2HP might be enough to run the headlights and A/C, but forget about it for moving 3000lbs of car + passengers.
No.
You would be very surprised to know how little power is required to maintain, say, 50 mph in a (small) car like the Prius. About 6 kW., which is about 8 hp. Of that, 1-2 kW is very significant, and could be used (for example) to improve highway mileage significantly, which is currently not a strength of hybrids.
This isn't bogus. Even in the very early 1980's, Chrysler made a big deal that their new "K-car" (Phymouth Reliant/Dodge Aries) could cruise at 55 mph with something like 12-15 hp. That was before gas prices fell, and people stopped caring. Today's small cars are a lot lighter and more aerodynamic than the Reliant/Aries was. So the Prius estimate at 6-9 kW (at 50-60 mph) seems very reasonable.
Admittedly, the full 1-2kW of solar power would only be available in rather idealized circumstances. And accelerating an automobile to 55 mph in a timescale acceptable to most (American) drivers requires a LOT more power for a few seconds. Nevertheless, the common notion that you need ~100 hp to keep a car moving is simply flat-out untrue.
In fact Dillion help fixed the vm bug in Linux 2.4.
Huh? FreeBSD's current VM (which Dillon played a very significant role in developing) was indeed the inspiration for the (Rik van Riel) VM in 2.4.0 through 2.4.9. But this VM was largely replaced in the 2.4 mainline kernels starting with 2.4.10. The replacement was principally a balance between Rik's multiqueue approach and Andrea Arcangeli's classzone-based patches. AFAIK, this "fix" had nothing to do with Dillon, besides it was a departure from the existing FreeBSD approach.
Rik's VM has continued to develop -- it was maintained in Alan Cox's -ac tree for a long time and today it is very successful in the form of Rik's reverse mapping patches, a feature that (in a somewhat different form) is also in FreeBSD.
The minimal reverse mapping feature itself has been integrated into kernel 2.5/2.6-test, though the page replacement algorithms in 2.5/2.6 are much closer to the original Arcangeli classzone approach than Rik's. We'll see if new page replacement schemes, and object-based rmap (akin to FreeBSD) show up in kernel 2.7.
Point being, I know Dillon has admitted interest in Linux VM development, but I'm unaware of any direct involvement (i.e. patches).
It stated in the article that chips like this have been in use in astronomy for a long time now for these exact reasons. (CMOS chips rather than CCD chips).
Uhhhh no, it says the opposite I believe:
"Specialized long-exposure CCDs have long proved invaluable in astronomy..."
Cooled CCD's are far and away the preferred detectors for astronomical imaging and spectroscopy at optical (visible) wavelengths.
This is only true if the image resolution is actually limited by the resolution of the CCD. If the CCD is properly matched to the optics of the telescope, the CCD pixels will properly sample the natural resolution (imaging quality) of the telescope, which typically is more limited by the distortions induced by the atmosphere -- and guiding errors and vibrations caused by tracking the motion of the sky. It may not be a "resolution" win, especially if the X3 pixels themselves are physically bigger!
Another reason: astronomical CCD's don't do the filter "mosaicing" that digital cameras do anyway -- rather, a single color filter is placed over all pixels simultaneously (or no filter at all; sometimes you don't want the color information). So you take three exposures, getting high-resolution red, green, and blue images -- and combine them in software to make the images you see published in magazines and on the Web.
So potentially, these new imaging arrays could save time because you wouldn't have to take separate red, green, and blue images. But that will happen only if the overall quantum efficiency (what fraction of the incoming photons is detected) is comparable to existing CCD's (the best of which can detect up to around 90% of the incident light). If the Foveon/X3 arrays are much less sensitive (or have high dark current, or exhibit color sensing behavior that changes as a camera is cooled to cryogenic temperatures, etc.) it will be faster/easier to take 3 exposures with a CCD.
The technology sounds very interesting, but I expect (given the available info) that a few iterations might be necessary to make these new imaging arrays better than CCD's for professional and amateur astronomers for a while. I hope it works out though!
Dark matter has been invoked to explain puzzling observations, within a theoretical framework that has proven pretty trustworthy for everything else so far. Either dark matter is real, or the framework needs a facelift somewhere. Jury's still out on that one.
Let's highlight one example. Consider first the motion of the planets around the Sun, which itself comprises most of the mass of the Solar System. The innermost planets like Mercury and Venus are whizzing around at high speeds to maintain their close orbits. By the time you get out to Neptune or Pluto, they're just crawling along, since they are far away from the Sun. This is all nice and reasonable.
Now when you look at (spiral) galaxies, most of their luminosity (and presumably mass) comes from their inner regions, so you might expect the same kind of rotation pattern; the inner regions would have high rotational velocities, and the outer regions would rotate slowly. But this doesn't happen -- in fact the outer regions tend to rotate every bit as fast as the inner regions! But there's no luminous matter out there to support such rapid motion! There must be something out there that has huge gravitational impact, but without emitting any kind of light. Hence one invocation of dark matter. There are others, but the theme is largely the same.
Some folks understandably find the notion of dark matter distasteful, and are working on modified theories of gravity that have 'appropriate' characteristics on the scale of an entire galaxy. But many consider those "modified gravity" arguments to be unconvincing, ad hoc and distasteful in their own way -- so far.
What is important is that we are learning more and more about our place in the Universe, and uncovering new puzzles that we don't yet understand. And that's what makes it fun.:)
Dude!:) Ahhh, quake haikus. A fine, fine pasttime.
Waitaminnit. Cornell still? Aren't you done yet? [ducking];)
Now if you'll excuse me, I have to go stand in a corner and write on the chalkboard a thousand times:
I will preview my posts before submitting them.
I will preview my posts before submitting them.
I will preview my posts before submitting them.
I will preview my posts before submitting them. ...
Could you provide some sort of support for that claim?
Aside: Really, it's not about the egos of scientists, or the perfection of our telescopes and instruments. Goodness knows, if they were so perfect, we wouldn't be begging for money to build new and better ones!:)
The link that pyrrho mentioned describes the basic reasons why baryons can't be all of the hypothesized dark matter. And since 1996 (when the article was written), the evidence has become vastly more convincing. I'll attempt to summarize.
Sure, we could hypothesize that the Universe is filled with "dim, normal stuff" like brown dwarfs, white dwarfs, lost airline luggage, missing socks, dryer lint... but we're just not able to see them. Fair enough. But there is a limit to this argument for numerous reasons.
There are not enough baryons in the Universe. The Big Bang only made so many baryons, and this is something we can measure. The limits on the number of baryons in the Universe are quite tight -- only, say, 5% of the mass needed to give the Universe an uncurved geometry.
Okay, so maybe we just live in an empty, open Universe! But numerous measurements of the curvature of the Universe, in particular recent observations of the cosmic microwave background itself suggest that the curvature is not open but uncurved. So we live in a Universe with plenty of gravitational matter of some form or another. Aside: we are gathering a huge amount of information by looking at the angular sizes of the bumps and dips in the cosmic microwave background, which is fossil radiation from the Big Bang and a few percent of the static you see on your TV when tuned to a blank UHF channel. This page shows what the CMB power spectra (that is, how many inhomogeneities occur at a given angular size) look like, and how changing various cosmological parameters has an effect on the spectrum you'd expect to see. Try out changing the baryon density -- the effect is quite pronounced. It also says that the Universe has the number of baryons that Big Bang theory says it should have.
Even if we can't see brown dwarfs (or basketballs, for that matter) by their reflected light, we CAN infer their existence by their gravitational interactions with light, i.e. gravitational (micro)lensing. It's not that we "haven't looked hard enough" -- but rather that "if the Universe was full of brown dwarfs, there'd be tons of observable microlensing events". But microlensing events are exceedingly rare. In this case, the null result is interesting, because it highlights that baryonic matter is not as prolific as we want/need!
So this makes us all feel a bit uncomfortable, because either some of the fundamental tenets of cosmology are flawed (even though they explain nearly all of the observable Universe, right down to the abundances of the elements and the large scale structure of galaxies and the cosmic microwave background, the recession of galaxies etc.)... OR... the Universe is mostly filled with matter what is unlike anything we yet know how to explain.
No, it's how the error estimates are reported. The HST key project that estimates an age around 13 billion years also stipulates +/- 10%, corresponding at most to ONE standard deviation, i.e. the 68% confidence level. This study is reporting their error bars at the 95% confidence level, which corresponds to two standard deviations, so the errors appear twice as large. The "13-14 billion year" age you report would have uncertainties of almost 3 billion years in either direction at the 95% confidence level. We have to compare apples to apples here!
There is another very important point to recognize here. The HST Key project results (based upon Cepheid variable stars) is independent of the measurement/modeling of the ages of the oldest stars of Milky Way halo stars and clusters. Sure, both measurements each have significant systematic errors, but their uncertainties come from entirely different things! So the fact that they agree is quite reassuring. It also means that the measurements can be combined, at least to some degree.
With the newest generation of instruments and telescopes observing the Universe from radio waves to gamma rays, there will be new, independent methods of measuring the age and fate of the Universe. Already measurements from Type 1a supernovae are narrowing the uncertainties in some cosmological parameters. Other methods that currently yield very large error bars, but will be pivotal in the next few years are gravitational lensing (a detailed description here) and the Senyaev-Zeldovich effect (some details here).
When and if we get to the point where all methods yield the same result, we'll have our answer. In the meantime, if you just quote the formal results from just a single group, from a single type of argument/measurement, the systematic errors are going to be large, particularly when you're dealing with anything on cosmological scales!
Sure, if you're only comparing benchmark scores, the comparison is uneven, and the 1.8GHz T-series wins hands down. If you're buying a desktop replacement, that's relevant. But for the mobile market, it doesn't really matter. All of the CPUs concerned are sufficiently fast for 95 percent of the software people actually spend their time using.
My point is that 2.5 kilobucks buys a nice T-series, it also can get one a nice PowerBook, and the only significant difference spec-wise is CPU speed. And who cares?:) 5 hours of battery life is a lot more useful to me, and some of the connectivity options are not available for the Thinkpad at any price. And I own a Thinkpad, and I really like it. 'Course, it runs Linux.:)
"If you judge a laptop by its specs only you'll miss what's most important. You won't know that the screen is hard to read in a well lit room. That the fan roars. That it feels flimsy. That it's sucking air in from a vent in the bottom and will melt-down if left running on a bed. That the speakers are bad or unbalanced. That the adapter takes 24 hours to recharge the batteries while the unit is on. That the DVD playback is seriously flawed. That there's no hardware volume control to silence it in a awkward moment. That the BIOS is non-standard. That it gets burning hot sitting on your lap. That the case is so thin you can damage the screen just carrying it wrong. Those annoyances add up in every day use." (Greg Moulton)
I have no argument with your points, but am merely suggesting an alternative point of view.
GNUStep is not a window manager. It is an implementation of the OpenStep API as published by NeXT several years ago, now known in its current form as Cocoa, the development environment that Apple is pushing for OS X. In principle, GNUstep allows one to write applications that build and work under Mac OS X and Linux/BSD. One example of such an application is GNUmail, which is excellent. On a side note, I can't emphasize enough how many things this OO development framework gets right, in comparison to certain other Linux development environments, which are only now random-walking their way toward sanity.:) So helping out GNUstep is NOT a redundant waste of time by any means.
Apple products carry a price premium, but it's not nearly as awful as you describe. In fact, their notebook line is probably (IMHO) where they shine brightest. If you compare a T-series Thinkpad (a far better comparison), the margin narrows. 256MB/40GB/1.8GHzP4 for the Thinkpad, and $2500. For the same price, you get the PowerBook with 256MB/30GB/667MHzG4. The CPU speed on the Apple suffers, even when Altivec optimizations are included, but on a laptop this is a much less important concern. More critically, the Apple gets an easy 5 hours of battery life, and the Thinkpad only 3 hours. The PowerBook has additional features, such as twice the video memory, solid Firewire support, and full gigabit ethernet -- which are important to keep in mind.
When it all comes down to it, you get what you pay for -- and it doesn't matter much if you buy from a high-end PC manufacturer or from Apple. So pick what suits you best!
Well, here's one published example of one such naming scheme: the Super Huge Interferometric Telescope, courtesy of a few astronomy graduate students at the University of Arizona'sSteward Observatory.
:)
References for your amusement:
link to ADS abstract, Bulletin of the American Astronomical Society, Vol. 31, p.1504
"I pledge impertinence to the flag waving, of the unindicted co-conspiritors of America, and to the Republicans for which I can't stand, one abomination, underhanded fraud, indefensible, with Liberty and Justice... Forget it."
from "Life in Hell"
See, it could be worse.:)
Given that the original 1954 ruling adding "Under God" to the Pledge was made for mostly political reasons at a time when the US was concerned about being clearly distinguishable from "godless communists", it should not be seen as a horrible action if those two words were removed. Not as if it were an attack on the Pledge itself; merely a rollback to the way it was in 1953. As far as I know, this in no way threatens the state of the Union...
On the other hand, I'm not entirely convinced that the change is really necessary either. We surely should be working harder to preserve the freedoms themselves, than worry about a Pledge that people regurgitate from instinct, rather than understanding and belief.
What *is* alarming is the furor, akin to a lynch-mob, that has erupted over this issue. The irony of protecting a piece of cloth or a Pledge at all cost, while slowly selling out the freedoms that they represent, is frightening.
Think about that very carefully as you spend tonight or tomorrow morning stuck in rush hour traffic. How free are you, really? Do you feel that you are really in a representative democracy, and that your voice will matter? Or just a rat, indoctrinated by corporate marketing since age 5, running in someone else's arbitrary maze?
It shouldn't surprise anyone that 2.4.10 VM performs better than 2.4.9. Even in terms of the "traditional" 2.4 VM from Rik, the Linus and Alan trees deviated starting around kernel 2.4.7. There were numerous complaints about the Linus tree missing important patches, and having contradicting patches applied. It ended up quite a mess, and VM performance reflects this. Alan's tree was much more conservative in this regard.
If you compare 2.4.11 to anything, please compare it to the latest -ac kernels from Alan, where the traditional 2.4 VM is actually working very well. There's NO sense in comparing 2.4.11 to 2.4.9; the VM in 2.4.9 and its kin -- it was just plain broken.
Side note: In Rik's VM, please remember to not just look at swap used as a gauge of whether you're swapping or not. All anonymous pages are mapped to swap, so the space is simply allocated. You can create a huge image in GIMP and lots of swap will be allocated, but without a drop of disk I/O! Use vmstat and look at the 'si' and 'so' columns to see if you're actually writing pages to swap. Or look in/proc/meminfo and subtract "SwapCached" from the amount of swap you think you're using. That's the amount of *written* swap you're using (a better comparison to 2.4.10).
This needs to be made sensible in 2.5, if this VM is to be resurrected.
Andrea's work has cleaned up the handling of inactive pages (which could have been done under the old system), and the new "classzone" approach and VM balancing isn't documented anywhere outside the code itself. In addition, there are very normal loads where it performs badly compared to the -ac tree. Here is a test suite that tests different aspects of aging and swapping, and the results as provided to linux-kernel. 2.4.10 (patched with Andrea's VM tweaks) swapped more pages, took longer, and had to swap more pages back in when the tests completed (i.e. it could have chosen better pages to swap out). It also caused XMMS to skip mp3 playback throughout the tests, whereas -ac didn't.
Nothing's perfect of course; a process that randomly walks through pages performs better in 2.4.10 since it's more streamlined and not trying to be as "intelligent" about page handling. Rik's code could no doubt be improved here.
That's the great thing about open source: let the best idea win! No doubt in 2.5 we'll see these two VM schemes hash it out in much more complete form (i.e. lose the remaining kernel 2.2-isms, maybe add physical page mapping, almost certainly swapfs -- mostly for Rik's scheme; I'm not sure what the next steps for Andrea's VM should be).
Slashdot Mirror
No.
You would be very surprised to know how little power is required to maintain, say, 50 mph in a (small) car like the Prius. About 6 kW., which is about 8 hp. Of that, 1-2 kW is very significant, and could be used (for example) to improve highway mileage significantly, which is currently not a strength of hybrids.
This isn't bogus. Even in the very early 1980's, Chrysler made a big deal that their new "K-car" (Phymouth Reliant/Dodge Aries) could cruise at 55 mph with something like 12-15 hp. That was before gas prices fell, and people stopped caring. Today's small cars are a lot lighter and more aerodynamic than the Reliant/Aries was. So the Prius estimate at 6-9 kW (at 50-60 mph) seems very reasonable.
Admittedly, the full 1-2kW of solar power would only be available in rather idealized circumstances. And accelerating an automobile to 55 mph in a timescale acceptable to most (American) drivers requires a LOT more power for a few seconds. Nevertheless, the common notion that you need ~100 hp to keep a car moving is simply flat-out untrue.
Huh? FreeBSD's current VM (which Dillon played a very significant role in developing) was indeed the inspiration for the (Rik van Riel) VM in 2.4.0 through 2.4.9. But this VM was largely replaced in the 2.4 mainline kernels starting with 2.4.10. The replacement was principally a balance between Rik's multiqueue approach and Andrea Arcangeli's classzone-based patches. AFAIK, this "fix" had nothing to do with Dillon, besides it was a departure from the existing FreeBSD approach.
Rik's VM has continued to develop -- it was maintained in Alan Cox's -ac tree for a long time and today it is very successful in the form of Rik's reverse mapping patches, a feature that (in a somewhat different form) is also in FreeBSD.
The minimal reverse mapping feature itself has been integrated into kernel 2.5/2.6-test, though the page replacement algorithms in 2.5/2.6 are much closer to the original Arcangeli classzone approach than Rik's. We'll see if new page replacement schemes, and object-based rmap (akin to FreeBSD) show up in kernel 2.7.
Point being, I know Dillon has admitted interest in Linux VM development, but I'm unaware of any direct involvement (i.e. patches).
Uhhhh no, it says the opposite I believe:
"Specialized long-exposure CCDs have long proved invaluable in astronomy..."
Cooled CCD's are far and away the preferred detectors for astronomical imaging and spectroscopy at optical (visible) wavelengths.
This is only true if the image resolution is actually limited by the resolution of the CCD. If the CCD is properly matched to the optics of the telescope, the CCD pixels will properly sample the natural resolution (imaging quality) of the telescope, which typically is more limited by the distortions induced by the atmosphere -- and guiding errors and vibrations caused by tracking the motion of the sky. It may not be a "resolution" win, especially if the X3 pixels themselves are physically bigger!
Another reason: astronomical CCD's don't do the filter "mosaicing" that digital cameras do anyway -- rather, a single color filter is placed over all pixels simultaneously (or no filter at all; sometimes you don't want the color information). So you take three exposures, getting high-resolution red, green, and blue images -- and combine them in software to make the images you see published in magazines and on the Web.
So potentially, these new imaging arrays could save time because you wouldn't have to take separate red, green, and blue images. But that will happen only if the overall quantum efficiency (what fraction of the incoming photons is detected) is comparable to existing CCD's (the best of which can detect up to around 90% of the incident light). If the Foveon/X3 arrays are much less sensitive (or have high dark current, or exhibit color sensing behavior that changes as a camera is cooled to cryogenic temperatures, etc.) it will be faster/easier to take 3 exposures with a CCD.
The technology sounds very interesting, but I expect (given the available info) that a few iterations might be necessary to make these new imaging arrays better than CCD's for professional and amateur astronomers for a while. I hope it works out though!
That's essentially it, yes.
:)
Dark matter has been invoked to explain puzzling observations, within a theoretical framework that has proven pretty trustworthy for everything else so far. Either dark matter is real, or the framework needs a facelift somewhere. Jury's still out on that one.
Let's highlight one example. Consider first the motion of the planets around the Sun, which itself comprises most of the mass of the Solar System. The innermost planets like Mercury and Venus are whizzing around at high speeds to maintain their close orbits. By the time you get out to Neptune or Pluto, they're just crawling along, since they are far away from the Sun. This is all nice and reasonable.
Now when you look at (spiral) galaxies, most of their luminosity (and presumably mass) comes from their inner regions, so you might expect the same kind of rotation pattern; the inner regions would have high rotational velocities, and the outer regions would rotate slowly. But this doesn't happen -- in fact the outer regions tend to rotate every bit as fast as the inner regions! But there's no luminous matter out there to support such rapid motion! There must be something out there that has huge gravitational impact, but without emitting any kind of light. Hence one invocation of dark matter. There are others, but the theme is largely the same.
Some folks understandably find the notion of dark matter distasteful, and are working on modified theories of gravity that have 'appropriate' characteristics on the scale of an entire galaxy. But many consider those "modified gravity" arguments to be unconvincing, ad hoc and distasteful in their own way -- so far.
What is important is that we are learning more and more about our place in the Universe, and uncovering new puzzles that we don't yet understand. And that's what makes it fun.
Waitaminnit. Cornell still? Aren't you done yet? [ducking]
Now if you'll excuse me, I have to go stand in a corner and write on the chalkboard a thousand times: .
...
I will preview my posts before submitting them.
I will preview my posts before submitting them.
I will preview my posts before submitting them
I will preview my posts before submitting them.
Dangit.
Aside: Really, it's not about the egos of scientists, or the perfection of our telescopes and instruments. Goodness knows, if they were so perfect, we wouldn't be begging for money to build new and better ones! :)
The link that pyrrho mentioned describes the basic reasons why baryons can't be all of the hypothesized dark matter. And since 1996 (when the article was written), the evidence has become vastly more convincing. I'll attempt to summarize.
Sure, we could hypothesize that the Universe is filled with "dim, normal stuff" like brown dwarfs, white dwarfs, lost airline luggage, missing socks, dryer lint... but we're just not able to see them. Fair enough. But there is a limit to this argument for numerous reasons.
Okay, so maybe we just live in an empty, open Universe! But numerous measurements of the curvature of the Universe, in particular recent observations of the cosmic microwave background itself suggest that the curvature is not open but uncurved. So we live in a Universe with plenty of gravitational matter of some form or another. Aside: we are gathering a huge amount of information by looking at the angular sizes of the bumps and dips in the cosmic microwave background, which is fossil radiation from the Big Bang and a few percent of the static you see on your TV when tuned to a blank UHF channel. This page shows what the CMB power spectra (that is, how many inhomogeneities occur at a given angular size) look like, and how changing various cosmological parameters has an effect on the spectrum you'd expect to see. Try out changing the baryon density -- the effect is quite pronounced. It also says that the Universe has the number of baryons that Big Bang theory says it should have.
So this makes us all feel a bit uncomfortable, because either some of the fundamental tenets of cosmology are flawed (even though they explain nearly all of the observable Universe, right down to the abundances of the elements and the large scale structure of galaxies and the cosmic microwave background, the recession of galaxies etc.) ... OR ... the Universe is mostly filled with matter what is unlike anything we yet know how to explain.
It's going to be a fun ride! :)
There is another very important point to recognize here. The HST Key project results (based upon Cepheid variable stars) is independent of the measurement/modeling of the ages of the oldest stars of Milky Way halo stars and clusters. Sure, both measurements each have significant systematic errors, but their uncertainties come from entirely different things! So the fact that they agree is quite reassuring. It also means that the measurements can be combined, at least to some degree.
With the newest generation of instruments and telescopes observing the Universe from radio waves to gamma rays, there will be new, independent methods of measuring the age and fate of the Universe. Already measurements from Type 1a supernovae are narrowing the uncertainties in some cosmological parameters. Other methods that currently yield very large error bars, but will be pivotal in the next few years are gravitational lensing (a detailed description here) and the Senyaev-Zeldovich effect (some details here).
When and if we get to the point where all methods yield the same result, we'll have our answer. In the meantime, if you just quote the formal results from just a single group, from a single type of argument/measurement, the systematic errors are going to be large, particularly when you're dealing with anything on cosmological scales!
Sure, if you're only comparing benchmark scores, the comparison is uneven, and the 1.8GHz T-series wins hands down. If you're buying a desktop replacement, that's relevant. But for the mobile market, it doesn't really matter. All of the CPUs concerned are sufficiently fast for 95 percent of the software people actually spend their time using.
My point is that 2.5 kilobucks buys a nice T-series, it also can get one a nice PowerBook, and the only significant difference spec-wise is CPU speed. And who cares? :) 5 hours of battery life is a lot more useful to me, and some of the connectivity options are not available for the Thinkpad at any price. And I own a Thinkpad, and I really like it. 'Course, it runs Linux. :)
"If you judge a laptop by its specs only you'll miss what's most important. You won't know that the screen is hard to read in a well lit room. That the fan roars. That it feels flimsy. That it's sucking air in from a vent in the bottom and will melt-down if left running on a bed. That the speakers are bad or unbalanced. That the adapter takes 24 hours to recharge the batteries while the unit is on. That the DVD playback is seriously flawed. That there's no hardware volume control to silence it in a awkward moment. That the BIOS is non-standard. That it gets burning hot sitting on your lap. That the case is so thin you can damage the screen just carrying it wrong. Those annoyances add up in every day use." (Greg Moulton)
I have no argument with your points, but am merely suggesting an alternative point of view.
Two points.
When it all comes down to it, you get what you pay for -- and it doesn't matter much if you buy from a high-end PC manufacturer or from Apple. So pick what suits you best!
References for your amusement:
Haven't seen this posted yet. :)
:)
"I pledge impertinence
to the flag waving,
of the unindicted co-conspiritors of America,
and to the Republicans
for which I can't stand,
one abomination,
underhanded fraud,
indefensible,
with Liberty and Justice...
Forget it."
from "Life in Hell"
See, it could be worse.
Given that the original 1954 ruling adding "Under God" to the Pledge was made for mostly political reasons at a time when the US was concerned about being clearly distinguishable from "godless communists", it should not be seen as a horrible action if those two words were removed. Not as if it were an attack on the Pledge itself; merely a rollback to the way it was in 1953. As far as I know, this in no way threatens the state of the Union...
On the other hand, I'm not entirely convinced that the change is really necessary either. We surely should be working harder to preserve the freedoms themselves, than worry about a Pledge that people regurgitate from instinct, rather than understanding and belief.
What *is* alarming is the furor, akin to a lynch-mob, that has erupted over this issue. The irony of protecting a piece of cloth or a Pledge at all cost, while slowly selling out the freedoms that they represent, is frightening.
Think about that very carefully as you spend tonight or tomorrow morning stuck in rush hour traffic. How free are you, really? Do you feel that you are really in a representative democracy, and that your voice will matter? Or just a rat, indoctrinated by corporate marketing since age 5, running in someone else's arbitrary maze?
There has to be hope. How do we make this better?
If you compare 2.4.11 to anything, please compare it to the latest -ac kernels from Alan, where the traditional 2.4 VM is actually working very well. There's NO sense in comparing 2.4.11 to 2.4.9; the VM in 2.4.9 and its kin -- it was just plain broken.
Side note: In Rik's VM, please remember to not just look at swap used as a gauge of whether you're swapping or not. All anonymous pages are mapped to swap, so the space is simply allocated. You can create a huge image in GIMP and lots of swap will be allocated, but without a drop of disk I/O! Use vmstat and look at the 'si' and 'so' columns to see if you're actually writing pages to swap. Or look in /proc/meminfo and subtract "SwapCached" from the amount of swap you think you're using. That's the amount of *written* swap you're using (a better comparison to 2.4.10).
This needs to be made sensible in 2.5, if this VM is to be resurrected.
Andrea's work has cleaned up the handling of inactive pages (which could have been done under the old system), and the new "classzone" approach and VM balancing isn't documented anywhere outside the code itself. In addition, there are very normal loads where it performs badly compared to the -ac tree. Here is a test suite that tests different aspects of aging and swapping, and the results as provided to linux-kernel. 2.4.10 (patched with Andrea's VM tweaks) swapped more pages, took longer, and had to swap more pages back in when the tests completed (i.e. it could have chosen better pages to swap out). It also caused XMMS to skip mp3 playback throughout the tests, whereas -ac didn't.
Nothing's perfect of course; a process that randomly walks through pages performs better in 2.4.10 since it's more streamlined and not trying to be as "intelligent" about page handling. Rik's code could no doubt be improved here.
That's the great thing about open source: let the best idea win! No doubt in 2.5 we'll see these two VM schemes hash it out in much more complete form (i.e. lose the remaining kernel 2.2-isms, maybe add physical page mapping, almost certainly swapfs -- mostly for Rik's scheme; I'm not sure what the next steps for Andrea's VM should be).