Quote: Microsoft says it will not search a user's email or other Microsoft service "unless the circumstances would justify a court order, if one were available."

In other words, they are saying that they are the judicial review, the judge, and the jury, and then the executioner -- they decide the process, they determine who will review the case, they decide who will make the judgement, and then they will read your email.

The first three bullet points in that list of reform processes basically says, "We will either use an employee, or a paid contractor, to review the situation to decide if this will continue". And if the reviewer says "Stop", well, they might use a different reviewer next time.

There is no independence. No double checking. No review. No safety at all.

As bad as Google might be claimed to be, this is Microsoft's bare expose: Even in the face of admitting a problem, they won't actually do anything to fix it.

Did Google mess up with the predecessor to Google Plus -- their first attempt at social networking with the ability to make real comments, real content, no silly 140 character limit, etc.? Yep -- and they had a fix in their ToS and code for that program within two days. (Sorry I don't remember the name. I actually liked it better than this new bleep(*) they have.

The difference: Google may be aware that datamining can break privacy. Microsoft says from page 1, they will break your privacy.

(*): Due to the courts and the FCC, my right of free speech has been revoked.

It's about strange sleep patterns.
It's not about "All nighters".

The article says:

...

Yes, it's mice, not people. And yes, it says that once doesn't cause harm. It's after days of this that the protection mechanism shuts down.

I agree that we need to tighten up federal aid.
I disagree that we can tell ahead of time of who will be a burden or not.

The real issue is the granting of loans in the first place.

The real issue: As the costs of a degree go up, the size of the loans go up as well. If the maximum size of the loan were to stay down -- or perhaps only go to people getting an education where the costs were limited -- then there would be pressure on the universities to keep the cost down.

In other words:
In a balanced market, if you price your produce -- an education -- too high, fewer people will buy it => you lower your price.
In a distorted market, no matter what price you put on it, someone will fund it's purchase -- and then demand repayment.

To fix the student loan problem, remove the "We'll fund your education at any price", and replace it with "We'll fund your education if it is likely to be repaid".

But that's only part of the answer. The other part is loan insurance.

"We'll fund your education, if it is likely to be repaid, if you agree to pay back the cost, plus a premium for loan insurance, along with the guarantee that if you can't pay it back within N years of graduation, your loan is forgiven". (N should be around 7-10).

In other words, since you know ahead of time that it is impossible to repay every student loan, that you limit the size of the loans, and expect those who succeed will cover the costs of those who fail.

What is the long-term result of that?

1. More people getting loans, and going to school.
2. More people getting degrees, and yet not having the jobs for all of them to get the high-paying jobs that will repay the debts.
3. Partial repayments from most, full repayments from others, and lots of discharged debts
4. As the discharge rate goes up (as fewer graduates, by percentage, can repay), the amount of loan goes down, and the amount of insurance goes up, reducing what can be spent on schools
5. Which in turn forces the cost of education down.
6. Which runs the risk of reducing the quality of education.
7. Which either causes other problems in the future, or forces the education system to fundamentally change.

7 is the fun part. With thousands of schools, and thousands of experiments, someone will succeed. Someone will figure out how to make schooling work better.

And then, either lots and lots of people will copy it, or it will be copyrighted, trademarked, patented, etc, and locked up for just one single group, with extension after extension, until ...

Wait, actually, our current system doesn't sound so bad after all :-)

Imagine what some of the most influential minds of our time could accomplish with an extra hundred years, or even an extra sixty. Regardless of whether or not their ideas were good ones.

I'm not saying that longer life is better or worse; I happen to like it.
I'm just saying we have a lot of maturity to do as a species before we could manage that.

Would you rather make a fortune in basic based on code stolen from someone else?

Alright, I am curious. What, exactly, stops you from flipping the switch back to "on"?

Yes, I am ignorant. But it's a serious question.

It was running, right? So you know that it works.

So,
1. What is so hard about starting it up, and
2. If you knew you could not turn it back on after turning it off, why turn it off?

Don't forget the model 16 -- a true multiuser system that ran Xenix (a unix variant).

68010, I think. My first unix system (v7 based initially).

Ahh... the days

Internally, I believe it was an upgraded model 2.

Genetics give three different types of insulin response to carbohydrate intake: low, medium, high. They are roughly (warning: 30+ year old data) 25, 50, and 25%. Diabetics (again, 30+ year old data, and incomplete/initial when it was taken) (insulin resistance, not pancreatic failure) correspond only to the high insulin response.

Any attempt to document the response of food/diet to population results that fails to account for which type of insulin response you are testing on is a fail. Basic, simple, first step of peer review fail.

When I see articles like this (and I read the article), I can only think that a non-peer reviewed preliminary study is getting as much press coverage as should go to well-established, peer-reviewed, properly control-tested studies.

Essentially:

1. Different levels of insulin response will have different effects on body behavior, as well as survivability in times of famine.
2. Low insulin response basically makes it impossible for your body to have the same behavior as high insulin response
3. Diets that are healthy for high insulin response is probably good for everyone, but low insulin response can safely eat things that high insulin response cannot.
4. High insulin response means a bigger store of fat, and survival through times of food scarcity.

None of that was addressed by this, or even most of the food studies that get lots of media attention. It's almost as if the people covering the news have no knowledge of what they are covering.

Oh, wait, what did I just say?

I'll just train the butterflies, and let their wings do the work for me.

Err, is that "obligatory XKCD"?
http://xkcd.com/378/

Nope.

All you have to do is be born or naturalized in the USA, and subject to its jurisdiction.

Unless, you are the courts: Then, you say all you have to do is be born or naturalized, and then you are subject to its jurisdiction.

Or, if you are someone who actually paid attention to older rulings: then, all you need is for a state to consider you a citizen; once one state says you are a citizen of that state, then all states have to accept you as a citizen.

Believe it or not, that was re-affirmed in a case from (memory ... might be off) 1999, against the state of california, for trying to make new arrivals to the state get less welfare support from california.

I think the quote was something like, "Citizens of the united states, whether rich or poor, have the right to choose their own state; states do not have the right to reject citizens".

The issue of being a citizen of a state or not actually dates back to old court rulings that had to deal with someone that was born, and lived entirely in Washington DC, and had never lived in any state. Prior to the 14th amendment, being a citizen or not was entirely up to the states to declare; some gave that to blacks, and others did not.

It is my understanding that the _early_ supreme court rulings after the 14th amendment actually made this clear: states still could issue citizenship, and if they did not, you could claim citizenship from the central government, and then get all the federal rights in state courts.

Don't ask me when the courts started messing up -- I don't know.
But now, it's more "If you are born here, then you are automatically subject to federal jurisdiction, and the restrictions of article 1 no longer apply".

Yes.

If you ever make a mistake -- and we are human -- then yes you need source control.

A better question: Do you need the ability to do lots of cheap branches and easy merging? I find it really makes things much, much easier, but changes the nature of "spagetti code" to the process of keeping track of all the branches.

EDIT: Brownian motion. Brownian motion. I can't actually find an "edit post" button.

That is the business man approach. "Your value, your contribution to society, the quality of what you do is determined entirely by your ability to make money from what you do".

No.

Skill at an art is not the same as skill at marketing that art. And it's different yet from skill at making a profit from marketing something that someone else did.

Skill at business isn't the same as skill at _blank_
Skill at marketing isn't the same as skill at _blank_

When did Einstein's paper on the photoelectic effect, or his paper on random walk of atoms (sorry, the name eludes me this morning) become "skill"? He got his Nobel for some other paper he wrote, where all he did was point out that the three basic assumptions of physics that people used were contradictory, and what happened when you only assumed two and let the third go by the wayside.

Well, in fairness, pointing out that common assumptions are contradictory probably is Nobel worthy. I'm sure Godel got a Nobel for doing that to math and logic, right?

(actually, I don't know if Godel got a Nobel for his incompleteness work.)

---
Side note: How do you get angle brackets in this text? I had to change those to "_blank_" because I could not use angle brackets around the word "blank".

But if you pay attention to modern evolution of OS's, everything is a directory makes more sense / a better map.

Files have extended attributes, various data/resource/etc forks; real directories cannot be read, but have a bunch of index'd names; etc.

People who say that hardware devices need special parallel communication: each such communication is just another entry in the directory.

And then there's the whole "This program contains different versions for different architectures" is just straight multiple single streams.