You cannot base any real analysis on figures take by looking at an artists rendering of the site.

The article says that they will have 85 windmills because there are 85 windmills in the picture. This is garbage. It is an artists rendering!

If you want to have a serious discussion, you have to wait until there is some actual real info to discuss.

Note that net metering is not running your plant completely off renewables. It's running it off renewables some of the time.

(I still do things the classic way: filesystem on lvm on luks on mdadm. not using ZFS yet.) I'm not sure it's exactly about what's required.

Consider wear leveling on SSDs. Only the filesystem really understands which blocks need to preserve data and which ones are don't-care. So to do SSDs right, it needs to pass info about unallocated storage down to the volume manager, whch then passes it to the encryption, which then passes it to the RAID, which then gives it to old-school "real" block device (which then passes it to the wear-leveling firmware, I guess). Sure, that can work. But when the filesystem can talk to the physical block device, it's easier. If you're writing block devices that implement things like volumes and encryption and RAID, from your PoV, things that are allocated vs not-allocated are totally different than how the filesystem sees it. To you, a block is just a block and a whole bunch of ioctls are totally irrelevant and not related to what you're working on. You're going to find this type of information to be pesky and you might not handle it right (or more likely, it takes a long time before you handle it at all). And in fact that has happened a few times, where certain block devices' feature set lagged a bit, behind what people with SSDs needed.

I suppose another easily-contrived example would be if you have a few gigabytes of data on a few terabytes of RAID, and need to [re]build the RAID. If your RAID doesn't know which blocks actually have data, then it'll need to copy/xor a few terabytes. If it's a unified system, then it can be complete after copying/xoring a few gigabytes.

..contracts requiring NDA's that now allows customers to review secret details of products or company practices on public forums.

Can someone who favors this, explain why this might be a good thing instead of a bad thing? Maybe an example? It sounds to me like endangering such a (seemingly, to me) bad practice might be an intended consequence, not an unintended one.

I can't even see how a review made under an NDA might be useful. The premise is that the reviewer is withholding information. "The spaghetti was excellent. [censored]I am prohibited from saying anything about the sauce.[/censored]"

or MIPS/Watt.

work per Watt makes no sense.

work/Joule is how "cheaply" you can get something done. MIPS/Watt is how fast you can get something done given a restricted power supply (or power envelope).

I started trying to think of situations where a person can have a wrist-worn PC but cannot have a handheld PC with them -- situations where people are constrained for some reason.

The obvious thing most people come up with, is where it's a natural or convenient constraint. You don't want to be holding something extra while you're swimming or swinging an axe or climbling a cliff. I think the related applications are already well-discussed.

What about when it's an artificial constraint? I initially drew a blank on how such a constraint would emerge, until I considered situations where the served parties by the two PCs are different, so that the handheld (if one is present) might serve the user (or manufacturer) but the wrist-worn serves someone else.

Once you start thinking of situations where the user is in an adversarial (or seemingly or potentially adversarial) relationship with the owner then it gets easier to see the applications.

Prisoners, parolees, etc. It's not so much that you let them wear the Pebble or iWatch, as you make them wear it. And your prisoner doesn't need to be surfing the web or otherwise doing things where the PC needs to communicate things to the user, so many of the disadvantages relative to handhelds, become totally irrelevant. The application, of course, is monitoring: being an open spy for the government.

Somewhat similarly: children. Mom wants to know where you are, but isn't really interested in giving you Yet Another porn terminal. Quit fapping and get back to your homework at the libra-- your friend's house?!? Get back to the library!

Marketing. Get 'em cheap enough, and these could replace your "frequent shopper" cards as your cookie. Wear our wrist PC as you walk around our store and check out, for a 2% discount. The application is spying, again. And I guess as long as it has a speaker, it can play location-triggered ads. "Whoa, you just walked right by our delicious canned spoo and instant flarn. Are you sure you don't want some?" The idea here is that you could perform the application with a handheld, but the existing handheld PC would be too pro-user so it might not really play the ads out loud and it might report false travel data. So you want the pro-store computer to be a physically different one. Then it becomes a wrist-worn simply because that's smaller and cheaper ($10 instead of $100).

Sweatshops. The Slurm factory employees are spending too much time on bathroom breaks, and texting their friends. Well, the employee wearable PC doesn't do texts, and it delivers a shock after 90 seconds in the bathroom. If a supervisor ever sees you without your wearable, you're fired.

Jealous spouses. Hubby's "Love Watch" chemical sensors are picking up interesting volatiles: perfume? My, he sure is breathing hard and the GPS has him in a residential neighborhood, not at the mid-town office. Oh, those are just fringe use cases: everyone knows the real purpose of the Love Watch is that it instantly relays every time you speak "I love you" into it. (OMG, that last part is so sickening that I bet a variant of this product already exists today.)

Think in terms of why you might want to "plant" (though not necessarily with subterfuge) your computer on someone else, to be your agent rather than the wearer's. Those may be the best applications for wrist-worn PCs.

I forgot to do the 30% part.

2900MWh times 61 * 0.3 or 53GWh, 53M kWh. 530,000 packs or $2.66B worth of packs (apparently I misplaced a decimal point before). 1 year of entire plant output.

A lot closer to workable, but still unworkable.

This is why grid-scale electricity storage is considered a nascent technology instead of a solved one.

And those batteries cannot hold a charge for 6 months anyway.

Even if they could, you're talking about a deficit of about 1/3rd at the peak of winter and a corresponding surplus in the summer. So let's assume you have a 1/3rd total energy surplus for 2 months in the summer and have to hold it 6 months until winter where you use it up.

That'd be 2900MWh times 61 or 177GWh. that's 177M kWh. A Tesla pack holds 85kWh, let's assume it's about to become 100kWh. And the pack costs over $10K, we'll assume it costs $5K.

That would mean they need 1,770,000 packs, at $5K a piece or $89B worth of packs. It's also the entire output of the plant for 3.5 years.

Does this seem workable to you?

I think you're not getting a good grip on the actual size of the problem.

> Is it getting major tax breaks?

Yes. It's getting huge tax breaks here. It got a nearly free auto plant from California.

It gets $7500/car in subsidy from the feds. Many states give $1500 to $5000 on top of that. Some countries they sell into give tens of thousands equivalent.

And this is beyond the emissions trading money it gets, which is a subsidy, but not directly from governments, just enforced by the government.

>becasue they are high end luxury vehicles. Do you send letter to Mercedes telling them their care a rudely expensive?

If they charged $90K for a car which is luxury equivalent to a Hyundai Sonata I would. It's a nice car, but it doesn't measure up to other $90K cars on luxury.

From the article:

"Reno gets an average of five peak sun hours per day."

Remember, as soon as you say the word "average" you are counting on a huge amount of storage so that you get the average amount of energy every day, even if that day is below average. And even if every day for the last two weeks has been below average.

In in fact, if you are using solar, you have to understand that nearly every day between the autumn equinox and the spring equinox is below average. That means you need enough storage to store up electricity all summer so you can use it in the winter! This is not at all realistic. More realistic is to make sure you produce more than you need in the summer and enough in the winter.

This does use more than solar though. However, I can't believe this guy counted the windmills in a PR picture.

Anyway, buying and erecting a 3MW windmill costs about $10M. That would mean Tesla would spend $850M on windmills. You cannot seriously think that Tesla is going to spend $850M on windmills before the plant even opens.

That was a TV show.

A work of fiction.

You shouldn't take what you saw seriously. It's not real.

What I'm trying to delicately explain to you, is that Heisenberg is still alive.

The difference is that phones are small and you only need to stock a dozen models to serve most clients.

Only a dozen? Let's see... within the iPhone 5S range in the US, we have 3 different storage capacities (16, 32, 64 Gb) in 3 different colour schemes, with 4 different network setups (Sprint, AT&T, Verizon, T-Mobile/unlocked). That's 27. 40 more for the iPhone 5c (5 colours, 2 sizes, 4 networks). The iPhone 4S is still for sale, but "only" 8 options there because it's 8 Gb only now - except the 3 larger ones are still under warranty, so make that another 32. Then we get the 4 and 3GS - but I'll stop there, because we're already at 99 different handsets for the iPhone alone, before we get into Android handsets! Call them $333 each on average, that's $33k of handsets you're mandated to store but not sell. That's insane - just to avoid a 24 hour wait to Fedex a replacement handset to you!? (Not to mention I'd rather have the replacement shipped to me next-day anyway rather than spend hours travelling just to collect it myself.)

Also, I seem to recall some of the Apple fan sites actually monitor stock levels in Apple's own stores - it usually takes a while for stores to have stock on hand after a launch (while customers buy up stock as soon as it arrives), then once a model is old they start running down stocks to avoid being left holding old kit. So, even Apple themselves don't actually carry stock on the scale the poster seems to be demanding, let alone 3rd party repair shops/vendors!

A few years ago, my MacBook Pro's Superdrive failed. Standard part they'd been using for years ... the Store would have a spare in stock surely? No, I had to wait a week for them to get one shipped from Panasonic ... then I was told I'd have to leave the MBP with them for up to another a week to fit it. Of course, by the time I add up the travel costs alone for 3 visits, I'd have spent the price of a brand new external drive, even before factoring in the c 10 hours of my time spent going to and from the Apple Store, so I told them to install the replacement drive somewhere it wouldn't fit easily and bought an external drive.

So, this legislation would be a hugely expensive "solution" to a trivial problem - and, of course, there's no guarantee that on the day your pink 32 Gb iPhone Verizon 5c happens to need replacing, someone else won't already have claimed that one replacement unit, so you can't have it anyway. Would the legislation somehow guarantee a quick replacement of the replacement by Apple, too? Or it would have to mandate everywhere having two of every handset, in case the first one's already taken ...

Yep, once we hit 6 Mb per screen we should be all set... that's the speed compressed HD video moves at at most. Anything that moves faster than a 1080p screen can't be represented by humans that fast, until we get to 4Kx1080.

That's something that hit me quite recently: with an 80/20 Mbps VDSL2 line, I can honestly say it's "fast enough" for everything I do. My satellite STB has a semi-streaming mode, where it downloads a whole show and will start playing once it has enough to play through without pausing for buffering - and it always starts in seconds. Big downloads, I'm almost always limited by the far end anyway; they tend to take longer to install than to download anyway - so if I had the option of a faster link, would I actually get any benefit at all? Previously, I jumped on every increase: from dialup (pricey, even a 30 Mb download was an ordeal) to half-megabit cable modem (yay, I could grab hundred meg files overnight), to ADSL, ADSL2+ (nothing big enough to leave downloading overnight any more), 50 Mbps cable modem, getting a big improvement each time. (I even have the option of upgrading to 330/30 if I want, for a price ... and it really isn't tempting, in the way I would have jumped at any chance of better speeds before.)

The funny thing is, I recall a Labour MP here in the UK saying something comparably dim back in 1998! At that point, all we had was expensive metered dialup; I was in a group campaigning to change that, and happened to meet him at some event. His reply was that he "didn't think the Internet was a good idea". One of those moments I really, really wish I could have recorded!

The choice [of a Broadcom SoC] by the RPi-team was utterly stupid and can only be attributed to incompetence.

Well, Eben Upton's job working for Broadcom was probably a factor there... Personally, I'd trace the idea back before he had that job - I recall a discussion about the Gameboy Advance developer kit in the summer of 2002, and the lack of affordable programmable devices at the time. I suspect he'd have had a real struggle getting anywhere close to the Pi's target price without getting discounted access to the Broadcom SoC he used, though. I haven't spoken to him recently, but my impression was that far from "RPi Foundation pressed Broadcom to stop selling BCM2835 to competing projects" as claimed, it was more "Eben twisted arms and got Broadcom to give RPF a special cut-price deal so they could afford it".

If anyone were to bring out a rival device from a "significantly superior" competitor, I'd be delighted to see it - and I suspect most if not all of the RPF people would too, since it wasn't about making money by selling lots of systems. (Of course, Broadcom didn't buy up the remains of ARM's parent company for nothing, so I'd be surprised to see something much better from a rival!) I was happy to see the Pi being ARM based, as a fan of ARM as far back as the ARM2 I first programmed, but I'm also happy to see rivals like the MIPS32 one mentioned recently: I like ARM, but I also like having a choice of platform, both hardware and software!

But acting like they don't, or even loosing a minor case might set other criminals at ease about the security of their data within the Microsoft infrastructure.

Which, of course, could be exactly what the NSA want - police and FBI priorities may differ, but I suspect the NSA would rather have access to more information, thanks to a false sense of security, even at the expense of not being able to use it in court easily. If they "win", they get to use evidence this time - and they just warned the next hundred criminals to avoid MS servers, because of this case. "Lose", and they can keep reading it all in secret, using the information behind the scenes instead.