The part about people going to Hell for asking question sounds a lot like a snide remark.
Oops. I just saw the part elsethread where someone reveals that
this is a reference to something St. Augustine wrote. So, I
withdraw my assertion that this is a snide remark. It's clearly
just a literary reference that I didn't catch, although if you
don't know that it's a reference, it does sound like a
snide remark.
In my defense I do have City of God on my bookshelf and
have at least been meaning to read it.
As someone with a fairly good training in physics, I read this statement to be a commentary on Hawking's annoyance with the question of what came before "time" began. Many religious people have attempted to reconcile the Big Bang with Judeo-Christian beliefs by having God be responsible for the Big Bang. I think that such an allusion should not be taken as necessarily antagonistic.
OK, that explains the first half of Hawking's comment. Now, what is
the explanation for the "was He preparing hell for people who asked such questions" part of the comment? I would think that if the
only goal were to make the point about "beginning of time" really
meaning beginning, that simply asking, "What was God
doing before He made the world?" would be sufficient. The
part about people going to Hell for asking question sounds
a lot like a snide remark.
How about the Audi A3 2.0T? Direct injection from Audi's R8 racing technology and turbocharged too.
Congrats to MIT for discovering already in use technology!
As the (proud) owner of an Audi A3 2.0T, I can confidently say
that the technology described in the article is different.
The A3 2.0T is one of the first production cars to combine
turbocharging and direct injection, but it does not do the
ethanol thing described in the article.
The point of the
article seems to be that, while direct injection has
already been done, and while mixing in alcohol (or water)
to combat
knocking has already been done, direct injection of alcohol
to combat knocking had not already been done, and
it apparently has dramatic effects that either of the two
alone does not have. Assuming this is the case, that
would then allow you to
increase compression with impunity.
Getting back to the A3, it's clear it doesn't achieve
its power by increasing compression with impunity. If it
did, it would not require the high-octane fuel that it does.
This is not to say that exercise isn't useful. BMR is determined by lean body mass, which is determined by your muscle mass, which is determined by genetics and exercise. Exercise does help you lose weight, but it takes a lot longer than diet. Exercise also has independent benefits on cardiovascular health and a host of other health measures.
I have to grudgingly admit that I agree, based on my own
personal experiences. For almost 2 years, I have been
regularly exercising, and for most of that time I have
stayed at the same weight. I kept hoping that the weight
loss would eventually kick in after I'd built some muscle
and raised my metabolic rate. I wanted to believe that
exercising regularly would buy me the freedom to mostly
eat what I want. But it didn't happen.
About a month ago, I gave in and decided to
adjust my diet as well. I made some reasonable changes
in my diet: reduced portions, slightly healthier foods
(only slightly), a complete ban on beer and on soft
drinks (except diet ones), and no snacks allowed.
Since then, the weight has pretty much effortlessly
dropped off.
Exercise seems to be playing a significant part, because
last time I had a major effort to lose weight, I did not
exercise, and it did not happen nearly this easily. But,
exercise alone didn't do it.
My conclusion is that if you want to lose weight, and if
you had to choose between only dieting or only exercising,
dieting would be the clear choice. It can work by itself.
But having tried all four possible combinations of diet
and/or exercise, I can say that doing both is
the best of all.
To summarize, based on personal experience, it goes like
this:
Right. Because after World War II, it became illegal to build them... the country went totally bonkers over a dream of car-based sprawl, and nearly everywhere zoning regulations were changed to demand low-density development. So there's a shortage of places like San Francisco, with an urban density that was grandfathered in, and because no-one can build more of them, the prices are getting bid up.
I'm not sure how you get that it's illegal to build
pedestrian-friendly areas. I think it's just an economic
thing: most people (i.e. potential customers) expect
things to be convenient for driving, so developers build
things that way. It's not impossible to make things both
pedestrian-friendly and car-friendly, but it's not
easy, so if they have to choose, developers will choose
car-friendly.
For what it's worth, the mayor of the City of Austin (Texas)
has
declared that it his goal to redevelop downtown to radically
increase the residential usage. Presently there are about
5000 people living in downtown Austin and
his goal is
25,000 residents over the next decade or so. Apparently as a result of this
(maybe through changes to zoning laws, or through some kind
of incentives), there are condos going in everywhere.
There are so many projects going on at once you can't keep
track of them. High-rise, low-rise, whatever -- you name it,
and it's going in.
And, there has been a general trend towards mixed-use
development in the last several years. It doesn't automatically
make things pedestrian-friendly, but it helps. At least you
have some chance of having corner shops you can easily walk to.
There are people (I knw this is unbelievable to the average american) that can't drive a car, for example I can't. For that reason I avoided seeking a job in the USA.
There are some cities and parts of certain cities in the
US where you can get around with a car pretty well. Just because
there are many places where you can't doesn't mean it's impossible
to find a place where you can. The thing about the US is that
the pedestrian-friendly areas tend to be more expensive to live
in.
Am I the only person who doesn't want perks? I want three things from work: the ability to do my job, more pay, and less time there. If an employer wants to show their appreciation, they can increase my pay, let me work fewer hours, or both.
I'm mostly the same way. I don't mind the perks, but push it
even slightly too far and it becomes more of a negative than
a positive. Maybe I'm just too cynical, but often I regard
perks as an attempt to buy my loyalty with trinkets.
At best, perks are usually something I could buy for myself
at a similar cost to what the company pays. At worst, the
company is spending money on something I don't want and won't
use and I'd rather have the money given to me as a slight
increase in salary or used to improve benefits (like lowering
the deductible on the health insurance) or even invested in the
long-term future of the company (like increasing the R&D
budget).
On the other hand, there are some perks that can be a
win-win situation. As another person mentioned here,
shuttles are a win-win because in the SF Bay area, they
get to use carpool lanes and thus save the employees
time. That saved time can be split between home (more
personal time) and work (more productivity), so it makes
sense. But now we're talking about something which is
almost not a perk anymore; instead, it's a way to make
more efficient use of time.
Americans want Socialism without giving up their illusion of the "self made man". They want medicaire, medicaid, social security, guaranteed jobs, government loads for businesses, lots of cops, etc. But DON'T tell them it is Socialism.
I agree, but I think it's part of a larger problem. Americans
want solutions to all problems that are as easy and instantaneous
as solving the problem of hunger by ordering a Big Mac. No wait,
not ordering a Big Mac, ordering a Combo #1, because ordering a
Big Mac, a side of fries, and a drink is too complicated and
takes too much time and effort.
That is why we have Ritalin for misbehaving children, soaring
popularity of obesity surgery, mountains of credit card debt
and other consumer debt, pathetic personal savings rates, and
piss-poor electronic voting machines. And it's also why we
want things like guaranteed jobs.
We are Veruca Salt, we want it NOW, and furthermore, we
certainly don't want to have to work to get it.
Why is it Americans always brings up this "Socialism" rubbish? Why do you still live in the 50s? Socialism in Europe has long since disappeared.
Well, you still have stunningly high taxes, social programs
everywhere you look, and a general attitude of "we should put
the government in charge of this" when there's a problem.
You have a tax on televisions, for crying out loud (or at
least the UK does). To American eyes, that looks like
socialism, because one of the key distinctive things about
the American identity is that, historically, most of us have
believed that, by their nature, governments mostly screw up
whatever they're involved in, and therefore they should be
involved in as few things as practical. (Many Americans
would say "as few things as possible".)
The European ideal seems a little different. In college, I
had a government class where the professor said the difference
between the US and Europe is that, in Europe, being a
government bureaucrat is a respectable or even prestigious
job, and people would aspire to have a position in the
government, whereas in the United States, being a government
worker will only earn you scorn because working for the
government is just about the least respectable thing you
could do.
So, I'm not saying Europe is necessarily socialist (although
maybe it is, just a lot less socialist than previously), but
keep in mind the perspective that Americans are viewing
Europe from. Americans are idealists, and the ideal is that
government should stay out of most matters. Europe doesn't
share the same ideal, so it looks different to Americans.
"Socialism" might not be the best, most precise term to use
to describe the difference, but that's the reason the term
gets used.
How much effect did one million Brits protesting in the streets have on Blair's policy towards Iraq?
We didn't elect these people: the Conservative party polled more votes overall than Labour, but Labour won a substantial majority of seats due both to skewed boundaries and the skewed first-past-the-post system.
If we went into the streets to protest against every hair-brained authoritarian scheme they enact, let alone propose, we wouldn't have time to earn a living wage.
Hmm, apparently hanging on in quiet desperation really is
the English way.
I'm intentionally being a little harsh here, but if you Brits
want anything done about this, you are going have to be
the ones who do it. The government won't spontaneously
reform itself, and other countries' citizens have problems
of their own. And making up a list of reasons why it won't
work, when you don't really know that it can't, is not going
to help anything.
I have my 'Mips Pro Auto Parallellizing Option 7.2.1' cd sitting right next to my Irix 6.5 machine... and I know it's YEARS old
Oh, are we having a contest for who can name the earliest
auto-parallelizing C compiler? If so, I nominate the
vc compiler on the Convex
computers. The Convex C-1
was released in 1985 and I believe had a vectorizing
compiler from the start, which would make sense since
it had a single, big-ass vector processor (one instruction,
crap loads of operands -- can't remember how many, but
it was something like 64 separate values being added to
another 64 separate values in one single instruction).
I personally remember watching somebody compile something
with it. It was really neat to watch -- required no
special pragmas or anything, just plain old regular C
code, and it would produce an
annotated copy of your file telling you which lines
were fully vectorized, partly vectorized, etc. You
could, of course, tweak the code to make it easier
for the compiler to vectorize it, but even when you
did, it was still plain old C code.
Alternately you can point out that she's about to go to her 40th high school reunion and should be retiring anyway.
Boy, I do not follow that reason at all. Most people graduate
from high school at age 18 or so, so a 40th reunion would make
someone 58 years old. I see no reason at all why someone who
is only 58 necessarily should be retiring. It's a
perfectly reasonable age to retire if you've already saved
up enough money not to need to work, but then so is age 35,
but there is no reason someone should be required to retire
just because they have reached the ripe "old" age of 58.
I think your point might've been that, at 58, one is probably
past their prime, but that's far from being a fair assumption
as well. I had an excellent calculus teacher in college
who must have been in his 70's. His mind was certainly sharp
enough to teach calculus at a college level; in fact, he was
sharper than most of the other college professors I've ever had.
And he certainly had his teaching style perfected by then.
If you had the proper background and simply came to class
and paid attention, it was almost impossible not to learn
the material. As a matter of fact, I myself never did any
homework (he assigned it but did not require you to turn it
in), but his lectures were so clear that I managed to get
near perfect scores on all the tests simply by sitting there
in class and listening closely to what he said, and I had
failed the same calculus course prior to taking it from him.
I don't understand how this has a -1 score. Just look for "The IT Crowd" on youtube to see how funny of a show it is. It is sad that Americans cannot see this legally except with a pricey BBC cable package.
I have a pricey BBC cable package, and I've never seen it on there.
Do I just not know where to look?
I've only ever seen the first two episodes, the ones which the BBC
directly made available for download themselves.
The person who brute force discovers and uses someone else's code is not the one causing their Copy of Windows to be invalidated. Microsoft is doing that.
It's a fairly weak argument that the person who discovers the key
is not causing the other person's copy of Windows to be invalidated.
Yes, it would not happen if Microsoft hadn't programmed its
activation servers to work that way, but (a) Microsoft did, and
(b) this is common knowledge, and (c) if the person didn't discover
and use the key, nothing would have happened, and (d) if the person
does discover and use the key, something will happen.
Yes, you can certainly argue that nothing would happen if Microsoft
hadn't come up with this licensing scheme, and that's true. So,
Microsoft is part of the cause of someone's copy of Windows being
deactivated. But in just the same way, the deactivation would
not have happened if the person hadn't discovered and used the
key. So, that person is part of the cause too. Both
parties (Microsoft, and the person who discovered someone
else's key) are causes of the event.
So perhaps the fair thing to do would be to spread the blame
equally. However, there is an argument that Microsoft should
share less of the blame than the person who discovered a key:
one way of assigning blame is to look at the person who had
the last reasonable opportunity to avoid causing the problem,
and in this case, clearly that is the person running the
keygen and discovering someone else's key, because they did
it after the activation policy was known, and they did it
intentionally (which means they did not try to avoid causing
the problem).
I'm all for reasonable rules about copyright, and I hate Microsoft
as much as the next guy (probably more), but it just isn't
reasonable to pretend someone
can take what they know is quite possibly already in use by
someone else and not be even part of the cause of problems
that person experiences. At best, it's overly simplistic.
At worst, it's disingenuous and stupid.
One cute feature of radix sort for very large data sets is that you only need linear access to the data, so it can be done with non-random-access external storage. The one practical implementation I've ever heard of used four 9-track tape drives.
You can do that with merge sort as well, right? First you
write n/2 unsorted records to tape A and the other n/2
unsorted records to tape B. Now you say each of those
tapes has n/2 sorted lists of length 1. Then, you read
both A and B at once, and you merge the first two sorted
lists on A and B and write them to tape C. You take the
second two sorted lists on A and B, merge them, and write
them to tape D. Continue in this manner, alternating your
writes between C and D, until you run out of data on tapes
A and B.
Now, tapes C and D each contain a sequence of sorted lists,
but each one is twice as long as the lists on tapes A and B.
So you repeat the step and read from C and D and write to
A and B. Every time you do this, the sorted lists get twice
as long and there are half as many of them, until eventually
you have one list with all the sorted elements.
For extra added efficiency, take the original input and sort
it using whatever sort you choose (including mergesort) in
RAM before you do the initial write to tapes A and B. This
reduces the number of passes to log2(data size / ram size)
instead of log2(data size).
He has a TRULY AWFUL implementation of merge sort. For example, every time he goes to split a sub-list in half, he does a linear search from the current node to the end the sub-list. Having determined this value he then does ANOTHER LINEAR SEARCH to find the half-way point.
So he has basically made an O(N^2) time complexity process just to divide the list for the merge sort. This is inside his n*log(n) merge sort. So he has an O(n^3) time complexity mergesort and trumps up how fast his modified radix sort is. Come on! Bubble sort would have beat his mergesort.
That is maybe a bad implementation, but it actually does not make it O(n^3). In fact, it doesn't make
it as bad as even O(n^2). Here's why:
The first thing I want to establish is that linearly iterating
halfway through the sublist a second time (to find the middle
of it) has no effect on the asymptotic analysis. It just means
going through the sublist 3/2 times, which is still a constant.
The second thing is this:
as you go deeper and deeper into the recursion, the size of
the sublist is repeatedly cut in half. Thus, at every level
as you go deeper into the recursion you are doing twice as many
loops as the previous level (one loop for the first level,
when the sublist equals the original list; two loops for the
second, when you are breaking halves into quarters; four loops
in the third level of recursion; and so on). BUT, although you
are doing more loops, the size of the sublist is always half
what it was at the previous level of recursion. So have you
twice as many loops but half as many loop iterations.
That means at every level of recursion (think of this size of
the call stack if that helps in visualizing it), you are looping
over the entire list once (well actually 3/2 times). And there
are O(log n) levels of recursion, because every time you
recurse, you are working on a list half as big as before.
The result is that the total extra added time of this unnecessary
linear traversal is O(n * log n) for the entire algorithm.
Since mergesort is already O(n * log n), the extra looping
to find the middle and end of the list has no effect on the
asymptotic behavior of the algorithm.
These people are way behind the curve. The Mac community
did this
years ago, running OS X 10.3 on an old 25 MHz Mac.
Because of the software emulation required to run the PowerPC
code on a 68k machine, the person who did the experiment estimated
that booting up should take about 7 days.:-)
I don't like this because it forces young girls to get vaccinated against a disease that they can prevent by simply not having sex. It's not like measles, which can be transmitted innocently and anonymously. You have to actually have sex to get the virus.
You must mean something different than what you are actually saying,
because what your statement implies is that you expect females to
never have sex. Not just young females, but all
females.
After all, if you get vaccinated at age 14, won't you still be immune
at 30 or 40? How many women do you expect to reach 40 without having
sex? This vaccine is not just meant to protect a girl when she's
a teenager. It's meant to "last throughout a woman's
reproductive years". The point of doing it early is that if
you're going to do it at some point, you might as well do it
earlier because statistically you get more benefit out of doing
it earlier.
Now, you may say that if people are in monogamous relationships
(i.e. marriage), they don't stand much chance of getting a
sexually-transmitted disease. But even if people believe they
should still be virgins when they get married and should never
divorce and they do their best to stick to that, they still
have some chance of getting a sexually-transmitted
disease. A woman could be a virgin when she gets married
and have a husband that cheats on her
without her knowledge. Or, a virgin woman could marry a man
who had one previous sexual partner, who happened to have
the disease. You don't have to be woman of highly
questionable moral character to get the disease.
Now, here's the next question: you may believe that having
sex as a teenager (or having premarital sex, or whatever)
is wrong. Let's say we all agree about that. But, what
should the penalty for giving in to temptation be? If we
don't give the vaccine to kids solely
because we think they shouldn't be
having sex, aren't we saying that we think the penalty for
making a poor decision about sex should (potentially) be
cancer? I thought the whole point of encouraging people
to be wise about who they have sex with and when they start
doing it and whether they take it seriously was to help them
have a better life because you care about them and love them.
But withholding a vaccine seems to amount to saying, "Haha,
you had sex and you got cancer! I guess that wouldn't have
happened to you if you were a good person!" That
doesn't seem like love. That seems like a hateful, damaging
form of self-proclaimed moral superiority.
Our women control us - I can't have a car with a stick shift because then my wife can't drive it:(
That is the #1 reason to buy a stick shift, at least when
it comes to friends wanting to borrow your car. It gives you the
perfect way to say "No, you moron, you'll only wreck it or
get me a parking ticket" without actually saying it
and being impolite.
Instead, you just say, "Wow, I didn't know you could drive stick,"
and then they say, "Oh damn, I forgot about that.
Hmm, I wonder if Bob would let me borrow his car...".
(Note that you only use the fake-surprised "Wow, you can drive
stick?" comment if you already happen to know that they can't.)
That said, this point from the article is interesting, making me believe researchers should (?) have incentives to disclose security bugs to Mozilla first and to the public only when the fix is distributed
There already is an incentive in place: not having people think
you are an arrogant dick.
I'm still running 1.5.0.9 and it works a treat. Am I missing something besides, apparently, h4x?
Yes: when the app crashes for whatever reason, Firefox 2.x
automatically offers you the opportunity to reload the
pages (and tabs) that you had open before the crash. I can't
think of any other compelling features of Firefox 2.x, but
to me, this alone is worth it. It's very handy, also, when
the browser hasn't completely crashed but is just mildly
wedged.
I believe you may be able to get basically the same feature
for Firefox 1.5.x with a plugin, but it's nice having it
built in.
Patent #5845280, "Method and apparatus for transmitting a file in a network using a single transmit request from a user-mode process to a kernel-mode process". Compare this with the Linux (and BSD) SendFile() API.
I'm not sure I buy that sendfile() (not SendFile(), by
the way) is really a violation
of that patent. In particular, the patent abstract says this:
the requested file is retrieved from the secondary data storage device and placed in kernel-mode accessed memory (e.g., cache memory). After the requested file information is stored in the kernel-mode accessed memory, a kernel-mode data transmission procedure transmits the requested file information directly
The way I read that, this patented mechanism always loads the entire
file into RAM before starting the transmission. Does sendfile()
really operate in this manner? The way I understand it, in effect
it just pushes the read()/write() loop into the
kernel. But the virtual memory subsystem will typically fault in
data from the filesystem (secondary storage) in a lazy manner.
There may be some read-ahead caused by the disk driver or the
filesystem, but if you open() a 500 MB file and then call
sendfile() on it, the kernel is not going to read the entire
500 MB of data into RAM first.
I realize that's picky, but I'm assuming patent interpretation
is inherently picky. Also, another technical point: sendfile()
doesn't necessarily read from disk and write to the network.
It just takes two file descriptors, so I don't see any reason
why it couldn't be used to copy from one network connection to
another (e.g. for a proxy server), or from disk to disk, or from/proc to/dev/tty, for all I care. Therefore,
it seems to me that sendfile() by itself cannot be an
infringement. You would have to write code that opens a disk
file and a network connection and then calls sendfile()
on the two in order to have infringement. (Not that there isn't
code out there that does this; I think that's why sendfile()
exists in the first place, after all...)
"Rewriting the code" is nowhere near so easy as the site makes it sound. Software patents are often granted for particular concepts - not just ways of doing them. What if some core kernel routine were found to be infringing? That can't just be ripped out and replaced, many years of development and testing have gone into it!
Why not? For a time, big parts of the virtual memory subsystem seemed to
get replaced every 6 months. And then static/dev got replaced by devfs
which then got replaced by udev before devfs was even widely adopted.
Also, many or most of the major functions of Linux were invented
and done on computers before Microsoft released Windows 1.0.
Virtual memory, privilege systems, filesystems, and all that
are basically not very new technology, most of them having
originated in the 1960's and 1970's. If Linux had to go back
and only use operating systems ideas that were invented in the
1970's or earlier, it could still be a pretty darned good
operating system.
Slashdot Mirror
[Quoting myself...]
Oops. I just saw the part elsethread where someone reveals that this is a reference to something St. Augustine wrote. So, I withdraw my assertion that this is a snide remark. It's clearly just a literary reference that I didn't catch, although if you don't know that it's a reference, it does sound like a snide remark.
In my defense I do have City of God on my bookshelf and have at least been meaning to read it.
OK, that explains the first half of Hawking's comment. Now, what is the explanation for the "was He preparing hell for people who asked such questions" part of the comment? I would think that if the only goal were to make the point about "beginning of time" really meaning beginning, that simply asking, "What was God doing before He made the world?" would be sufficient. The part about people going to Hell for asking question sounds a lot like a snide remark.
As the (proud) owner of an Audi A3 2.0T, I can confidently say that the technology described in the article is different. The A3 2.0T is one of the first production cars to combine turbocharging and direct injection, but it does not do the ethanol thing described in the article.
The point of the article seems to be that, while direct injection has already been done, and while mixing in alcohol (or water) to combat knocking has already been done, direct injection of alcohol to combat knocking had not already been done, and it apparently has dramatic effects that either of the two alone does not have. Assuming this is the case, that would then allow you to increase compression with impunity.
Getting back to the A3, it's clear it doesn't achieve its power by increasing compression with impunity. If it did, it would not require the high-octane fuel that it does.
I have to grudgingly admit that I agree, based on my own personal experiences. For almost 2 years, I have been regularly exercising, and for most of that time I have stayed at the same weight. I kept hoping that the weight loss would eventually kick in after I'd built some muscle and raised my metabolic rate. I wanted to believe that exercising regularly would buy me the freedom to mostly eat what I want. But it didn't happen.
About a month ago, I gave in and decided to adjust my diet as well. I made some reasonable changes in my diet: reduced portions, slightly healthier foods (only slightly), a complete ban on beer and on soft drinks (except diet ones), and no snacks allowed. Since then, the weight has pretty much effortlessly dropped off.
Exercise seems to be playing a significant part, because last time I had a major effort to lose weight, I did not exercise, and it did not happen nearly this easily. But, exercise alone didn't do it.
My conclusion is that if you want to lose weight, and if you had to choose between only dieting or only exercising, dieting would be the clear choice. It can work by itself. But having tried all four possible combinations of diet and/or exercise, I can say that doing both is the best of all.
To summarize, based on personal experience, it goes like this:
I'm not sure how you get that it's illegal to build pedestrian-friendly areas. I think it's just an economic thing: most people (i.e. potential customers) expect things to be convenient for driving, so developers build things that way. It's not impossible to make things both pedestrian-friendly and car-friendly, but it's not easy, so if they have to choose, developers will choose car-friendly.
For what it's worth, the mayor of the City of Austin (Texas) has declared that it his goal to redevelop downtown to radically increase the residential usage. Presently there are about 5000 people living in downtown Austin and his goal is 25,000 residents over the next decade or so. Apparently as a result of this (maybe through changes to zoning laws, or through some kind of incentives), there are condos going in everywhere. There are so many projects going on at once you can't keep track of them. High-rise, low-rise, whatever -- you name it, and it's going in.
And, there has been a general trend towards mixed-use development in the last several years. It doesn't automatically make things pedestrian-friendly, but it helps. At least you have some chance of having corner shops you can easily walk to.
There are some cities and parts of certain cities in the US where you can get around with a car pretty well. Just because there are many places where you can't doesn't mean it's impossible to find a place where you can. The thing about the US is that the pedestrian-friendly areas tend to be more expensive to live in.
I'm mostly the same way. I don't mind the perks, but push it even slightly too far and it becomes more of a negative than a positive. Maybe I'm just too cynical, but often I regard perks as an attempt to buy my loyalty with trinkets. At best, perks are usually something I could buy for myself at a similar cost to what the company pays. At worst, the company is spending money on something I don't want and won't use and I'd rather have the money given to me as a slight increase in salary or used to improve benefits (like lowering the deductible on the health insurance) or even invested in the long-term future of the company (like increasing the R&D budget).
On the other hand, there are some perks that can be a win-win situation. As another person mentioned here, shuttles are a win-win because in the SF Bay area, they get to use carpool lanes and thus save the employees time. That saved time can be split between home (more personal time) and work (more productivity), so it makes sense. But now we're talking about something which is almost not a perk anymore; instead, it's a way to make more efficient use of time.
I agree, but I think it's part of a larger problem. Americans want solutions to all problems that are as easy and instantaneous as solving the problem of hunger by ordering a Big Mac. No wait, not ordering a Big Mac, ordering a Combo #1, because ordering a Big Mac, a side of fries, and a drink is too complicated and takes too much time and effort.
That is why we have Ritalin for misbehaving children, soaring popularity of obesity surgery, mountains of credit card debt and other consumer debt, pathetic personal savings rates, and piss-poor electronic voting machines. And it's also why we want things like guaranteed jobs.
We are Veruca Salt, we want it NOW, and furthermore, we certainly don't want to have to work to get it.
Well, you still have stunningly high taxes, social programs everywhere you look, and a general attitude of "we should put the government in charge of this" when there's a problem. You have a tax on televisions, for crying out loud (or at least the UK does). To American eyes, that looks like socialism, because one of the key distinctive things about the American identity is that, historically, most of us have believed that, by their nature, governments mostly screw up whatever they're involved in, and therefore they should be involved in as few things as practical. (Many Americans would say "as few things as possible".)
The European ideal seems a little different. In college, I had a government class where the professor said the difference between the US and Europe is that, in Europe, being a government bureaucrat is a respectable or even prestigious job, and people would aspire to have a position in the government, whereas in the United States, being a government worker will only earn you scorn because working for the government is just about the least respectable thing you could do.
So, I'm not saying Europe is necessarily socialist (although maybe it is, just a lot less socialist than previously), but keep in mind the perspective that Americans are viewing Europe from. Americans are idealists, and the ideal is that government should stay out of most matters. Europe doesn't share the same ideal, so it looks different to Americans. "Socialism" might not be the best, most precise term to use to describe the difference, but that's the reason the term gets used.
Hmm, apparently hanging on in quiet desperation really is the English way.
I'm intentionally being a little harsh here, but if you Brits want anything done about this, you are going have to be the ones who do it. The government won't spontaneously reform itself, and other countries' citizens have problems of their own. And making up a list of reasons why it won't work, when you don't really know that it can't, is not going to help anything.
Oh, are we having a contest for who can name the earliest auto-parallelizing C compiler? If so, I nominate the vc compiler on the Convex computers. The Convex C-1 was released in 1985 and I believe had a vectorizing compiler from the start, which would make sense since it had a single, big-ass vector processor (one instruction, crap loads of operands -- can't remember how many, but it was something like 64 separate values being added to another 64 separate values in one single instruction).
I personally remember watching somebody compile something with it. It was really neat to watch -- required no special pragmas or anything, just plain old regular C code, and it would produce an annotated copy of your file telling you which lines were fully vectorized, partly vectorized, etc. You could, of course, tweak the code to make it easier for the compiler to vectorize it, but even when you did, it was still plain old C code.
Boy, I do not follow that reason at all. Most people graduate from high school at age 18 or so, so a 40th reunion would make someone 58 years old. I see no reason at all why someone who is only 58 necessarily should be retiring. It's a perfectly reasonable age to retire if you've already saved up enough money not to need to work, but then so is age 35, but there is no reason someone should be required to retire just because they have reached the ripe "old" age of 58.
I think your point might've been that, at 58, one is probably past their prime, but that's far from being a fair assumption as well. I had an excellent calculus teacher in college who must have been in his 70's. His mind was certainly sharp enough to teach calculus at a college level; in fact, he was sharper than most of the other college professors I've ever had. And he certainly had his teaching style perfected by then. If you had the proper background and simply came to class and paid attention, it was almost impossible not to learn the material. As a matter of fact, I myself never did any homework (he assigned it but did not require you to turn it in), but his lectures were so clear that I managed to get near perfect scores on all the tests simply by sitting there in class and listening closely to what he said, and I had failed the same calculus course prior to taking it from him.
I have a pricey BBC cable package, and I've never seen it on there. Do I just not know where to look? I've only ever seen the first two episodes, the ones which the BBC directly made available for download themselves.
It's a fairly weak argument that the person who discovers the key is not causing the other person's copy of Windows to be invalidated. Yes, it would not happen if Microsoft hadn't programmed its activation servers to work that way, but (a) Microsoft did, and (b) this is common knowledge, and (c) if the person didn't discover and use the key, nothing would have happened, and (d) if the person does discover and use the key, something will happen.
Yes, you can certainly argue that nothing would happen if Microsoft hadn't come up with this licensing scheme, and that's true. So, Microsoft is part of the cause of someone's copy of Windows being deactivated. But in just the same way, the deactivation would not have happened if the person hadn't discovered and used the key. So, that person is part of the cause too. Both parties (Microsoft, and the person who discovered someone else's key) are causes of the event.
So perhaps the fair thing to do would be to spread the blame equally. However, there is an argument that Microsoft should share less of the blame than the person who discovered a key: one way of assigning blame is to look at the person who had the last reasonable opportunity to avoid causing the problem, and in this case, clearly that is the person running the keygen and discovering someone else's key, because they did it after the activation policy was known, and they did it intentionally (which means they did not try to avoid causing the problem).
I'm all for reasonable rules about copyright, and I hate Microsoft as much as the next guy (probably more), but it just isn't reasonable to pretend someone can take what they know is quite possibly already in use by someone else and not be even part of the cause of problems that person experiences. At best, it's overly simplistic. At worst, it's disingenuous and stupid.
You can do that with merge sort as well, right? First you write n/2 unsorted records to tape A and the other n/2 unsorted records to tape B. Now you say each of those tapes has n/2 sorted lists of length 1. Then, you read both A and B at once, and you merge the first two sorted lists on A and B and write them to tape C. You take the second two sorted lists on A and B, merge them, and write them to tape D. Continue in this manner, alternating your writes between C and D, until you run out of data on tapes A and B.
Now, tapes C and D each contain a sequence of sorted lists, but each one is twice as long as the lists on tapes A and B. So you repeat the step and read from C and D and write to A and B. Every time you do this, the sorted lists get twice as long and there are half as many of them, until eventually you have one list with all the sorted elements.
For extra added efficiency, take the original input and sort it using whatever sort you choose (including mergesort) in RAM before you do the initial write to tapes A and B. This reduces the number of passes to log2(data size / ram size) instead of log2(data size).
That is maybe a bad implementation, but it actually does not make it O(n^3). In fact, it doesn't make it as bad as even O(n^2). Here's why:
The first thing I want to establish is that linearly iterating halfway through the sublist a second time (to find the middle of it) has no effect on the asymptotic analysis. It just means going through the sublist 3/2 times, which is still a constant.
The second thing is this: as you go deeper and deeper into the recursion, the size of the sublist is repeatedly cut in half. Thus, at every level as you go deeper into the recursion you are doing twice as many loops as the previous level (one loop for the first level, when the sublist equals the original list; two loops for the second, when you are breaking halves into quarters; four loops in the third level of recursion; and so on). BUT, although you are doing more loops, the size of the sublist is always half what it was at the previous level of recursion. So have you twice as many loops but half as many loop iterations.
That means at every level of recursion (think of this size of the call stack if that helps in visualizing it), you are looping over the entire list once (well actually 3/2 times). And there are O(log n) levels of recursion, because every time you recurse, you are working on a list half as big as before.
The result is that the total extra added time of this unnecessary linear traversal is O(n * log n) for the entire algorithm. Since mergesort is already O(n * log n), the extra looping to find the middle and end of the list has no effect on the asymptotic behavior of the algorithm.
These people are way behind the curve. The Mac community did this years ago, running OS X 10.3 on an old 25 MHz Mac.
Because of the software emulation required to run the PowerPC code on a 68k machine, the person who did the experiment estimated that booting up should take about 7 days. :-)
You must mean something different than what you are actually saying, because what your statement implies is that you expect females to never have sex. Not just young females, but all females. After all, if you get vaccinated at age 14, won't you still be immune at 30 or 40? How many women do you expect to reach 40 without having sex? This vaccine is not just meant to protect a girl when she's a teenager. It's meant to "last throughout a woman's reproductive years". The point of doing it early is that if you're going to do it at some point, you might as well do it earlier because statistically you get more benefit out of doing it earlier.
Now, you may say that if people are in monogamous relationships (i.e. marriage), they don't stand much chance of getting a sexually-transmitted disease. But even if people believe they should still be virgins when they get married and should never divorce and they do their best to stick to that, they still have some chance of getting a sexually-transmitted disease. A woman could be a virgin when she gets married and have a husband that cheats on her without her knowledge. Or, a virgin woman could marry a man who had one previous sexual partner, who happened to have the disease. You don't have to be woman of highly questionable moral character to get the disease.
Now, here's the next question: you may believe that having sex as a teenager (or having premarital sex, or whatever) is wrong. Let's say we all agree about that. But, what should the penalty for giving in to temptation be? If we don't give the vaccine to kids solely because we think they shouldn't be having sex, aren't we saying that we think the penalty for making a poor decision about sex should (potentially) be cancer? I thought the whole point of encouraging people to be wise about who they have sex with and when they start doing it and whether they take it seriously was to help them have a better life because you care about them and love them. But withholding a vaccine seems to amount to saying, "Haha, you had sex and you got cancer! I guess that wouldn't have happened to you if you were a good person!" That doesn't seem like love. That seems like a hateful, damaging form of self-proclaimed moral superiority.
Wow, I didn't know aerospace borrowed drivetrain technology from cars. I thought it was only the other way around.
That is the #1 reason to buy a stick shift, at least when it comes to friends wanting to borrow your car. It gives you the perfect way to say "No, you moron, you'll only wreck it or get me a parking ticket" without actually saying it and being impolite. Instead, you just say, "Wow, I didn't know you could drive stick," and then they say, "Oh damn, I forgot about that. Hmm, I wonder if Bob would let me borrow his car...". (Note that you only use the fake-surprised "Wow, you can drive stick?" comment if you already happen to know that they can't.)
There already is an incentive in place: not having people think you are an arrogant dick.
Yes: when the app crashes for whatever reason, Firefox 2.x automatically offers you the opportunity to reload the pages (and tabs) that you had open before the crash. I can't think of any other compelling features of Firefox 2.x, but to me, this alone is worth it. It's very handy, also, when the browser hasn't completely crashed but is just mildly wedged.
I believe you may be able to get basically the same feature for Firefox 1.5.x with a plugin, but it's nice having it built in.
Oh sure... completely at random... Come on, admit it: you're a horrible, evil serial killer, right?
I'm not sure I buy that sendfile() (not SendFile(), by the way) is really a violation of that patent. In particular, the patent abstract says this:
The way I read that, this patented mechanism always loads the entire file into RAM before starting the transmission. Does sendfile() really operate in this manner? The way I understand it, in effect it just pushes the read()/write() loop into the kernel. But the virtual memory subsystem will typically fault in data from the filesystem (secondary storage) in a lazy manner. There may be some read-ahead caused by the disk driver or the filesystem, but if you open() a 500 MB file and then call sendfile() on it, the kernel is not going to read the entire 500 MB of data into RAM first.
I realize that's picky, but I'm assuming patent interpretation is inherently picky. Also, another technical point: sendfile() doesn't necessarily read from disk and write to the network. It just takes two file descriptors, so I don't see any reason why it couldn't be used to copy from one network connection to another (e.g. for a proxy server), or from disk to disk, or from /proc to /dev/tty, for all I care. Therefore,
it seems to me that sendfile() by itself cannot be an
infringement. You would have to write code that opens a disk
file and a network connection and then calls sendfile()
on the two in order to have infringement. (Not that there isn't
code out there that does this; I think that's why sendfile()
exists in the first place, after all...)
Why not? For a time, big parts of the virtual memory subsystem seemed to get replaced every 6 months. And then static /dev got replaced by devfs
which then got replaced by udev before devfs was even widely adopted.
Also, many or most of the major functions of Linux were invented and done on computers before Microsoft released Windows 1.0. Virtual memory, privilege systems, filesystems, and all that are basically not very new technology, most of them having originated in the 1960's and 1970's. If Linux had to go back and only use operating systems ideas that were invented in the 1970's or earlier, it could still be a pretty darned good operating system.