The story is odd because it doesn't mention the National Virtual Translation Center http://www.nvtc.gov/index.html, which serves a similar mission, although in a more structured format.
Your post is odd because you suggest top-notch [Arabic] translators are recent graduates. Perhaps junior translators, but for a language like Arabic, even that is pushing it. Good translators take a while to grow.
And you should look at where that two trillion goes with respect to Arabic language education. Consider, for instance, the National Middle East Language Resource Center http://nmelrc.org/ which "undertakes and supports projects such as teacher training, materials development, testing and assessment, integration of pedagogy and technology, study abroad, and K-12 programs" for critical ME languages such as Arabic. They are underfunded by the underfunded Dept of Education, and see hardly a trickle of that huge $2T.
Those universities with good Arabic programs have been oversubscribed for years. There are consistently more students than can be accomodated. And yet those same departments suffer budget cuts and are forced to shrink rather than expand.
When it comes to fielding more Arabic language expertise, I'd say the USA is in dire need of a "deft ploy" by the American government.
Reminds me a lot of a handheld system currently in limited use with the military called the Phrasealator, developed by Applied Data Systems. It's got a bunch of various language packs available for it, and the gist of it is that you speak a question into it (e.g, "Have you seen insurgents nearby?") and after a little A/D conversion and fuzzy matching, would find a corresponding phrase in the target language. The handheld then plays a voice recording of that phrase.
The device is geared for yes/no questions, and communicating unambiguous tasks (e.g., "Please point on this map as to where they are now"), but of course doesn't parse responses. But if you can potentially get a job done with someone via yes/no phrases, this sort of thing would be better than the existing alternative, which looks a lot like people playing Pictionary.
I asked one veteran E.R. doc about the language situation in hospitals and he said it just wasn't that big a deal because often times the whole family would show up in the E.R., and someone in the bunch would be the terp. But that's just one data point.
That's one more helping of misfortune for all the guys with that popular name (It's one of the seven names for "lion" taught to Arabic-speaking children: Laith, Sab, Asad, Qaswara, Ghadanfar, Dirgham and Usama). It's the new Adolf, I guess.
I am trying to imagine the conversation at Yahoo where some bright spark proposed this whole setup:
"What else can we do to cave in to any and all outside pressures?"
"I've got it! By blocking the substring 'allah' while ignoring most every other controversial string, all of our problems will be solved."
"You've cracked it, old boy! Get those pointy heads down in engineering to make it so."
Speaking of changing God's word, I find it fascinating that so many people's lives are based on a Bible that is, in essence, a collection of translations from languages that most Christians cannot read. And those translations were made a long time ago, and were paid for by someone. I'm not qualified to make a judgement on the accuracy of the translations, but how bizarre that so many people predicate their entire lives on a translated Bible without bothering to question its accuracy. Islam, on the other hand, places a huge premium on following the source code, as it were, in the original Arabic.
Here in the Slashdot crowd, we have a whole sub-population of people who do not trust source code that they cannot see, touch, read, compile, or at least compare MD5's. It would be interesting (admittedly, perhaps just to me) to see how much of that sub-population worships a holy book based on source code they can't read, compiled/interpreted by people they may well not have trusted.
While I am not Muslim, I am working my way through the Qur'an in Arabic because I would really like to read this book, and several Arabists have claimed that the English translations are flawed. Onlookers often make some comment to me about how foreign it must be to read right-to-left. When I mention that really important bits of the Bible were actually written in a similar fashion, little squiggles going the wrong way and so on, more often than not I get a skeptical look!
Sure, here's a link to the book on Amazon: On Killing.
As for references, the book is foot-noted throughout. I suppose it comes as no surprise that the military has measured everything, measures everything today, and will likely always measure everything, right down to the last round fired.
The same guy seems to have also written a book called "Stop Teaching Our Kids to Kill : A Call to Action Against TV, Movie and Video Game Violence", which I imagine is more of an elaboration on what he just touches on at the end of On Killing, and which is no doubt even more relevant to the thread topic.
I never put much faith in the idea that voilent video games help make kids into killers until I read Lt. Col. Dave Grossman's book On Killing, which discusses in a systematic and well-referenced manner exactly what the armed forces have done since the Civil War to increase the firing rate of their infantrymen.
Firing rate? Contrary to what you may think of the typical Civil War battlefield, most soldiers did not fire their weapons. On a big field running with blood, cannons booming and everyone screaming, most soldiers would not fire a single shot. Battles would end with literally thousands upon thousands of loaded muskets on the ground. Fast forward to WWII, where we have the image of brave American soliders firing automatic weapons under terrible conditions. The nonfiring rate among infantrymen was 80-85%. Further, only 1% of airmen accounted for over 40% of all downed enemy aircraft. Most pilots did not shoot anyone down or even try to.
The Army decided to look into this. What they found out is that people generally don't want to kill anybody, and would often rather die themselves, even in battle when they are scared to death, than shoot someone. Not that the soldiers were cowards. On the contrary, the same soldiers that would not fire a shot would repeatedly take terrible risks to rescue a wounded comrad. But the Army wanted them to pull the trigger and hit something, and they figured out how. The only way someone that scared would be able to do anything in that situation is if they had been subject to operant conditioning. They would need to program the soldier's midbrain to fire the weapon, since the forebrain is no longer in use under that much stress. They began to make training as realistic as possible in terms of exposure to violence, and make the thought/action of killing part of a soldier's reflex, so that when the bullets started flying, the American soldier would respond.
It worked. During Korea the nonfiring rate among infantrymen rose to about 55%, and by Vietnam it was an amazing 90-95%. The American infantryman was a killer on the battlefield, and only later did the Army realize that fully 98% of soldiers who experience close combat and pull the trigger would be psychiatric casualties. The 2% that weren't mentally crippled are people who, outside the military, would be locked up.
The author makes an excellent study of how this sort of operant conditioning for violence exists outside the military, in movies and video games. Before you knee-jerk and say that violent video games have no impact on the children who play them hours and hours a day, and who then go watch violent movies and television, you should check out this book. It's hard to dismiss the data out of hand.
SGI was my first non-government job, and my first time exposed to the Bay Area, back in the early 90's. SGI was just on a tear then, with Jurassic Park and virtual reality and so on, and it was a blast to work there. In fact, looking back, I'd say I was happier when I was at the office than when I wasn't. The people were brilliant, the products were dead sexy, and the environment was all about balance. For instance, while the group I worked in taught me a lot about what can be done with a polygon, they also introduced me to sumo wrestling (those padded costumes), windsurfing, motorcycle riding, a Grateful Dead concert (one of Jerry's last ones), and strip clubs (bachelor party for a team member).
Debian Stable? With numbers like that, you're talking Duke Nukem Forever II, which is scheduled to release around the time our sun goes nova.
It's funny to see people lump all software into the same category of criticality, as this journalist does. Sometimes, software development should be treated with much more care than it is. But sometimes, much less. Before one goes about writing any software of any kind, it makes sense to find out where you are on the spectrum and get that squared away, because ultimately it will answer every question the development process puts to you.
You are so not alone in this sentiment. Apple, please, give us a real Quicken replacement.
And yes, I have tried Moneydance. It's just missing way too much still.
I have a WinXP box sitting next to my Mac now. I turn it on once a month to fire up Quicken, and then it gets turned back off. Could this be the year Apple rescues us?
I don't know about a cyberterrorist, per se, but there sure are a lot of compromised machines out there. Anyone remember the article that quoted an estimated 200,000 zombies added every day?
Alan Cox said it best in this interview http://www.oreillynet.com/pub/a/network/2005/09/12 /alan-cox.html: "We are still in a world where an attack like the slammer worm combined with a PC BIOS eraser or disk locking tool could wipe out half the PCs exposed to the internet in a few hours."
The 10.4.3 description page reports "Improves the handling of Hebrew and Arabic text." I am not noticing any differences with Arabic and I'm curious what they changed. The little artifacts that are sometimes left when characters are deleted are still there. And I have often wondered why is the default font size for Arabic so teeny? I have to Cmd-+ about 5 times before the Arabic is as visible to me as the default Latin characters.
I wonder if this improved handling is tied to any particular bug reports that we can see?
Deb: What are you drawing? Napoleon Dynamite: A liger. Deb: What's a liger? Napoleon Dynamite: It's pretty much my favorite animal. It's like a lion and a tiger mixed...
Speaking from my own experience, tired + Quicksilver is not a good combo. Having finished the trilogy, I'm actually rereading Quicksilver now trying to figure out who the hell everybody was.
Although it never really occurred to me until now, I guess my philosophy over the past few years has been to get contracts at places where folks with your philosophy have left their mark. When I spend a few weeks optimizing their bloaty app so that it runs easily on 2 servers instead of sweating on 15, the customers are measurably happier. Not only do they save on hardware and have a better performing system, but they do not need as many people to look after the hardware either.
Wherever I go, I encounter this strange allergy to system profiling and I don't really understand how so many people continue to get away with it.
My philosophy is to make the hardware fast and reliable and to keep speed optimizations as far away from the code as possible. This btw is usually more sufficent and measurabley cheaper for the customers aswell.
SFLan.org is a great idea and I applaud the folks behind it, but your statement that it's a decent wireless network in SF is a bit of a stretch. According to the Nagios map, there are 30 active nodes in SF, but mousing over them shows that only about half are deemed reachable. And even if all 30 were reachable, there is not a single one within over a mile from my place in North Beach. Living in one of the most populalated neighborhoods and still being over 1 mile from an SFLan access point in a city that is only 7x7 miles across means SF does not have a decent wireless network.
SFLan's TOS doesn't matter to me because there's no "S" in it for me. I would be happy with whatever TOS the city came up with--- archive my packets for future inspection for all I care--- so long as I can get decent WiFi and don't have to rely on Starbucks to be my ISP.
Slashdot Mirror
The story is odd because it doesn't mention the National Virtual Translation Center http://www.nvtc.gov/index.html, which serves a similar mission, although in a more structured format.
Your post is odd because you suggest top-notch [Arabic] translators are recent graduates. Perhaps junior translators, but for a language like Arabic, even that is pushing it. Good translators take a while to grow.
And you should look at where that two trillion goes with respect to Arabic language education. Consider, for instance, the National Middle East Language Resource Center http://nmelrc.org/ which "undertakes and supports projects such as teacher training, materials development, testing and assessment, integration of pedagogy and technology, study abroad, and K-12 programs" for critical ME languages such as Arabic. They are underfunded by the underfunded Dept of Education, and see hardly a trickle of that huge $2T.
Those universities with good Arabic programs have been oversubscribed for years. There are consistently more students than can be accomodated. And yet those same departments suffer budget cuts and are forced to shrink rather than expand.
When it comes to fielding more Arabic language expertise, I'd say the USA is in dire need of a "deft ploy" by the American government.
I wrote about this in another thread not long ago.
The reference I cite is Lt. Col. Dave Grossman's book On Killing.
Reminds me a lot of a handheld system currently in limited use with the military called the Phrasealator, developed by Applied Data Systems. It's got a bunch of various language packs available for it, and the gist of it is that you speak a question into it (e.g, "Have you seen insurgents nearby?") and after a little A/D conversion and fuzzy matching, would find a corresponding phrase in the target language. The handheld then plays a voice recording of that phrase.
The device is geared for yes/no questions, and communicating unambiguous tasks (e.g., "Please point on this map as to where they are now"), but of course doesn't parse responses. But if you can potentially get a job done with someone via yes/no phrases, this sort of thing would be better than the existing alternative, which looks a lot like people playing Pictionary.
I asked one veteran E.R. doc about the language situation in hospitals and he said it just wasn't that big a deal because often times the whole family would show up in the E.R., and someone in the bunch would be the terp. But that's just one data point.
That's one more helping of misfortune for all the guys with that popular name (It's one of the seven names for "lion" taught to Arabic-speaking children: Laith, Sab, Asad, Qaswara, Ghadanfar, Dirgham and Usama). It's the new Adolf, I guess.
I am trying to imagine the conversation at Yahoo where some bright spark proposed this whole setup:
"What else can we do to cave in to any and all outside pressures?"
"I've got it! By blocking the substring 'allah' while ignoring most every other controversial string, all of our problems will be solved."
"You've cracked it, old boy! Get those pointy heads down in engineering to make it so."
Speaking of changing God's word, I find it fascinating that so many people's lives are based on a Bible that is, in essence, a collection of translations from languages that most Christians cannot read. And those translations were made a long time ago, and were paid for by someone. I'm not qualified to make a judgement on the accuracy of the translations, but how bizarre that so many people predicate their entire lives on a translated Bible without bothering to question its accuracy. Islam, on the other hand, places a huge premium on following the source code, as it were, in the original Arabic.
Here in the Slashdot crowd, we have a whole sub-population of people who do not trust source code that they cannot see, touch, read, compile, or at least compare MD5's. It would be interesting (admittedly, perhaps just to me) to see how much of that sub-population worships a holy book based on source code they can't read, compiled/interpreted by people they may well not have trusted.
While I am not Muslim, I am working my way through the Qur'an in Arabic because I would really like to read this book, and several Arabists have claimed that the English translations are flawed. Onlookers often make some comment to me about how foreign it must be to read right-to-left. When I mention that really important bits of the Bible were actually written in a similar fashion, little squiggles going the wrong way and so on, more often than not I get a skeptical look!
Sure, here's a link to the book on Amazon: On Killing.
As for references, the book is foot-noted throughout. I suppose it comes as no surprise that the military has measured everything, measures everything today, and will likely always measure everything, right down to the last round fired.
The same guy seems to have also written a book called "Stop Teaching Our Kids to Kill : A Call to Action Against TV, Movie and Video Game Violence", which I imagine is more of an elaboration on what he just touches on at the end of On Killing, and which is no doubt even more relevant to the thread topic.
I never put much faith in the idea that voilent video games help make kids into killers until I read Lt. Col. Dave Grossman's book On Killing, which discusses in a systematic and well-referenced manner exactly what the armed forces have done since the Civil War to increase the firing rate of their infantrymen.
Firing rate? Contrary to what you may think of the typical Civil War battlefield, most soldiers did not fire their weapons. On a big field running with blood, cannons booming and everyone screaming, most soldiers would not fire a single shot. Battles would end with literally thousands upon thousands of loaded muskets on the ground. Fast forward to WWII, where we have the image of brave American soliders firing automatic weapons under terrible conditions. The nonfiring rate among infantrymen was 80-85%. Further, only 1% of airmen accounted for over 40% of all downed enemy aircraft. Most pilots did not shoot anyone down or even try to.
The Army decided to look into this. What they found out is that people generally don't want to kill anybody, and would often rather die themselves, even in battle when they are scared to death, than shoot someone. Not that the soldiers were cowards. On the contrary, the same soldiers that would not fire a shot would repeatedly take terrible risks to rescue a wounded comrad. But the Army wanted them to pull the trigger and hit something, and they figured out how. The only way someone that scared would be able to do anything in that situation is if they had been subject to operant conditioning. They would need to program the soldier's midbrain to fire the weapon, since the forebrain is no longer in use under that much stress. They began to make training as realistic as possible in terms of exposure to violence, and make the thought/action of killing part of a soldier's reflex, so that when the bullets started flying, the American soldier would respond.
It worked. During Korea the nonfiring rate among infantrymen rose to about 55%, and by Vietnam it was an amazing 90-95%. The American infantryman was a killer on the battlefield, and only later did the Army realize that fully 98% of soldiers who experience close combat and pull the trigger would be psychiatric casualties. The 2% that weren't mentally crippled are people who, outside the military, would be locked up.
The author makes an excellent study of how this sort of operant conditioning for violence exists outside the military, in movies and video games. Before you knee-jerk and say that violent video games have no impact on the children who play them hours and hours a day, and who then go watch violent movies and television, you should check out this book. It's hard to dismiss the data out of hand.
SGI was my first non-government job, and my first time exposed to the Bay Area, back in the early 90's. SGI was just on a tear then, with Jurassic Park and virtual reality and so on, and it was a blast to work there. In fact, looking back, I'd say I was happier when I was at the office than when I wasn't. The people were brilliant, the products were dead sexy, and the environment was all about balance. For instance, while the group I worked in taught me a lot about what can be done with a polygon, they also introduced me to sumo wrestling (those padded costumes), windsurfing, motorcycle riding, a Grateful Dead concert (one of Jerry's last ones), and strip clubs (bachelor party for a team member).
If there's ever a funeral for SGI, I'd show up.
Debian Stable? With numbers like that, you're talking Duke Nukem Forever II, which is scheduled to release around the time our sun goes nova.
It's funny to see people lump all software into the same category of criticality, as this journalist does. Sometimes, software development should be treated with much more care than it is. But sometimes, much less. Before one goes about writing any software of any kind, it makes sense to find out where you are on the spectrum and get that squared away, because ultimately it will answer every question the development process puts to you.
Yep, good to keep the economics aspect in mind.
Reminds me of one of Murphy's Laws of Combat:
"Remember, your weapon was made by the lowest bidder."
You are so not alone in this sentiment. Apple, please, give us a real Quicken replacement.
And yes, I have tried Moneydance. It's just missing way too much still.
I have a WinXP box sitting next to my Mac now. I turn it on once a month to fire up Quicken, and then it gets turned back off. Could this be the year Apple rescues us?
I don't know about a cyberterrorist, per se, but there sure are a lot of compromised machines out there. Anyone remember the article that quoted an estimated 200,000 zombies added every day?
2 /alan-cox.html:
Alan Cox said it best in this interview http://www.oreillynet.com/pub/a/network/2005/09/1
"We are still in a world where an attack like the slammer worm combined with a PC BIOS eraser or disk locking tool could wipe out half the PCs exposed to the internet in a few hours."
We have different books. In mine (hardcopy), it's pp. 620-621.
Note to self: don't casually pick up Cryptonomicon mid-day if you actually have things to do. I just re-read that whole chapter.
Do you remember whereabouts in the book this passage is?
The 10.4.3 description page reports "Improves the handling of Hebrew and Arabic text." I am not noticing any differences with Arabic and I'm curious what they changed. The little artifacts that are sometimes left when characters are deleted are still there. And I have often wondered why is the default font size for Arabic so teeny? I have to Cmd-+ about 5 times before the Arabic is as visible to me as the default Latin characters.
I wonder if this improved handling is tied to any particular bug reports that we can see?
That is bizarre. I am not seeing either of these problems after my upgrade.
I remember this fellow from 15 years ago at my alma mater http://www.umbc.edu/. Here's his bio http://www.umbc.edu/engineering/me/appa.htm.
Reminds me of something an old professor of mine would quote whenever someone would wish disproportionate punishment upon a person.
"Heat not a furnace for your foe so hot that it do singe yourself." - Shakespeare, King Henry VIII
If we wish murder upon this guy, how medieval should we act toward actual threats to society?
Is it just me, or is browsing Apple's store really slow and painful in Safari, yet zippy with Firefox?
Deb: What are you drawing?
Napoleon Dynamite: A liger.
Deb: What's a liger?
Napoleon Dynamite: It's pretty much my favorite animal. It's like a lion and a tiger mixed...
I have a Motorola RazrV3, and the author does not think the Clicker will ever work on it (per his email to me).
The phone is so thin and sleek, but it's so impoverished when it comes to software.
Speaking from my own experience, tired + Quicksilver is not a good combo. Having finished the trilogy, I'm actually rereading Quicksilver now trying to figure out who the hell everybody was.
CL rocks.
Would melting the north pole really cause the seas to rise? The ice would be replaced by an equal mass of water, right?
Now if the south pole were to melt, and all that ice slid off the continent there and into the seas, that seems like it would cause the seas to rise.
Not advocating a new shipping lane or anything...just saying.
Wherever I go, I encounter this strange allergy to system profiling and I don't really understand how so many people continue to get away with it.
My philosophy is to make the hardware fast and reliable and to keep speed optimizations as far away from the code as possible. This btw is usually more sufficent and measurabley cheaper for the customers aswell.
SFLan.org is a great idea and I applaud the folks behind it, but your statement that it's a decent wireless network in SF is a bit of a stretch. According to the Nagios map, there are 30 active nodes in SF, but mousing over them shows that only about half are deemed reachable. And even if all 30 were reachable, there is not a single one within over a mile from my place in North Beach. Living in one of the most populalated neighborhoods and still being over 1 mile from an SFLan access point in a city that is only 7x7 miles across means SF does not have a decent wireless network.
SFLan's TOS doesn't matter to me because there's no "S" in it for me. I would be happy with whatever TOS the city came up with--- archive my packets for future inspection for all I care--- so long as I can get decent WiFi and don't have to rely on Starbucks to be my ISP.