Kdrive has been distributed with XFree86 forever. It actually has some support for quite a number of display adapters, including a "good enough" VESA mode for generic usage. I use Kdrive on my older laptops.
It looks like the future of X.org, though, is not Kdrive, but yet another X server. This one is being written to run on top of OpenGL. This means no more worries about support on new video cards: they all run OpenGL, so they'll all just work. It also means that the new Render and Compositing stuff will go scary fast. I'm looking forward to it.
Recall that primality testing is now in polytime. It's currently impractical, since the polynomial is order 12 or thereabouts. But expect the search for the largest known prime to get much more boring once someone figures out how to get this algorithm down to a reasonable running time.
Submitter wrote "...The Dalles, Oregon (a little hick town east of Portland)." Geez. I don't recommend a career in journalism, or especially PR, for anyone who would post that in a national forum. On behalf of my friends who live in The Dalles, may I suggest something more like "...The Dalles, Oregon, a rural community of around 15,000 people 80 miles east of Portland?"
Oops---sorry if you took that the wrong way. I was referring to myself in the grandparent:-).
Re:PostGIS
on
Open Maps?
·
· Score: 4, Informative
Yeah, this is the kind of/. question that drives me nuts. "I want some map data, with a whole bunch of constraints on what kind it is, and I want it to be free. Oh, by the way, I found exactly that from the USGS. However, in spite of the fact that there are tens or maybe hundreds of open source projects that use it just fine, I can't figure out how. So that's no good."
The first page of freshmeat.net after searching for "tiger" contains a link to this open source TIGER map server.
Maybe that would be a good starting point. Further down the page are getmap and geotools, which also support TIGER.
I wish submitters and especially editors would realize that when they don't do their homework, they're wasting the time of literally hundreds of thousands of people. Sometimes a lot of time, like when the idiots actually waste extra time writing a long-winded reply.
There's a famous essay by a former Senator (Daniel Moynihan or SI Hayakawa, I think---I tried to track it down online, but didn't spot it offhand) arguing that some of the increase in crimes, prosecution, and jailing comes from creep in the social norms constituting criminal behavior. There is always a tendency to criminalize borderline behavior, which then makes formerly non-borderline behavior borderline...
As cool as this antenna is, it isn't very expensive to buy a nice prebuilt wireless antenna these days. Pacific Wireless dishes, for example, are about US$50 for 19dBi or $70 for 24dBi. I've used their products, and they are very nice. I've given up on building 802.11 antennas: it's too much work vs the cheap commercial antennas for me.
I currently have 3 different wireless solutions working well for my laptop:-).
Lucent Orinoco PCMCIA card. Works great with stock kernel stuff, Kismet understands it, etc. Disadvantages: not super cheap, 802.11b only, 64-bit WEP only (doesn't matter for security, but some APs are set up 128b).
Airlink+ B/G Cardbus card. Got ndiswrapper installed and working, and it all seems to work fine. Disadvantages: need to set up ndiswrapper, need to get the right windows driver, doesn't do some of the advanced wireless extensions.
Centrino built-in Intel Pro Wireless on my Powerbook T-41. Works with the recently posted sourceforge IPW driver. Would probably also work with ndiswrapper, but I couldn't figure out how to make it coexist with (2) above. Disadvantages: B only. Had to find and compile the software to use it. Built-in, so I can't loan it out.
Probably the biggest problem at this point is that the configuration is primitive and awful. I finally have it so that my hard ethernet and all 3 wireless cards are autodetected and conform to the current networking "scheme". It wasn't easy. I really need to do the HOWTO sometime, but it is possible.
I run a Windows XP box at home. My wife uses it for random stuff. I've had this incarnation of XP for about 6 months, since my previous Windows install got too flaky and I decided it was time to upgrade. Had to upgrade the motherboard to get XP onto the box, too.
Recently, the box started hanging mysteriously. Would just freeze up so that even a hard reset wouldn't touch it. Suspected a HW problem. Poked around a lot. Finally uninstalled a server app downloaded from TiVO.com (the "Home Media Option") and the box seems to have quit hanging.
So I got a bad app---so what?
It is quite difficult for a bad Linux app to lock up the box, even if run as root. Run as an unprivileged user, it's nearly impossible. I replaced the Windows app with an open source Linux app. It runs as an unpriviliged user just fine now. I had to fix a bug to stop the Linux server from crashing in an unusual configuration I tried. But its death never bothered the rest of the machine it was running on in any way.
Consider the usage pattern you describe. "I boot up in the morning, switch between video and photo editing software hundreds of times throughout the day with regular use of MSIE and Eudora as well, and then I shut it down at night."
This is an incredibly OS-friendly pattern. You only use a few apps, and you reboot the system once a day. Yet you're downright proud of how XP doesn't crash. Methinks you protest too much. My Linux boxes stay up for 6 months at a time, run all kinds of random garbage as root, have no spyware or virus protection. My internet firewall box is a Linux box. Yet my Linux boxes never crash either, and I don't even think about it, because this is how computing is supposed to be.
The low-hanging fruit for preventing Windows crashes has all been plucked. XP does indeed seem better than 98 or even 2K as far as crash-proofing. But the problems to solve to make it truly crash-resistant are fundamental. As long as most Windows apps insist on being able to scribble in a world-writable registry also used by the OS, there's going to be trouble. As long as many Windows apps insist on being able to scribble in memory that has not been allocated to them by the OS, there's going to be trouble. As long as Windows apps need to bring along a bunch of their own customized libraries and install them with or over Windows libraries, there's going to be trouble. With the installed base of Windows apps, that's a hard set of problems to fix. Ditching that installed base leaves Windows head-to-head with Linux, which arguably has more (and more functional) apps than Windows if you count only stuff less than two years old.
I've rarely talked to a real-world user of Windows outside strictly controlled corporate environments who didn't complain about frequent OS crashes and lockups. (Don't even get me started on the Windows malware problem.) I'm not sure which "real world" you're talking to Windows users in, but it's apparently a different one from mine.
Why target Opera and not an open source browser? Because Opera can be driven out of business.
An OSS browser will keep developing and improving as long as there's a dozen people using it: Opera needs to sell hundreds of thousands of copies a year to pay the bills. Microsoft understands how to deal with traditional business competition, and does it well. It's harder to know what to do with OSS: high-risk (of negative publicity) low-return (doesn't do much) attacks are probably a bad idea.
Aside from the fact that Cobalts were actually based on MIPS parts, as several other posters have noted...
The generally accepted reason for the failure of ARM to move up-scale so far is that the main company producing high-end ARMs these days has been Intel. Oddly, they seem to have issues with creating a competitor to their flagship CPUs, so they keep leaving the FPU off the part. In 2004, a CPU sans FPU is a pretty unlikely desktop box.
I'm very excited about the ARM Ltd part for this reason. Not only an FPU: a vector FPU! Build me a board, and I'll buy one tomorrow.
X has been a thorn in the side of desktop development for two long. The Y-Windows [y-windows.org] paper describes why, and why they are creating a replacement from scratch. It will also be network-transparent and integrated. This hack of emulating a desktop on top of a library on top of a window manager on top of a graphics server is completely amateur and unprofessional.
Yeah, it's hard to imagine why, in the year since its announcement, no one has made any significant progress on Y. Or not. Maybe developing a new window system from scratch is both hard and pointless at this point in history?
You completely misdescribe the X architecture, which does make it easier to criticize. In fact, the X "desktop" window system runs directly on the graphics server, the window manager is just another client app, and all the client apps run on top of libraries. Aside from the fact that this works quite well, and has for "two long" decades, it is also pretty much exactly what Microsoft and Apple do. The only noticeable distinction is that the window management functions are integrated into the graphics server on these OSes: this doesn't appear to offer much advantage in practice, and makes it very difficult to change desktop behavior.
You don't need to hear individual bits. You just need to see a statistical difference in speed over iterated modular exponentiations. See Kocher's seminal 1996 paper on timing attacks for a good explanation of this type of analysis. Note that Kocher is measuring operation times on the order of 10ms. This is well within-bounds for an audio measurement.
Not being a professional cryptographer myself, I'm really not qualified to say for sure how practical this attack is. However, I can't rule it out on the basis of my amateur knowledge. I'd say there's a good chance it could work.
Uh, no. Your analysis runs contrary to cryptanalytic principles and the history of these sorts of attacks.
If you spot me 1 bit of key information, you have by definition halved the work for an attack. In this specific analysis, I need only consider those settings of key bits (in this case, bits of p and q) that correspond to observed behavior for an interval of the spectogram. This means that I can potentially crack the key in time almost linear in the size of the key, rather than completely exponential.
The work on timing attacks and power attacks uses very similar sorts of information, and the anlysis used here will likely be similar also.
This is why Shamir, who is certainly qualified to evaluate the work at this point, describes it as "proof of concept": it would be surprising if the observed information fails to extend to a practical attack. It's just that in science, you publish when you have anything interesting to report, so that folks know you got there first.
Just now I'm working on an econometric model for the Panama Canal (they're trying to make it bigger and need to figure out if it's worth the effort/investment) and playing with over 300 variables and 100 parameters to simulate dozens of different scenarios can make any server beg for more cycles, and any user beg for a crystal ball.
As a practicing SE with 20 years experience as well as a University prof who teaches SE and AI for a living, I suspect you're wrong. Talk to someone at your local Uni who is an expert in "combinatorial optimization". What you will find out is that (1) there are good algorithms that might help you a lot (only 100 parameters? hah:-) and that (2) because this class of problem tends to scale exponentially with effective problem size, any performance gain you get from slicing off constant factors of 3-20 or so will be swamped by even slight improvement in the search algorithm used.
I'm not there and YMMV, of course, but check it out---it will probably be worth it.
I took HS calculus and then took college calc. Twice, at two different schools. Did well grade-wise all three times. By the third time through, I finally got it pretty well:-).
The worst teacher/course I ever had (and stayed in) in 13 years (not a typo) of college was a freshman calc course. One of the best courses I ever had was also a freshman calc course. It kind of depends who is teaching you and what the situation is (duh).
Point is, I never regretted taking HS calc, and recommend it to my HS interns (I'm a CS prof now). Just plan on taking it again when you get to college. What're you going to take in HS instead, anyhow? I took 3/4 of the electives in our large curriculum, and had plenty of time for math classes.
BTW, if your HS offers a course in logic (few do), jump at it. If not, take it first thing when you get to college. For an undergrad CS major, this is the most important thing they never bother to teach you, or at least not until too late.
But when windows XP SP2 comes out, the worms will die away a bit, and it will only be social engineering attachment trojans in outlook.
Muahaha. Can I quote you on that about 2 months after the glorious Windows XP SP2 solves all of MS's security and reliability problems in one fell swoop?
No, I can't because you wisely cowered in anonymity. Nice troll.
Anyone with the least degree of responsibility and clue working outside their world knows to get help from an expert when getting started. I'm doing a job involving in-store market research right now: think I'm just figuring it out on my own? I don't buy the excuse that Googling or hitting the obvious website is so hard: my non-techie friends seem to be able to do it fine with no help from me. But at the very least, the reviewer should have asked someone who knew the program where the information he needed was.
Besides, if you look under the Help menu in GIMP 2.0, the last entry is entitled "The GIMP Online", and will bring up a browser pointing at the web site. I'd say that ought to be enough of a cue for anyone who was actually trying to figure the thing out, rather than just bash it.
Yes, the author of this "review" claims that there is no manual for the GIMP. I guess it would have been too difficult to hit gimp.org and click on the link labeled "Documentation". Heck, the author couldn't even be bothered to look up the fact that GIMP is capitalized as I am doing and is an acronym for GNU Image Manipulation Program. That information is the first sentence on gimp.org, for pity's sake.
I call BS on those who say this review is more than just OSS bashing. Apparently the reviewer is some professional graphics bigshot. Good for them: given the quality of this review, I wouldn't hire them. There are plenty of professional graphics designers in my hometown with ethics and a brain as well as good artistic skills. I'll hire one of them instead, thanks much---and yes, they're welcome to use whatever tool they find effective.
As someone who teaches SE at University, I say "amen" to the parent:-). Grandparent's teacher is either being misunderstood (likely) or an idiot. A post farther down in this thread points out that distinguishing functional from non-functional requirements is, uh, kind of important. So is understanding the idea of a "design constraint".
Of course, "requirements are the enemy of design" sounds pretty cool, and it is true in a way. The important thing to notice is that at the end of the day, requirements always win. That is, no matter how cool the design is, if the product doesn't do what it is supposed to, it is useless. If the product meets requirements, it really doesn't matter what design got it to that point: it's fine. Given that, you might think that it would be a good idea to know what you need to build before you start designing it. You'd be right.
Grandparent, make sure you aren't completely misinterpreting your teacher's position. If you are not, drop the course. In grad school, I dropped my SE course because my (mighty famous) teacher was so ignorant. It didn't preclude me from learning the field to the point where I teach regularly in a professional SE MS program.
Slashdot Mirror
Kdrive has been distributed with XFree86 forever. It actually has some support for quite a number of display adapters, including a "good enough" VESA mode for generic usage. I use Kdrive on my older laptops.
It looks like the future of X.org, though, is not Kdrive, but yet another X server. This one is being written to run on top of OpenGL. This means no more worries about support on new video cards: they all run OpenGL, so they'll all just work. It also means that the new Render and Compositing stuff will go scary fast. I'm looking forward to it.
Recall that primality testing is now in polytime. It's currently impractical, since the polynomial is order 12 or thereabouts. But expect the search for the largest known prime to get much more boring once someone figures out how to get this algorithm down to a reasonable running time.
Submitter wrote "...The Dalles, Oregon (a little hick town east of Portland)." Geez. I don't recommend a career in journalism, or especially PR, for anyone who would post that in a national forum. On behalf of my friends who live in The Dalles, may I suggest something more like "...The Dalles, Oregon, a rural community of around 15,000 people 80 miles east of Portland?"
Oops---sorry if you took that the wrong way. I was referring to myself in the grandparent :-).
Yeah, this is the kind of /. question that drives me nuts. "I want some map data, with a whole bunch of constraints on what kind it is, and I want it to be free. Oh, by the way, I found exactly that from the USGS. However, in spite of the fact that there are tens or maybe hundreds of open source projects that use it just fine, I can't figure out how. So that's no good."
The first page of freshmeat.net after searching for "tiger" contains a link to this open source TIGER map server. Maybe that would be a good starting point. Further down the page are getmap and geotools, which also support TIGER.
I wish submitters and especially editors would realize that when they don't do their homework, they're wasting the time of literally hundreds of thousands of people. Sometimes a lot of time, like when the idiots actually waste extra time writing a long-winded reply.
I own a TMD fan. Bought it at Fry's: forget the brand. Still uses conventional bearings, though. Check out this page for more info.
There's a famous essay by a former Senator (Daniel Moynihan or SI Hayakawa, I think---I tried to track it down online, but didn't spot it offhand) arguing that some of the increase in crimes, prosecution, and jailing comes from creep in the social norms constituting criminal behavior. There is always a tendency to criminalize borderline behavior, which then makes formerly non-borderline behavior borderline...
As cool as this antenna is, it isn't very expensive to buy a nice prebuilt wireless antenna these days. Pacific Wireless dishes, for example, are about US$50 for 19dBi or $70 for 24dBi. I've used their products, and they are very nice. I've given up on building 802.11 antennas: it's too much work vs the cheap commercial antennas for me.
I currently have 3 different wireless solutions working well for my laptop :-).
Probably the biggest problem at this point is that the configuration is primitive and awful. I finally have it so that my hard ethernet and all 3 wireless cards are autodetected and conform to the current networking "scheme". It wasn't easy. I really need to do the HOWTO sometime, but it is possible.
I run a Windows XP box at home. My wife uses it for random stuff. I've had this incarnation of XP for about 6 months, since my previous Windows install got too flaky and I decided it was time to upgrade. Had to upgrade the motherboard to get XP onto the box, too.
Recently, the box started hanging mysteriously. Would just freeze up so that even a hard reset wouldn't touch it. Suspected a HW problem. Poked around a lot. Finally uninstalled a server app downloaded from TiVO.com (the "Home Media Option") and the box seems to have quit hanging.
So I got a bad app---so what?
I've rarely talked to a real-world user of Windows outside strictly controlled corporate environments who didn't complain about frequent OS crashes and lockups. (Don't even get me started on the Windows malware problem.) I'm not sure which "real world" you're talking to Windows users in, but it's apparently a different one from mine.
Why target Opera and not an open source browser? Because Opera can be driven out of business.
An OSS browser will keep developing and improving as long as there's a dozen people using it: Opera needs to sell hundreds of thousands of copies a year to pay the bills. Microsoft understands how to deal with traditional business competition, and does it well. It's harder to know what to do with OSS: high-risk (of negative publicity) low-return (doesn't do much) attacks are probably a bad idea.
Aside from the fact that Cobalts were actually based on MIPS parts, as several other posters have noted...
The generally accepted reason for the failure of ARM to move up-scale so far is that the main company producing high-end ARMs these days has been Intel. Oddly, they seem to have issues with creating a competitor to their flagship CPUs, so they keep leaving the FPU off the part. In 2004, a CPU sans FPU is a pretty unlikely desktop box.
I'm very excited about the ARM Ltd part for this reason. Not only an FPU: a vector FPU! Build me a board, and I'll buy one tomorrow.
Thanks very much for your suggestion: I took you up on it. Should have done it a long time ago.
X has been a thorn in the side of desktop development for two long. The Y-Windows [y-windows.org] paper describes why, and why they are creating a replacement from scratch. It will also be network-transparent and integrated. This hack of emulating a desktop on top of a library on top of a window manager on top of a graphics server is completely amateur and unprofessional.
Yeah, it's hard to imagine why, in the year since its announcement, no one has made any significant progress on Y. Or not. Maybe developing a new window system from scratch is both hard and pointless at this point in history?
You completely misdescribe the X architecture, which does make it easier to criticize. In fact, the X "desktop" window system runs directly on the graphics server, the window manager is just another client app, and all the client apps run on top of libraries. Aside from the fact that this works quite well, and has for "two long" decades, it is also pretty much exactly what Microsoft and Apple do. The only noticeable distinction is that the window management functions are integrated into the graphics server on these OSes: this doesn't appear to offer much advantage in practice, and makes it very difficult to change desktop behavior.
Damn you, X trolls. You get me every time.
You don't need to hear individual bits. You just need to see a statistical difference in speed over iterated modular exponentiations. See Kocher's seminal 1996 paper on timing attacks for a good explanation of this type of analysis. Note that Kocher is measuring operation times on the order of 10ms. This is well within-bounds for an audio measurement.
Not being a professional cryptographer myself, I'm really not qualified to say for sure how practical this attack is. However, I can't rule it out on the basis of my amateur knowledge. I'd say there's a good chance it could work.
Uh, no. Your analysis runs contrary to cryptanalytic principles and the history of these sorts of attacks.
If you spot me 1 bit of key information, you have by definition halved the work for an attack. In this specific analysis, I need only consider those settings of key bits (in this case, bits of p and q) that correspond to observed behavior for an interval of the spectogram. This means that I can potentially crack the key in time almost linear in the size of the key, rather than completely exponential.
The work on timing attacks and power attacks uses very similar sorts of information, and the anlysis used here will likely be similar also. This is why Shamir, who is certainly qualified to evaluate the work at this point, describes it as "proof of concept": it would be surprising if the observed information fails to extend to a practical attack. It's just that in science, you publish when you have anything interesting to report, so that folks know you got there first.
"...or so says their website."
Uh, no. Their website doesn't say anything anymore. Indeed, it's apparently powered by something called "PostNuke".
/. --- because the net needed a gratuitous Heisenberg effect.
Just now I'm working on an econometric model for the Panama Canal (they're trying to make it bigger and need to figure out if it's worth the effort/investment) and playing with over 300 variables and 100 parameters to simulate dozens of different scenarios can make any server beg for more cycles, and any user beg for a crystal ball.
As a practicing SE with 20 years experience as well as a University prof who teaches SE and AI for a living, I suspect you're wrong. Talk to someone at your local Uni who is an expert in "combinatorial optimization". What you will find out is that (1) there are good algorithms that might help you a lot (only 100 parameters? hah :-) and that (2) because this class of problem tends to scale exponentially with effective problem size, any performance gain you get from slicing off constant factors of 3-20 or so will be swamped by even slight improvement in the search algorithm used.
I'm not there and YMMV, of course, but check it out---it will probably be worth it.
I took HS calculus and then took college calc. Twice, at two different schools. Did well grade-wise all three times. By the third time through, I finally got it pretty well :-).
The worst teacher/course I ever had (and stayed in) in 13 years (not a typo) of college was a freshman calc course. One of the best courses I ever had was also a freshman calc course. It kind of depends who is teaching you and what the situation is (duh).
Point is, I never regretted taking HS calc, and recommend it to my HS interns (I'm a CS prof now). Just plan on taking it again when you get to college. What're you going to take in HS instead, anyhow? I took 3/4 of the electives in our large curriculum, and had plenty of time for math classes.
BTW, if your HS offers a course in logic (few do), jump at it. If not, take it first thing when you get to college. For an undergrad CS major, this is the most important thing they never bother to teach you, or at least not until too late.
But when windows XP SP2 comes out, the worms will die away a bit, and it will only be social engineering attachment trojans in outlook.
Muahaha. Can I quote you on that about 2 months after the glorious Windows XP SP2 solves all of MS's security and reliability problems in one fell swoop?
No, I can't because you wisely cowered in anonymity. Nice troll.
Dr. FTC: We will fine him almost...[holds pinky to corner of mouth] one million dollars!
Anyone with the least degree of responsibility and clue working outside their world knows to get help from an expert when getting started. I'm doing a job involving in-store market research right now: think I'm just figuring it out on my own? I don't buy the excuse that Googling or hitting the obvious website is so hard: my non-techie friends seem to be able to do it fine with no help from me. But at the very least, the reviewer should have asked someone who knew the program where the information he needed was.
Besides, if you look under the Help menu in GIMP 2.0, the last entry is entitled "The GIMP Online", and will bring up a browser pointing at the web site. I'd say that ought to be enough of a cue for anyone who was actually trying to figure the thing out, rather than just bash it.
I'll stand by my original assessment.
Yes, the author of this "review" claims that there is no manual for the GIMP. I guess it would have been too difficult to hit gimp.org and click on the link labeled "Documentation". Heck, the author couldn't even be bothered to look up the fact that GIMP is capitalized as I am doing and is an acronym for GNU Image Manipulation Program. That information is the first sentence on gimp.org, for pity's sake.
I call BS on those who say this review is more than just OSS bashing. Apparently the reviewer is some professional graphics bigshot. Good for them: given the quality of this review, I wouldn't hire them. There are plenty of professional graphics designers in my hometown with ethics and a brain as well as good artistic skills. I'll hire one of them instead, thanks much---and yes, they're welcome to use whatever tool they find effective.
Yes :-). Should have written M.Sc., but it never occurred to me.
As someone who teaches SE at University, I say "amen" to the parent :-). Grandparent's teacher is either being misunderstood (likely) or an idiot. A post farther down in this thread points out that distinguishing functional from non-functional requirements is, uh, kind of important. So is understanding the idea of a "design constraint".
Of course, "requirements are the enemy of design" sounds pretty cool, and it is true in a way. The important thing to notice is that at the end of the day, requirements always win. That is, no matter how cool the design is, if the product doesn't do what it is supposed to, it is useless. If the product meets requirements, it really doesn't matter what design got it to that point: it's fine. Given that, you might think that it would be a good idea to know what you need to build before you start designing it. You'd be right.
Grandparent, make sure you aren't completely misinterpreting your teacher's position. If you are not, drop the course. In grad school, I dropped my SE course because my (mighty famous) teacher was so ignorant. It didn't preclude me from learning the field to the point where I teach regularly in a professional SE MS program.