What's interesting here is that, for all other food, the body has mechanisms to stop us from eating more when we've had enough. Lustvig claims that for fructose, this mechanism doesn't work properly.

This means that, the more fructose we eat, the more we will eat beyond what is good for us. I would personally put addictive substances which you can't get enough of in a class of its own.

I've heard this repeated many times, but in this case it's absolutely irrelevant. Interesting fact, sure, but irrelevant. Lustvig says that because it's a substance that can only be metabolized by the liver, it's a poison. I can't find that definition on wikipedia, but seems reasonable. Maybe that class of substances has a more technical name, but sometimes using trivial names is a good thing.

This isn't about LD50. Alcohol has an LD50 too - like everything - but we don't claim that nothing bad happens if you stay under LD50 at any given time. Lustvig makes some pretty specific claims about the long term effects of fructose, and that's what's interesting here, not wether it's toxic or not, or wether that term has no semantic value.

I'm wondering about how the moderation "insightful" is used here on Slashdot. What you say is interesting, but adds nothing to the debate. So, insightful in general perhaps, but not in this instance.

This article seems to indicate so:

http://www.computerworld.com/s/article/9214002/Safari_IE_hacked_first_at_Pwn2Own

"But the Safari patches still had a part to play in Vupen winning. If the vulnerability used by Vupen to hack Safari had been fixed in 5.0.4, TippingPoint would not have awarded the $15,000 prize."

Huh. Never had those problems, but I guess there are still some issues to work out, for sure.

iOS should close apps when it's low on memory, so if they keep running because they're holding on to the camera or something that sounds like a bug. The only apps that should keep running indefinitely is the currently running media app and maybe an active VoIP app, as far as I understand.

Have you tried iOS 4.2? Heard it should improve performance on 3G/3GS.. maybe they've worked out some of these things.

"No, it closes the applications."

Only if they're running. Most of them are not. My point is that it's pointless to clear the list.

"No that's wasn't the whole point, but had they succeeded on that aspect there wouldn't be a task manager. There is."

It depends on your definition. If you call it a task manager because you can use it to switch between apps, then sure. If you call it a task manager because you can tell iOS "if this application is open then close it", then sure... But from an average users perspective it shouldn't be considered anything but a recent app list, since that is how it behaves, there's no reason for a user to remove an app from it unless the app has hung somehow (can also be done with the sleep button) or they want to hide that they've used the app.

I'm aware that there's been some performance issue with iPhone 3G, so maybe it may serve a purpose to clear it there, but I noticed no performance degredation on 3GS.

The OS could be more intelligent about this if apps were classified according to what kind of application it is. An example: some apps on Mac OS X (Pages, Word, Browsers, etc.) doesn't do anything unless you have a document window open. If I haven't been using one for a while, a "document" application could just as well be closed. (In Windows/Linux it kinda works like this already though).

You'd have to find out what classes of applications you need, specify desired behaviour and perhaps create an API for them though, which is exactly what Apple has done with iOS. Personally I think it works quite well, for instance audio player applications work exactly as I would expect (works with media keys, doesn't compete for audio with other media apps, quits when I'm not using it anymore).

In a real desktop OS you'd obviously need to allow traditional management of apps though, but an unexperienced user shouldn't be required to use it.

Windows, Linux and Mac do this to some degree already I guess (especially for system services.. think xinetd and launchd), but I think it could be done better and more systematically. But then again.. everything could be done better or differently, so I'm not complaining. But it's not unreasonable to discuss how things might be better is it?

If you're going into the recent apps list (double clicking home) and remove every item, then what've you done is clear a recent app list, not close running applications. iOS will close applications running in the background if it's out of memory or the apps have been idle for a while. It's a common misconception that the list is a running apps list, leading some to think they have to close them. I think the only apps that will run continuously in the background no matter what is the media player app and some Apple apps. The whole point of the way iOS multitasking was design was that you shouldn't have to use a task manager. Task managing is something that is utterly pointless for a human being to do, and it's insanely stupid that we're actually still doing it in Windows/Mac OS/Linux.

If Apple has declared war on tinkering, how is it that a 13-year old can freely and successfully compete with mega-corporations as John Gruber pointed out? How is this less tinker-friendly than the Apple 2? Sure.. it's on another level. We can't really tinker with the circuits anymore.. but do we want to? You couldn't tinker with the inside of the memory chips or CPU in the Apple 2 either, because technology had gotten so far that it was better to produce these circuits as integrated ones. But few (none?) complained about that, because it was recognized that it was useful to move things up a level.

The future of computing looks more and more bright. Hopefully all alternative will not perish (My dad owns a modern car with a tinker-hostile blocky engine and an old tinker-friendly veteran. I own an physically tinker-hostile iPhone and a tinker-friendly PC.. deal with it).

> All browsers should support Theora, and publishers can host H.264 (or any other codec) seamlessly on top of that if they want. As of right now it's just as impossible for all browsers to support Theora as it is for all browsers to support H.264 (well, I'd argue it's not impossible for Firefox to support H.264 in practice, but that's another discussion). Mobile browser just doesn't have the hardware support it needs for decent support for Theora video. Personally I'd prefer it if H.264 became the standard. It's partly developed by the same standardization grouped that developed JPEG (ITU-T), and in my opinion JPEG is a pretty good example of a successful standard. I just don't think that Theora will be as successful as H.264 without proper industry backing. And even though the licensees aren't legally exempt from patent litigation with H.264, I think you'd be more protected in practice (it's in ITU-T/MPEG's interest to protect the standard).