If I say that I don't have a prescription, but claim that my eye doctor is Dr. Daff E Duck, Warner Bros Clinic, Los Angeles, I'm providing false information, but I can't see how it would be considered forgery. (not sure about fraud...)
In the hard sciences there are few companies that make money off of the success or failure of an academic research project.
That doesn't explain sociology; however searching through retractionwatch.com I'm not sure how much misconduct there is in that field (at least within the US and western Europe) as opposed to just plain errors. Which shouldn't be surprising in a field with small sample sizes, poor funding, and high noise.
Sorry, I'm used to my area of research - computer science - which is a hell of a lot less shady, at least in the US. (probably because almost all the money is being made totally outside of academic and research settings)
The article you link makes me wonder whether there's a way that publishers (e.g. NEJM and JAMA) could force clinical trial data out into the open. If it were a requirement for publication in the top venues, then the drug companies would have to either 'fess up or work with second-rate researchers. (unfortunately there's probably an option 3, which they'll find and we won't like...)
Chinese scientists also blame what they call the skewed incentives they say are embedded within their nation's academic system.
It sounds like they have a similar problem to the US's collapsing "publish or perish" paradigm. People should be less focused on what the scientists are doing and focus on the cause of such behavior.
To change the behavior of a group you must correct the feedback loops that control them.
In the US, gross misconduct (like impersonating other scientists in order to review your own papers) is a career death sentence, in part because "publish or perish" is administered by a tenure vote of the people you work with (and compete with), instead of a bean-counting administrator somewhere. There are lots of incentives to do semi-unsavory things - e.g. splitting your work into "least-publishable units", or "P-hacking", where you try every combination of data to see if one of them supports your conclusion - however if you cross the line and start doing things your colleagues aren't willing to do, they'll be happy to come down on you like a ton of bricks.
If you look at the demo video on their website - http://www.bitmanagement.com/en/demos/geo - it appears to be a browser-based plugin; the actual "demo" link tries to download a file called "BS_Contact_VRML-3DX.exe from www.bitmanagement.com. The demo shows a user-controlled fly-over of an urban area, and seems highly applicable to a lot of military uses. (the software to *create* the model seems to be different - the software named in the suit is the viewer for the 3D models)
It's quite possible that the 500K number refers to the number of machines which downloaded and installed the plugin. It's also possible that the Navy "disabled tracking of installations" (as the FA states) by putting a copy on their own server, and that the vendor was tracking installations by looking at their web server logs.
That's all speculation. It's definitely a job for a lawyer at this point, and it also points out the risks of per-user licenses when you might not have control over the number of users.
All things considered, is that *really* such a bad plan? Is it any worse than what we have now, which is a government that mostly listens to big corporations?
Unfortunately he's going to open his mouth from time to time, and his advisors may not remain advisors any longer if they don't back him up on what he says. One is reminded of the beginning of Game of Thrones...
Check the map. Not that it matters, but Clinton has 28 states and Trump 22. It's even easier to have a majority when you have most of the big states *and* most of the states overall.
If he can't count states on a map (28 for Clinton, 22 for Trump when the post was made), subtleties like state populations are going to go entirely over his head...
What I get from that map is that, despite a clear majority of states voting for Trump, their model still for some reason "predicts" a Clinton win. Almost as if the election has been rigged. Somehow.
The astute reader will notice that 22 states on that map are colored red or pink. I'll leave you to draw your own conclusions.
I have a great idea - let's allow developers to settle their disagreements by physical combat, just like in the Middle Ages. Set up a conference room properly, have a ref to enforce MMA rules, and if someone calls someone else out, set a time and they can have at it. If you complain that this would be unfair to women, short people, or fat people, I can trot out dozens of counter-examples - just google short / female / fat MMA fighter for a bunch of them.
The problems with this are the same as with the caustic social environment in some open source groups - (a) certain groups are more vulnerable than others; in other words, there are attacks which work against one group and don't work against another, and defenses available to some groups and not others, and (b) the ability to deliver and withstand this sort of abuse has NOTHING to do with engineering ability or how good your ideas are.
When you reach a certain level of seniority you start realizing that one of the roles you need to take at meetings is to defend the shy junior person with a good idea against the outspoken jerk who is opposing it reflexively. Unfortunately we live in a world where four times out of five that person is going to be female, and the jerk is a lot less likely to oppose the same idea proposed by a junior, shy, but male engineer.
From the original article: In contrast, software development is inherently fair. If you write it correctly, your program runs. Otherwise, it doesn’t. Your computer doesn’t get offended if you don’t state your message well. It doesn’t hold a grudge. It just waits until you write it correctly.
And *this* is the distinction between Open Source and commercial software development. To begin with, a job of almost any sort is inherently a social environment. Not only that, but any large engineering project lives or dies by communication between team members. How many of you have worked with a "lone wolf" who wanted to go and write his (in my experience, always 'his') own code, without talking to anyone else or figuring out whether it would work with everyone else's? Or worse yet, someone who arbitrarily decided that all their code needed to be re-written just when you needed to get a release out, because it failed their internal criteria for perfection? And unless that person was truly brilliant, how many of you would willingly work with that person again?
Open source projects often subscribe to the myth of the individual contributor at the expense of the team, a myth which is sustained by not only people's desire to believe it, but also by the rare examples of people who really *can* accomplish amazing things on their own. Part of this myth is the idea that social relationships, and by extension social niceties, are irrelevant in the project. This myth, in turn, can shape online interactions in ways which are uncomfortable for normally-socialized white males, and positively vicious for women and minorities.
As another commenter pointed out, the 1.5Tbit/in^2 number in the posting (which is taken from the original article) is pretty bogus. Seagate's 2TB 7mm 2.5" drive has an areal density of 1.32Tbit/in^2, and it's probably a safe bet that they (and WD) can wring another 15% density improvement out of SMR technology in the next year or two.
For those commenters bemoaning the fact that the highest density drives today are SMR rather than "regular" drives, get over it - the odds of conventional non-HAMR, non-shingled drives getting much denser than the roughly 1TByte per 3.5" platter we see today are slim to none:
To get smaller bits, you need a smaller write head. That smaller write head has a weaker magnetic field. The weaker field means the media has to be more easily magnetizable (i.e. has lower coercivity). The lower coercivity media needs to have a bigger grain size (size of the individual magnetic domains), so that grains don't flip polarity too often due to thermal noise.
Since a bit can't be smaller than a grain, that means that smaller your write head is, the larger your minimum bit size is. Eventually those two lines cross on the graph, and it's game over.
Two ways of getting out of this are SMR (shingled magnetic recording) and HAMR (heat-assisted magnetic recording):
SMR - stop making the write head smaller, but keep making the bits smaller. Overlap tracks like clapboards on the side of the house (where'd this "shingle" nonsense come from?), allowing small bits with large write heads. Of course this means that you can't re-write anything without wiping out adjacent tracks, which means you need something like a flash translation layer inside the drive, and because of that, random writes might be *really* slow sometimes. (I've seem peak delays of 4 seconds when we're really trying to make them behave badly)
HAMR - Write your bits on low-coercivity media with a tiny, wimpy head, and store them on high-coercivity media with tiny magnetic grains. How do you do this? By heating heating a high-coercivity media with a laser (say to 450C or so) to reduce its coercivity to reasonable levels, then letting it cool down afterwards. But you need a big laser (20mw?) on each head, which causes a whole bunch of problems. Which is probably why they're delaying them.
Oh, and you can overlap tracks on HAMR drives, creating an SMR HAMR drive, with even higher density but the performance problems of both technologies. Which they'll probably do as soon as HAMR hits the market, because with today's SSDs the market for fast HDDs is dying a very quick death.
I'm reminded of the old bag of glass SNL skit - some products (or product features) are just plain dangerous, and saying "but we explain the risks in page 17 of the manual" isn't a good excuse.
How much effort would it take to set defaults that (1) disable anonymous FTP for addresses outside of the local subnet, and (b) inject a fake robots.txt that prevents search engine indexing? And then add an explanation of the risks if you try to disable those defaults?
Drive performance is kind of like airplane legroom - people gripe about it, but in the end they ignore it and buy the cheap ticket.
Shingled drives aren't better - they're bigger, and that's what people pay for. WD's 10TB helium drive is shingled, and I would guess that every drive over 10TB will be shingled for the foreseeable future. By the time HAMR and BPM come out, SSDs will probably have killed off the high-performance drive market, so those technologies will probably be released as capacity-optimized shingled drives, too.
in Massachusetts (and probably other places) use Bluetooth phone tracking: http://www.mass.gov/governor/pressoffice/pressreleases/2014/0411-governor-patrick-announces-go-time-expansion.html
"The GO Time real time traffic system measures travel times between two points by anonymously tracking the Bluetooth enabled devices carried by motorists and their vehicles. The system complies with new federal legislation that requires real time traffic information to be provided to the public."
The author of the study makes a lot of arguments based on factors that are easily changed, like the configuration of an SSD. However there are a few basic technological trends:
1. Disks and NAND flash are both getting more dense at fairly comparable speeds - disk has been getting cheap faster than flash lately, but may have a hiccup in the next few years. Where flash has conclusively replaced disk is in applications like iPods and mobile where "enough" storage is cheaper than a single disk. (the iPod went flash when 2GB of flash reached $50, which is the price of a micro-disk) It's not going to replace disk for high volume data storage anytime soon.
2. With today's disks and chips, a hard disk drive has a relatively fixed cost (the cost of the factory amortized over the number of drives produced) and similarly flash has a relatively fixed cost (cost of fabrication plant over the number of chips produced in its useful lifespan). The number of bits on each doesn't really matter - that's why packing them more tightly makes the bits cheaper.
3. Disk bandwidth for 7200K drives isn't going to go over say 300MB/s anytime soon with today's perpendicular recording technology - if the disk is moving past the head at a constant speed, the only way to get more bits through per second is to pack them more closely on the platter. And the best you can do by spinning faster is a factor of 2, at 15K. (and those are very low capacity and very expensive)
2 and 3 mean that flash can easily supply cheaper bandwidth than disk - it's the SSD maker's choice how widely they want to stripe data over the chips in the drive. (64 ways isn't unreasonable) There's a huge advantage today, and it will stay the same (see #2) if flash chips don't get faster, and get bigger if they do. (at some point getting that speed may require paying for more flash than you need, but at that point a single disk will be bigger than you need, too)
For years flash was getting slower and less reliable (requiring more complex error correcting codes) as it got denser - that's partly why it got cheap so much faster than e.g. RAM, where you can't cut those corners. The next generation of flash (3D NAND) may reverse that for a while; in addition SSDs are finally a noticeable fraction of the market so there's an incentive for vendors to make faster flash. (3 years ago SSDs were 3% of the flash market, and the rest went into iPods, phones, and removable drives and cards - SSD vendors had to make do with flash that was designed for systems where you don't care about performance)
Does anyone else remember when bubble memory was supposed to replace hard drives? There's a long road between the current state of post-NAND technologies (Phase Change Memory, spin-torque-transfer magnetic RAM, Resistive RAM, and a few others) and mass-market high-volume chips. If one of them becomes good enough for someone to risk a $5 billion fab on, and it gives more bits per dollar than flash, then it will probably replace flash almost instantly. If no one bets a cutting edge fab, however, it doesn't matter how promising the technology is. (in particular, the "10x better" is based on assumptions that e.g. PCM can be built in sizes vastly smaller than today's flash - of course we don't know how to build the fabrication plants to do that yet. No one has a story for something 10x better at the same feature size.)
The paper is from Steve Swanson's group at UCSD, *not* Microsoft Research. And the reasons for slowdown with more bits per cell: (a) writing is done in incremental steps, which have to be smaller for the more precise levels needed for 8 or 16 levels per bit, requiring more steps, and (b) the charge on a flash cell can't be measured directly; instead the chip can measure which cells in a page are above (or below) a threshold voltage, so sensing 16 levels requires 15 separate read operations.
There's a tweet from a WBZ reporter to that effect here: http://www.sbnation.com/2013/4/15/4228130/boston-marathon-explosion-cell-phone-shutdown I don't know if it's true, though.
Remember that most scientific papers are vehicles to describe work which has been done by the authors. It makes no sense for someone else to modify it - they typically don't have enough information about the work the paper is based on, and when they're done it's no good to them, as it still describes someone else's work.
There are probably people - e.g. tech journalists - who could make use of my writing if I used a liberal license, but you don't have the same reciprocity you have in open-source software. People who contribute to open source software also benefit from others' contributions, while in the case of scholarly writing the benefits would primarily flow in one direction.
We academics understand perfectly well that other people in the world have hard jobs too.
Well, as another academic, I think *some* of us are aware of that. In general it seems that a lot of people with desk jobs seem to feel that their profession is uniquely difficult, and that the reason the guy who cleans their office in the middle of the night gets paid less is because he doesn't work as hard. Academics seem just as likely to believe that as anyone else.
All we ask is that other people recognize that our jobs are, first and foremost, jobs, like anyone else's.
All SSDs have a bit more storage than their rating. Partitioning a little less space on a vendor-fresh drive can double or triple the extra storage available to the SSD's internal wear leveling algorithms.
This won't actually work - partitions don't exist from a disk's point of view, but are just bytes in sector 0. The SSD will religiously preserve the useless data in the sectors outside of the partition you create, using up space that could otherwise be put to good use.
As other posters have explained in bits and pieces, flash chips can be written in pages (2KB or 4KB, usually) but have to be erased in blocks (64 or 128 pages). If you overwrite 10 pages in the middle of a block, the writes will go into fresh pages somewhere else, and the original 10 will be useless until you get around to erasing the entire page. Since those other pages are holding data, you can't erase them until they either (a) get replaced by additional new writes, or (b) are moved somewhere else. If you end up having to copy say 3 pages of data for every new page you write, your write performance is going to go down by a factor of 4.
The more free space you have, the more likely it is that even with totally random writes there will be some blocks that are entirely empty and can be erased without having to copy any data. That's why the 32GB Intel X25-E (the enterprise drive) has 40GB of flash chips inside it. On the other hand, just about every consumer drive has 6.7% or so free space, because that lets them use say 64GB (64 * 2^30) of flash chips and legally advertise it as 64GB (64 * 10^9).
Typically your file system has a fair amount of free space (compared to 6.7%), because performance suffers and you run the risk of running out of space when you get close to 100% usage. Without TRIM, however, the SSD can't make use of that space, and carefully preserves the contents of every block on the file system free list. In theory TRIM should allow the OS to identify the file system free list to the SSD, which will then have much more space available for garbage collection, resulting in reduced copying and better performance. In practice, your mileage may vary.
If you look closely you'll notice that (a) the benchmarks were run on a Thinkpad T60 laptop, and (b) there were significant differences on some benchmarks like RAM bandwidth that should have little or no OS components.
This sounds to me like the power management was dialing down the CPU on the later releases...
Slashdot Mirror
If I say that I don't have a prescription, but claim that my eye doctor is Dr. Daff E Duck, Warner Bros Clinic, Los Angeles, I'm providing false information, but I can't see how it would be considered forgery. (not sure about fraud...)
In the hard sciences there are few companies that make money off of the success or failure of an academic research project.
That doesn't explain sociology; however searching through retractionwatch.com I'm not sure how much misconduct there is in that field (at least within the US and western Europe) as opposed to just plain errors. Which shouldn't be surprising in a field with small sample sizes, poor funding, and high noise.
Sorry, I'm used to my area of research - computer science - which is a hell of a lot less shady, at least in the US. (probably because almost all the money is being made totally outside of academic and research settings)
The article you link makes me wonder whether there's a way that publishers (e.g. NEJM and JAMA) could force clinical trial data out into the open. If it were a requirement for publication in the top venues, then the drug companies would have to either 'fess up or work with second-rate researchers. (unfortunately there's probably an option 3, which they'll find and we won't like...)
Gotta love this quote from one of the linked articles:
"When a lot of the fake peer reviews first came up, one of the reasons the editors spotted them was that the reviewers responded on time"
Chinese scientists also blame what they call the skewed incentives they say are embedded within their nation's academic system.
It sounds like they have a similar problem to the US's collapsing "publish or perish" paradigm. People should be less focused on what the scientists are doing and focus on the cause of such behavior.
To change the behavior of a group you must correct the feedback loops that control them.
In the US, gross misconduct (like impersonating other scientists in order to review your own papers) is a career death sentence, in part because "publish or perish" is administered by a tenure vote of the people you work with (and compete with), instead of a bean-counting administrator somewhere. There are lots of incentives to do semi-unsavory things - e.g. splitting your work into "least-publishable units", or "P-hacking", where you try every combination of data to see if one of them supports your conclusion - however if you cross the line and start doing things your colleagues aren't willing to do, they'll be happy to come down on you like a ton of bricks.
If you look at the demo video on their website - http://www.bitmanagement.com/en/demos/geo - it appears to be a browser-based plugin; the actual "demo" link tries to download a file called "BS_Contact_VRML-3DX.exe from www.bitmanagement.com. The demo shows a user-controlled fly-over of an urban area, and seems highly applicable to a lot of military uses. (the software to *create* the model seems to be different - the software named in the suit is the viewer for the 3D models)
It's quite possible that the 500K number refers to the number of machines which downloaded and installed the plugin. It's also possible that the Navy "disabled tracking of installations" (as the FA states) by putting a copy on their own server, and that the vendor was tracking installations by looking at their web server logs.
That's all speculation. It's definitely a job for a lawyer at this point, and it also points out the risks of per-user licenses when you might not have control over the number of users.
All things considered, is that *really* such a bad plan? Is it any worse than what we have now, which is a government that mostly listens to big corporations?
Unfortunately he's going to open his mouth from time to time, and his advisors may not remain advisors any longer if they don't back him up on what he says. One is reminded of the beginning of Game of Thrones...
Check the map. Not that it matters, but Clinton has 28 states and Trump 22. It's even easier to have a majority when you have most of the big states *and* most of the states overall.
If he can't count states on a map (28 for Clinton, 22 for Trump when the post was made), subtleties like state populations are going to go entirely over his head...
What I get from that map is that, despite a clear majority of states voting for Trump, their model still for some reason "predicts" a Clinton win. Almost as if the election has been rigged. Somehow.
The astute reader will notice that 22 states on that map are colored red or pink. I'll leave you to draw your own conclusions.
I have a great idea - let's allow developers to settle their disagreements by physical combat, just like in the Middle Ages. Set up a conference room properly, have a ref to enforce MMA rules, and if someone calls someone else out, set a time and they can have at it. If you complain that this would be unfair to women, short people, or fat people, I can trot out dozens of counter-examples - just google short / female / fat MMA fighter for a bunch of them.
The problems with this are the same as with the caustic social environment in some open source groups - (a) certain groups are more vulnerable than others; in other words, there are attacks which work against one group and don't work against another, and defenses available to some groups and not others, and (b) the ability to deliver and withstand this sort of abuse has NOTHING to do with engineering ability or how good your ideas are.
When you reach a certain level of seniority you start realizing that one of the roles you need to take at meetings is to defend the shy junior person with a good idea against the outspoken jerk who is opposing it reflexively. Unfortunately we live in a world where four times out of five that person is going to be female, and the jerk is a lot less likely to oppose the same idea proposed by a junior, shy, but male engineer.
From the original article: In contrast, software development is inherently fair. If you write it correctly, your program runs. Otherwise, it doesn’t. Your computer doesn’t get offended if you don’t state your message well. It doesn’t hold a grudge. It just waits until you write it correctly.
And *this* is the distinction between Open Source and commercial software development. To begin with, a job of almost any sort is inherently a social environment. Not only that, but any large engineering project lives or dies by communication between team members. How many of you have worked with a "lone wolf" who wanted to go and write his (in my experience, always 'his') own code, without talking to anyone else or figuring out whether it would work with everyone else's? Or worse yet, someone who arbitrarily decided that all their code needed to be re-written just when you needed to get a release out, because it failed their internal criteria for perfection? And unless that person was truly brilliant, how many of you would willingly work with that person again?
Open source projects often subscribe to the myth of the individual contributor at the expense of the team, a myth which is sustained by not only people's desire to believe it, but also by the rare examples of people who really *can* accomplish amazing things on their own. Part of this myth is the idea that social relationships, and by extension social niceties, are irrelevant in the project. This myth, in turn, can shape online interactions in ways which are uncomfortable for normally-socialized white males, and positively vicious for women and minorities.
As another commenter pointed out, the 1.5Tbit/in^2 number in the posting (which is taken from the original article) is pretty bogus. Seagate's 2TB 7mm 2.5" drive has an areal density of 1.32Tbit/in^2, and it's probably a safe bet that they (and WD) can wring another 15% density improvement out of SMR technology in the next year or two.
For those commenters bemoaning the fact that the highest density drives today are SMR rather than "regular" drives, get over it - the odds of conventional non-HAMR, non-shingled drives getting much denser than the roughly 1TByte per 3.5" platter we see today are slim to none:
To get smaller bits, you need a smaller write head. That smaller write head has a weaker magnetic field. The weaker field means the media has to be more easily magnetizable (i.e. has lower coercivity). The lower coercivity media needs to have a bigger grain size (size of the individual magnetic domains), so that grains don't flip polarity too often due to thermal noise.
Since a bit can't be smaller than a grain, that means that smaller your write head is, the larger your minimum bit size is. Eventually those two lines cross on the graph, and it's game over.
Two ways of getting out of this are SMR (shingled magnetic recording) and HAMR (heat-assisted magnetic recording):
SMR - stop making the write head smaller, but keep making the bits smaller. Overlap tracks like clapboards on the side of the house (where'd this "shingle" nonsense come from?), allowing small bits with large write heads. Of course this means that you can't re-write anything without wiping out adjacent tracks, which means you need something like a flash translation layer inside the drive, and because of that, random writes might be *really* slow sometimes. (I've seem peak delays of 4 seconds when we're really trying to make them behave badly)
HAMR - Write your bits on low-coercivity media with a tiny, wimpy head, and store them on high-coercivity media with tiny magnetic grains. How do you do this? By heating heating a high-coercivity media with a laser (say to 450C or so) to reduce its coercivity to reasonable levels, then letting it cool down afterwards. But you need a big laser (20mw?) on each head, which causes a whole bunch of problems. Which is probably why they're delaying them.
Oh, and you can overlap tracks on HAMR drives, creating an SMR HAMR drive, with even higher density but the performance problems of both technologies. Which they'll probably do as soon as HAMR hits the market, because with today's SSDs the market for fast HDDs is dying a very quick death.
I'm reminded of the old bag of glass SNL skit - some products (or product features) are just plain dangerous, and saying "but we explain the risks in page 17 of the manual" isn't a good excuse.
How much effort would it take to set defaults that (1) disable anonymous FTP for addresses outside of the local subnet, and (b) inject a fake robots.txt that prevents search engine indexing? And then add an explanation of the risks if you try to disable those defaults?
Drive performance is kind of like airplane legroom - people gripe about it, but in the end they ignore it and buy the cheap ticket.
Shingled drives aren't better - they're bigger, and that's what people pay for. WD's 10TB helium drive is shingled, and I would guess that every drive over 10TB will be shingled for the foreseeable future. By the time HAMR and BPM come out, SSDs will probably have killed off the high-performance drive market, so those technologies will probably be released as capacity-optimized shingled drives, too.
in Massachusetts (and probably other places) use Bluetooth phone tracking: http://www.mass.gov/governor/pressoffice/pressreleases/2014/0411-governor-patrick-announces-go-time-expansion.html
"The GO Time real time traffic system measures travel times between two points by anonymously tracking the Bluetooth enabled devices carried by motorists and their vehicles. The system complies with new federal legislation that requires real time traffic information to be provided to the public."
The author of the study makes a lot of arguments based on factors that are easily changed, like the configuration of an SSD. However there are a few basic technological trends:
1. Disks and NAND flash are both getting more dense at fairly comparable speeds - disk has been getting cheap faster than flash lately, but may have a hiccup in the next few years. Where flash has conclusively replaced disk is in applications like iPods and mobile where "enough" storage is cheaper than a single disk. (the iPod went flash when 2GB of flash reached $50, which is the price of a micro-disk) It's not going to replace disk for high volume data storage anytime soon.
2. With today's disks and chips, a hard disk drive has a relatively fixed cost (the cost of the factory amortized over the number of drives produced) and similarly flash has a relatively fixed cost (cost of fabrication plant over the number of chips produced in its useful lifespan). The number of bits on each doesn't really matter - that's why packing them more tightly makes the bits cheaper.
3. Disk bandwidth for 7200K drives isn't going to go over say 300MB/s anytime soon with today's perpendicular recording technology - if the disk is moving past the head at a constant speed, the only way to get more bits through per second is to pack them more closely on the platter. And the best you can do by spinning faster is a factor of 2, at 15K. (and those are very low capacity and very expensive)
2 and 3 mean that flash can easily supply cheaper bandwidth than disk - it's the SSD maker's choice how widely they want to stripe data over the chips in the drive. (64 ways isn't unreasonable) There's a huge advantage today, and it will stay the same (see #2) if flash chips don't get faster, and get bigger if they do. (at some point getting that speed may require paying for more flash than you need, but at that point a single disk will be bigger than you need, too)
For years flash was getting slower and less reliable (requiring more complex error correcting codes) as it got denser - that's partly why it got cheap so much faster than e.g. RAM, where you can't cut those corners. The next generation of flash (3D NAND) may reverse that for a while; in addition SSDs are finally a noticeable fraction of the market so there's an incentive for vendors to make faster flash. (3 years ago SSDs were 3% of the flash market, and the rest went into iPods, phones, and removable drives and cards - SSD vendors had to make do with flash that was designed for systems where you don't care about performance)
Does anyone else remember when bubble memory was supposed to replace hard drives? There's a long road between the current state of post-NAND technologies (Phase Change Memory, spin-torque-transfer magnetic RAM, Resistive RAM, and a few others) and mass-market high-volume chips. If one of them becomes good enough for someone to risk a $5 billion fab on, and it gives more bits per dollar than flash, then it will probably replace flash almost instantly. If no one bets a cutting edge fab, however, it doesn't matter how promising the technology is. (in particular, the "10x better" is based on assumptions that e.g. PCM can be built in sizes vastly smaller than today's flash - of course we don't know how to build the fabrication plants to do that yet. No one has a story for something 10x better at the same feature size.)
The paper is from Steve Swanson's group at UCSD, *not* Microsoft Research.
And the reasons for slowdown with more bits per cell: (a) writing is done in incremental steps, which have to be smaller for the more precise levels needed for 8 or 16 levels per bit, requiring more steps, and (b) the charge on a flash cell can't be measured directly; instead the chip can measure which cells in a page are above (or below) a threshold voltage, so sensing 16 levels requires 15 separate read operations.
I can confirm that texts were getting through just fine to runners farther up the course (in my case, near BC) when cell calls weren't going through.
There's a tweet from a WBZ reporter to that effect here: http://www.sbnation.com/2013/4/15/4228130/boston-marathon-explosion-cell-phone-shutdown
I don't know if it's true, though.
Remember that most scientific papers are vehicles to describe work which has been done by the authors. It makes no sense for someone else to modify it - they typically don't have enough information about the work the paper is based on, and when they're done it's no good to them, as it still describes someone else's work.
There are probably people - e.g. tech journalists - who could make use of my writing if I used a liberal license, but you don't have the same reciprocity you have in open-source software. People who contribute to open source software also benefit from others' contributions, while in the case of scholarly writing the benefits would primarily flow in one direction.
We academics understand perfectly well that other people in the world have hard jobs too.
Well, as another academic, I think *some* of us are aware of that. In general it seems that a lot of people with desk jobs seem to feel that their profession is uniquely difficult, and that the reason the guy who cleans their office in the middle of the night gets paid less is because he doesn't work as hard. Academics seem just as likely to believe that as anyone else.
All we ask is that other people recognize that our jobs are, first and foremost, jobs, like anyone else's.
Amen to that.
All SSDs have a bit more storage than their rating. Partitioning a little less space on a vendor-fresh drive can double or triple the extra storage available to the SSD's internal wear leveling algorithms.
This won't actually work - partitions don't exist from a disk's point of view, but are just bytes in sector 0. The SSD will religiously preserve the useless data in the sectors outside of the partition you create, using up space that could otherwise be put to good use.
As other posters have explained in bits and pieces, flash chips can be written in pages (2KB or 4KB, usually) but have to be erased in blocks (64 or 128 pages). If you overwrite 10 pages in the middle of a block, the writes will go into fresh pages somewhere else, and the original 10 will be useless until you get around to erasing the entire page. Since those other pages are holding data, you can't erase them until they either (a) get replaced by additional new writes, or (b) are moved somewhere else. If you end up having to copy say 3 pages of data for every new page you write, your write performance is going to go down by a factor of 4.
The more free space you have, the more likely it is that even with totally random writes there will be some blocks that are entirely empty and can be erased without having to copy any data. That's why the 32GB Intel X25-E (the enterprise drive) has 40GB of flash chips inside it. On the other hand, just about every consumer drive has 6.7% or so free space, because that lets them use say 64GB (64 * 2^30) of flash chips and legally advertise it as 64GB (64 * 10^9).
Typically your file system has a fair amount of free space (compared to 6.7%), because performance suffers and you run the risk of running out of space when you get close to 100% usage. Without TRIM, however, the SSD can't make use of that space, and carefully preserves the contents of every block on the file system free list. In theory TRIM should allow the OS to identify the file system free list to the SSD, which will then have much more space available for garbage collection, resulting in reduced copying and better performance. In practice, your mileage may vary.
If you look closely you'll notice that (a) the benchmarks were run on a Thinkpad T60 laptop, and (b) there were significant differences on some benchmarks like RAM bandwidth that should have little or no OS components.
This sounds to me like the power management was dialing down the CPU on the later releases...