Yes, at PSAS we're using 802.11b for telemetry, so 2.4GHz video would be bad. Besides, good cheap off-the-shelf gear is available for the 1.277GHz amateur TV band and it propagates well. That's what we went with this time around. We have used the audio channel with a modem as a telemetry channel in the past: this time we have an overlay board.
Here's a page with some onboard video from our April 1999 launch. Be gentle to our server:-).
Allow me to recommend an article from Annals of Improbable Research, most easily available in one of their "Best Of" collections:
David P. Cann and Phillip Pruna
Xerox Enlargement Microscopy
Annals of Improbable Research (1:2), March/April 1995
This wonderful article describes how to image down to the level of single atoms or even subatomic particles, using nothing more than an ordinary photocopier!
Too bad the film-scanner folks missed this: could have saved themselves a lot of work.
However, this lawsuit [against McDonald's, over a hot coffee spill] was, IMHO, stupid, because of the sum that was paid out (can't remember the exact figure, but it was huge).
All folks ever remember is the initial judgement ($2.7M here), and they almost never hear the facts of the case. (McDonald's knew for a long time their coffee was incredibly dangerously hot and refused to do anything about it. They also refused to compensate Plaintiff for her medical costs, which is why she chose to sue.) Lectlaw describes the facts of the McDonald's coffee case.
Note that the judgement was reduced by 5x on appeal: this is the norm. But especially note that to avoid another round of appeals and pay her medical bills (reference here), the plaintiff accepted a sealed out-of-court settlement. We will never know the amount, but I would guess offhand that it mostly covered her substantial medical and legal costs.
The American justice system does occasionally run amok. But if the story is that some ordinary individual or small business has beaten up a multinational corporation in court, you can usually bet that justice was seriously on the side of the little guy.
(Flamebait: And I don't know why he's talking about "his" genre. The Callahan books aren't SF; they're Chicken Soup for the Geek's Soul.)
Huh? OK, I'll flame. Spider Robinson has written a couple of things other than the Callahan series: here's the first bibliography Google hit that looked promising. His novels and short stories are almost all unarguably science fiction. If you're going to comment on Robinson's qualifications, you might want to find out about his work first.
Heh. Sadly, you're too sane to play this game. "Copyright shall not exceed 10 years"? I'd say one year, maximum. (10 years actually strikes me as a semi-reasonable compromise.) "All campaign donations from individuals"? Add "and no individual citizen may donate more than $100/year total to all political causes" and I'm in. (Remember, the fact that the bill won't pass constitutional muster is meaningless: you never expect to see it enacted anyhow.) "Mandatory minimum" (which is what you meant) "sentences for crimes by government employees"? Sure: such crimes should automatically become "A" felonies, with the mandatory minimum being 1 year jail time in a medium-security correctional facility.
The problem with the "big scare bill" tactic is that the kind of people willing to fully employ it tend to be, uh, what's the word I'm looking for? "Evil", maybe? "Evil always wins, because good is dumb." --Dark Helmet, Spaceballs
I find this sort of bill among the more reprehensible things our legislature does. This bill has no chance of passage, and the authors undoubtedly know this. Further, if it were to pass it would be the target of a million legal challenges.
The purpose of this bill is almost certainly to force a "compromise" bill that achieves the achievable portion of these effects. By staking out an extreme position, the sponsors paint opponents as staking out the opposite extreme, and suggest that the difference be split.
Honest congressfolk: don't give in. There is no honorable compromise here. P2P is just folks communicating via computer---to restrict the medium of the net is the beginning of the end of free speech in America and around the world. I would rather see our civil liberties go down fighting than turn to the dark side voluntarily.
Most of the examples of code generation I've seen around are basically kludges around insufficiently powerful high-level programming languages. Datatypes and representations should be able to be generic in the programming language; this enables directly coding at the desired level of abstraction. Check out ML or Haskell for one cut on these ideas.
That said, I've been doing CG for a long time:-). The M4 macro preprocessor is a fine tool for this task. See e.g. my student's M4 CG for XCB: a classic example, as we were stuck with C as the implementation language (indeed that was the whole point).
I think they missed a bit of a hacker market by deliberately disabling this possibility.
The typical concern with toys is that the hackers might propagate content inappropriate for children. I'm guessing that this sort of content control was a factor here.
I recently switched to Namazu as a full-text search engine for my nmh e-mail. The combination seems to do pretty much what I want. Searches are reasonably fast and flexible, indexing is incremental, and I can script things massively.
I'd still like to get a machine-learning classifier going someday, perhaps by adapting the fine-looking dbacl, but my current combination seems like a good first cut.
And if his punishment is harsh, so what? If he's found guilty, he's a criminal. He deserves whatever he gets at that point.
We all "deserve whatever we get" for one thing or another. By the grace of God most of us, most of the time, receive mercy instead of "perfect justice".
Are you seriously saying that I should have no mercy to spare for an 18-year-old who committed an entirely abstract and trivially easy crime? Are you seriously suggesting that anything short of jail time would make a travesty of justice for this person?
If so, I pity your family, and suspect that there are few who count themselves your friend. Count me as a "script kiddie sympathizer".
Gord Adams, a Toronto music engineer, set up the sound system at last summer's Harley-Davidson 100th Anniversary Open Road Tour, which featured artists such as Journey and April Wine. He witnessed autotuners being employed there by about half the major recording artists.
Heck yeah, I'll bet "If You See Kay" would really suck without pitch correction. WTF?!?
See our PSAM project site for a refereed paper evaluating several machine learning spam filtering techniques (although not specific filters). This site also contains large standardized corpora for evaluation. The paper contains a number of tips on evaluating ML spam filters.
The/.-referenced article has some good ideas about evaluation. I particularly liked the explicit discussion of the false positives. The recommendations at the end are excellent. On the other hand, the evaluation isn't across a broad or obviously representative corpus, many of the tests are a bit odd, the ROC tradeoffs are not discussed. In particular, the evaluation set for the tests did not include enough ham to be able to accurately estimate the false positive rate: consider what would happen to the precision estimates if 0.5 were added to each of the numbers in the false positive table.
Overall, though, this was an interesting evaluation, and I'm glad that the author published it.
Let's keep in mind that Moore's Law was more an observation than a predictive law of nature, despite how people treat it that way.
Not entirely. The folks designing FooCorp's next generation of e.g. chip fabs generally use Moore's Law to tell them where the competition will be by the time the fab is built: FooCorp needs to be competitive at that point in the future. Then the folks designing e.g. PDAs use Moore's Law to tell them what processor power, memory capacity, etc will be available to them by the time their next PDA is in production.
In short, Moore's Law is something of a self-fulfilling prophecy.
This sort of effect, BTW, is one of the big causes of the dot-bomb. Companies that had been counting on communication tech being there to support their business went under themselves when the telecom companies went under. Companies counting on building a business on middleware went under when their customers did. It only took a few dominos tipping over to start the whole cascade. Thankfully, this hasn't happened in the CPU, memory, or storage business. Yet.
Re:Multi-Channel motion control
on
Mirror, Mirror
·
· Score: 1
Servo motors are probably not necessary for this project: the precision doesn't have to be high, and cheap is important. There's lots of ways to use 10c motors: an optical interrupter wheel would be best, but it would probably be about as good to just cal the mirrors up somehow. Over time, they would drift: one would need some way of running them to their end stop so that they could be set right again.
Most commercial robot toys (Furby etc) dont bother with servo motors, for similar reasons.
Me (use mh) too. (BTW, mh is technically a MUA, not an MTA, but WTF.)
I've also scripted the living daylights out of it. I've written scripts to filter my incoming e-mail, handle multiple mail drops, mass e-mail homework grades, etc. This kind of scripting is really easy with MH, and would be basically impossible with any other MUA I've ever used: you'd either have the functionality already, or you'd never get it.
Many learning filters are Bayesian in character, but by no means all. Indeed, it is arguable that other approaches have advantages. For a good discussion and comparison of approaches, see the paper at my spam-filtering site
I have been the pround owner of a VPI Aires Scout [audiophile turntable] for almost a year now....the same music [classical, available on CD] is available on vinyl, and it's available for as little as a dollar.
Absolutely. With the Scout turntable + JMW tonearm at a mere $1600, you break even after only your 100th album!
I also find my vinyl listening session are less iritating on my ears and last longer.
You bet. Vinyl has the nice property that it reproduces high and low frequencies badly to begin with, and continuously degrades each time you play it rather than working perfectly for a while. The rapid loss of those irritating high frequencies can really make harsh music much more pleasant to listen to!
The problem with claiming Einstein as a misunderstood genius from outside the scientific establishment is that his ideas were widely and rapidly accepted by the scientific mainstream. Examine the famous 1905 volumes 17-18 of Annalen der Physik: many people feel that any of the four unrelated papers Einstein published in these volumes would have been sufficient to net him a Nobel Prize.
Clearly, special relativity was the most controversial of the four ideas, but it was taken seriously enough that immediate plans were made to test its predictions. It is true that there was much argument about the validity of special relativity, but this argument actually tended to be mostly among the less distinguished scientists and "science popularizers".
This whole line of development is in sharp contrast to Lynds, who as far as I know has not proposed a testable scientific theory that makes realistic predictions. If he were to do so on such an important subject as the flow of time, and if his theory made sense, I feel pretty confident that the theory would be widely publicized, and the tests quickly performed.
The font sizes, icon sizes, etc are all user-configurable in the stock KDE and GNOME environments. I have vision problems, and run 1600x1200 on a 19" display routinely with no problem.
<OFFTOPIC>I wish folks would at least spend 15 minutes investigating on their own before asking Slashdot. I also wish the editors would enforce this. Booting off a Knoppix CD would have answered the question in advance.</OFFTOPIC>
I'm a CS Prof, not a CEO/CIO. If I were the latter, I would be sending out the following letter:
Dear Competitor of My Company:
The Gartner Group has recently come out with a report that you should definitely be aware of [url]. Gartner recommends that you "go slow" with Linux. As your competitor, I urge you to seriously consider that approach.
We currently plan to ignore the Gartner Group advice, and move to Linux and related Free/Open software as quickly as possible. Please believe us when we say that the likely resulting legal problems should put us at a huge competitive disadvantage relative to your more cautious organization. Trust us when we say that the putative low cost, high quality, and ability to control risk attributed to the Linux platform is over-hyped. By avoiding Linux, you aren't missing much---honest.
We're about to make the mistake of using freely available source code maintained by dedicated teams of volunteer developers to power our enterprise. We encourage you, dear competitor, not to make the same mistake: we'd hate to see you miss the opportunity to beat us out.
Not to mention the fact that neither the numbers in the article or at dvdrhelp.com are particularly decisive as to format. 85-90% vs 90-95% compatibility is probably not the most important factor in the -R/+R decision, and the -RW/+RW compatibility data is inconclusive. I found the media cost data in the article more interesting: that's a major factor in my purchasing decision right now.
Slashdot Mirror
Yes, at PSAS we're using 802.11b for telemetry, so 2.4GHz video would be bad. Besides, good cheap off-the-shelf gear is available for the 1.277GHz amateur TV band and it propagates well. That's what we went with this time around. We have used the audio channel with a modem as a telemetry channel in the past: this time we have an overlay board.
Here's a page with some onboard video from our April 1999 launch. Be gentle to our server :-).
Allow me to recommend an article from Annals of Improbable Research, most easily available in one of their "Best Of" collections:
This wonderful article describes how to image down to the level of single atoms or even subatomic particles, using nothing more than an ordinary photocopier!Too bad the film-scanner folks missed this: could have saved themselves a lot of work.
However, this lawsuit [against McDonald's, over a hot coffee spill] was, IMHO, stupid, because of the sum that was paid out (can't remember the exact figure, but it was huge).
All folks ever remember is the initial judgement ($2.7M here), and they almost never hear the facts of the case. (McDonald's knew for a long time their coffee was incredibly dangerously hot and refused to do anything about it. They also refused to compensate Plaintiff for her medical costs, which is why she chose to sue.) Lectlaw describes the facts of the McDonald's coffee case.
Note that the judgement was reduced by 5x on appeal: this is the norm. But especially note that to avoid another round of appeals and pay her medical bills (reference here), the plaintiff accepted a sealed out-of-court settlement. We will never know the amount, but I would guess offhand that it mostly covered her substantial medical and legal costs.
The American justice system does occasionally run amok. But if the story is that some ordinary individual or small business has beaten up a multinational corporation in court, you can usually bet that justice was seriously on the side of the little guy.
(Flamebait: And I don't know why he's talking about "his" genre. The Callahan books aren't SF; they're Chicken Soup for the Geek's Soul.)
Huh? OK, I'll flame. Spider Robinson has written a couple of things other than the Callahan series: here's the first bibliography Google hit that looked promising. His novels and short stories are almost all unarguably science fiction. If you're going to comment on Robinson's qualifications, you might want to find out about his work first.
Heh. Sadly, you're too sane to play this game. "Copyright shall not exceed 10 years"? I'd say one year, maximum. (10 years actually strikes me as a semi-reasonable compromise.) "All campaign donations from individuals"? Add "and no individual citizen may donate more than $100/year total to all political causes" and I'm in. (Remember, the fact that the bill won't pass constitutional muster is meaningless: you never expect to see it enacted anyhow.) "Mandatory minimum" (which is what you meant) "sentences for crimes by government employees"? Sure: such crimes should automatically become "A" felonies, with the mandatory minimum being 1 year jail time in a medium-security correctional facility.
The problem with the "big scare bill" tactic is that the kind of people willing to fully employ it tend to be, uh, what's the word I'm looking for? "Evil", maybe? "Evil always wins, because good is dumb." --Dark Helmet, Spaceballs
I find this sort of bill among the more reprehensible things our legislature does. This bill has no chance of passage, and the authors undoubtedly know this. Further, if it were to pass it would be the target of a million legal challenges.
The purpose of this bill is almost certainly to force a "compromise" bill that achieves the achievable portion of these effects. By staking out an extreme position, the sponsors paint opponents as staking out the opposite extreme, and suggest that the difference be split.
Honest congressfolk: don't give in. There is no honorable compromise here. P2P is just folks communicating via computer---to restrict the medium of the net is the beginning of the end of free speech in America and around the world. I would rather see our civil liberties go down fighting than turn to the dark side voluntarily.
Most of the examples of code generation I've seen around are basically kludges around insufficiently powerful high-level programming languages. Datatypes and representations should be able to be generic in the programming language; this enables directly coding at the desired level of abstraction. Check out ML or Haskell for one cut on these ideas.
That said, I've been doing CG for a long time :-). The M4 macro preprocessor is a fine tool for this task. See e.g. my student's M4 CG for XCB: a classic example, as we were stuck with C as the implementation language (indeed that was the whole point).
I think they missed a bit of a hacker market by deliberately disabling this possibility.
The typical concern with toys is that the hackers might propagate content inappropriate for children. I'm guessing that this sort of content control was a factor here.
I recently switched to Namazu as a full-text search engine for my nmh e-mail. The combination seems to do pretty much what I want. Searches are reasonably fast and flexible, indexing is incremental, and I can script things massively.
I'd still like to get a machine-learning classifier going someday, perhaps by adapting the fine-looking dbacl, but my current combination seems like a good first cut.
And if his punishment is harsh, so what? If he's found guilty, he's a criminal. He deserves whatever he gets at that point.
We all "deserve whatever we get" for one thing or another. By the grace of God most of us, most of the time, receive mercy instead of "perfect justice".
Are you seriously saying that I should have no mercy to spare for an 18-year-old who committed an entirely abstract and trivially easy crime? Are you seriously suggesting that anything short of jail time would make a travesty of justice for this person?
If so, I pity your family, and suspect that there are few who count themselves your friend. Count me as a "script kiddie sympathizer".
I think I've just been trolled. Oh well.
Its main duties will be atmospheric chemistry, systems biology, catalysis and materials science.
IOW, studies in dealing with the power consumption of 2000 Itanium processors.
Gord Adams, a Toronto music engineer, set up the sound system at last summer's Harley-Davidson 100th Anniversary Open Road Tour, which featured artists such as Journey and April Wine. He witnessed autotuners being employed there by about half the major recording artists.
Heck yeah, I'll bet "If You See Kay" would really suck without pitch correction. WTF?!?
Keith Packard's midiplay may do the right thing if you can figure out how to obtain and build it. I don't recall. I know it displays lyrics.
See our PSAM project site for a refereed paper evaluating several machine learning spam filtering techniques (although not specific filters). This site also contains large standardized corpora for evaluation. The paper contains a number of tips on evaluating ML spam filters.
The /.-referenced article has some good ideas about evaluation. I particularly liked the explicit discussion of the false positives. The recommendations at the end are excellent. On the other hand, the evaluation isn't across a broad or obviously representative corpus, many of the tests are a bit odd, the ROC tradeoffs are not discussed. In particular, the evaluation set for the tests did not include enough ham to be able to accurately estimate the false positive rate: consider what would happen to the precision estimates if 0.5 were added to each of the numbers in the false positive table.
Overall, though, this was an interesting evaluation, and I'm glad that the author published it.
Let's keep in mind that Moore's Law was more an observation than a predictive law of nature, despite how people treat it that way.
Not entirely. The folks designing FooCorp's next generation of e.g. chip fabs generally use Moore's Law to tell them where the competition will be by the time the fab is built: FooCorp needs to be competitive at that point in the future. Then the folks designing e.g. PDAs use Moore's Law to tell them what processor power, memory capacity, etc will be available to them by the time their next PDA is in production.
In short, Moore's Law is something of a self-fulfilling prophecy.
This sort of effect, BTW, is one of the big causes of the dot-bomb. Companies that had been counting on communication tech being there to support their business went under themselves when the telecom companies went under. Companies counting on building a business on middleware went under when their customers did. It only took a few dominos tipping over to start the whole cascade. Thankfully, this hasn't happened in the CPU, memory, or storage business. Yet.
Servo motors are probably not necessary for this project: the precision doesn't have to be high, and cheap is important. There's lots of ways to use 10c motors: an optical interrupter wheel would be best, but it would probably be about as good to just cal the mirrors up somehow. Over time, they would drift: one would need some way of running them to their end stop so that they could be set right again.
Most commercial robot toys (Furby etc) dont bother with servo motors, for similar reasons.
Me (use mh) too. (BTW, mh is technically a MUA, not an MTA, but WTF.)
I've also scripted the living daylights out of it. I've written scripts to filter my incoming e-mail, handle multiple mail drops, mass e-mail homework grades, etc. This kind of scripting is really easy with MH, and would be basically impossible with any other MUA I've ever used: you'd either have the functionality already, or you'd never get it.
Many learning filters are Bayesian in character, but by no means all. Indeed, it is arguable that other approaches have advantages. For a good discussion and comparison of approaches, see the paper at my spam-filtering site
.Please don't say things in the headlines like "Metroworks suit". I'm thinking to myself, first SCO, and now this?
(It's spelled "suite", and pronounced sweet, not soot.)
I have been the pround owner of a VPI Aires Scout [audiophile turntable] for almost a year now. ...the same music [classical, available on CD] is available on vinyl, and it's available for as little as a dollar.
Absolutely. With the Scout turntable + JMW tonearm at a mere $1600, you break even after only your 100th album!
I also find my vinyl listening session are less iritating on my ears and last longer.
You bet. Vinyl has the nice property that it reproduces high and low frequencies badly to begin with, and continuously degrades each time you play it rather than working perfectly for a while. The rapid loss of those irritating high frequencies can really make harsh music much more pleasant to listen to!
The problem with claiming Einstein as a misunderstood genius from outside the scientific establishment is that his ideas were widely and rapidly accepted by the scientific mainstream. Examine the famous 1905 volumes 17-18 of Annalen der Physik: many people feel that any of the four unrelated papers Einstein published in these volumes would have been sufficient to net him a Nobel Prize.
Clearly, special relativity was the most controversial of the four ideas, but it was taken seriously enough that immediate plans were made to test its predictions. It is true that there was much argument about the validity of special relativity, but this argument actually tended to be mostly among the less distinguished scientists and "science popularizers".
This whole line of development is in sharp contrast to Lynds, who as far as I know has not proposed a testable scientific theory that makes realistic predictions. If he were to do so on such an important subject as the flow of time, and if his theory made sense, I feel pretty confident that the theory would be widely publicized, and the tests quickly performed.
Especially right after a story about snopes.com.
The font sizes, icon sizes, etc are all user-configurable in the stock KDE and GNOME environments. I have vision problems, and run 1600x1200 on a 19" display routinely with no problem.
<OFFTOPIC>I wish folks would at least spend 15 minutes investigating on their own before asking Slashdot. I also wish the editors would enforce this. Booting off a Knoppix CD would have answered the question in advance.</OFFTOPIC>
I'm a CS Prof, not a CEO/CIO. If I were the latter, I would be sending out the following letter:
Not to mention the fact that neither the numbers in the article or at dvdrhelp.com are particularly decisive as to format. 85-90% vs 90-95% compatibility is probably not the most important factor in the -R/+R decision, and the -RW/+RW compatibility data is inconclusive. I found the media cost data in the article more interesting: that's a major factor in my purchasing decision right now.