Not rare but they produce a ton of toxic waste being processed. It all went to China because they could do it cheaper and eat the toxic waste, too.
When China were embargoing exports (or talking about it) there was talk of granting exceptions to a closed mine in California to re-open as a strategic hedge.
I think taking out Kim is probably easier than Trump or Putin simply because his security and military defenses aren't able to stop some threats, like a cruise missile strike or some kind of air strikes. His secretive nature, though, probably contradicts this advantage -- you have to know where he is to hit him successfully.
The problem is they've had 60-odd years to multiply and dig-in their artillery emplacements capable of hitting Seoul. I don't think there's a scenario where Seoul doesn't experience significant damage, no matter how effective counter-battery or airstrikes are at silencing those guns. Even pessimistically, 1500 guns getting off 10 rounds each is a lot of artillery strikes for a modern city to absorb. The ability to hit Seoul with artillery is a greater deterrent than nukes, and really Kim should have invested in dumb rocket launchers and even more artillery.
I think the only way Seoul escapes is some kind of decapitation strike that kills Kim and his immediate circle so convincingly that the rest of his military doesn't react and surrenders.
IMHO, this isn't entirely far-fetched -- my speculation is that in a country so paranoid, field commanders are scared witless and almost trained *not* to make decisions. I would question how many of them are existentially committed to fighting to the bitter end to protect and restore a new Kim-style dynasty in the event of the inner circle's untimely demise.
Trouble is, that decapitation strike is tactically difficult and the consequences of not doing it perfectly.
In one way, this could help religion by providing people with a religious experience or even a very realistic recreation of religious events, connecting people to the origins and mythology.
In another way, religions could view it as a threat. I've read several times that psychedelic drugs were often suppressed by religions because they provided people with a transcendental experience not controlled by the religion. I can see someone producing a slick VR religion program that's not endorsed or controlled by mainstream religions being seen as a big threat.
I think some of the issues with smartphones is somehow tied to both the close proximity of the device and the fact that it involves an additional sensory interaction -- the touch screen, which I think is somehow different than a simple button.
I'm sort of willing to grant them leeway on the batteries, especially if its a mod that uses 18650s. Finding legitimate ones isn't easy. I actually bite the bullet and buy them from established brick and mortar vape shops because of all the fakes.
But even with the fixed batteries most common in the 50-100 watt regulated mods I'm sorta skeptical, especially with vape gear brand identity kind of sketchy outside of a handful of brands that have been around (Kanger, Smok, etc).
My guess is that it's a result of the endless race to see who can pump the most watts into a sub-ohm coil with the cheapest cells they can find. There's a point at which it turns into something like shorting out a lithium battery.
Whether it's worthwhile seems to come down to something like how many square meters of collector does it take to create 300 liters of water per day and is it all done with just solar power or does it require extra inputs and how much does it cost.
Management's policy process lack's the knowledge and (ugh, sorry) "agility" to adapt what they want to the ever-changing landscape of what and where data is and how it's accessed.
Whatever policy is on paper is likely woefully vague or out of date relative to technology. Much of the time the organization itself is willfully non-compliant as various centers within the company store and access data in various public clouds, social media sites, on personal or hand-held devices, and so on.
Even when everyone kind of has their shit together, the technology industry is subverting "corporate data" by turning themselves into personal technology companies, like Apple and now Microsoft, where they've figured out that if you sell to the individual end users as consumers you can essentially *make* corporations support (and sometimes buy) your product.
1) Cut off the spine and sheet feed books that were common enough that a copy could be destroyed in the process (or hell, it could have been rebound and resold -- perhaps they could have advanced a technique for unbinding-rebinding, too).
2) Automate page flipping for books that couldn't be spine-cut or sheet fed. This seems like it would have been a major side benefit of this project, a page flipping book scanner. It would seem like something that could almost be a niche product business unto itself -- I'd imagine there are many organizations or even people with stacks of bound materials they would like scanned but where the reality of scanning bound material in traditional ways is just too labor intensive.
3) Some small number of books may need to manually scanned because they are fragile or rare, but I'd think automated page flipping would cover the vast majority.
I thought the summary was a little bit hyped, but in my opinion municipal data networks aren't inherently anti-competitive and actually encourage competition.
Let some free-standing but government-owned corporation run the municipal network at layer 2 and lower. Mandate they run at only enough profit level to cover their operating costs and future investment/expansion. They provide nothing more than link layer connectivity and operate a CO data center open to all comers -- schools, businesses, for-profit companies, etc, that want to sell data services to subscribers.
In short, regulate the shit out of them but make them independent enough that they can't be easily bullied politically. It might even make sense to bid running the entire infrastructure out to some private company for a 5-10% profit ceiling with all the expected rewards and penalties for mismanagement or cost overruns.
At this point, there is no real state "competition" to business. No business wants to compete on last mile connectivity -- it's a money loser for sure, and its why so few businesses, including Google, have made much of an effort or traction in building out new last mile networks in the face of existing last mile networks. Comcast, et al, only want to be in that business to the extent its a practical monopoly to extract rents from.
In a municipal network, the data providers (ISPs, video, phones) whatever you can pump over a data network can actually compete. Of course Comcast hates this because now they're having to justify a $100 TV package versus $50 for a half-dozen streaming accounts. No rents to be extracted for an inferior product.
At least a human has, somewhere in it's limbic system, some evolutionary instinct that says if you hoard all of the food and don't kill all the other people, you will either get your head bashed in by your own tribe or you will lack a tribe at all and the next tribe over the hill will bash your head in.
Of course AI's answer to that problem is Skynet, so there's that.
What's kind of stupid is how we used to have that very same network, but it was called the phone system.
I'm curious if any telephone system, especially smaller ones if they still exist (county or regional co-ops), actually continuously upgraded their networks to the point where they were running a FTTH home network.
IE, phase one would have been a skeleton fiber network to neighborhood-level DSLAMs to provide as-good-as-it-got DSL over the existing copper runs to houses. And then over time doing the harder work of distributing fiber to the individual address level (perhaps still finishing this phase), but using the skeleton fiber network for the DSLAMs as the uplink.
A common CO would allow anyone to buy into the network at the top -- ISP, etc, and then provide a data channel to their subscribers. The school system could be a CO provider, maybe even some businesses with multiple sites could be CO providers using the municipal network as their WAN.
But somehow, despite being literally decades ahead of anyone else, the phone companies just didn't do this. They got to DSL, and mostly local exchange only, so lots of pairs were too long or too shitty, for more than awful throughput. Or they milked their T1 business to business areas not yet wired for cable.
But it's like they had every chance to continuously upgrade and literally own the network but they didn't. They got their capital drained dry in mergers and acquisitions and are now a spent shell, like the bad debt of a failed bank.
A container is a package containing your application and designating how much of which version of the OS, libraries, file system, utilities, etc. it gets to see. It looks to the app like it's running on its own little machine, just like in a VM. But it's actually running (along with everything else) under the native Linux kernel, which is using several compartmentalization mechanisms to give the app its own, limited and tuned, view of task numbers, file system, tables, etc.
This is where I get lost with containers vs. virtualization. How does a container choose what version of the OS it gets to use if it runs under a given OS? The library aspects I think I get, assuming you're able to install multiple copies of the libraries or apps in question in the OS.
Or is that the part I don't get -- it's more like an app build process, where you essentially compile the app and install its binaries and linked libraries, including system libraries into the container? I guess this makes sense, but then I don't get how you're able to obtain OS portability for containers without essentially throwing every bit of the OS into the container it might need. Or do they not have OS portability, and the container is more or less locked to the OS it was built under?
At some point I'm curious how containers aren't just basically a method of obtaining what amounts to a statically linked binary with FS jails and networking baked into the container host.
My guess is they have some model that says being "proactive" reduces scams by $x but has a side effect of reduced $y legitimate sales, too, along with the risk of some big negative publicity when a legitimate seller has his account cancelled or something.
I don't know, but I suspect a major growth sector for Amazon is basically competing with the flea market over at Ebay and not creating a bunch of ill will against that type of seller means something to them.
IMHO, Amazon should have fewer flea market sellers and the crappy intermingling of substandard SKUs, but apparently Amazon doesn't think so.
I do give them credit, though, I've sold one thing through Amazon (spare, unused and unopened 512GB SSD) and it was a pretty slick process.
I gave up on a miles credit card when the only tickets I could get were on 3-layover flights to Omaha in February. The airlines took all the value out of the miles.
A switch to US Bank was a slight improvement -- their rewards points could buy tickets about 2/3rds of the value of the way it used to be with airline points, or you could buy stupid shit from the catalog.
Switching to the Costco card was the best deal ever -- the reward rates are great, and I get $400 in cash nearly every year.
The solar installs I've seen on boats amounted to flush mounted over cockpits or other roofline enclosures. Nothing to catch the wind.
Other than the one Greenline with 1.3kw solar, for most purposes solar on a boat is only really useful for offsetting DC consumption for house batteries -- a couple hundred watts -- not propulsion, unless you're doing some really unusual purpose-built solar boat.
Shore charging is fine, but you'd really want diesel generation for both higher running speeds than batteries can supply and for distance. Plus you'd need it anyway for house power.
I think a battery array install along the keel line in the bilge would be ideal, and a lot of larger fiberglass hulls end up with lead there anyway for ballast. They wind up being sort of tall for their typical running weight and more mass at the keel line in the bilge does a lot to help counter rolling at anchor and makes stabilizers much more effective.
I was thinking of a specific use case -- inland lake recreational marine. A plane is a different animal, as is ocean-running boats.
The lake I boat on is pretty big -- 22 square miles comprised of a dozen or so interlocking bays, but you would be hard pressed to run more than 30 miles of traveled distance in a single day.
For this specific application, I don't see why a single engine hybrid, diesel-battery-electric application wouldn't work. Half the boats on the lake are single engine and a total engine failure isn't exactly life threatening, either.
Most days we don't travel more than 10 miles, a range a Tesla-size battery could provide about 6 knots of cruise on a 30 foot hull pretty easily. Many slips have 30A service for charging and an on-board diesel generator for charging and diesel-electric propulsion would extend the range to fuel on board.
I'm only aware of one commercial option, Greenline Hybrid, which uses a single diesel and batteries but it puts the prop, combination motor/generator and diesel on a single shaft with a hydraulic clutch to operate as prime mover and generator as well as having about 6 hours of cruise at 6 knots on battery alone. They also have a model with 1.3 kW solar panels which they claim can actually drive the entire thing (albeit slowly, like 3 knots).
If I had the money I would be inclined to find an good hull with worn out engines and roll my own for this use, but see if I could make it pure electric drive with the diesel only for electric generation.
It was accessible to program and I think the fact that all it did was boot to a command prompt made it tempting to program, versus the endless distraction of a graphic/web environment of today.
Plus you could get down to actual programming logic immediately, without endless distractions of display management in a GUI.
I think this is part of the problem, the strong desire, masking as need, for relentless modularity and configuration that the vast majority of users don't care about. The result is a tangle of code that needs to be a tangle to keep satisfying a core group of influencers with obscure features that most people don't care about and don't know how to configure.
Meanwhile, people on the Windows side laugh and wonder what the big deal is, they have had a useful GUI forever and Remote Desktop solves remote execution (albeit differently than X), too.
It sure hasn't hit the recreational marine market, though. You can get pod drives, but they are mechanically driven off of diesel engines.
I always wondered if they could do a hybrid drive on a boat with a single engine and batteries driving electric motors. I'm only aware of one (newish) maker doing it, but I would think for a lot of the inland boating market where ranges under 25 miles and cruise speeds of 10 knots are common.
The batteries would be a wash if you removed one of the engines and less fuel might offset a heavier generator.
I'm guessing the turboshaft is meant to provide take-off power and it can cruise on batteries in a kind of winged flight mode? Otherwise I'm not seeing the advantage of a turboshaft given the energy losses associated with electric generation.
I think the general modem pool for the timeshare system used in CompSci 3104 might have been 1200 baud capable in 1986 at the University I attended. I know for a fact in high school it wasn't -- there were a handful of 1200 baud lines restricted to admin logins, and an admin I knew used to gripe a lot about wasted money on a Hayes 1200 modem that seldom could get the 1200 baud lines.
FWIW, I think the AppleCat had a Bell 212 "normal" 1200 baud option, but it was nearly twice the price of the non-212 version. And the usual problem was almost nothing was 1200 baud capable, especially BBSes.
The prices sure seemed to come down fast, though. Within about 5 years, it seemed like 14.4k was pretty cheap.
Slashdot Mirror
Not rare but they produce a ton of toxic waste being processed. It all went to China because they could do it cheaper and eat the toxic waste, too.
When China were embargoing exports (or talking about it) there was talk of granting exceptions to a closed mine in California to re-open as a strategic hedge.
I think taking out Kim is probably easier than Trump or Putin simply because his security and military defenses aren't able to stop some threats, like a cruise missile strike or some kind of air strikes. His secretive nature, though, probably contradicts this advantage -- you have to know where he is to hit him successfully.
The problem is they've had 60-odd years to multiply and dig-in their artillery emplacements capable of hitting Seoul. I don't think there's a scenario where Seoul doesn't experience significant damage, no matter how effective counter-battery or airstrikes are at silencing those guns. Even pessimistically, 1500 guns getting off 10 rounds each is a lot of artillery strikes for a modern city to absorb. The ability to hit Seoul with artillery is a greater deterrent than nukes, and really Kim should have invested in dumb rocket launchers and even more artillery.
I think the only way Seoul escapes is some kind of decapitation strike that kills Kim and his immediate circle so convincingly that the rest of his military doesn't react and surrenders.
IMHO, this isn't entirely far-fetched -- my speculation is that in a country so paranoid, field commanders are scared witless and almost trained *not* to make decisions. I would question how many of them are existentially committed to fighting to the bitter end to protect and restore a new Kim-style dynasty in the event of the inner circle's untimely demise.
Trouble is, that decapitation strike is tactically difficult and the consequences of not doing it perfectly.
In one way, this could help religion by providing people with a religious experience or even a very realistic recreation of religious events, connecting people to the origins and mythology.
In another way, religions could view it as a threat. I've read several times that psychedelic drugs were often suppressed by religions because they provided people with a transcendental experience not controlled by the religion. I can see someone producing a slick VR religion program that's not endorsed or controlled by mainstream religions being seen as a big threat.
I think some of the issues with smartphones is somehow tied to both the close proximity of the device and the fact that it involves an additional sensory interaction -- the touch screen, which I think is somehow different than a simple button.
I'm sort of willing to grant them leeway on the batteries, especially if its a mod that uses 18650s. Finding legitimate ones isn't easy. I actually bite the bullet and buy them from established brick and mortar vape shops because of all the fakes.
But even with the fixed batteries most common in the 50-100 watt regulated mods I'm sorta skeptical, especially with vape gear brand identity kind of sketchy outside of a handful of brands that have been around (Kanger, Smok, etc).
My guess is that it's a result of the endless race to see who can pump the most watts into a sub-ohm coil with the cheapest cells they can find. There's a point at which it turns into something like shorting out a lithium battery.
It made me so mad, I quit having sex with virgins!
Whether it's worthwhile seems to come down to something like how many square meters of collector does it take to create 300 liters of water per day and is it all done with just solar power or does it require extra inputs and how much does it cost.
I'm always amazed at the number of people on Instagram who have actual pictures of themselves standing in front of a mirror.
I just can't quite wrap my head around that level of narcissism.
Management's policy process lack's the knowledge and (ugh, sorry) "agility" to adapt what they want to the ever-changing landscape of what and where data is and how it's accessed.
Whatever policy is on paper is likely woefully vague or out of date relative to technology. Much of the time the organization itself is willfully non-compliant as various centers within the company store and access data in various public clouds, social media sites, on personal or hand-held devices, and so on.
Even when everyone kind of has their shit together, the technology industry is subverting "corporate data" by turning themselves into personal technology companies, like Apple and now Microsoft, where they've figured out that if you sell to the individual end users as consumers you can essentially *make* corporations support (and sometimes buy) your product.
Jeeze, no wonder it wasn't a huge success.
I would have assumed they could done:
1) Cut off the spine and sheet feed books that were common enough that a copy could be destroyed in the process (or hell, it could have been rebound and resold -- perhaps they could have advanced a technique for unbinding-rebinding, too).
2) Automate page flipping for books that couldn't be spine-cut or sheet fed. This seems like it would have been a major side benefit of this project, a page flipping book scanner. It would seem like something that could almost be a niche product business unto itself -- I'd imagine there are many organizations or even people with stacks of bound materials they would like scanned but where the reality of scanning bound material in traditional ways is just too labor intensive.
3) Some small number of books may need to manually scanned because they are fragile or rare, but I'd think automated page flipping would cover the vast majority.
I thought the summary was a little bit hyped, but in my opinion municipal data networks aren't inherently anti-competitive and actually encourage competition.
Let some free-standing but government-owned corporation run the municipal network at layer 2 and lower. Mandate they run at only enough profit level to cover their operating costs and future investment/expansion. They provide nothing more than link layer connectivity and operate a CO data center open to all comers -- schools, businesses, for-profit companies, etc, that want to sell data services to subscribers.
In short, regulate the shit out of them but make them independent enough that they can't be easily bullied politically. It might even make sense to bid running the entire infrastructure out to some private company for a 5-10% profit ceiling with all the expected rewards and penalties for mismanagement or cost overruns.
At this point, there is no real state "competition" to business. No business wants to compete on last mile connectivity -- it's a money loser for sure, and its why so few businesses, including Google, have made much of an effort or traction in building out new last mile networks in the face of existing last mile networks. Comcast, et al, only want to be in that business to the extent its a practical monopoly to extract rents from.
In a municipal network, the data providers (ISPs, video, phones) whatever you can pump over a data network can actually compete. Of course Comcast hates this because now they're having to justify a $100 TV package versus $50 for a half-dozen streaming accounts. No rents to be extracted for an inferior product.
At least a human has, somewhere in it's limbic system, some evolutionary instinct that says if you hoard all of the food and don't kill all the other people, you will either get your head bashed in by your own tribe or you will lack a tribe at all and the next tribe over the hill will bash your head in.
Of course AI's answer to that problem is Skynet, so there's that.
What's kind of stupid is how we used to have that very same network, but it was called the phone system.
I'm curious if any telephone system, especially smaller ones if they still exist (county or regional co-ops), actually continuously upgraded their networks to the point where they were running a FTTH home network.
IE, phase one would have been a skeleton fiber network to neighborhood-level DSLAMs to provide as-good-as-it-got DSL over the existing copper runs to houses. And then over time doing the harder work of distributing fiber to the individual address level (perhaps still finishing this phase), but using the skeleton fiber network for the DSLAMs as the uplink.
A common CO would allow anyone to buy into the network at the top -- ISP, etc, and then provide a data channel to their subscribers. The school system could be a CO provider, maybe even some businesses with multiple sites could be CO providers using the municipal network as their WAN.
But somehow, despite being literally decades ahead of anyone else, the phone companies just didn't do this. They got to DSL, and mostly local exchange only, so lots of pairs were too long or too shitty, for more than awful throughput. Or they milked their T1 business to business areas not yet wired for cable.
But it's like they had every chance to continuously upgrade and literally own the network but they didn't. They got their capital drained dry in mergers and acquisitions and are now a spent shell, like the bad debt of a failed bank.
A container is a package containing your application and designating how much of which version of the OS, libraries, file system, utilities, etc. it gets to see. It looks to the app like it's running on its own little machine, just like in a VM. But it's actually running (along with everything else) under the native Linux kernel, which is using several compartmentalization mechanisms to give the app its own, limited and tuned, view of task numbers, file system, tables, etc.
This is where I get lost with containers vs. virtualization. How does a container choose what version of the OS it gets to use if it runs under a given OS? The library aspects I think I get, assuming you're able to install multiple copies of the libraries or apps in question in the OS.
Or is that the part I don't get -- it's more like an app build process, where you essentially compile the app and install its binaries and linked libraries, including system libraries into the container? I guess this makes sense, but then I don't get how you're able to obtain OS portability for containers without essentially throwing every bit of the OS into the container it might need. Or do they not have OS portability, and the container is more or less locked to the OS it was built under?
At some point I'm curious how containers aren't just basically a method of obtaining what amounts to a statically linked binary with FS jails and networking baked into the container host.
My guess is they have some model that says being "proactive" reduces scams by $x but has a side effect of reduced $y legitimate sales, too, along with the risk of some big negative publicity when a legitimate seller has his account cancelled or something.
I don't know, but I suspect a major growth sector for Amazon is basically competing with the flea market over at Ebay and not creating a bunch of ill will against that type of seller means something to them.
IMHO, Amazon should have fewer flea market sellers and the crappy intermingling of substandard SKUs, but apparently Amazon doesn't think so.
I do give them credit, though, I've sold one thing through Amazon (spare, unused and unopened 512GB SSD) and it was a pretty slick process.
I gave up on a miles credit card when the only tickets I could get were on 3-layover flights to Omaha in February. The airlines took all the value out of the miles.
A switch to US Bank was a slight improvement -- their rewards points could buy tickets about 2/3rds of the value of the way it used to be with airline points, or you could buy stupid shit from the catalog.
Switching to the Costco card was the best deal ever -- the reward rates are great, and I get $400 in cash nearly every year.
The solar installs I've seen on boats amounted to flush mounted over cockpits or other roofline enclosures. Nothing to catch the wind.
Other than the one Greenline with 1.3kw solar, for most purposes solar on a boat is only really useful for offsetting DC consumption for house batteries -- a couple hundred watts -- not propulsion, unless you're doing some really unusual purpose-built solar boat.
Shore charging is fine, but you'd really want diesel generation for both higher running speeds than batteries can supply and for distance. Plus you'd need it anyway for house power.
I think a battery array install along the keel line in the bilge would be ideal, and a lot of larger fiberglass hulls end up with lead there anyway for ballast. They wind up being sort of tall for their typical running weight and more mass at the keel line in the bilge does a lot to help counter rolling at anchor and makes stabilizers much more effective.
I was thinking of a specific use case -- inland lake recreational marine. A plane is a different animal, as is ocean-running boats.
The lake I boat on is pretty big -- 22 square miles comprised of a dozen or so interlocking bays, but you would be hard pressed to run more than 30 miles of traveled distance in a single day.
For this specific application, I don't see why a single engine hybrid, diesel-battery-electric application wouldn't work. Half the boats on the lake are single engine and a total engine failure isn't exactly life threatening, either.
Most days we don't travel more than 10 miles, a range a Tesla-size battery could provide about 6 knots of cruise on a 30 foot hull pretty easily. Many slips have 30A service for charging and an on-board diesel generator for charging and diesel-electric propulsion would extend the range to fuel on board.
I'm only aware of one commercial option, Greenline Hybrid, which uses a single diesel and batteries but it puts the prop, combination motor/generator and diesel on a single shaft with a hydraulic clutch to operate as prime mover and generator as well as having about 6 hours of cruise at 6 knots on battery alone. They also have a model with 1.3 kW solar panels which they claim can actually drive the entire thing (albeit slowly, like 3 knots).
If I had the money I would be inclined to find an good hull with worn out engines and roll my own for this use, but see if I could make it pure electric drive with the diesel only for electric generation.
It was accessible to program and I think the fact that all it did was boot to a command prompt made it tempting to program, versus the endless distraction of a graphic/web environment of today.
Plus you could get down to actual programming logic immediately, without endless distractions of display management in a GUI.
I think this is part of the problem, the strong desire, masking as need, for relentless modularity and configuration that the vast majority of users don't care about. The result is a tangle of code that needs to be a tangle to keep satisfying a core group of influencers with obscure features that most people don't care about and don't know how to configure.
Meanwhile, people on the Windows side laugh and wonder what the big deal is, they have had a useful GUI forever and Remote Desktop solves remote execution (albeit differently than X), too.
It sure hasn't hit the recreational marine market, though. You can get pod drives, but they are mechanically driven off of diesel engines.
I always wondered if they could do a hybrid drive on a boat with a single engine and batteries driving electric motors. I'm only aware of one (newish) maker doing it, but I would think for a lot of the inland boating market where ranges under 25 miles and cruise speeds of 10 knots are common.
The batteries would be a wash if you removed one of the engines and less fuel might offset a heavier generator.
I'm guessing the turboshaft is meant to provide take-off power and it can cruise on batteries in a kind of winged flight mode? Otherwise I'm not seeing the advantage of a turboshaft given the energy losses associated with electric generation.
I think the general modem pool for the timeshare system used in CompSci 3104 might have been 1200 baud capable in 1986 at the University I attended. I know for a fact in high school it wasn't -- there were a handful of 1200 baud lines restricted to admin logins, and an admin I knew used to gripe a lot about wasted money on a Hayes 1200 modem that seldom could get the 1200 baud lines.
FWIW, I think the AppleCat had a Bell 212 "normal" 1200 baud option, but it was nearly twice the price of the non-212 version. And the usual problem was almost nothing was 1200 baud capable, especially BBSes.
The prices sure seemed to come down fast, though. Within about 5 years, it seemed like 14.4k was pretty cheap.