My understanding is that the HDD market (by TB, at least) is mostly hyperscale operators - Google, Facebook, Dropbox, Microsoft Azure and OneDrive, etc. They store most of their data on HDD, because HDD capacity is cheaper, and do most of their reads/writes from SSDs, because SSD performance is cheaper. Unaccessed files get moved to slower and cheaper parts of the system over time.

Note also that NVMe SSDs cost a lot to deploy. A normal (non-Google) person can buy a bare system with 45 3.5" HDD slots for maybe $5K, adding about $100 to the cost of each deployed drive. A system you can plug 45 NVMe drives into is going to cost enormously more. You can deploy SATA or SAS HDDs for about the same per-drive cost overhead as HDDs, but you sacrifice a lot of performance.

"Rust" that's flat at the sub-nanometer scale, with a head that flies about 1nm above the surface and seeks to the middle of a 50nm-wide track, is a bit more expensive. (for comparison, green light wavelength is about 500nm)

Hoist with their own petard, as it were.

The judge's opinion is pretty short (12 pages) and to the point, and has way more information than TFA.

There's a pumped storage facility in Massachusetts that's been around half a century, that delivers 3 times the power (1200MW) for 4 times as long (8 hours): https://www.wbur.org/news/2016/12/02/northfield-mountain-hydroelectric-station

And yes it works below freezing, just like hydro dams work in the winter.

Of course in a state where most of the high points are old landfills (https://www.sun-sentinel.com/local/broward/fl-reg-mount-trashmore-grows-20180712-story.html) it might be a bit tricky...

"5-" or 10-year battery. And I hesitate to call an original iPhone SE "modern", but a newer phone would draw even more power.

As a rule I don't watch endless Youtube videos. However a bit of (possibly incorrect) math seems to indicate that the power density from C14 decay is somewhere around 2 to 7 W/kg.

As a comparison, my iPhone SE has an 1800 mAh battery @ 3.7v, or 6.6Wh, so discharging over a day is about 0.25W. It ain't gonna power a modern cell phone, but it seems like a great replacement for anywhere you need a 6 or 10-year battery. (a 9v cell weighs about 50g, and is rated for 500mAh - let's be generous and call that 4.5Wh, although the voltage droops a lot at the end. Getting 5 years out of it in a smoke detector means you have to limit your average current to about 10 micro-amps. If you could get 1W/kg of electric energy out of the C14, and get 25g of it in a 50g package, that would give you a continuous current of about 10mA, or about a kilowatt-hour over the course of 5 years)

[don't know why I bothered writing a carefully reasoned post for slashdot - that's so 20th-century...]

This was never a problem in the past - in 2010, a 2002 laptop would have been laughable, fit only for a doorstop. (going back farther, in 1992 I had a 25MHz MIPS machine on my desk at work; in 2000 I had a mid-range PC with a 500-MHz CPU) The slow increase we've seen in CPU speeds in the last 8 years (and the flat price of DRAM from 2011 to 2019) is unprecedented in the history of computing, although there's a good chance that it's our future from now on.

When you couple that with good mechanical design (at least compared with the other laptops I've owned) it means that MacBooks from the 2012 era are pretty much the first 8-year-old laptops that significant numbers of people actually care about keeping alive.

The trouble is that it's a *pain* to support lots of old versions of something. I've got a 1994 motorcycle, and although technically they still make spare parts, I've had to wait for them to be shipped from Japan. I tried finding spare parts to a Delta drill press I have, and discovered that the company was sold a few times and the only spare parts still available are the screws holding the power switch in front.

People care about fixing old motorcycles and cars, which is why I finally got my parts after a long wait. In contrast a cheap drill press is kind of like a cheap consumer appliance - it's not cost-effective to pay someone to fix it, and most people won't do it themselves, so there's probably not much of a market for parts even when the thing is new.

8-year-old laptops used to be like that drill press - most owners wouldn't even try to get them fixed, but would have already junked them and bought a new one. It costs money for a vendor to support a large number of old models, and if no one cares about the older ones they might as well forget about them. As the effective lifespan of a laptop increases people are starting to care; hopefully vendor policies will change in response.

In the mean time there's a pretty thriving 3rd-party repair ecosystem for Apple products, and if your hardware is old enough that it's not supported under newer versions of MacOS you can use tricks from the Hackintosh world to keep it running. And be sure to support consumer Right To Repair legislation in your state, so the repair places don't get sued out of existence.

AWS is hugely profitable, and most companies save money using AWS instead of running their own computers.

The reason really has to do with scale. Everyone knows that scale matters for the physical aspects (power and machines and buildings), but it matters for people possibly even more so.

If you're willing to build and staff a data center in the middle of nowhere it can probably be close to as efficient as Amazon's data center's - by "close" I mean well under 2x, not 5% or something. This is based on mghpcc.org, built by the 5 big Massachusetts universities - 15MW, $100M for the building, competitive power costs, room for 20-30K machines. Based on the public numbers, amortized over 20 years, the building costs less than $200 per 1U per year. At $0.10/kWhr (public tariff for MW rate * 1.2 PUE) power costs about $1/W/yr; our servers seem to draw in the 300W range or less. Since servers seem to last 3-5 years, that means the machines themselves cost considerably more than the building or the power, but not a huge amount more.

The MGHPCC costs about $5M/yr in depreciation; you could probably make the building smaller and cheaper, but the cost per computer would undoubtably go up, and there's not a huge amount of room before it starts adding a lot of per-computer cost - my wild-ass guess is that below half this price it would start getting really expensive. (I'd really need an industrial-scale A/C expert to make that prediction)

The other place where scale comes into account is your workforce, because people are expensive and they come in integer units. If you want to use serverless, and elastic mapreduce, and block and s3-like storage, and virtual machines, and a couple of database solutions, and a load balancer, and a few other of the services that are trivial to select on your AWS dashboard, then you are going to have to hire somewhere between a dozen and a few dozen highly-qualified people to set them up and run them. (among other reasons, because unless you're Google or Amazon the very few people who can do all of these won't want to work for you) At this point you're talking a payroll of millions of dollars per year.

Add the payroll to the depreciation costs of the building, and you've got your fixed costs; to achieve vaguely Amazon-like efficiency this probably needs to be 25% of your total costs or less. Working it all out, if you've got more than maybe $50 million worth of computers (preferably lots more), and can hire good people, you can run an operation close to as efficiently as Amazon and save lots of money. I would think Bank of America falls in that category. In contrast, for non-huge companies Amazon is a bargain, because you're in effect buying a small fraction of not only a large data center, but of each of the folks who maintain the services you use.

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"

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.