a lot of work going into making input devices much more tactile to help with this over the next few years - braille keyboards for hardcore VR gamers anyone?
Perhaps. By not applying Full Disclosure to the identity of the "insider" that has resulting in this you could accuse John Cartright of breaching his and the list's principles, but without knowing the details of the threat (and the list has resistant many such threats in the past) it's difficult to know what the consequence of that might be. Or maybe there is no really significant threat other than some inconvenience, but this is just the straw that broke the camel's back. If not taking down this list would result in the breach of a court order, then this is almost certainly the right tack to take, regardless of how painful it might seem, unless we are expecting John to potentially become another fugitive from justice, like Edward Snowden?
Sure,it's a sad day for freedom of information, and will no doubt have negative consequences due to more information being known only those with malicious intentions and companies sweeping issues under the rug due to lack of exposure, but even so I don't think it's ont that is worth compromising your life over, let alone expecting someone else to do so.
Ahh, yes! That low, low, price doesn't include the price of building renewable-power plants to replace those coal fired ones that are to be shutdown, nor does it even include any budget for sending out lots of blankets to prevent millions from freezing to death during the next cold spell. Nope, that part is entirely left down to others, and specifically to the government and thus taxpayers money. So, yeah, $50b to buy and shutdown the plants, and then what, maybe $500b of pork on a good day to replace them with renewables? Sure, sounds like a good deal to me...
I think I'll just chalk this up as another ill thought out scheme that'll never work.
It's really either this or tape. Just be sure to verify your backups, make sure that you can actually restore, and keep a copy off the data off site at a suitably remote family member/friend's place. Personally, I use external USB3 disks as they are cheap and can be left copying data overnight, especially so if you structure the data so that it's easy to segment data on the RAID across multiple external drives, and replace them long before any MTBF kicks in. Also unlike some tape systems, don't require particular software to backup/restore - any directory copy/sync tool of choice will do, and for restoring specific files you can even use a file manager or command line copy.
It also reduces the time it takes to write the file out to disk or memory card. That could have a small knock on effect in a number of areas like the burst length on cameras and battery life on mobile devices (assuming that the new codec isn't much more CPU intensive). If the extra 10% compression improvement mentioned in the summary is from photographic images rather than illustrations then that could be quite a significant difference.
I'm guessing FB sees it more about data mining and the ad revenue (the product) than the actual users (the raw material). If they tack an brief ad on the end of each message and charge for a premium ad-free service, then it becomes more a case of how many messages do WhatsApp users send to each each other that we can make money off. Even charging the advertiser something ridiculous like 0.01c/message, given the rate typical teens messages each other that's going to add up pretty quickly, although I doubt it's going to ever add up to $16b though - especially if they really are shedding users at the rates implied in the tech press.
I think they mean it's something that would need to be pushed out by each of the hardware vendors as a 4.2 OS update, not something that Google could patch via the Play Store update mechanism as would be the case if the issue was with one of their apps built on top of GMS. Kind of like expecting Microsoft to fix a bug in a PC's BIOS. Given how badly vendors are doing at upgrading to new versions of the OS, I suspect that getting them to go back and patch a version that is already out of date is going to be an even harder mountain to climb.