metageek

IIRC, it takes just as much energy to reach L4 (or L5) as to reach the Moon--they're in the same orbit.

If you've got settlers on the Moon, or beyond the Earth-Moon system, L4 is probably a good place to build a commercial and industrial outpost; but, for your only space installation, something lower down is much cheaper.

That makes more sense, thanks.

Back at Netscape, Jim Barksdale hated the term "browser war", on the grounds that a war justifies extreme behavior. He didn't want us believing the browser war meant we should be breaking the normal rules.

>Seriously if AT&T doing this, there is no reason to believe that other providers do not.

The article says AT&T has the data for any telco that uses AT&T switches. This is a little odd, given that the company currently known as AT&T is not the same company that used to manufacture switches; but, if this has been going on for 26 years, I suppose the government would have insisted it continue across corporate splits and mergers.

ARM doesn't make anything physical--but neither do most software companies. ARM makes products that their customers actually want. Patent trolls don't.

If LR operated as described in the indictment, it didn't really have information about its customers, so those records might not have been all that useful.

paddy_m said browser engine, so Chromium doesn't count. And WebKit wasn't actually new; it was based on KHTML.

Isn't JS injection a copyright violation, since it creates a derived work? Or has that idea been shot down before?

Or as soon as the states close the loophole so that this program is treated the same as a lease.

It did have one, but it ablated.

"Neesh" is taught as Correct in the US, though it's not universally used.

I agree, except for the "focus" bit. The engineers that worked on this alleged feature would not have been much help signing new content distribution deals.

(Well. I'm generalizing. I'm an engineer, and I know I wouldn't be.)

My main Netflix player is a Blu-Ray player from 2009. I don't think Samsung is going to push an update to add the Don't Share This button. I'm just not going to connect my accounts.

Doesn't that run afoul of the same problem as Righthaven? There is no bare "right to sue" that can be transferred without transferring the copyright.

>I'd apply the same copyright principle as phone books

So would I—phone books are not subject to compilation copyright, because there's no editorial choice going on. That was settled in the _Feist_ case.

And presumably the autistic community knows about ASL, and the children who can learn it, do—it'd be greatly preferable to carrying around a multi-thousand-dollar device. I think we can assume that this child's problems leave her unable to learn ASL.

Yes, but that sort of regulation doesn't require medallions.

>One may argue that Europe went too far with their "social" expenses (hence, the current financial meltdown).

I'm not sure Greece in particular went too far, as such; isn't the problem that they've got a massive epidemic of tax evasion, and no political will to tackle it? If they were enforcing the tax laws on the books, maybe they'd have the money to pay for the services they're providing.

>So, we "saved" on a silly, cheap follow-up, but ended up spending a huge amount later. And guess what? These patients can't pay their bill anyhow, so we have to write the expense off.

Hear, hear.

Nor is "how not to be interrupted" covered in the CS curriculum.

Interesting; I've encountered "yak shaving" as meaning something slightly different: it's when your main task depends on a subordinate task which is long, difficult, and stupid--e.g., in programming, it's something which the tools should be doing for you. (Bonus points if you know that, on a different OS, or in a different language, the tools would be doing it for you.)

The original MIT meaning is closer to what I recently heard described as meta-yak-shaving, which is when, before you can start shaving the yak, you discover that each hair of the yak contains a subyak, and each subyak has to be shaved individually.

They're not even using the Android sharing system. The only way to share from the Instagram app is to upload to Instagram, and optionally tell it to forward the photo to one or more of the sites they support (Facebook, Twitter, Foursquare, Tumblr). If you want real Android sharing, you have to switch to the Gallery and share from there.

>And is there anything wrong with people feeling like they are creative?

Yes, there is, when they aren't being creative. When they're just choosing a cookie cutter, telling them they should feel creative is cheating them of the real thing.

I think ianterrell's idea was that, if you're going from, say, version 2 to version 3, you release version 3 for free, but all the new version 3 features are a single in-app purchase. No nickel and diming.

One result of this approach is that everybody gets bugfixes for free, which users are sure to like better than a model where they have to pay for fixes.

One thing you have to decide is what to do if you then release version 4, and someone never paid for 3. Does he have to pay for 3 and 4 separately, or can he just buy 4 and get the 3 features for free? (I'm pretty sure you don't want to let him buy 4 and not have 3; that's where you start getting configuration explosion.) The latter has precedent in the packaged world, with vendors that would give you the same discount when upgrading from any earlier version.

Yeah, but up here we're more likely to slip on ice and drop ourselves.

I'm getting the right cert.

You'd probably have to write new code; but you could port the algorithms, which are the distillation of their trading expertise. That'd boost you several years ahead in domain knowledge—and, as GS pointed out, you'd also know how GS was going to trade, which would help you predict market movements.

I suspect the latter wouldn't actually be all that accurate, though, since all those split-second decisions are going to be highly dependent on how fast everything runs, which means your GS emulator can't make the same decisions unless you port everything exactly.

I just ran into it for the first time. I bought an MHL cable from Monoprice, and it was DOA. I requested an RMA; they asked one question to make sure it wasn't pilot error, then said they'd send a new cable, and I didn't need to send the old one back. I was pleased; my gold standard has been Amazon, where they pay the shipping, but I'm still out the time it takes to go to the post office.

Even if you're not putting the RString struct on the stack, the embedded string optimization means calling malloc() just once instead of twice.

And are they also good engineers? That's the hard part; the kind of obsessive focus usually required to be a good geek tends to impair the social skills required to be a good manager. Not always, I admit, but usually.

I did work in Silicon Valley, but not for very long; and it was in the late 90s, at Netscape, so my experience was probably atypical.

>I've always assumed that e-books are expensive because their buyers are known to be affluent.

That's probably part of it, yeah. Maybe we'll see that eroding with the spread of smartphones.

The problem there is that generating multiple versions of the ebook throws away some of the economic appeal. The segmentation between hardcovers and paperbacks is natural in the format: once the hardcovers sell out, and nobody's willing to pay hardcover prices any more, you need to make a cheaper edition, so you make paperbacks, which require you to spend effort on laying out the book again. With ebooks, once you generate the initial edition, there's no inherent need to invest in the book ever again. Once it stops selling at $20, you just cut the price.

To yield segmentation like this, there'd have to be a strong economic incentive, or else a cheap way of generating the multiple editions. Maybe if the high-end version had pictures, and the low-end version didn't; then you could build them both from the same source file.