And that's important. Specs are a basis for _talking_about_ things. But they are _not_ a basis for implementing software.
I think this quote gives a better insight into what he thinks than your quote. Also you need to understand the context in which he considers the spec "useless".
The summary is twisting the discussion into something very flamebait worthy, and it gives a false impression of what Linus really said.
From my impression of the discussion it isn't specs he is against, but rather following those specs without being flexible enough to take reality into account. Would you really want to fly the plane built to spec, if some of those specs turned out to not accurately reflect reality? In that case you'd change the specs, but that's not always possible.
In this specific case, the problem was someone using the abstraction layering described in the spec, while the kernel would according to the others be better off using another design. If the software behaves equivalently, it should not matter how it is designed internally.
Hmm... Why is it more daring to choose the bad solution?
The amount of effort put into a bad solution could have been spendt on something else, like an even more optimal refinement of the better design. Or they might have spendt it on other projects, projects that would give better returns than the lessons learned from the shuttle. Certain resources are in short supply, especially talent.
I belive you misunderstood the real problem with the shuttle. It isn't the technical challenge, but rather the whole design concept being just plain wrong. They went for something overly complex and inefficient solution, instead of using a simple and safe one.
Consider a gun; do you fire small airplane shaped objects, or something that looks more like a dildo?
There's an interesting fantasy book by Gene Wolfe where an empire has declared that one may only speak using sentences from the holy books. This was supposed to prevent anyone from communicating anything improper.
So a captive from this empire would tell a story by selecting passages from this holy book that matched what he wanted to say, it was not as efficient and required the listener to interpret more. But still it did not prevent him from telling a story unrelated to the holy books.
Banning words will only make the communication channel less efficient, and somewhat more ambigious, but people will still be able to say what they want.
Actually, or models do *not* currently suggest they form by their own gravity alone. Or rather, that's no how it normally happens. Surrounding stars warm up the molecular clouds, and only those below 10K (iirc) in temperature will be able to form stars. The gas "pressure" usually counteracts the gravitational pull.
So in addition to this, you often need some outside force to compress the molecular cloud, like supernovas. These give birth to a large number of stars and since the stars that end up as supernovas are short lived, it usually causes a cycle of star formations.
Your right in that the model you presented is wrong, but that's cause it's a strawman rather than the real one.
If you have broken tests you either fix or deprecate them. What you absolutely do *not* do is break the code being tested, just so it can pass the test.
Well, you can... But then they are a hinderance, not a benefit.
A philosopher would have spendt all his time trying to find a definition of "Base" in the LSB. I belive what you meant was that he should put forward some hypothesis and bolster those.
And why shouldn't he be allowed to whine if, as he claims, the bugs are in the *tests*? If he's the maintainer of glibc, I'd assume he knows more about this domain than the average hacker.
Well, we can always find the "tipping point" by pushing until it does the tipping. Although this might be fun with cows, I'd rather not we did this to the climate.
Let's say we detonate all the worlds nuclear weapons i a grid pattern over the amazon. If we use one per 350 km we should be able to cover the whole of the amazon rainforest. What isn't incinerated by the initial blast would propably burn down later.
Tell me, would it have an effect on the global climate and if so are we still just flattering ourselves? If you don't agree with the former then there's really no point discussing this with you. It might be a slighly extreme example, but the point is that it isn't just hubris to say we might be causing the global warming.
But we don't really need big and flashy weapons to do this, all you need is some gasoline and a match. Using these simple tools people might start burning down alot of vegetation, imagine that. Ahh, but then you'd argue that vegetation doesn't affect the global climate?
64bit _code_ is usually 15% larger than 32bit, and I'd expect the larger address pointers to require comparable increase in the amount of memory for data structures.
Now you got several different reports, you can now calculate the median. Thus we know know that the TOC of Linux equals that of MS Windows, with Solaris being some 50% more expensive.
Slashdot Mirror
That piece of evidence should be dismissed as it is too embarrassing to the defendant.
I much preferred Yuki Seto's Accelerando and it's treatment of the 'rapture of the nerds'.
And that's important. Specs are a basis for _talking_about_ things. But they are _not_ a basis for implementing software.
I think this quote gives a better insight into what he thinks than your quote. Also you need to understand the context in which he considers the spec "useless".
The summary is twisting the discussion into something very flamebait worthy, and it gives a false impression of what Linus really said.
From my impression of the discussion it isn't specs he is against, but rather following those specs without being flexible enough to take reality into account. Would you really want to fly the plane built to spec, if some of those specs turned out to not accurately reflect reality? In that case you'd change the specs, but that's not always possible.
In this specific case, the problem was someone using the abstraction layering described in the spec, while the kernel would according to the others be better off using another design. If the software behaves equivalently, it should not matter how it is designed internally.
Any sufficiently sarcastic post is indistinguissable from flamebait.
Hmm... Why is it more daring to choose the bad solution?
The amount of effort put into a bad solution could have been spendt on something else, like an even more optimal refinement of the better design. Or they might have spendt it on other projects, projects that would give better returns than the lessons learned from the shuttle. Certain resources are in short supply, especially talent.
I belive you misunderstood the real problem with the shuttle. It isn't the technical challenge, but rather the whole design concept being just plain wrong. They went for something overly complex and inefficient solution, instead of using a simple and safe one.
Consider a gun; do you fire small airplane shaped objects, or something that looks more like a dildo?
There's an interesting fantasy book by Gene Wolfe where an empire has declared that one may only speak using sentences from the holy books. This was supposed to prevent anyone from communicating anything improper.
So a captive from this empire would tell a story by selecting passages from this holy book that matched what he wanted to say, it was not as efficient and required the listener to interpret more. But still it did not prevent him from telling a story unrelated to the holy books.
Banning words will only make the communication channel less efficient, and somewhat more ambigious, but people will still be able to say what they want.
Please define what life is, I've always pondered that one.
It's 10 Kelvin jobs, the guys wearing jeans.
Actually, or models do *not* currently suggest they form by their own gravity alone. Or rather, that's no how it normally happens. Surrounding stars warm up the molecular clouds, and only those below 10K (iirc) in temperature will be able to form stars. The gas "pressure" usually counteracts the gravitational pull.
So in addition to this, you often need some outside force to compress the molecular cloud, like supernovas. These give birth to a large number of stars and since the stars that end up as supernovas are short lived, it usually causes a cycle of star formations.
Your right in that the model you presented is wrong, but that's cause it's a strawman rather than the real one.
If you have broken tests you either fix or deprecate them. What you absolutely do *not* do is break the code being tested, just so it can pass the test.
Well, you can... But then they are a hinderance, not a benefit.
A philosopher would have spendt all his time trying to find a definition of "Base" in the LSB. I belive what you meant was that he should put forward some hypothesis and bolster those.
And why shouldn't he be allowed to whine if, as he claims, the bugs are in the *tests*? If he's the maintainer of glibc, I'd assume he knows more about this domain than the average hacker.
Clearly, there's no way any well thought-out reasoning can possibly be interesting.
It's a meritocracy, your vote doesn't count.
Well, we can always find the "tipping point" by pushing until it does the tipping. Although this might be fun with cows, I'd rather not we did this to the climate.
Ahh, i had completely forgotten about my player having an FM tuner...
Uhm... Does this count as an convincing argument?
I was under the impression that this urge to rip apart items inquisitively was a sign of intelligence, rather than the opposit as you seem to suggest.
Let's say we detonate all the worlds nuclear weapons i a grid pattern over the amazon. If we use one per 350 km we should be able to cover the whole of the amazon rainforest. What isn't incinerated by the initial blast would propably burn down later.
Tell me, would it have an effect on the global climate and if so are we still just flattering ourselves? If you don't agree with the former then there's really no point discussing this with you. It might be a slighly extreme example, but the point is that it isn't just hubris to say we might be causing the global warming.
But we don't really need big and flashy weapons to do this, all you need is some gasoline and a match. Using these simple tools people might start burning down alot of vegetation, imagine that. Ahh, but then you'd argue that vegetation doesn't affect the global climate?
1) Plant a tree. ...
2)
3) Profit.
This ofcourse assumes you don't burn them later on.
It does when Microsoft implements the 64bit OS.
64bit _code_ is usually 15% larger than 32bit, and I'd expect the larger address pointers to require comparable increase in the amount of memory for data structures.
Cthulhu is stiring, but i fear he'll take a disliking to fast-food.
A bit of ambiguity here, which half of the audience gasped?
The ecological activists on the mention of lithium fission, or the physicists on the mention of "breeding"?
You don't see the stars as the propbe has punched through the shell on which they are painted.
Now you got several different reports, you can now calculate the median. Thus we know know that the TOC of Linux equals that of MS Windows, with Solaris being some 50% more expensive.