Not settling for attaching a cubit of information to each electron, Nigerian scientists have found a way to attach an entire 255 character text string. The system uses a network of Mobile Electron Monitors, made out of pixie dust, who's existence was predicted by Jamie Zawinski's Theory of Open Source in 1999. Originally developed for conversationalists to keep in contact using their thumbs, the Simple Message Service allows electrons to share a significant amount of information all at once and the ability to communicate with the scientists using the familiar cell phone network. SMS is already in use by a large number of gregarious teens to bypass note passing laws in many western classrooms.
I love pointing out to people that javascript and activex are not web standards and any useful functionality of the two [eg, drop down menus] can be done using html and css. It is so much easier to make a site W3C compatible than to do everything via javaschit.
Though not exactly the topic at hand, the MS forums require JS, but using firebug and being quick to tell it to stop loading you can navigate it without, ie, like most sites, you don't need the javascript. I find this isn't to hard to do when I'm not welcome somehow.
I don't know if I really understand what you're saying. Making Rails and other interpreters take advantage of smp and the like is not trivial, but it's easier than having to convert all the ruby, python, perl, lisp etc apps, you've just got to convert the engine. If the low level functions, especially those with big loops like map and reduce, search, and regexp processing, can take advantage of more cores, you've got all the performance benefit you need. IMHO I'd like to see better vectorisation options to make this sort of low level parallelism easier, rather than all this threading stuff. I understand it isn't always the best way, but on your average dual or quad core machine it'd be worth it.
So I'm not sure if there's all that much work in it, I guess.
since we've already got to write somewhat parallel code [because cores are pipelined what, 20 deep now?], it'd be nice if that parallelism extended easily to multi cores. For example, when dividing up some function that reduces [large array -> scalar], we've already got to split execution 20 ways. It'd be nice if there was some way to split such things into the minimum arbitrary number of independent threads supported by the machine, at runtime.
How many software developers work on Windows and IE? Seven or eight years on [by OSX standards], and they still haven't caught up to where the competition was in terms of usability back then. They must have actually been trying to hold out on decent software, in order to keep some sort of market exploitable.
I do contract mechanical design when I need a little money, and I got an older lcd as a gift last October. How I did design before then, I'm not entirely sure. The main thing, I guess, is having the CLI on one screen, so you can read the output and type long commands, and your drawing area on another. it's great if you're working from.pdf documentation, bring up your reference and off you go. My office is practically dead-tree free now. Everything else I do only really needs one screen, but not having to flip between workspaces or programs, when you really do need to see both at the same time, is something every power user needs.
No, you've just got to rewrite the compilers. Considering the plethora of hardware Linux is available for, I'd say anyone interested in promoting a new hardware format can move up with a few weeks of rewriting a compiler. Once one has been done, it's usually pretty easy to make a new lexical analyzer and port all the other compilers too.
Here's my confusion with that RISC argument: faster clocks like that bring relativistic issues into moving data between the processor and the memory. Wherever processing is mostly local to a memory segment, high speed RISC machines are fantastic. But when you've got to access larger memory which may be further away in the machine, you've got to take that latency much more seriously. At 4GHz, Light travels around 75mm in a clock cycle, so a request for RAM access at that speed takes a couple of cycles [with today's FSB speeds, it's two cycles on that bus. Sure, that's four processor cycles, but the bus is pretty wide too]. If modern RISC processors were working at 16GHz and beyond, that latency would compound considerably. Getting the full power out of such a processor would be more work than it's worth.
I'd rather have a larger processor that's going to do more useful work in the same time, the SIMD instructions on the x86 are a pretty mediocre example but they are, I feel, a step in the right direction [vectorisation].
This is something I find most embarrassing concerning x86. Besides the clock speed increase, have we seen a 1000-fold performance increase with that increase in complexity? Sure, some of that complexity has bought about performance increases, we've gone from 8 to 64 bit, and we've got SIMD now. Still, there's a four-fold increase in complexity that isn't accounted for there. I'd rather see all those transistors do something useful, and with the advent of FOSS operating systems, there's some chance one of the big chip makers to turn that four-fold increase in transistors available into useful performance.
I concur- one more powerful computer you can use for your real tasks, and one for your background stuff. these guys have some interesting stuff. There was one company in a post on/. a while back that had a 1GHz ARM machine in a really, really tiny case. Can't seem to find the article.
There's a whole other set of atheistic theories that are usually pushed along with it, like the one you just mentioned. As a mathematician, I know better, but I still know the implications of saying "I believe in evolution". It's more than a theory, it's a mindset, and some have turned it into a strange cult, with its own dogma and clergy.
You sure can. Especially the simple buffer overflow stuff like this. Arguments about protected IE7 aren't relevant, because it's not IE7 that handles the animated cursor [it only loads it]. Only browsers that don't load animated cursors by default fix the problem. Further, the point that this is arbitrary code execution means that the code runs with full ring-4 privileges [unless this exploit runs on Alpha too, then I'm not sure what happens, although you were specifically referring to Vista which doesn't], which is the same level that system services run in. So really, you can't mitigate the effect much at all. I'm no expert on the ring-0 processes in Windows or the glue that binds them to the rest of the system, but I imagine there's some things an exploit of this kind can do [such as accessing your TCP/IP stack and file system] and things it can't [such as making Windows use a new TCP/IP stack of the cracker's creation- but if you've got the privileges above, why bother?].
So now that we've gotten rid of the "this is not really MS fault" suggestion and removed your false sense of security in "protected" mode, where does that leave us?
arguments concerning specific digits are irrelevant. Once you've decided what floating-point precision to use, the precision is fixed. weather 4 is equal to 4. or not depends on the rounding scheme. And honestly, I think Python gets it right. If the numbers are the same to the 53rd binary place, which they are certain to be in Python, they may as well be treated as the same number. If you've got to carry around accuracy values, they need to be in metadata anyway, because they [errors] combine differently under different operations.
Also, who was talking about comparing complex with float? Much as I hate saying this, TMTOWTDI. In Python and on paper. If you need to compare floats and complex numbers, use.abs(),.re() or.im().
Well, I find the filter technique they use very very interesting. But different materials are more or less opaque and reflective to different frequencies, and since radio waves permeate just about everything fairly well, but scatter better [are less directional] because of the longer wavelength. So it'd make sense that the physics you get from these frequencies are ok scattering but more materials appear opaque.
TFA was pretty good, I pretty much just responded to the slashdot article:P but yeah, physics of THz-frequency electromagnetic radiation doesn't feel that surprising somehow.
Slashdot Mirror
#include "bsod.h" main() { if(TRUE) { crash(horribly, messily); } return proprietary_blob(patented); }
Protected mode IE does nothing to even slow the attack, the code can be run not as part of IE but as part of the graphics handler.
Not settling for attaching a cubit of information to each electron, Nigerian scientists have found a way to attach an entire 255 character text string. The system uses a network of Mobile Electron Monitors, made out of pixie dust, who's existence was predicted by Jamie Zawinski's Theory of Open Source in 1999. Originally developed for conversationalists to keep in contact using their thumbs, the Simple Message Service allows electrons to share a significant amount of information all at once and the ability to communicate with the scientists using the familiar cell phone network. SMS is already in use by a large number of gregarious teens to bypass note passing laws in many western classrooms.
I love pointing out to people that javascript and activex are not web standards and any useful functionality of the two [eg, drop down menus] can be done using html and css. It is so much easier to make a site W3C compatible than to do everything via javaschit.
Though not exactly the topic at hand, the MS forums require JS, but using firebug and being quick to tell it to stop loading you can navigate it without, ie, like most sites, you don't need the javascript. I find this isn't to hard to do when I'm not welcome somehow.
I guess we will all need to re-tune our tin hats, too.
I don't know if I really understand what you're saying. Making Rails and other interpreters take advantage of smp and the like is not trivial, but it's easier than having to convert all the ruby, python, perl, lisp etc apps, you've just got to convert the engine. If the low level functions, especially those with big loops like map and reduce, search, and regexp processing, can take advantage of more cores, you've got all the performance benefit you need. IMHO I'd like to see better vectorisation options to make this sort of low level parallelism easier, rather than all this threading stuff. I understand it isn't always the best way, but on your average dual or quad core machine it'd be worth it.
So I'm not sure if there's all that much work in it, I guess.
since we've already got to write somewhat parallel code [because cores are pipelined what, 20 deep now?], it'd be nice if that parallelism extended easily to multi cores. For example, when dividing up some function that reduces [large array -> scalar], we've already got to split execution 20 ways. It'd be nice if there was some way to split such things into the minimum arbitrary number of independent threads supported by the machine, at runtime.
How many software developers work on Windows and IE? Seven or eight years on [by OSX standards], and they still haven't caught up to where the competition was in terms of usability back then. They must have actually been trying to hold out on decent software, in order to keep some sort of market exploitable.
This is why you never let Physicists do mathematics.
I do contract mechanical design when I need a little money, and I got an older lcd as a gift last October. How I did design before then, I'm not entirely sure. The main thing, I guess, is having the CLI on one screen, so you can read the output and type long commands, and your drawing area on another. it's great if you're working from .pdf documentation, bring up your reference and off you go. My office is practically dead-tree free now. Everything else I do only really needs one screen, but not having to flip between workspaces or programs, when you really do need to see both at the same time, is something every power user needs.
It'd be Graetzel if they'd check for novelty before posting on dashslot.
No, you've just got to rewrite the compilers. Considering the plethora of hardware Linux is available for, I'd say anyone interested in promoting a new hardware format can move up with a few weeks of rewriting a compiler. Once one has been done, it's usually pretty easy to make a new lexical analyzer and port all the other compilers too.
Here's my confusion with that RISC argument: faster clocks like that bring relativistic issues into moving data between the processor and the memory. Wherever processing is mostly local to a memory segment, high speed RISC machines are fantastic. But when you've got to access larger memory which may be further away in the machine, you've got to take that latency much more seriously. At 4GHz, Light travels around 75mm in a clock cycle, so a request for RAM access at that speed takes a couple of cycles [with today's FSB speeds, it's two cycles on that bus. Sure, that's four processor cycles, but the bus is pretty wide too]. If modern RISC processors were working at 16GHz and beyond, that latency would compound considerably. Getting the full power out of such a processor would be more work than it's worth.
I'd rather have a larger processor that's going to do more useful work in the same time, the SIMD instructions on the x86 are a pretty mediocre example but they are, I feel, a step in the right direction [vectorisation].
This is something I find most embarrassing concerning x86. Besides the clock speed increase, have we seen a 1000-fold performance increase with that increase in complexity? Sure, some of that complexity has bought about performance increases, we've gone from 8 to 64 bit, and we've got SIMD now. Still, there's a four-fold increase in complexity that isn't accounted for there. I'd rather see all those transistors do something useful, and with the advent of FOSS operating systems, there's some chance one of the big chip makers to turn that four-fold increase in transistors available into useful performance.
The most obvious, full-proof method is that they just lifted the rocks on their own. I suspect we were a lot, lot stronger all those eons ago.
I concur- one more powerful computer you can use for your real tasks, and one for your background stuff. these guys have some interesting stuff. There was one company in a post on /. a while back that had a 1GHz ARM machine in a really, really tiny case. Can't seem to find the article.
There's a whole other set of atheistic theories that are usually pushed along with it, like the one you just mentioned. As a mathematician, I know better, but I still know the implications of saying "I believe in evolution". It's more than a theory, it's a mindset, and some have turned it into a strange cult, with its own dogma and clergy.
"You can't stop vulnerabilities"
You sure can. Especially the simple buffer overflow stuff like this. Arguments about protected IE7 aren't relevant, because it's not IE7 that handles the animated cursor [it only loads it]. Only browsers that don't load animated cursors by default fix the problem. Further, the point that this is arbitrary code execution means that the code runs with full ring-4 privileges [unless this exploit runs on Alpha too, then I'm not sure what happens, although you were specifically referring to Vista which doesn't], which is the same level that system services run in. So really, you can't mitigate the effect much at all. I'm no expert on the ring-0 processes in Windows or the glue that binds them to the rest of the system, but I imagine there's some things an exploit of this kind can do [such as accessing your TCP/IP stack and file system] and things it can't [such as making Windows use a new TCP/IP stack of the cracker's creation- but if you've got the privileges above, why bother?]. So now that we've gotten rid of the "this is not really MS fault" suggestion and removed your false sense of security in "protected" mode, where does that leave us?
IAAM, IAAE, IAAP:
.abs(), .re() or .im().
arguments concerning specific digits are irrelevant. Once you've decided what floating-point precision to use, the precision is fixed. weather 4 is equal to 4. or not depends on the rounding scheme. And honestly, I think Python gets it right. If the numbers are the same to the 53rd binary place, which they are certain to be in Python, they may as well be treated as the same number. If you've got to carry around accuracy values, they need to be in metadata anyway, because they [errors] combine differently under different operations.
Also, who was talking about comparing complex with float? Much as I hate saying this, TMTOWTDI. In Python and on paper. If you need to compare floats and complex numbers, use
Well, I find the filter technique they use very very interesting. But different materials are more or less opaque and reflective to different frequencies, and since radio waves permeate just about everything fairly well, but scatter better [are less directional] because of the longer wavelength. So it'd make sense that the physics you get from these frequencies are ok scattering but more materials appear opaque. TFA was pretty good, I pretty much just responded to the slashdot article :P but yeah, physics of THz-frequency electromagnetic radiation doesn't feel that surprising somehow.
more importantly, will it carry the patented crazy dyson pricetag feature?
We normally call those frequencies microwave. Microwave transmission is nothing new?
Perhaps even human EPROMs?
Or better yet, if you need better performance under load, why not Xen? I expected to see something in there on paravirtualisation, but nothing.
Plastic Polymer? My! What next?!?