Actually, for repetetive server tasks, dynamic meta level translation can be just the right thing.
I suspect that an implementation is at least several years off, but it should theoretically be possible to generate specialised execution paths with system calls turned into method calls (because the dynmaic translation software could proove no information would escape the specified code path). Inlining the system method calls, you could then potentially get rid of the copying between user and kernel space... potentially, your application could inline the parts of the kernel, all the way down through the networking stack.
As to whether this is likely to happen... who knows. IIRC cdrom.com maxes out an OC-3 on a dual pentium pro, so perhaps it would be overkill. But theoretically possible nonetheless
There already is an automated system. The suck/rule-someter using google is one such instance. Actually, the hot-oral-sex-ometer might be a good way to screen against porn. Figure most porn site will link to other porn sites, so the links ought to be fairly accurate.
Reminds me of that classic scene in Back To School where Rodney Dangerfield hires Kurt Vonnegut to write a bookreport on Slaughterhouse 5, and he gets an F...
A similar story by Asimov has a scientist bring Shakespear back from the past, but he fails a class on himself and commits suicide in a fit of depression. (IIRC: was more than a few years ago)
The problem is that the people who are willing to take this unsigning bonus are the ones who know they'll have an easy time getting work somewhere else -- exactly those people you WANT to hire. Those who'll refuse the money are those who lucked into getting a job and know it... the ones you'd rather avoid.
IBM found this out during the last recession, when all their best minds accepted the voluntary quit bonus and got a new job the next day.
A friend of mine once wondered why it seemed that functional programmers were more productive than others. He came to the conclusion that the cart was before the horse. Functional programming is unfortunately only popular in the hallowed halls of academia, and (yes generalising here) thus the people who were functional programmers thus tended to be more intelligent than your average code monkey.
Case in point is that he wrote much of his garbage collector (by necessity) in C, and managed to produce some clear and consise code in a language that seems designed to thwart that aim.
However, your post is correct in that it requires discipline to pull this off. The aforementioned code monkey has enough of that to fit in a box of matches without taking the matches out first. High level languages are no panacea: I've seen scheme code that is scary, and have, in flash of insanity, written Haskell that more resembled fortran than a proper language. However, they do make writing horrible code a bit more awkward while writing good code a bit less so.
great. So now you have diluted radioactive elements. Which would be great appart from the fact that in order to be viable as a storage option, you'd have to dump alot of stuff into the volcano. When it erupts, the resulting radioactive magma (was that a joke I didn't get?) lands ontop of everything.
great! so now the whole region is covered in radioactive rocks. This has helped the situation how?
Of course, GOOD advertising should also avoid the "familiarity" including an association with feelings of annoyance. ("Brand X? Hey, that's those annoying jerks with those irritating ads! I think I'll try brand Y instead..."). For THAT reason, I would expect "pop-up" and other "interruption based" ads would be even more of a failure (especially full-page ones) on the whole, no matter whether the "click-through" rate goes up or not.
We are seeing more and more of this. Bad ads are somewhat perversely driving people to buy TiVo like ad screeners. Which has the reinforcing effect of making ads even more annoying, because now they need ot eke out more views from the remaining people who don't have screeners.
Personally, I'm not at all annoyed by good ads, but will go out of my way to avoid bad ones. I BUY magazines just for the ads. This is merely another extension of demographics. Johan hates the hit-the-monkey ads. Don't show him those. Hit me twice with cool design shit. That's a GREAT demographic. One I would volunteer information to build.
Now, interruption based advertising is the pop-up of linear entertainment. As such, computer savvy people will apply technological solutions to avoid them. However, it is not clear to what extend a flash animation can be circumvented. Maybe hacked flash plugins that skip all the animation and just skip to the end?
Anyways, the possibilty is there. What the advertising industry haven't yet realised, it seems, is that the key to their success isn't getting as much attention as possible to an ad, but rather to not annoy me into blocking all ads, and then only as a secondary aim to attract my attention.
So a suggestion: Perhaps what is needed is a browser config that allows you to set your ad preferences. All you have to do is make it all sum up to 100 attention points. You get more points for accepting interruption based ads, or for giving better demographics...
The cooling device and the fuel preheater are the same? SWEET!
Just out of curiosity: if a manned plane were to lose power while hypersonic, would it be able to passively slow down to subsonic without melting, or 1) is this effectively impossible due to scramjet design or 2) a powered slowdown is absoultely required, so don't fail.
Let the binary be N bytes long. So we have N starting offsets, and each offset X has (N-X) possible lengths, for a grand total of N*(N+1)/2 possible entries in the challenge/response table.
Note that each dll or binary get its own table, so that you don't add up all the lengths to make a combined table (sum of squares, not square of sums!).
For a binary that is 500K = 5E5, the table needs to hold 25E10/2 = 125 gig. Clearly feasable. In comparison, a 56 bit key lookup table has 2^56 which is ALOT more (conversion to base 10 left as an excersise).
Furthermore, it is unlikely that AOL generates the challenges in a completely random manner. They likely have a cache on their end as well, to speed things up. Thus, the table will be [very] sparse, and thus will not need the full 125 gigs.
That's the often predicted trend; powerful realtime 3d graphics processing will offer generic product placement. So when archie bunker drinks a "beer" each market will be able to replace that in real time with whoever is sponsoring the local rebroadcast.
I guess the product placement people could then sign exclusivity agreements for even more money so that their "placed" products were never overlaid.
Sports have been doing this for a while, using blue-screening technology to replace billboards on televised soccer games.
back pointers? I suppose you mean a collection of weak references.
which languages implement this, outside of the DB world (where you could argue it is explict as well, being stored in a table) or UML (which so far has resisted all attempts at being executable)?
1) the oopsla 99 BOF on Squeak had a fun presentation on the new Jitter for squeak -- written in a weird self interpreting style, IIRC. It seemed pretty cool. (also one of the best stories on what to use C++ for. "C++ is a great language for implementing langauges to solve your problem". Interestingly, this is the same story many of the functional programming people sing, only they use bigger words like monads and combinators) How is this jitter coming along?
2) I also seem to remember a product that compiled smalltalk to JVM bytecode. They were very cagey about how they would get around the dispatch issues without resorting to reflection. Any word on this?
Ever use wily? (I should say acme, I guess). It uses extensive chording of mouse buttons to do away with pretty much all menus and control commands.
It's been a while since I used it (and then only briefly, so I'll make it up), but typically, you select a bit of text with B1, and then B2B1 chord on a word to execute that word with the selection as stdin. B1B2 chord selects then cuts, B1B3 pastes. jsut B3 opens a file.
I was unable to live without font-lock and programming modes from emacs so I quickly reverted, but it was suprising how quickly the idioms became subconcious (jsut like ctrl-a and ctrl-e are impossible for me to unlearn when using windows, to my enless select-alling and centering annoyance).
What efficency is that? Some back of the napkin calculations: earth orbit is r = 149,597,890,000 ~ 1.49 E11 meters from the sun (according to www.space.com). So the total surface area of that dyson-sphere is 4*pi*r*r ~ 2.23795 E22 m2.
A search for "power output of the sun" gives it as ~ 4 E 26 Watts, so we get an energy flux of ~ 1.7 E4 W/m2 = 17 KW/m2.
So I would be VERY suprised if they manage to get 100KW / m2.
So the claims of the space shuttle reaching Mach25 aren't that impressive? Does that mean that while in orbit it is effectively orbiting at Mach infinity (as sound has zero propagation speed in space...)?
For some reason I thought Mach was defined as the speed of sound at such a presure and humidity, analogously to c, which is constant, vs the Speed Of Light, which varies.
they did make some [at the time] incomprehensible changes to the S-Boxes that DES uses for its non-linear component. However, it was later discovered that these changes were 100 beneficial, in that they were specifically designed to protect against differential cryptanalysis (IIRC: if the non-linear transform isn't uniform, the bias can be statistically sampled and used to cut down the search space), that the original design would have been vunerable to.
So, they may move in strange ways, but that is not always against everyone else.
(mind you, at the time, they were possibly the only people who could have afforded to build a brute-forcer, so you could argue that this change WAS to their benefit in the long run)
Funny thing is that it is fixed now, so it appears that sig changes take place retroactively. For some reason I figured they'd be embedded into the posting.
Why does KDE care whether QT is GPLed or not? GPLled software can't be USED by closed software, but it fully able to USE closed software (presuming that QTs licence allows this).
can you perhaps shed some light on this: is there a major philosofical difference between KDE and Gnome that transcends implementational details (like vi's command mode and edit mode duality vs emacs's lisp programmability) or are there mainly surface differences (like pico's keybindings vs jove's)?
Slashdot Mirror
Actually, for repetetive server tasks, dynamic meta level translation can be just the right thing.
I suspect that an implementation is at least several years off, but it should theoretically be possible to generate specialised execution paths with system calls turned into method calls (because the dynmaic translation software could proove no information would escape the specified code path). Inlining the system method calls, you could then potentially get rid of the copying between user and kernel space... potentially, your application could inline the parts of the kernel, all the way down through the networking stack.
As to whether this is likely to happen... who knows. IIRC cdrom.com maxes out an OC-3 on a dual pentium pro, so perhaps it would be overkill. But theoretically possible nonetheless
There already is an automated system. The suck/rule-someter using google is one such instance. Actually, the hot-oral-sex-ometer might be a good way to screen against porn. Figure most porn site will link to other porn sites, so the links ought to be fairly accurate.
Reminds me of that classic scene in Back To School where Rodney Dangerfield hires Kurt Vonnegut to write a bookreport on Slaughterhouse 5, and he gets an F...
A similar story by Asimov has a scientist bring Shakespear back from the past, but he fails a class on himself and commits suicide in a fit of depression. (IIRC: was more than a few years ago)
You mean Schneier, surely.
IBM found this out during the last recession, when all their best minds accepted the voluntary quit bonus and got a new job the next day.
A friend of mine once wondered why it seemed that functional programmers were more productive than others. He came to the conclusion that the cart was before the horse. Functional programming is unfortunately only popular in the hallowed halls of academia, and (yes generalising here) thus the people who were functional programmers thus tended to be more intelligent than your average code monkey.
Case in point is that he wrote much of his garbage collector (by necessity) in C, and managed to produce some clear and consise code in a language that seems designed to thwart that aim.
However, your post is correct in that it requires discipline to pull this off. The aforementioned code monkey has enough of that to fit in a box of matches without taking the matches out first. High level languages are no panacea: I've seen scheme code that is scary, and have, in flash of insanity, written Haskell that more resembled fortran than a proper language. However, they do make writing horrible code a bit more awkward while writing good code a bit less so.
great. So now you have diluted radioactive elements. Which would be great appart from the fact that in order to be viable as a storage option, you'd have to dump alot of stuff into the volcano. When it erupts, the resulting radioactive magma (was that a joke I didn't get?) lands ontop of everything.
great! so now the whole region is covered in radioactive rocks. This has helped the situation how?
providers are in business. Nobody in a business wants to be redundant. Hence, the internet model is incompatible with business interests.
yes, but they were HUGE. You need to move large volumes of air to get bass, and if you don't have throw, you have to substitute surface area.
may electrostats come with subwoofers for this reason.
We are seeing more and more of this. Bad ads are somewhat perversely driving people to buy TiVo like ad screeners. Which has the reinforcing effect of making ads even more annoying, because now they need ot eke out more views from the remaining people who don't have screeners.
Personally, I'm not at all annoyed by good ads, but will go out of my way to avoid bad ones. I BUY magazines just for the ads. This is merely another extension of demographics. Johan hates the hit-the-monkey ads. Don't show him those. Hit me twice with cool design shit. That's a GREAT demographic. One I would volunteer information to build.
Now, interruption based advertising is the pop-up of linear entertainment. As such, computer savvy people will apply technological solutions to avoid them. However, it is not clear to what extend a flash animation can be circumvented. Maybe hacked flash plugins that skip all the animation and just skip to the end?
Anyways, the possibilty is there. What the advertising industry haven't yet realised, it seems, is that the key to their success isn't getting as much attention as possible to an ad, but rather to not annoy me into blocking all ads, and then only as a secondary aim to attract my attention.
So a suggestion: Perhaps what is needed is a browser config that allows you to set your ad preferences. All you have to do is make it all sum up to 100 attention points. You get more points for accepting interruption based ads, or for giving better demographics...
feedback, anyone?
see if I got this right:
The cooling device and the fuel preheater are the same? SWEET!
Just out of curiosity: if a manned plane were to lose power while hypersonic, would it be able to passively slow down to subsonic without melting, or 1) is this effectively impossible due to scramjet design or 2) a powered slowdown is absoultely required, so don't fail.
incorrect, or misleading at least.
Let the binary be N bytes long. So we have N starting offsets, and each offset X has (N-X) possible lengths, for a grand total of N*(N+1)/2 possible entries in the challenge/response table.
Note that each dll or binary get its own table, so that you don't add up all the lengths to make a combined table (sum of squares, not square of sums!).
For a binary that is 500K = 5E5, the table needs to hold 25E10/2 = 125 gig. Clearly feasable. In comparison, a 56 bit key lookup table has 2^56 which is ALOT more (conversion to base 10 left as an excersise).
Furthermore, it is unlikely that AOL generates the challenges in a completely random manner. They likely have a cache on their end as well, to speed things up. Thus, the table will be [very] sparse, and thus will not need the full 125 gigs.
That's the often predicted trend; powerful realtime 3d graphics processing will offer generic product placement. So when archie bunker drinks a "beer" each market will be able to replace that in real time with whoever is sponsoring the local rebroadcast.
I guess the product placement people could then sign exclusivity agreements for even more money so that their "placed" products were never overlaid.
Sports have been doing this for a while, using blue-screening technology to replace billboards on televised soccer games.
back pointers? I suppose you mean a collection of weak references.
which languages implement this, outside of the DB world (where you could argue it is explict as well, being stored in a table) or UML (which so far has resisted all attempts at being executable)?
A couple of Qs:
1) the oopsla 99 BOF on Squeak had a fun presentation on the new Jitter for squeak -- written in a weird self interpreting style, IIRC. It seemed pretty cool. (also one of the best stories on what to use C++ for. "C++ is a great language for implementing langauges to solve your problem". Interestingly, this is the same story many of the functional programming people sing, only they use bigger words like monads and combinators) How is this jitter coming along?
2) I also seem to remember a product that compiled smalltalk to JVM bytecode. They were very cagey about how they would get around the dispatch issues without resorting to reflection. Any word on this?
Ever use wily? (I should say acme, I guess). It uses extensive chording of mouse buttons to do away with pretty much all menus and control commands.
It's been a while since I used it (and then only briefly, so I'll make it up), but typically, you select a bit of text with B1, and then B2B1 chord on a word to execute that word with the selection as stdin. B1B2 chord selects then cuts, B1B3 pastes. jsut B3 opens a file.
I was unable to live without font-lock and programming modes from emacs so I quickly reverted, but it was suprising how quickly the idioms became subconcious (jsut like ctrl-a and ctrl-e are impossible for me to unlearn when using windows, to my enless select-alling and centering annoyance).
Is the loopback device fixed in the RH 2.4.2? I recall it being pretty much kaput in the vanilla 2.4.2.
No idea if you are a troll or not, but the compiler you ship with 7.0 fails to compile many sources. It is pretty broken.
Regression is a good thing when you've gone too far.
hrm.
What efficency is that? Some back of the napkin calculations: earth orbit is r = 149,597,890,000 ~ 1.49 E11 meters from the sun (according to www.space.com). So the total surface area of that dyson-sphere is 4*pi*r*r ~ 2.23795 E22 m2.
A search for "power output of the sun" gives it as ~ 4 E 26 Watts, so we get an energy flux of ~ 1.7 E4 W/m2 = 17 KW/m2.
So I would be VERY suprised if they manage to get 100KW / m2.
I'm thinking high altitude balloons dropping rains of ballbearings. Any jet flying into that should tear itself to pieces just using its OWN velocity.
Now that is aikido air defence.
So the claims of the space shuttle reaching Mach25 aren't that impressive? Does that mean that while in orbit it is effectively orbiting at Mach infinity (as sound has zero propagation speed in space...)?
For some reason I thought Mach was defined as the speed of sound at such a presure and humidity, analogously to c, which is constant, vs the Speed Of Light, which varies.
OT sig ref: I hate, you milk man dan!
yes,
they did make some [at the time] incomprehensible changes to the S-Boxes that DES uses for its non-linear component. However, it was later discovered that these changes were 100 beneficial, in that they were specifically designed to protect against differential cryptanalysis (IIRC: if the non-linear transform isn't uniform, the bias can be statistically sampled and used to cut down the search space), that the original design would have been vunerable to.
So, they may move in strange ways, but that is not always against everyone else.
(mind you, at the time, they were possibly the only people who could have afforded to build a brute-forcer, so you could argue that this change WAS to their benefit in the long run)
Funny thing is that it is fixed now, so it appears that sig changes take place retroactively. For some reason I figured they'd be embedded into the posting.
eh.
Why does KDE care whether QT is GPLed or not? GPLled software can't be USED by closed software, but it fully able to USE closed software (presuming that QTs licence allows this).
or am I confused again?
well,
can you perhaps shed some light on this: is there a major philosofical difference between KDE and Gnome that transcends implementational details (like vi's command mode and edit mode duality vs emacs's lisp programmability) or are there mainly surface differences (like pico's keybindings vs jove's)?