"Solar and wind is very cheap in Developing countries where Electric outages are simply a matter of fact. In the Developed world outages are largely unacceptable"
My wife works in a developing country. One of the primary handbrakes on industry there (and the ability to move up the development chain) is the unreliability of the power - and the shortage of it. Peak demand is at least 4 times higher than the supply capacity, resulting in rolling blackouts being the norm simply to stop the distribution system burning out.
The political decision to solve this? A 600MW coal plant built in the poorest section of the largest city (which is now almost a decade behind schedule as various shortcomings due to poor oversight are discovered and rectified), along with several other large coal plants in other cities. The immediate-area environmental effects will be bad and it doesn't solve the issue that we need wean ourselves off carbon burning as a matter of global priority.
"And yet we never hear of solving the climate change problem by making those poor countries into wealthy countries"
That depends who you talk to. Various climate change researchers have made this very proposal. The problem is that to get the required energy density needed, you _need_ nuclear plants and everyone's still doing the "oooh, radiation is Baaaaaad" routine.
Wind and Solar are intermittent and you need to overbuild by a factor of _at least_ 4 in order to maintain average output. The numbers quoted above are based on nameplate generation figures. In addition to that you need to geographically disperse your generation plant and overbuild your power distribution network by a substantial factor to handle power flows in relatively unpredictable directions. These figures are never included in the calculations of wind/solar output.
Nuclear tech is vastly overpriced due to the wildly inefficient (and fairly dangerous) technologies currently in use (60-70% of mined uranium is discarded before hitting the reactor. only 1% of the fuel which enters the reactor is currently used. The only positive about nuclear waste from this tech is that 60 years worth of waste will comfortably fit in an olympic-sized pool and is safe enough to handle in 300 years)
If/when molten salt tech hits commercial viability it should drop the cost by an order of magnitude simply because all the radiological safety issues imposed by having water in direct contact with nuclear materials (radioactive steam, hydrogen explosions, etc) will be eliminated. The factor of being able to "burn" existing "high level waste", not having massive supply inefficiencies (thorium is a waste product of rare earth mining 10 times more abundant that Uranium) and being able to use at least 90% of the input fuel are all just extra features.
I'll keep making the comparison that Thorium LFTR technology is akin to Watt's improved steam engine vs the Neucomen engines (current pressurised/boiling water tech) that preceeded it - because it was Watt's vastly more efficient engine which kickstarted the industrial revolution into high gear (about the only outfits which could afford to run a Neucomen engine were coal mines. For everyone else the fuel costs were simply too high to consider). LFTR has the potential to be the same kind of game changer that Watt's engine was.
It's worth noting that if coal plants were regulated like nuclear ones are, they'd all be shut down tomorrow due to excess radiation emissions and lax safety systems. Coal plants emit more radioactive materials (mainly radium) into the atmosphere each year than several Chernobyl events.
"Pinatubo's eruption more than offset the temp rise of the past 50 years all by itself and for up to 3 years."
You'd need more than just a few explosive charges around it to set it off again.
The amount of energy involved in a volcanic eruption even of the relatively small scale of Pinataubo is larger than the combined energies of all the worlds nuclear bombs dialled up to full power.
For that matter, a cold front passing over New Zealand's Southern Alps dissipates more energy than that.
We like to think we are masters of nature, but our best efforts are fairly puny compared to the day-to-day energies in the atmosphere or the lithosphere. At best, we can guide it, but only up to a point, as many river engineers have discovered. (When a river "decides" to change its course, nothing can stop it doing so)
I'm not confused. You need the generation where the demand is.
If you covered 1% of the sahara desert in solar panels, You'd lose 90% of the generated energy in transmission losses getting it to where it's needed - and that's without even going into the issue that any such transmission line project would be the largest single engineering endeavour in human history.
You'd also pretty much guarantee that the burgeoning environmental disaster around existing chinese solar PV factories will result in the deaths and/or forced relocation of at least 50 million people thanks to the watershed becoming unusable.
But let's say we take your 1% figure at face value. For starters it's only available for 10 hours/day (solar insolation is effectively zero until the sun is substantially above the horizon) so you need a storage system for the off hours and that has around 35% efficiency, so your space requirement just increased to a little over 3%. At this point you're still only matching existing energy demands, so to handle future expansion due to the drive to eliiminate as much carbon as possible, that's now 30%
Now factor in the issue that 90% of the world's existing electricity supply is consumed by 20% of the population. The other 80% of the planet's population need access to cheap energy in order to lift themselves out of poverty (richer people have fewer children) and stop burning so much biomass (deforestation) or oil (CO2...). That just pushed the requirement up by a factor of 5 again - or 150% of the Sahara's area - and whilst the vast majority of the Sahara is rocky desert floor (not dunes), it's still not suitable territory to build a solar PV plant because quite simply there's not that much "flat" space you can build on.
"Solar thermal" you say? - fine, but experience has already shown that the molten salt systems in the USA are using a _lot_ more gas than was originally planned to keep them warm enough overnight that they're usable in the morning (Deserts get COLD overnight. It routinely goes well below freezing in the Sahara each night despite being equatorial - and you need to account for the violent temperature swings in your environmental designs), such that the savings are only 25% over a gas-fired CCGT plant - and the point of alternative energy generation is that we need to _eliminate_ carbon emission, not just reduce it a little. Therefore if you need to keep the salt molten it's much easier to dispense with solar heating and use nuclear technology (LFTRs), at which point your location becomes pretty much irrelevant and the multi-trillion dollar, 100-year engineering effort distributing energy from an isolated source to the population centres is no longer necessary.
Bear in mind also that we don't have 100 years to implement this. CO2 levels have spiked so high, so quickly that an anoxic event seems on the cards.
I work with climate modellers. The factor of an anoxic event hasn't even crossed their collective radar. They're still focussed on ice melt and thermal levels changing sea levels (the ones I've spoken to don't even want to look at the possibility of an anoxic event on top of what they already see as impending catastrophic changes(*) - what's stated publicly is substantially toned down to avoid political attacks and accusations of scaremongering).
My opinion:
Judging from the geological record there's a fairly good chance that sea level rise doesn't particularly matter because the civilisations which will be flooded out would have ceased to exist by the time levels rise enough to flood them out - and in the worst case scenario every terrestrial animal species over ~45kg could be wiped out. With our big oxygen-hungry brains we are especially vulnerable to drops in atmospheric O2 levels, especially if it drops below 16% as it has in past events.
I hope I'm wrong, but we really have done a bang-up job of poisoning our biosphere. The planet and "nature" don't care. They'll carry on regardless, but we may not.
On top of that it's frequently the case that recording anywhere inside a courthouse (not just the courtrooms) is a criminal offence - even in countries with otherwise very open recording laws.
" Turns out it was overheating due to a bad fan, but nothing WARNED me about this"
I have specific daemons on my linux laptop to warn of overheating CPUs. Temperature monitoring in Windows is available but virtually noone makes use of it.
"Short answer: the coolest drives are 21.92 Celcius and the hottest drive was 30.54 degrees."
Based on the Google stats from a few years ago it was pretty clear that drive temperature was only a problem above 55C
I target 45C as allowable maximum and 35C as normal with no apparent increase in mortality over colder temperatures, but that saves a lot in terms of running the cooling plant. The batches of Seagates Constellations we had with stupidly high failure rates ran well under 30C
For home use my fileserver's drives have peaked out at 50C in hot weather (AC is rare in a UK house) and those drives currently have 42-48,000 hours on them (5-6 years) so it doesn't seem to have affected their reliability. OTOH Seagate ST2000DM drives in the same fileserver lasted less than 9 months (The DL series tended to run for 3 years before failing)
Based on years of experience the best advice for RAID work I can offer is "Don't use a raid composed of the same make/model drives and if you must do that, then FFS try to ensure the drives come from different batches. Otherwise one drive failure is the only warning you have of impending array doom" (RAIDZ3 is good, but multiple drive failures within the same batch is still a high risk thing)
"Sorry to hear about your loss. I hope you kept backup copies"
I haven't had a drive failure result in data loss in more than a decade. The backup tapes only tend to be pulled because someone deleted something critical and then went "oops"
It's still annoying and time consuming to replace the things. I have better things to do and so do my staff.
You need to temper the HGST fanboiism with the knowledge that the chinese competition authority have finally allowed WD to merge HGST into the mothership.
Whilst that means some HGST tech will show up in WD equipment, history shows that it is far more likely to mean that WD kit will start showing up with a HGST sticker on it. I regarded the $250k of HGST 4TB drives I purchased late last year as the last chance we had to obtain unsullied units.
Seagate was allowed to merge in the Toshiba unit they've been forced to operate as a separate unit too, with the same caveats being applied to any preference for Toshiba based on past reliability.
My personal opinion is that the chinese held off allowing those mergers (creating a world HDD duopoly in the process) until they decided SSDs are finally cheap enough to challenge them.
"Is there a real mechanical difference between "consumer" and "enterprise" drives, these days, at the bleeding edge of the storage-per-unit curve?"
Mechanically, no. But enterprise drives frequently have better ECC/processors and are programmed to be allowed to draw more power, which translates to more force being able to be used in the arm and better tracking (in many cases the CPU and driver chips are the the same, it's just the power constraints which are changed in firmware, and they might have larger onboard caps.)
That said, Seagate enterprise drives are just as trashy as their consumer cousins and have equally high failure rates - We had a 350% failure rate on our fleet of Seagate Constellations in raid arrays over their warranty period as one example.
5400 vs 7200rpm only makes any real difference for sequential reads/writes. As soon as you start seeking there's virtually nothing to choose between them in terms of IOPs (it's usually something like 115 vs 125 IOPs at best)
This must be getting close to the Last Hurrah for spinning drives in the enterprise, considering that read-optimised enterprise 4TB SSDs are now around 10x the price of 4TB consumer drives and still falling fast, with performance that blows spinning drives away no matter how you measure it.
"RAID friendly drives react quicker to failures, meaning they are less likely to fail the RAID over correctable error"
Your're referring to TLER - which used to be a tunable value until Seagate/WD started using it to differentiate enterprise/domestic drives (it dictates how hard a drive will try to recover sector errors before marking them bad and moving on)
On the other hand to your example, if you put a RAID-friendly drive in standalone use and there's a sector issue you're far more likely to lose data.
It would be interesting to know if the TLER is tuneable on these drives (it isn't on lower capacity STx000DM-x drives), but given a 200%+ failure rate in the warranty period on Seagates's DM001 drives (2 and 3TB) I would still be very wary.
" And I can't say that I recall hearing about any particular health problems, not that that necessarily means anything."
Health problems in the exclusion zone are those you'd expect to see with a lack of medical care.
Similarly the documented problems of chernobyl firefighters are more down to them being treated as pariahs and and denied decent medical care/accomodation than radiation exposure.
Interesting stat: When enhanced thyroid cancer screening was introduced in local areas in the wake of Chernobyl and Fukushima, detection rates went up 10-20fold. However exactly the same increase was seen in Korea when they introduced enhanced screening systems _before_ Fukushima happened. Correlation does not imply causality.
Gamma/Beta/Alpha radiation seldom causes cancer. It generally kills cells due to the energies involved.
On the other hand, various chemical toxins, ultraviolet light and heat are well proven to be carcinogenic to various extents. The cancer rates around Love Canal and Minimata Bay, vs those around Oak Ridge and downwind of the atmospheric nuke tests in Nevada are ample testament to that.
Why? Chernobyl isn't that hot. The remaining reactors were (and are) staffed by people right up to their shutdown. Unless you're planning on eating some of the Ukrainian fungii, you're perfectly safe in almost all areas.
Cold war mentality at work there. The reality of multi-megaton weapons was gone by the 1970s. Almost all nuclear weapons are under 1000kT and the vast majority of those are well under 100kT (most tactical and strategic weapons are dialable from 5-150kT, with some (depth charges) being as low as 0.3kT)
Nuclear winter was/is an unlikely scenario from the late 1960s onwards, but it was a hell of a good way of scaring the civilians into submission.
That said: As of 1979 no USA military commander authorised to use nuclear weapons in a battlefield simulation had ever done so twice. They'd play every possible tactic in order to avoid using them in subsequent simulations including unconditional surrender. Apparently it's better to be red than dead.
Even in close proximity to the plant the hotspots are pretty easily characterised and dealt with if there was a will to do so - mushrooms in particular have been shown to concentrate the contaminants.
The entire Chernobyl exclusion zone is about as radioactive as the Yorkshire Dales and less so than downtown Helsinki (granite) or Denver (altitude). For that matter the Yorkshire Dales are several times more radioactive than the entire Fukushima exclusion zone (including the contaminated water tanks) except for the areas inside the reactor buildings immediately adjacent to the damaged reactors.
Radioactive spots are easy to detect and cleanup from a distance. They also decay with time. Chemical poisons like mercury, lead, arsenic and depleted uranium(*) are much harder to detect and are around forever.
(*) DU oxide is as dangerous as lead oxides from a biological point of view. No radiation needed. When you see kids playing all over dead tanks in Iraq or people growing crops around the things, you should be worried.
The big folly of renewables plants is that if you carpeted the earth with them you could just about match overall existing electrical generation capacity, but current electricity generation only accounts for about 40% of global carbon emissions at best. By the time you eliminate carbon for heating and for most terrestrial transport systems your electrical requirements will increase by a factor of at least 10.
Add in the requirement for developing countries to have adequate electricity for developmment (the way to encourage poor people to have fewer children is to left them out of poverty) and you really start seeing why renewables are a diversion (mostly farming "subsidies" - not economically producing electricity, even compared to expensive nuclear systems) from the real issue - which is getting safer(**) nuclear systems online as fast as possible before we trigger an anoxic oceanic dieoff. (Some would say this has already started)
(**) Safer than we have now - molten salt systems are several orders of magnitude safer simply on the basis that you can't have a radioactive steam explosion, let alone a hydrogen one or the usual issues with nuke plants venting radioactive steam (TMI) or water (happens regularly). The problem with current nuclear technology isn't so much the nuke part, it's the water that almost all reactors are immersed in.
"The only way to build a solar power plant in Sahara and use the electricity to power Europe "
Would be not to do it. There's at least as much demand south of the Sahara as there is north of it and distribution lines are somewhat easier to run over land (the largest submarine connectors in existence are only about 2GW). On top of that it would be easy to characterise exporting electricity to europe as yet more colonial imperialism and theft from Africa.
Virtually no ad blockers will filter 1st party advertising (ie, adverts directly from the site you're viewing).
The problem isn't malvertising itself, it's that companies which used to closely vet what kind of ads went into their print/video/audio media are passing off the responsibility to 3rd parties who have repeatedly proven they aren't up to the task.
IE: malvertising is asymptom of the security problem, not the cause.
They also don't account for enough of the missing mass.
The principal component of the Solar System is the Sun, a G2 main-sequence star that contains 99.86% of the system's known mass and dominates it gravitationally. The Sun's four largest orbiting bodies, the giant planets, account for 99% of the remaining mass, with Jupiter and Saturn together comprising more than 90%. (https://en.wikipedia.org/wiki/Solar_System)
Planets really are down in the noise of measurements by most standards, even superjupiters.
Slashdot Mirror
"Solar and wind is very cheap in Developing countries where Electric outages are simply a matter of fact. In the Developed world outages are largely unacceptable"
My wife works in a developing country. One of the primary handbrakes on industry there (and the ability to move up the development chain) is the unreliability of the power - and the shortage of it. Peak demand is at least 4 times higher than the supply capacity, resulting in rolling blackouts being the norm simply to stop the distribution system burning out.
The political decision to solve this? A 600MW coal plant built in the poorest section of the largest city (which is now almost a decade behind schedule as various shortcomings due to poor oversight are discovered and rectified), along with several other large coal plants in other cities. The immediate-area environmental effects will be bad and it doesn't solve the issue that we need wean ourselves off carbon burning as a matter of global priority.
"And yet we never hear of solving the climate change problem by making those poor countries into wealthy countries"
That depends who you talk to. Various climate change researchers have made this very proposal. The problem is that to get the required energy density needed, you _need_ nuclear plants and everyone's still doing the "oooh, radiation is Baaaaaad" routine.
Wind and Solar are intermittent and you need to overbuild by a factor of _at least_ 4 in order to maintain average output. The numbers quoted above are based on nameplate generation figures. In addition to that you need to geographically disperse your generation plant and overbuild your power distribution network by a substantial factor to handle power flows in relatively unpredictable directions. These figures are never included in the calculations of wind/solar output.
Nuclear tech is vastly overpriced due to the wildly inefficient (and fairly dangerous) technologies currently in use (60-70% of mined uranium is discarded before hitting the reactor. only 1% of the fuel which enters the reactor is currently used. The only positive about nuclear waste from this tech is that 60 years worth of waste will comfortably fit in an olympic-sized pool and is safe enough to handle in 300 years)
If/when molten salt tech hits commercial viability it should drop the cost by an order of magnitude simply because all the radiological safety issues imposed by having water in direct contact with nuclear materials (radioactive steam, hydrogen explosions, etc) will be eliminated. The factor of being able to "burn" existing "high level waste", not having massive supply inefficiencies (thorium is a waste product of rare earth mining 10 times more abundant that Uranium) and being able to use at least 90% of the input fuel are all just extra features.
I'll keep making the comparison that Thorium LFTR technology is akin to Watt's improved steam engine vs the Neucomen engines (current pressurised/boiling water tech) that preceeded it - because it was Watt's vastly more efficient engine which kickstarted the industrial revolution into high gear (about the only outfits which could afford to run a Neucomen engine were coal mines. For everyone else the fuel costs were simply too high to consider). LFTR has the potential to be the same kind of game changer that Watt's engine was.
It's worth noting that if coal plants were regulated like nuclear ones are, they'd all be shut down tomorrow due to excess radiation emissions and lax safety systems. Coal plants emit more radioactive materials (mainly radium) into the atmosphere each year than several Chernobyl events.
"Pinatubo's eruption more than offset the temp rise of the past 50 years all by itself and for up to 3 years."
You'd need more than just a few explosive charges around it to set it off again.
The amount of energy involved in a volcanic eruption even of the relatively small scale of Pinataubo is larger than the combined energies of all the worlds nuclear bombs dialled up to full power.
For that matter, a cold front passing over New Zealand's Southern Alps dissipates more energy than that.
We like to think we are masters of nature, but our best efforts are fairly puny compared to the day-to-day energies in the atmosphere or the lithosphere. At best, we can guide it, but only up to a point, as many river engineers have discovered. (When a river "decides" to change its course, nothing can stop it doing so)
I'm not confused. You need the generation where the demand is.
If you covered 1% of the sahara desert in solar panels, You'd lose 90% of the generated energy in transmission losses getting it to where it's needed - and that's without even going into the issue that any such transmission line project would be the largest single engineering endeavour in human history.
You'd also pretty much guarantee that the burgeoning environmental disaster around existing chinese solar PV factories will result in the deaths and/or forced relocation of at least 50 million people thanks to the watershed becoming unusable.
But let's say we take your 1% figure at face value. For starters it's only available for 10 hours/day (solar insolation is effectively zero until the sun is substantially above the horizon) so you need a storage system for the off hours and that has around 35% efficiency, so your space requirement just increased to a little over 3%. At this point you're still only matching existing energy demands, so to handle future expansion due to the drive to eliiminate as much carbon as possible, that's now 30%
Now factor in the issue that 90% of the world's existing electricity supply is consumed by 20% of the population. The other 80% of the planet's population need access to cheap energy in order to lift themselves out of poverty (richer people have fewer children) and stop burning so much biomass (deforestation) or oil (CO2...). That just pushed the requirement up by a factor of 5 again - or 150% of the Sahara's area - and whilst the vast majority of the Sahara is rocky desert floor (not dunes), it's still not suitable territory to build a solar PV plant because quite simply there's not that much "flat" space you can build on.
"Solar thermal" you say? - fine, but experience has already shown that the molten salt systems in the USA are using a _lot_ more gas than was originally planned to keep them warm enough overnight that they're usable in the morning (Deserts get COLD overnight. It routinely goes well below freezing in the Sahara each night despite being equatorial - and you need to account for the violent temperature swings in your environmental designs), such that the savings are only 25% over a gas-fired CCGT plant - and the point of alternative energy generation is that we need to _eliminate_ carbon emission, not just reduce it a little. Therefore if you need to keep the salt molten it's much easier to dispense with solar heating and use nuclear technology (LFTRs), at which point your location becomes pretty much irrelevant and the multi-trillion dollar, 100-year engineering effort distributing energy from an isolated source to the population centres is no longer necessary.
Bear in mind also that we don't have 100 years to implement this. CO2 levels have spiked so high, so quickly that an anoxic event seems on the cards.
I work with climate modellers. The factor of an anoxic event hasn't even crossed their collective radar. They're still focussed on ice melt and thermal levels changing sea levels (the ones I've spoken to don't even want to look at the possibility of an anoxic event on top of what they already see as impending catastrophic changes(*) - what's stated publicly is substantially toned down to avoid political attacks and accusations of scaremongering).
My opinion:
Judging from the geological record there's a fairly good chance that sea level rise doesn't particularly matter because the civilisations which will be flooded out would have ceased to exist by the time levels rise enough to flood them out - and in the worst case scenario every terrestrial animal species over ~45kg could be wiped out. With our big oxygen-hungry brains we are especially vulnerable to drops in atmospheric O2 levels, especially if it drops below 16% as it has in past events.
I hope I'm wrong, but we really have done a bang-up job of poisoning our biosphere. The planet and "nature" don't care. They'll carry on regardless, but we may not.
On top of that it's frequently the case that recording anywhere inside a courthouse (not just the courtrooms) is a criminal offence - even in countries with otherwise very open recording laws.
" Turns out it was overheating due to a bad fan, but nothing WARNED me about this"
I have specific daemons on my linux laptop to warn of overheating CPUs. Temperature monitoring in Windows is available but virtually noone makes use of it.
"Short answer: the coolest drives are 21.92 Celcius and the hottest drive was 30.54 degrees."
Based on the Google stats from a few years ago it was pretty clear that drive temperature was only a problem above 55C
I target 45C as allowable maximum and 35C as normal with no apparent increase in mortality over colder temperatures, but that saves a lot in terms of running the cooling plant. The batches of Seagates Constellations we had with stupidly high failure rates ran well under 30C
For home use my fileserver's drives have peaked out at 50C in hot weather (AC is rare in a UK house) and those drives currently have 42-48,000 hours on them (5-6 years) so it doesn't seem to have affected their reliability. OTOH Seagate ST2000DM drives in the same fileserver lasted less than 9 months (The DL series tended to run for 3 years before failing)
Based on years of experience the best advice for RAID work I can offer is "Don't use a raid composed of the same make/model drives and if you must do that, then FFS try to ensure the drives come from different batches. Otherwise one drive failure is the only warning you have of impending array doom" (RAIDZ3 is good, but multiple drive failures within the same batch is still a high risk thing)
"Sorry to hear about your loss. I hope you kept backup copies"
I haven't had a drive failure result in data loss in more than a decade. The backup tapes only tend to be pulled because someone deleted something critical and then went "oops"
It's still annoying and time consuming to replace the things. I have better things to do and so do my staff.
You need to temper the HGST fanboiism with the knowledge that the chinese competition authority have finally allowed WD to merge HGST into the mothership.
Whilst that means some HGST tech will show up in WD equipment, history shows that it is far more likely to mean that WD kit will start showing up with a HGST sticker on it. I regarded the $250k of HGST 4TB drives I purchased late last year as the last chance we had to obtain unsullied units.
Seagate was allowed to merge in the Toshiba unit they've been forced to operate as a separate unit too, with the same caveats being applied to any preference for Toshiba based on past reliability.
My personal opinion is that the chinese held off allowing those mergers (creating a world HDD duopoly in the process) until they decided SSDs are finally cheap enough to challenge them.
"Is there a real mechanical difference between "consumer" and "enterprise" drives, these days, at the bleeding edge of the storage-per-unit curve?"
Mechanically, no. But enterprise drives frequently have better ECC/processors and are programmed to be allowed to draw more power, which translates to more force being able to be used in the arm and better tracking (in many cases the CPU and driver chips are the the same, it's just the power constraints which are changed in firmware, and they might have larger onboard caps.)
That said, Seagate enterprise drives are just as trashy as their consumer cousins and have equally high failure rates - We had a 350% failure rate on our fleet of Seagate Constellations in raid arrays over their warranty period as one example.
5400 vs 7200rpm only makes any real difference for sequential reads/writes. As soon as you start seeking there's virtually nothing to choose between them in terms of IOPs (it's usually something like 115 vs 125 IOPs at best)
This must be getting close to the Last Hurrah for spinning drives in the enterprise, considering that read-optimised enterprise 4TB SSDs are now around 10x the price of 4TB consumer drives and still falling fast, with performance that blows spinning drives away no matter how you measure it.
"Use dirt cheap commodity products that are good quality, have exceptional redundancy, throw them away as they implode."
Quite simply, there are two philosophies to consider:
One is to try and build everything as robust as possible and make it as reliable as possible at component/system level. (expensive hardware)
The other is simply to say "Everything fails sooner or later. Design accordingly" (complex software)
The best cost/benefit approach lies somewhere within the 2 extremes.
"RAID friendly drives react quicker to failures, meaning they are less likely to fail the RAID over correctable error"
Your're referring to TLER - which used to be a tunable value until Seagate/WD started using it to differentiate enterprise/domestic drives (it dictates how hard a drive will try to recover sector errors before marking them bad and moving on)
On the other hand to your example, if you put a RAID-friendly drive in standalone use and there's a sector issue you're far more likely to lose data.
It would be interesting to know if the TLER is tuneable on these drives (it isn't on lower capacity STx000DM-x drives), but given a 200%+ failure rate in the warranty period on Seagates's DM001 drives (2 and 3TB) I would still be very wary.
What do you call the clusterfuck of pollution around the chinese solar PV plants?
There's serious potential for watersheds servicing something north of 100 million people being rendered undrinkable.
" And I can't say that I recall hearing about any particular health problems, not that that necessarily means anything."
Health problems in the exclusion zone are those you'd expect to see with a lack of medical care.
Similarly the documented problems of chernobyl firefighters are more down to them being treated as pariahs and and denied decent medical care/accomodation than radiation exposure.
Interesting stat: When enhanced thyroid cancer screening was introduced in local areas in the wake of Chernobyl and Fukushima, detection rates went up 10-20fold. However exactly the same increase was seen in Korea when they introduced enhanced screening systems _before_ Fukushima happened. Correlation does not imply causality.
"so who cares if they'll develop cancer"
Gamma/Beta/Alpha radiation seldom causes cancer. It generally kills cells due to the energies involved.
On the other hand, various chemical toxins, ultraviolet light and heat are well proven to be carcinogenic to various extents. The cancer rates around Love Canal and Minimata Bay, vs those around Oak Ridge and downwind of the atmospheric nuke tests in Nevada are ample testament to that.
"I suspect you're not going to get too many people willing to staff a nuclear power plant 24/7 in the Exclusion Zone,"
You'd suspect wrong. There are staff on the reactor sites 24*7 and have been all along (they're bussed in/out from Kiev)
"such a place would have to be remotely managed"
Why? Chernobyl isn't that hot. The remaining reactors were (and are) staffed by people right up to their shutdown. Unless you're planning on eating some of the Ukrainian fungii, you're perfectly safe in almost all areas.
Cold war mentality at work there. The reality of multi-megaton weapons was gone by the 1970s. Almost all nuclear weapons are under 1000kT and the vast majority of those are well under 100kT (most tactical and strategic weapons are dialable from 5-150kT, with some (depth charges) being as low as 0.3kT)
Nuclear winter was/is an unlikely scenario from the late 1960s onwards, but it was a hell of a good way of scaring the civilians into submission.
That said: As of 1979 no USA military commander authorised to use nuclear weapons in a battlefield simulation had ever done so twice. They'd play every possible tactic in order to avoid using them in subsequent simulations including unconditional surrender. Apparently it's better to be red than dead.
"That's only in close proximity to the plant"
Even in close proximity to the plant the hotspots are pretty easily characterised and dealt with if there was a will to do so - mushrooms in particular have been shown to concentrate the contaminants.
The entire Chernobyl exclusion zone is about as radioactive as the Yorkshire Dales and less so than downtown Helsinki (granite) or Denver (altitude). For that matter the Yorkshire Dales are several times more radioactive than the entire Fukushima exclusion zone (including the contaminated water tanks) except for the areas inside the reactor buildings immediately adjacent to the damaged reactors.
Radioactive spots are easy to detect and cleanup from a distance. They also decay with time. Chemical poisons like mercury, lead, arsenic and depleted uranium(*) are much harder to detect and are around forever.
(*) DU oxide is as dangerous as lead oxides from a biological point of view. No radiation needed. When you see kids playing all over dead tanks in Iraq or people growing crops around the things, you should be worried.
The big folly of renewables plants is that if you carpeted the earth with them you could just about match overall existing electrical generation capacity, but current electricity generation only accounts for about 40% of global carbon emissions at best. By the time you eliminate carbon for heating and for most terrestrial transport systems your electrical requirements will increase by a factor of at least 10.
Add in the requirement for developing countries to have adequate electricity for developmment (the way to encourage poor people to have fewer children is to left them out of poverty) and you really start seeing why renewables are a diversion (mostly farming "subsidies" - not economically producing electricity, even compared to expensive nuclear systems) from the real issue - which is getting safer(**) nuclear systems online as fast as possible before we trigger an anoxic oceanic dieoff. (Some would say this has already started)
(**) Safer than we have now - molten salt systems are several orders of magnitude safer simply on the basis that you can't have a radioactive steam explosion, let alone a hydrogen one or the usual issues with nuke plants venting radioactive steam (TMI) or water (happens regularly). The problem with current nuclear technology isn't so much the nuke part, it's the water that almost all reactors are immersed in.
"The only way to build a solar power plant in Sahara and use the electricity to power Europe "
Would be not to do it. There's at least as much demand south of the Sahara as there is north of it and distribution lines are somewhat easier to run over land (the largest submarine connectors in existence are only about 2GW). On top of that it would be easy to characterise exporting electricity to europe as yet more colonial imperialism and theft from Africa.
HTML5 doesn't need Javascript. Or Java.
Virtually no ad blockers will filter 1st party advertising (ie, adverts directly from the site you're viewing).
The problem isn't malvertising itself, it's that companies which used to closely vet what kind of ads went into their print/video/audio media are passing off the responsibility to 3rd parties who have repeatedly proven they aren't up to the task.
IE: malvertising is asymptom of the security problem, not the cause.
> Planets aren't "visible"
They also don't account for enough of the missing mass.
The principal component of the Solar System is the Sun, a G2 main-sequence star that contains 99.86% of the system's known mass and dominates it gravitationally. The Sun's four largest orbiting bodies, the giant planets, account for 99% of the remaining mass, with Jupiter and Saturn together comprising more than 90%. (https://en.wikipedia.org/wiki/Solar_System)
Planets really are down in the noise of measurements by most standards, even superjupiters.
"For reasons that aren't clear to me, most UK ATMs have a noticeable delay between inserting the card and letting you enter your PIN."
They had that in the days of mag stripes too.