When you start comparing crime rates, violent crime rates, gun deaths, or any other socially important data, you really need to pay careful attention to terminology. It matters little that the UK may experience only 1% of our gun deaths, if they also experience 800% of our violent crime rate. After you are mutilated or dead, is it really going to matter to you that you were killed with a gun, or a knife, or a stone, or you were choked to death? Violent crime is violent crime.

You're half right. You are right in that you really need to pay attention to terminology. You are wrong when you say "violent crime is violent crime". Why? Terminology.

"Violent crime" in UK stats is a very wide term that covers a lot of things. "Violent crime" in USA stats is a very narrow term that doesn't cover a lot of things. The terminology means different things in the two countries, so what is being measured is different.

Read this for more details, including links to the definitions being used. The fact is that the UK is less violent than the USA once you look at what's being measured instead of assuming "violent crime" means the same thing in both cases.

What on earth is the point of publishing the story days before we know for sure what will happen?

That's nothing. In previous years, Slashdot has quite happily published stories about Apple products while the presenters were still on stage announcing them. Hence the discussion is useless because everybody is talking about things that are shown to be irrelevant five minutes later and the stories invariably leave a bunch of things out, necessitating updates and subsequent articles. It's a real clusterfuck sometimes.

The $649 iPhone 5S costs Apple about $199 to build. And of course, that doesn't account for things like the cost of developing the software, or operating the servers that supply service to these devices.

Also unaccounted for: royalties of around $120-$150. So in total, an iPhone doesn't cost Apple about $100, it costs them upwards of $350.

Apple sells phones with cheap hardware worth about $100 for $600.

The $649 iPhone 5S costs Apple about $199 to build. And of course, that doesn't account for things like the cost of developing the software, or operating the servers that supply service to these devices.

why is paying by phone so much better than with plastic?

One less thing to carry around. No need to hunt for the right card. Fingerprint sensors rather than having to enter a PIN. The ability to incorporate new features with a software update. The ability for your phone to keep track of your payment history instead of relying on what your bank tells you. All kinds of features that are possible with a proper CPU and data storage behind it.

Cards are essentially dumb custom hardware that do a job in the cheapest manner possible. If cards weren't already pervasive and somebody came along offering the choice between cards and phones, everybody would be questioning why on earth you would pick cards.

USB's installed base is in the billions. Thunderbolt's biggest problem is a relatively small installed base

If they are changing the connector type, there is absolutely no reason to consider the installed base of USB. USB-with-C-type-connectors has an installed base of zero, not billions.

Why are you ranting about Stallman's ideology preventing a stable API? Stallman's got no say over the Linux kernel. Torvalds decided on the Linux API policy, and it was for pragmatic reasons not ideological ones.

Any idea where the REST API documentation is? I can't seem to find it anywhere.

MailChimp? Campaign Monitor? Google Groups?

It may be the most popular open-source mailing list manager, but that's not what was claimed.

Actually iPhone sales have been falling.

They sold 37.4m iPhones in 2013Q2 and 43.7m iPhones in 2014Q2. That's a year-over-year increase of about 16%.

I'm afraid you fell for a classic misleading graph.

I didn't fall for anything, that article is just dumb. It makes no sense to compare quarterly reports with immediately preceding quarterly reports for highly seasonal products like the iPhone because different quarters perform differently.

Of course if you look at the numbers that way in early September 2013 you're going to see decline - that year's iPhone model was released in late September. iPhone sales spike after launch and during the holidays, tail off through the rest of the year, then spike again when the next model is released. You can only gauge trends properly if you compare year-over-year numbers. And the year-over-year numbers show that iPhone sales are continuing to grow.

Apple's entire business is based on breaking new ground with an innovative new product, exploiting that products uniqueness before the rest start copying them and flood the market with "me too" devices. Then Apple has to move on to something else.

The smartphone market has been flooded with iPhone copies for years now, yet iPhone sales continue to grow. Their Mac division is still profitable and growing, despite it being decades old.

I agree that Apple get a huge first-mover advantage - this is to be expected. But I think you're dead wrong about Apple being reliant upon it. Apple will still be making money hand over fist with the iPhone when it's a decade old. They don't need to move away from old products at all.

It would be interesting to know how the story went inside Apple HQ as they added things like in-app purchases, set minimum prices/price increments/etc. for the store, and so on. Did they fail to foresee the problem? Saw it coming but figured that so long as their platform and hardware remained nicer it wouldn't hurt them since it would happen to the competition as well? Felt forced into it? (if so, by Android? by online/partially online stuff that got money out of users on the desktop/browser side and offered free mobile clients? by concern over some other potential competitor?)

One thing that seems to have been forgotten - when in-app purchases first came to iOS, they were for paid apps only. Freemium was against the App Store rules. I know as an app developer, I had a lot of clients who were unhappy about this. I also know as an app developer that Apple really couldn't give a shit about my clients being unhappy about it.

I doubt they felt pressured, but I expect that they foresaw the problem but underestimated how bad it would be for games. There are signs they are making small changes to the App Store to compensate for this, e.g. marking free apps with in-app purchases in listings. The App Store is so large now that I doubt they'll want to make large sweeping changes to policy, so I expect to see regular small changes to steer it away from the more shitty freemium business models.

in my experience the bug reports and feedback you'll get from Joe Public will be next to worthless

Bug reports and feedback aren't the only valuable things that can come out of this. If an application crashes for a significant number of users at a particular point, it makes it easier to prioritise. It also makes it easier to detect problems that occur with real-world data and system rather than test data.

It seems to me that they are reinventing the <a> element, badly. Semantically, what they are trying to express is a series of related links. What they should be doing is relaxing the restrictions on nested <a> elements and defining the meaning of this, then defining a suitable URN for dated copies of documents. That way they don't need to replicate perfectly fine attributes such as rel in a DSL that isn't used anywhere else and the semantics of the relationship are more accurately described.

Being able to scale from 1 billion records a day to 10 billion a day does not a premature optimization make.

It does when you are currently averaging zero records per day.