Congratulations! Instead of a relatively low noise, moderately power hungry CCD sensor, I now have a relatively high noise, low power CMOS sensor that needs to be cooled to suppress dark current enough to get usable imagery. Thanks! I sure am looking forward to seeing digital cameras with TECs or cryo-coolers like my FLIR uses in them.
not duke per se, but certainly other high prestige universities who really should know better.
poster hasn't worked much with universities
on
12.8 Petabytes, You Say?
·
· Score: 2, Insightful
The fact that this is coming out of a university gives me hope that this technology won't turn out to be just so much vapor.
<rant>i don't think the poster has worked with many universities. my experience with using them as subs on r & d projects has been highly mixed — occasionally, you'll find a group that just rocks. the problem is that the remaining 7-9 out of 10 times, you end up just replacing the components that the university was supposed to deliver because they either (a) failed to deliver anything at all (not uncommon) or (b) delivered code that was so horrendously broken that it was less effort to redo their pieces than to shepherd them through the process of fixing things.</rant>
before i'm flamed to death: please note that i didn't say all universities suck (and in particularly, i didn't imply that your university sucks).
as it happens, $55-65 is about what it costs to have a two bedroom apartment cleaned (at least in the denver area). the choice wasn't quite as random as it seems;) by all means, make it in the $105-115 range (which should cover having laundry taken care of and possibly some shopping done) if $55 isn't enough.
give 'em some reasonable number of requests, and after that charge them $55-65 per incident (which should nicely cover the cost of having cleaners deal with at least some of the domestic stuff for you).
...teach them to communicate. This means consistency, clarity, coherency, cohesion, and completeness. Once they're able to maintain the five C's of communication,/then/ see about teaching them how to write. You'll find that once the core communications skills are in place, it will be a lot easier to get people to understand some of the more technical details (comma use, why parallel structure is a good thing, why active voice is a good thing, etc.)
Some thoughts on assignments: As a group, define $really_common_term. Make sure multiple groups get the same term. Once the groups are done, have them share their definitions with the class, then compare and contrast them. Ask them about their process, how they found their wording ("why did you phrase things like ______ rather than like ______") -- this is about getting your students thinking critically about their writing. Maybe do this on a regular basis (new groups and terms each time).
As a class, define characteristics of successful communication. Use these characteristics to define metrics for communication, then apply them (for example, to the above group exercise).
Ultimately, your goal should be to use these exercises to instill some kind of discipline into your students' approaches to communication. Show them how you can apply engineering principles to constructing their communications in an effective manner and let the writing center deal with punctuation and grammatical minutiae.
fantastic. you've now told spammers how to defeat basically every statistical spam filter. now i get to attempt to teach the generally tech-clueless people in my life about pgp or equivalent so that i can automatically block all non-signed email. except i can't, because there are no online vendors / banking services / etc. that sign their outbound email, to the best of my knowledge.
just because you know how to do something like essentially unbreakable steganography in video sequences doesn't mean that it's something you need to share with the rest of the world.
http://valgrind.org/ used to include a tool called hellgrind for finding just such problems. unfortunately, hellgrind has gone away for a bit (it broke when the VM was re-done to support non-x86 platforms), but julian & co are working hard to get it working again Real Soon Now (tm). if you're using x86, you can use an old release of valgrind (2.2.0 i think) and you should be fine.
personally, i can't say enough good things about valgrind. there are a couple non-obvious issues (support for sse/sse2/sse3 is still in the works, so if you get an inexplicable SIGILL, this is probably the problem), but it's saved me hundreds of hours over the past year (and i'm sure it'll save me even more in the future).
that all said, my (admittedly limited) experience with threading is that it's best to design the deadlocks away before you even touch the editor. i wonder if there are any design tools which support deadlock / contention checking at the model or design level?
finding a few people to kick in a couple thousand here and there isn't a big deal, the problem is that you need at least $x per league contestant + prizes to have those pro gamers have even a hope of earning a living (never mind a good living). if you let $x be twice the minimum wage in the us, you're talking on the order of $83k per year per team of four for salaries. add in another $20k for equipment, another $15k on travel, and $10k for tournament fees per team and you're up to $128k per year. suppose the league gives out $2mil in prizes over a season and has 20 teams, you're talking about $4.54mil per year to run the league. if your average sponsor kicks in $50k per year (which is about five half-page color ads in a major newspaper), you're going to need about 92 sponsors. convincing that many sponsors that $50k per year out of their advertising budgets spent on pro-gaming is going to be a hard sell until those gaming leagues generate several times that amount in revenue. i don't see that kind of serious revenue generation happening any time soon, and i suspect that most people who would consider sponsoring those leagues feel the same way as i do.
i think you're right -- i don't think the economics of pro-gaming as a spectator sport make any sense, other than advertising ("oooh, ($new_fps|$new_gaming_box|$new_mouse)... must.buy.now!"). i'm willing to wager that the vast majority of people (i.e. not 0.00001% that the slashdot crowd comprises) would rather play games than watch them, which means no ad revenue, which means no sponsors, which means no sustainable leagues that can support professional play.
step 1, like other posters have mentioned, is to get him to understand that he has a problem (ideally before he gets fired). this involves some sort of intervention. you'll want to plan it before hand for two reasons. firstly, you want to make sure that you've got the right mix of people talking to him; secondly, you as the interveners are going to need some practice. expect that you might get any reaction from "shit, you're right." to "you're just jealous" to overt hostility.
there are several tactics you can use to get your point across. if your friend doesn't let challenges go unanswered, challenge him to put himself in a situation where he can't play the game for a month or two. tell your friend how his addiction is affecting your relations with him (this is particularly relevant if his family or significant other is at the intervention). you need to be careful that when you do this, you're making "i" statements --- "i feel like _____ when you ditch me" rather than "you ditched me, jackass". you know your friend a lot better than i do, so you've got a better feel for what may or may not work well for him. while you're having the intervention, it's really important that you all make it clear that a) you're there for him, b) you're not judging him, c) you're going to help him pull through when he asks for your help, d) this is not something that will be discussed outside of the people in the room. your goal is to make the room a safe space (much easier said than done).
good luck -- you're about to go through a really rough patch.
suppose that m$ did find some kind of infringing ip in linux and they attempt to sue. i expect that if they were to win, they'd see a second slew of anti-trust lawsuits making the claim that microsoft's enforcement of certain patents demonstrably stifles innovation and generally screws the public because it is now impossible to work on $package. personally, i think that companies which have been convicted of anti-trust violations need to have some limits on their ability to enforce their ip rights as a way of offsetting the damage they did. just a thought...
The charges arise from Heller's alleged disclosure two years ago of legal papers from the Los Angeles office of international law firm Jones Day, which represented Diebold at the time. Heller was under contract as a word processor at Jones Day.
The documents included legal memos from one Jones Day attorney to another regarding allegations by activists that Diebold had used uncertified voting systems in Alameda County elections beginning in 2002.
emphasis my own. the memos in question (at least according to tfa) make it pretty clear that they are discussing allegations that the voting machines weren't certified. they did not (as far as i can tell from the article) state that the voting machines were, in fact, not certified. big, big difference. in the best case, he's a fool for misunderstanding the content of the documents and living under the misapprehension that he's a whistleblower. in the worst case, he stole internal documents in an attempt to win his moment in the sun.
i seriously doubt that you are a scientist of any calibre, because if you were you would realize that the following truths apply: 1) funding has to come from somewhere, it does not magically appear ex nihilo; 2) science is usually hard to do without some substantial capital expenses and a constant revenue of funding to pay for lab techs and folks doing the grunt work; 3) if you love what you do, you're excited about it and it's something that you want to share with people and help them get excited about too.
getting funding requires that you get the funding agency as excited about what it is that you do as you are, i.e. it requires that you be a "pop scientist". no pop scientist == no funding == no science.
i believe that witten showed two classes of these theories to be equivalent. as i recall, it's an open question as to whether all classes are equivalent. kind of the physicist's version of the P?=NP problem as i recall (i am most certainly not a physicist, so feel free to take anything i say with a suitably large grain of salt)
but patents (like guns) are not a priori evil. if they really do bother you so much, why don't you do something about it by becoming a patent examiner with a clue?
... i imagine the dod may have a thing or two to say about this, since wireless networks are a perennial favorite sbir / sttr topic. in fact, i could see some really nasty fallout for boeing from this if they use their patent to stifle small business innovation in the areas that the dod cares about.
how exactly using linux in violation of the gpl is a violation of sarbanes oxley? the article does an awful lot of handwaving but doesn't actually explain any of the hows or whys.
i'll have to read again, but it looks like this is f/oss trying its hand at the fud game.
webcams under linux break into roughly two camps: firewire (aka DCAM or IIDC) cameras and anything that can be handled by V4L or V4L2. which software you use depends very much on what camera you have (sucks, but that's how it is).
if you have an iidc camera (a la apple's isight, an orange micro ibot, etc.) then you really should be using coriander which can be compiled with support for ftp upload of the images. you can also use coriander to set things up for using ffmpeg (see below).
otherwise, i'd suggest looking into using ffserver and ffmpeg. when compiled correctly, it can handle both dv, iddc/dcam, and v4l cameras. i've had rather good results with them, and they're a lot cheaper (i.e. free) than an axis box.
if neither of these works for you, i wish you the best of luck -- you're probably going to need it.
Slashdot Mirror
Congratulations! Instead of a relatively low noise, moderately power hungry CCD sensor, I now have a relatively high noise, low power CMOS sensor that needs to be cooled to suppress dark current enough to get usable imagery. Thanks! I sure am looking forward to seeing digital cameras with TECs or cryo-coolers like my FLIR uses in them.
not duke per se, but certainly other high prestige universities who really should know better.
The fact that this is coming out of a university gives me hope that this technology won't turn out to be just so much vapor.
<rant>i don't think the poster has worked with many universities. my experience with using them as subs on r & d projects has been highly mixed — occasionally, you'll find a group that just rocks. the problem is that the remaining 7-9 out of 10 times, you end up just replacing the components that the university was supposed to deliver because they either (a) failed to deliver anything at all (not uncommon) or (b) delivered code that was so horrendously broken that it was less effort to redo their pieces than to shepherd them through the process of fixing things.</rant>
before i'm flamed to death: please note that i didn't say all universities suck (and in particularly, i didn't imply that your university sucks).
as it happens, $55-65 is about what it costs to have a two bedroom apartment cleaned (at least in the denver area). the choice wasn't quite as random as it seems ;) by all means, make it in the $105-115 range (which should cover having laundry taken care of and possibly some shopping done) if $55 isn't enough.
give 'em some reasonable number of requests, and after that charge them $55-65 per incident (which should nicely cover the cost of having cleaners deal with at least some of the domestic stuff for you).
...teach them to communicate. This means consistency, clarity, coherency, cohesion, and completeness. Once they're able to maintain the five C's of communication, /then/ see about teaching them how to write. You'll find that once the core communications skills are in place, it will be a lot easier to get people to understand some of the more technical details (comma use, why parallel structure is a good thing, why active voice is a good thing, etc.)
Some thoughts on assignments:
As a group, define $really_common_term. Make sure multiple groups get the same term. Once the groups are done, have them share their definitions with the class, then compare and contrast them. Ask them about their process, how they found their wording ("why did you phrase things like ______ rather than like ______") -- this is about getting your students thinking critically about their writing. Maybe do this on a regular basis (new groups and terms each time).
As a class, define characteristics of successful communication. Use these characteristics to define metrics for communication, then apply them (for example, to the above group exercise).
Ultimately, your goal should be to use these exercises to instill some kind of discipline into your students' approaches to communication. Show them how you can apply engineering principles to constructing their communications in an effective manner and let the writing center deal with punctuation and grammatical minutiae.
Good luck!
fantastic. you've now told spammers how to defeat basically every statistical spam filter. now i get to attempt to teach the generally tech-clueless people in my life about pgp or equivalent so that i can automatically block all non-signed email. except i can't, because there are no online vendors / banking services / etc. that sign their outbound email, to the best of my knowledge.
just because you know how to do something like essentially unbreakable steganography in video sequences doesn't mean that it's something you need to share with the rest of the world.
http://valgrind.org/ used to include a tool called hellgrind for finding just such problems. unfortunately, hellgrind has gone away for a bit (it broke when the VM was re-done to support non-x86 platforms), but julian & co are working hard to get it working again Real Soon Now (tm). if you're using x86, you can use an old release of valgrind (2.2.0 i think) and you should be fine.
personally, i can't say enough good things about valgrind. there are a couple non-obvious issues (support for sse/sse2/sse3 is still in the works, so if you get an inexplicable SIGILL, this is probably the problem), but it's saved me hundreds of hours over the past year (and i'm sure it'll save me even more in the future).
that all said, my (admittedly limited) experience with threading is that it's best to design the deadlocks away before you even touch the editor. i wonder if there are any design tools which support deadlock / contention checking at the model or design level?
finding a few people to kick in a couple thousand here and there isn't a big deal, the problem is that you need at least $x per league contestant + prizes to have those pro gamers have even a hope of earning a living (never mind a good living). if you let $x be twice the minimum wage in the us, you're talking on the order of $83k per year per team of four for salaries. add in another $20k for equipment, another $15k on travel, and $10k for tournament fees per team and you're up to $128k per year. suppose the league gives out $2mil in prizes over a season and has 20 teams, you're talking about $4.54mil per year to run the league. if your average sponsor kicks in $50k per year (which is about five half-page color ads in a major newspaper), you're going to need about 92 sponsors. convincing that many sponsors that $50k per year out of their advertising budgets spent on pro-gaming is going to be a hard sell until those gaming leagues generate several times that amount in revenue. i don't see that kind of serious revenue generation happening any time soon, and i suspect that most people who would consider sponsoring those leagues feel the same way as i do.
i think you're right -- i don't think the economics of pro-gaming as a spectator sport make any sense, other than advertising ("oooh, ($new_fps|$new_gaming_box|$new_mouse)... must.buy.now!"). i'm willing to wager that the vast majority of people (i.e. not 0.00001% that the slashdot crowd comprises) would rather play games than watch them, which means no ad revenue, which means no sponsors, which means no sustainable leagues that can support professional play.
step 1, like other posters have mentioned, is to get him to understand that he has a problem (ideally before he gets fired). this involves some sort of intervention. you'll want to plan it before hand for two reasons. firstly, you want to make sure that you've got the right mix of people talking to him; secondly, you as the interveners are going to need some practice. expect that you might get any reaction from "shit, you're right." to "you're just jealous" to overt hostility.
there are several tactics you can use to get your point across. if your friend doesn't let challenges go unanswered, challenge him to put himself in a situation where he can't play the game for a month or two. tell your friend how his addiction is affecting your relations with him (this is particularly relevant if his family or significant other is at the intervention). you need to be careful that when you do this, you're making "i" statements --- "i feel like _____ when you ditch me" rather than "you ditched me, jackass". you know your friend a lot better than i do, so you've got a better feel for what may or may not work well for him. while you're having the intervention, it's really important that you all make it clear that a) you're there for him, b) you're not judging him, c) you're going to help him pull through when he asks for your help, d) this is not something that will be discussed outside of the people in the room. your goal is to make the room a safe space (much easier said than done).
good luck -- you're about to go through a really rough patch.
suppose that m$ did find some kind of infringing ip in linux and they attempt to sue. i expect that if they were to win, they'd see a second slew of anti-trust lawsuits making the claim that microsoft's enforcement of certain patents demonstrably stifles innovation and generally screws the public because it is now impossible to work on $package. personally, i think that companies which have been convicted of anti-trust violations need to have some limits on their ability to enforce their ip rights as a way of offsetting the damage they did. just a thought...
The charges arise from Heller's alleged disclosure two years ago of legal papers from the Los Angeles office of international law firm Jones Day, which represented Diebold at the time. Heller was under contract as a word processor at Jones Day.
The documents included legal memos from one Jones Day attorney to another regarding allegations by activists that Diebold had used uncertified voting systems in Alameda County elections beginning in 2002.
emphasis my own. the memos in question (at least according to tfa) make it pretty clear that they are discussing allegations that the voting machines weren't certified. they did not (as far as i can tell from the article) state that the voting machines were, in fact, not certified. big, big difference. in the best case, he's a fool for misunderstanding the content of the documents and living under the misapprehension that he's a whistleblower. in the worst case, he stole internal documents in an attempt to win his moment in the sun.
fox isn't enough.
i seriously doubt that you are a scientist of any calibre, because if you were you would realize that the following truths apply: 1) funding has to come from somewhere, it does not magically appear ex nihilo; 2) science is usually hard to do without some substantial capital expenses and a constant revenue of funding to pay for lab techs and folks doing the grunt work; 3) if you love what you do, you're excited about it and it's something that you want to share with people and help them get excited about too.
getting funding requires that you get the funding agency as excited about what it is that you do as you are, i.e. it requires that you be a "pop scientist". no pop scientist == no funding == no science.
impossible. dvorak is a scalar.
too bad there's that whole 5th amendment nonsense :-P
i believe that witten showed two classes of these theories to be equivalent. as i recall, it's an open question as to whether all classes are equivalent. kind of the physicist's version of the P?=NP problem as i recall (i am most certainly not a physicist, so feel free to take anything i say with a suitably large grain of salt)
but patents (like guns) are not a priori evil. if they really do bother you so much, why don't you do something about it by becoming a patent examiner with a clue?
false. consider new amsterdam, aka new york.
... i imagine the dod may have a thing or two to say about this, since wireless networks are a perennial favorite sbir / sttr topic. in fact, i could see some really nasty fallout for boeing from this if they use their patent to stifle small business innovation in the areas that the dod cares about.
... they're getting one salary for essentially two (or more) full time jobs. at least the poor fools working at E.A. (unless E.A. has changed...)
how exactly using linux in violation of the gpl is a violation of sarbanes oxley? the article does an awful lot of handwaving but doesn't actually explain any of the hows or whys.
i'll have to read again, but it looks like this is f/oss trying its hand at the fud game.
when launching a new product, make sure there's enough to go around. even if said product is nominally produced by a bunch of right bastards.
webcams under linux break into roughly two camps: firewire (aka DCAM or IIDC) cameras and anything that can be handled by V4L or V4L2. which software you use depends very much on what camera you have (sucks, but that's how it is).
if you have an iidc camera (a la apple's isight, an orange micro ibot, etc.) then you really should be using coriander which can be compiled with support for ftp upload of the images. you can also use coriander to set things up for using ffmpeg (see below).
otherwise, i'd suggest looking into using ffserver and ffmpeg. when compiled correctly, it can handle both dv, iddc/dcam, and v4l cameras. i've had rather good results with them, and they're a lot cheaper (i.e. free) than an axis box.
if neither of these works for you, i wish you the best of luck -- you're probably going to need it.