Yeah +1 for spideroak. But you still do need to trust them.

The source is closed. So you can't inspect and build the client yourself. So you have no way of knowing whether its really zero knowledge or not; or whether the client can or is sending the keys to the server etc.

They also specifically disclaim zero knowledge for web based access and mobile. The former should be obvious, but the latter is a bit of a surprise/disappointment.

Still I -do- generally trust them; and recommend them. Their business model isn't advertising and harvesting data.

But the fact that I've decided to trust them is a far cry from it being a provably trustworthy system.

Yeah, me too. I think part of the problem with recyling is lack of education. I honestly don't know what actually is and is not ok to put into what...

For example I recently bought a mcdonalds meal...

What about a macdonalds bag? Is that ok to put in the paper?
What about Unused napkins? Used napkins?

What about the 'cardboard' thing the bigmac was in? Is that paper or cardboard or is it just garbage?

Can I recycle the the plastic fork? The little plastic bag the fork came in? or the straw? The plastic lid on the cup?

What about the wax paper cup?

Would I need to wash all these things? or does the recyling processes itself mean that a bit of salad dressing on the fork, or a bit cola on the cup is completely irrelevant?

And what the hell am I supposed to do with a pringles can or the containers Ice Tea powder comes in? The ones with the cardboard cylinder (although maybe some sort of foil coating on it?) plus it has a metal ring at the top lid, and a metal base.

Is the plastic lid recycleable? The ice tea has the #4 recyle symbol on it... but the pringles can doesn't have any symbol that I can see... but surely its recycleable? isn't it?

Should I err on the side of caution, and toss anything I'm not 100% sure of in the garbage, or should i err on the side of recycling?

I think most people, like me, simply don't know the answers to these questions and we make a lot of mistakes we'd avoid because of it.

Actually, one of my premises, is that as the government is paying for it, we ARE paying for it.

When that someone is the governent, then yes? Isn't that what a library is?

Again, we are the paying customers.

But Microsoft set a price for doing so, and then it was paid. Its not taking away from anything, its a whole extra revenue stream.

Yes. They set a price to compensate themselves for the time it would take to do that, with enough profit built in to motivate them to do it. Then we paid it.

:eyerolls:

We're all learners. But anyone with more than a passing familiarity with C++ already knows everything in that puff piece of a Dice article.

They never said that. Either of them. A few blowhards and talking heads have made such comments over the years; and the press eats it up because that's what the press does.

The country as a whole is grappling with rather intractable problem. It simultaneously wants it to be possible for the NSA to break encrypted data belonging to terrorists and criminals in general with a proper warrant. But it doesn't want encryption itself to be compromised with a back door so that they can get in whenever they want without leaving so much as a trace. Unfortunately meeting those two objectives right now is a fantasy. So we get loonies at both ends making nonsensical statements.

As long as saner heads prevail who cares. Linux isn't a terrorist front and any honest investigation not run with the same mentality as the spanish inquisition will promptly discern this to be the case.

The thing that irks me is that once various governments and organizations have "sucked-it-up" and ponied up the "ransom" to keep XP going -- why cant the public at large benefit from this. Especially given that we are the ones literally paying for it.

Once the patches are written, tested, and released why aren't they available on Windows update?

Don't get me wrong, I want XP to die in a fire. Cutting over to Vista onward, embracing 64 bit*, leaving the days of "administrator by default" behind, etc were all good things. But still if my government dropped 9 million bucks to get MS to develop some more security patches for XP; it'd be nice if the lathes at work could have them too.

* (yes, yes, i know xp 64 bit existed. shut up. :)

I'm pretty sure those are the sort of things domestic terrorists would actually have... so what exactly is your issue with it?

A bunch of food and a gun doesn't get you classified as a terrorist; its a flag that the group might be worth investigating.

An investigation isn't persecution.

Even a little safer is better. Roundabouts generally bring the speed of traffic approaching the intersection down.

They put one in by the school near where I live, and the new roundabout doesn't take any more space than the old intersection did.

And that's a residential simple one lane going each way intersection, with parking along the curb. At the roundabout the curbs bulb out preventing parking, and create the room for the circular space.

Previously, there was effectively 2 lanes of room at the intersection, so someone waiting to turn left could be driven around on the right side, or cars could make right turns while people were waiting to go through or turn right. This complexity is what often led to accidents both between vehicles and involving pedestrians.

Now, its always one car entering from each side at a time; and it never really stops except for pedestrians. Honestly its been working well.

It seems to scale well to 2 lanes as well We have several 2 lane round abouts and those also work well.

No you don't. You can use a local account; and still use the app store. I've both 8.1 and the 10 tech preview setup like that.

Using an function where its undefined is fundamentally invalid.

lets say I define function:

f(x,y) = 1+ x/y; for all y in R such that y>0; for all x in R such that x>=0

If I call this function f(-4,2) same thing. Its not defined, its NOT defined for xbased on preceding values

Its not hard to completely break this.

Consider sin(1/x)

http://www.wolframalpha.com/in...

See what that does? It doesn't merely shoot off to infinity; its clearly bounded between 1 and -1. But nor does it have a clear value at x=0; its limit is undefined at 0.

http://www.wolframalpha.com/in...

Limit does not exist.

It does go towards a finite number, it doesn't go to plus or minus infinity. Its JUST not defined. Don't ask for a value there because there isn't one is the ONLY correct answer.

The point is, a function is defined on an interval. If there is somewhere where its undefined, it is an error to ask it for a value there. If you are interested in where the function is going, then don't ask it for a value, as it where it going (ie take a limit). But don't be too surprised when you find out there isn't a limit either.

And don't forget that lots of real world functions have values that don't line up with their limits.

f(x) = { 2x for all x in R, such that x != 5; 5 if x = 5 } this is a perfectly valid function. The limit of f(x) as x->5 = 10 from the left, and the right; but f(5) = 5.

I can't imagine a scenario in financial where division by zero should ever be fudged.
I can't imagine a scenario in graphing where division by zero should ever be fudged.
In amimation; yes; sometimes you take shortcuts to correctness for performance, but honestly handling division by zero properly tends to be an optimization. You are usually dealing with a vertical line (infinite slope), or a the intersection of a line parallel to a plane or something; and you can and should handle it correctly easily.

Can you cite an example where not handling a case where division by zero is going to occur properly makes sense to you? Rather than just hand waving that examples exist, can you acutally provide one? Because I can't think of one.

Anyone who thinks Snowden is the first and only person who had the access, ability, and inclination to take the data he took is as high as a fucking kite.

Snowden is just the only one who went public.

Snowden didn't have special access or magical powers. Thousands of other people work there just like he did; do you really think its inconceivable none have them have sold out? or have been compromised and are under foreign leverage? or outright work for a foreign government?

Think about that for half a second. Your just substituting 1/0 for some "E" symbol. How is the plot going to plot "E" on an x,y plane?

It'll still need special handling, and special data types. (you can't sqrt(-1) on a double; you have to use special complex classes everywhere for that to be allowed;

I don't generally use complex numbers when writing algorithms that plot circles or parabolas, because I'm not usually interested in complex results. If I'm plotting a circle on an x,y screen; using y = +/- sqrt(r^2 - x^2)+h; I'm not interested in complex results; and I'm using doubles in the algorithm. So sqrt(-1) is still an error to workaround, not i. (And if I was using complex numbers, then I'd still have to test for i and discard it.

So what have we gained with 'zeplex' numbers? Unless it's actually useful for some other aspect of mathematics.

No. You aren't.

http://www.wolframalpha.com/in...

What do you think it should do at 0?

Certainly not here. And in general no. Dividing by "nearly zero" is generally just as bad as dividing by zero. If X is approaching zero division by x is approaching an infinity.

No matter what data type you are using you are going to blow past its MIN/MAX; you are propagating error into the significant digits, etc. 10^80

Unless you are literally trying to plot a hyperbola or something, odds are if you are dividing something large by 0.0000000001 you are probably doing something wrong somewhere.

And there are countless other examples where its just nonsensical. Lets say you are doing a test and have a running tally of pass vs fail... and you haven't made any fails yet... so 5 pass, 0 fail... 5/0 is an error; so we'll just display 5/0.00000001 because that's a 'reasonable facsimile' right? :)

You could define anything you like.

1) Are you sure that wouldn't simply make the whole system inconsistent? ie Can we use 1/0 = E to prove 0 = 1 ?

2) What would that actually give us that would be useful?

Complex numbers are useful.