As a former Newton developer, I found developing for iPhone so tedious and needlessly complex that it just wasn't fun. It's great that the iPhone SDK is free, but I'd rather pay hundreds of dollars for it in exchange for a decent development experience. The Newton Toolkit was something like $700 but worth every penny in terms of the amount of effort it saved developers; it Just Worked. It made the simple things easy and the hard things possible. Whereas the iPhone SDK environment is powerful almost entirely at the expense of ease-of-use; simple apps are far more complex than they need to be. In particular, both the Interface Builder and the DRM stuff seem very poorly integrated with XCode, providing many opportunities for things to break in confusing ways.

If paying these guys $99 means I can make a usable iPhone app without spending much time in XCode, Objective C or Interface Builder, that seems like a huge win to me.

IPCC corrected the error relying on one person's speculation in some paper.

How, exactly, did they "correct" it? If you download the report right now, will that mistake have been fixed? No? Okay, is there an officially-maintained errata somewhere you'll find the correction in? No? Then what do you mean by the claim "IPCC corrected the error"?

I remember once reading about a system that was supposed to have existed about a century ago - I can't remember the name associate with it - but it was a school system that was mostly aimed at teaching the poor. It was a lot like the montessori system except it was really cheap by virtue of having about a thousand-to-one student-teacher ratio. It scaled much the way karate schools still scale today - there's one master teacher who teaches a group of senior students, who teach less senior students, and so on down the line. In this system, you weren't judged to have really mastered a subject until you've taught it to somebody else, so it taught teaching as well as learning. That would be a great system to experiment with but I suspect the teacher's unions would never allow it in a public school. (Near as I can tell, public schools are mostly run for the benefit of the teachers, not the kids.)

Almost a polar opposite approach is the Sudbury Method, where you just let the kids learn what they want from who they want in an organic fashion. But age-mixing seems like a big part of the secret sauce for all of these. It just strikes me there's something fundamentally *wrong* about the idea that every kid needs to learn the same subjects in the same order at the same time as every other kid the same age. Kids aren't interchangeable parts. They'll have different aptitudes and interests; they'll want to race ahead in some areas and fall behind in others and *that's okay* as far as I'm concerned. A good school system should embrace the ability to do that. Even if it means teachers are a little less powerful.

What we have to ensure is that this program doesn't fall prey to overzealous parents - especially in the "everyone is a winner" mentality that we currently possess in America. I guarantee that if this gets passed there will be an outcry of "my child shouldn't be discriminated against. (S)he should be able to head to college too at this grade!"

Which...would be awesome! It might be the start of a trend towards everyone wasting fewer years being babysat before they go get real credentials and real jobs and a real life. Then maybe we can slice a couple years off grade school too while we're at it. I'm really failing to see a down side here. Ultimately we could have mixed-age classrooms with a fair bit of self-paced learning and students helping their peers keep up. From which kids leave "when they've learned enough" rather than because the earth has orbited the sun a specified number of times since they arrived.

Maybe you think that wouldn't work as well as what we've got now, but isn't it worth trying?

>>there's been a rising cost of disasters due to more CO2

>This is based on the story by Jonathan Leake, I'm guessing. The IPCC report said that one study indicated there was an increase in costs due to AGW while other studies did not detect a trend. So the IPCC report was balanced. Isn't that what people want?

The IPCC based their claim on a preprint, unpublished, non-peer-reviewed article which, when eventually published, did not show a trend. The IPCC ignored actual peer-reviewed articles that showed no trend to do this. And maintained this position despite complaints by the "expert reviewers", going so far as to misrepresent the view of one of the scientists (Pielke) whose work had found no trend. So the IPCC report was *not* balanced on this subject. "balanced" would have been to show the *actual* consensus view of the peer-reviewed literature at the time: that there was no trend.

I'm not basing any of this on stories by Jonathan Leake. Leake wrote some stories based on what had been uncovered in the blogosphere, of which there's a lot more where those came from; the IPCC has responded with weak apologetics that included such silly claims as "only one error has been found".

> And here's the thing: when the IPCC is found to be in error, they correct it.

Your evidence for this is what, exactly? Like I said, there is no provision for correcting errors other than waiting for the next report to come out year later and hoping it gets fixed then. Even if they admit an error has been made, they don't republish the report fixing the errors and don't publish an errata listing them. Do they?

>That's something I don't see from skeptics.

Maybe you're following the wrong skeptics. My impression has been the reverse of yours. When people like Steve McIntyre or Craig Loehle or Ross McKitrick make a mistake, they admit it and fix it and redo the work to see if it made a difference. And they make their data public so anybody can check it. When people like Michael Mann or Gavin Schmidt or Phil Jones make a mistake, they deny it and hide their data from critics and pretend the error doesn't matter or doesn't exist.

> The IPCC document is something like 2400 pages and so far there has only been one error found.

That is nonsense on stilts. There have certainly been many dozens of errors "found". Maybe you mean "publicized in national papers", in which case there were - whst? - four recently? At least four. More if you count sub-parts, like the claims about glaciers were wrong in many specific aspects, not just one. The errors I can think of off the top of my head:
(1) Himalayan glaciers could be gone by 2030
(2) there's been a rising cost of disasters due to more CO2
(3) 40% of amazon forest could be destroyed by warming
(4) 55% of Holland is below sea level.

The fundamental problem is that there is no defined process by which errors found in the IPCC reports can be reported, collected, and fixed. So if I find a problem, there's nothing for me to *do* with that information such that it would show up in anyone's tally of how many errors have been found. Bunches of people have been finding bunches of errors but it wasn't until after the tide turned with climategate that this became newsworthy. Now newspapers are willing to write stories critical of the IPCC. But still, when somebody say there's only been one error found they mean "...that I've heard of. This week. And am willing to admit to."

The idea that jurors need to be entirely ignorant of the case is a relatively recent invention and arguably a bad idea. If you go back to a time when people lived in such small towns that everybody was likely to know everybody, you find a different notion: that it was good for jurors to know not just the facts but the people involved, because already knowing a witness made it easier to accurately judge the credibility of that person. Turning courts into fact-free zones like they are today makes lawyers and judges more powerful but it's not clear it produces better verdicts.

I'm sure it has, but why would you expect public school to be a significant factor in that improvement? The best way to learn grammar and vocabulary is to read a lot for fun. I suspect the parent poster was reading and writing far above the level school was teaching to in any given year.

I thought school was a waste of time largely to the degree that it prevented me from learning more by reading up independently on the subjects I was interested in. (Though I still did a lot of that, stealthily, during class lectures.) Given that I tested as "reading at a 12th grade level" when I was in 9th grade, are you going to claim whatever "working vocabulary" improvements I made during high school are due to having taken standard 10th, 11th, and 12th-grade English? I question that assertion.

That sounds a little unclear on the concept of unschooling. If you're consistently teaching "a half-day worth of lessons", that's just normal homeschooling. Note that the most efficient way to do normal homeschooling is to get self-paced homeschooling workbooks and have your kid work through them; that is how home-schooled kids usually come to know more than their parents had the time/competence/energy to teach.

The best way to learn Grammatical English is to read a lot of books until you internalize the rules. Music and Art suggests signing up for private music or art lessons on the side - those are easy to find and you'll get better, more personalized instruction than one would in the public schools.

No, we're not. That is not the premise. One premise here is that some (most?) kids can learn in a way that better suits their interests and desires and better preserves their intrinsic motivation if they have the opportunity to exercise their intrinsic motivation. Think of motivation as a muscle that can waste away. If all kids get to practice is learning what somebody else has told them to learn in the exact manner and according to the exact schedule set by others, their ability to set goals for themselves and maintain interest in a subject for its own sake is likely to suffer.

Some kids can learn more or deeper or more efficiently or in a way better tuned to their particular needs and interests. It does not matter why they can do this. Maybe they had good genes, maybe they had good parenting, maybe they were just lucky enough not to have the curiosity beaten out of them at a young age. Unschoolers tend think that most kids could do well in this environment but especially the ones that "don't fit in" so either the top or the bottom performers are likely to be good candidates.

It's worth noting in this context that if you've ever seen the movie "Stand and Deliver", one way Jaime Escalante got those great results was by giving his students access to self-paced homeschooling math workbooks rather than using the standard school-provided curriculum.

Kids are interested in different things at different times. Forcing everyone to learn the same things at the same time as their peers is efficient in teacher time but inefficient in student-time. In an average class a third of the students are behind and lost, a third are ahead and bored, so at most a third could be learning productively, but most of those aren't particularly interested and aren't necessarily paying close attention. So the information transfer rate is very low, and the system makes up for this with lots of review and by taking a lot of time to cover a very very small amount of material. One thing people want is for kids to have the ability to learn at their own pace rather than at the pace of the class without being penalized for this. Whether that means "zooming ahead" in some subjects or "falling behind" in others and catching up later. When kids have the opportunity to do this they tend to learn more in less time and enjoy the process more.

Many smart kids still need to be told what to learn, and forced to learn it

Do they? What do you mean by "need"? Do you mean that if they learn what they choose to learn they won't learn the things you think they should on the timescale you think is appropriate? That's probably true, but they'll learn other things instead and it's not clear a priori that the other things they're learning are less valuable than the things you had in mind. Maybe they'll learn more about the joy of finding things out for themselves. Maybe they'll learn more about the joy of setting and achieving their own goals in a self-directed manner.

Actually it does, really, go quite a lot faster than you can walk. Unless you can walk over 12 mph, which I rather doubt. But your impression it doesn't does reveal another possible reason people scoff at it. In trying to make it a mass market device, they bent over backwards to make it safe. The segway is, quite frankly, too safe. Too few people have hurt themselves by using one. The multiple keys and speed limiters make it inconvenient to go fast; removing any element of skill in staying upright makes it hard to swerve out of control. So unlike with a motorcycle or a skateboard - two forms of transport it might otherwise seem to resemble - there's no illicit flirting-with-danger cachet. A segway rider isn't risking death and isn't demonstrating skill because the machine won't let you go faster than can be easily stabilized.

Riding a segway is like riding a tricycle slowly wearing a huge helmet, full pad, and full yellow reflectors. Who needs to be *that* safe?

So I recommend they design an "extreme" version of the segway with different styling which includes an "overdrive" gear with no speed limiter so you can drive fast enough that it actually takes skill not to die. Or perhaps let people know how to *hack* their segways to get rid of the speed limiter. Once a few thrill-seekers have died from going over 100mph on a segway and crashing into a tree, it might start to seem a little cool.