Nowadays, readers assume that "golden boys and girls" are basically "princes and princesses," in the metaphorical sense...
Interesting stuff! I've actually never read or seen the play, but I'm familiar with that passage (it was set to music by Loreena McKennitt at one point, and I could be described as a drooling fanboy of hers.) I certainly had no idea there was any ambiguity in those lines.
(Parenthetically, it'd be interesting to know the meaning behind another line: "The sceptre, learning, physic must all follow this, and come to dust". A simple equation of kings, sages, and men of medicine? If so, why the weird verb tense in "learning"? Just to pass on my confusion, I stuck it on a plaque in the middle of a dungeon in Eye of the Beholder III, but nobody ever recognized it or commented on it as far as I could tell.:)
Anyway, you have a point, but it may actually serve to back up my own in this case. Regardless of the literal meaning behind the terms he used, it's obvious that Shakespeare was saying that we're all equal in the eyes of Death. He may have found a clever way to express the same idea in two different ways recognizable as such by contemporary audiences who knew about the dandelion business, but the effect is the same: the communication (not just the expression, but the transmission and reception) of the idea that death is equally inevitable to all.
Subtlety is usually wasted on machines, so I'm not sure I can think of a good use for double entendre on the programming side of things. But I suppose when you're writing for people it can make the difference between being dismissed as a wannabee and being admired for your cleverness for the next 400 years.:) Humans are susceptible to a broader range of "effects" than machines, which is why natural languages seem so diverse and ambiguous compared to machine-oriented languages. But I still say the differences are more by degree than by design.
Unfortunately, natural languages have almost nothing in common with computer languages. Computer languages are for the most part 1:1 codes - the same command means the same thing in whatever context it appears in a particular language. Natural languages are not codes; an idiom means different things in different contexts. That's part of the problem comparing the two.
Not that it has anything to do with grade inflation, but that's an overbroad assertion on your part. You're focusing on details of meaning and context when you contrast programming languages with human languages... but neither good programmers nor good writers are concerned with meaning, when you get right down to it. They're concerned with effect. How can I cause my target machine to process the most information in the fewest machine cycles? There's always an endless variety of semantically-interchangeable expressions available to the programmer, no less than to the writer. But they're not all equally efficient. Practitioners of both professions are judged by the efficiency of the expressions they choose.
Whether the language is C++ or English, you're issuing instructions to one or more processors: Think this way. There are compelling parallels between the mechanisms of natural languages and computer languages. When Strunk & White tells us to "Keep related words together," is this any different from what an x86 assembly programmer does to optimize register usage? When we're told to "Avoid fancy words," it's hard not to recall Abrash's admonition against microcoded instructions -- those that seem erudite to the novice but consume far more machine cycles on modern processors than the prosaic RISC-like instruction sequences they replace. Sending the reader to the dictionary to look up one of Strunk's famous "twenty-dollar words" is no less a crime against efficiency than an unnecessary L2 cache miss on an Athlon XP!
I know plenty of humanists who would stupidly assume that programming doesn't require any brains; after all, "it's just writing down instructions for machines. What's so hard about that."
Be careful that you don't make a similar mistake by confusing the meaning of individual terms used by an author (or a programmer) with the effect she's trying to achieve with her text. If context-dependent idioms are in fact exclusive to natural languages, it's only because the designer of an artificial language -- whether intended for use by humans or not -- would probably be loath to include them. There may be a 1:1 mapping of terms to meanings in a typical programming language, but that fact alone does nothing to constrain the programmer's expression space vis-a-vis the writer's.
It's really the same job, except that the programmer is blessed with an entirely-apolitical audience.
CRTs are still better than most of the crappy LCDs sold at CompUSA, but they have been soundly thrashed by the higher-end LCDs -- the kind that sell for around 3X the price per square inch of a good CRT display.
What Sony's doing here is acknowledging that customers who are after a high-quality display are probably also looking for a large display. There's no point making an expensive, high-quality, small CRT monitor anymore, unless you're selling video reference monitors (an entirely different market).
... although I don't tend to agree with Ed's assertion that 802.11's core technology is some sort of deep voodoo that could never have come out of anyone's ham shack.
Nowadays, the 'interesting' part of the RF world is UWB (ultra-wideband wireless) technology. UWB occupies the embryonic-technology niche that spread-spectrum occupied 20 years ago. And if I can just scrape together whatever Picosecond Pulse Labs wants for their new 100-GHz samplers and the time to fool around with them, well... there ya go.
(begin quote of parent message from Ed Hare)
> But, no, instead what I see here is the ham > radio organizations trying (hopelessly, I might > add) to kill or restrict one of the best things > to happen to public communications since > Marconi flew a kite.
ARRL is not trying to kill or restrict the best thing to happen since Marconi, it is trying to use the 802.11b technology within amateur radio. Although I am sure that amateurs will make some progress with pushing the state of the art, the concept that the 802.11b standard could have been developed by a handful of experimenters is ludicrous. It took an industry millions and millions of dollars to make it happen.
802.11 is really not even very good spread spectrum. One of the future goals of the HSMM group is to develop ways that amateurs can experiment with non-standard spread spectrum and possibly make some real improvements along the way. Please don't feel that these amateurs see the Part 15 users as the enemy, because they don't. Some are even professionals in the field.
Just as there is overlap in our spectrum, there is also a natural separation, and the goals and intent of unlicensed operation are not necessarily the same as Part 97. Those that want to build a network so that anyone in their community can access xxx.com should do so under Part 97. Those that want to experiment with protocols and equipment and antennas and amplification should do so under Part 97, with the clear understanding that the result can only be used by licensed amateurs and the prohibition against business use and the content of xxx.com are clearly spelled out in Part 97. I expect that, just as is happening now, no small number of the Part 15 users will continue to be hams, and those who want to do more experimentation than allowed under Part 15 will become licensed under Part 97. The HSMM group is sure hoping that the latter happens.
Those who are interested in amateur radio and its potential can vistit the ARRL Web site at http://www.arrl.org.
I see some of your points, I guess, but I'm a little unclear as to how "This does not allow for any content provider to enter the market without an affiliation." Any site would be free to set up its own Salon-like subscription model, or continue begging on the streetcorner for PayPal/Amazon donations just like many do now. Who is doing the "allowing" that you're referring to here?
How else do you suggest that small sites overcome the obscurity and inertia that limits their financial models these days? And how am I supposed to support my 10 favorite sites while spending less than $500 a year? Those aren't rhetorical questions -- I'd really be curious if there are any answers besides syndication.
I don't see how the RIAA is "not a syndicate." Which definition are you using?
Well, you and some other people have raised a (very valid) point in this thread: syndicates that start out representing artists often don't end up that way.
I don't think many genuine artists (as opposed to synth-pop replicants) would argue with a straight face that the RIAA actually represents them. Today, the RIAA is purely a legislative mouthpiece for the music industry's corporate interests, one that began as an innocuous-sounding standards organization (ever heard of the "RIAA equalization curve" for phonograph records?).
In the case of a Web content syndicate, such an organization's reach would (ideally) be contractually constrained from the very first click. What I'd like to see is simply a way for ISPs to put a list of check boxes on their member pages, with labels like Slashdot, Kuro5hin, Salon, BoingBoing, Joe's Lame-Ass Weblog, whatever. As a subscriber, I check the boxes for the sites I want to read. A "total subscription amount" field above the boxes would go up by $10.00 for a year of Salon (which employs professional journalists and editors), $2.50 for a year of Slashdot (which does not), and $0.25 for BoingBoing (which is primarily a hobbyist weblog with no significant costs or original content to pay for). The final amount would be added to the ISP bill I already pay. Most of the revenue would then be divided proportionally among the content providers, with the ISP keeping the rest (and competing with other ISPs to offer me the most content for my money).
Similarly, when Joe and his crappy weblog join the syndicate, he's asked "What do you want to be paid for your content?" If Joe says "I'd like to be paid $0.05 per viewer-year," that's what the syndicate charges me to subscribe. If he says "I'd like to be paid $50.00 per viewer-year," that's fine too, from the syndicate's point of view. It doesn't cost anything for them to add Joe's check box to their web page, after all. In both cases, the users are the ones who decide what sites are worth paying for based on their operators' asking prices. That's already a pretty big departure from the way most big, bad, evil syndicates work, right?
At the end of the day, the big, bad, evil syndicate should really be nothing more than a clearinghouse that keeps the good sites like kuro5hin.org from having to beg on the street corner for financial patronage. Right now -- excluding paid text ads -- K5's business model relies on selling a small number of relatively-expensive memberships to a (necessarily) small number of readers. That sucks. It would be much better if K5 could lower their subscription prices by 10X and pick up 20X the subscriber base as a result. But not only is that too expensive due to financial-processing overhead, it's just too much of a pain in the neck for large numbers of users to bother with. A useful syndication model would (1) make life easy for both consumers and providers; (2) result in revenue streams that are much finer-grained and more intimately tied to the value (or lack thereof) of the content being provided; and (3) stick to its core mission rather than trying to become a power-drunk cartel like the RIAA and MPAA.
Maybe the latter course is inevitable, assuming the syndication concept succeeds spectacularly enough. But we need to at least try. Currently, Web content providers and have no mojo to abuse in the first place, which is no better for us all in the long run than the appearance and domination of the next RIAA-like organization. Either way, we, the consumers of content, risk losing out on some good stuff.
... is the right way to address this, not micropayments (which will never be economically viable without a syndicate-style clearinghouse that insulates the participants from contact with actual financial institutions).
Paying $50/year to subscribe to a site sounds like a lot of money, because it is. But if I could pay $100/year for ad-free access to all of my favorite sites on an a la carte basis, it'd sound like a bargain. That's where commercial Web content will have to go eventually. I can't imagine any alternatives that will meet the needs of both consumers and site operators in the long term.
It'd be nice if one or more of the major ISPs would offer a pilot program along these lines. Not necessarily MSN or TW/AOL; even someone like Speakeasy could make a credible effort at syndicating content for their members, IMHO.
1. Make nitroglycerin from a recipe written by someone who can't spell the names of the chemicals ("sulferic acid") required. 2. Light fuse. 3. Get away. 4. (Your heirs) profit!
"Engineering is the only profession where your value to the company goes down the older you get."
... is that in our line of work, experience doesn't count for as much as it does in other fields of endeavor. That's the sign of a rapidly-growing and (yes) immature industry where progress often takes place via change and mutation rather than simple growth.
But that line of reasoning often turns into a psychological crutch for chronic whiners. How many posts on Slashdot read something like, "Dammit, I know Logo, BASIC, Pascal, VB, FORTRAN, assembly, Java, C++, and C#... and I still got laid off!" Sure, but how good were you at solving problems? Should an auto shop manager be impressed when a job applicant claims to have worked on Pintos, Novas, Malibus, Mustangs, Explorers, Cavaliers, and Excursions? How many of those cars drove away from the applicant's garage bay with their lugnuts cross-threaded?
Quality software engineering is more than a resume full of hip languages and buzzwords from the Gamma book. The best software engineering is usually done by people who got into the business because computers seemed like a really powerful and enjoyable way to solve engineering or (in the games biz) aesthetic problems. Those folks -- not the language lawyers, tool fetishists, and epicene gnomes of Unix who still have their home page set to schwab.com -- are the ones who tend to have the best answers to the question, "OK, why should I hire you?"
Good point, if that's actually when the artifacts first showed up. (I actually never played through any of the Ultimas after V, so what I wrote shouldn't be taken as canon.)
Actually the 'O' was just a stylized sphere interposed between the cube ('E') and pyramid ('A') symbols that comprised the EA logo. It made for a memorable if misleading trademark.
The cube, sphere, and pyramid objects in Ultima VIII were originally supposed to be evil artifacts in Ultima VI, if I remember correctly. Richard Garriott had a bit of a grudge against Trip Hawkins ("Pirt Snikwah") back in those days, as Origin was less than thrilled with their treatment as an EA affiliated label in the late Eighties. Richard had painstakingly planned a whole boatload of EA-as-the-embodiment-of-ultimate-evil allusions for the Ultimas beyond V, but EA skillfully dodged his wrath by offering to buy the company.:)
Because we spend $I_don't_know_how_much_money_but_it's_a_boatload on bandwidth giving away the RAD Video Tools for free.:)
I'm not involved with that particular product and can't rattle any stats off the top of my head, but I remember being shocked to hear how much download traffic our site sees per month. The donation button shows up only when you're downloading a freebie like the video toolset that's of primary interest to (non-paying) end users.
Re:Spielberg Over the Hill?
on
Taken?
·
· Score: 2
I made the same mistake ("aliens? WTF?!?") the first time I saw AI.
If there's a time on every project schedule to shoot the engineer, and a time on every movie schedule to shoot the director, there was DEFINITELY an overlooked time on AI at which the VFX lead should have been shot. Those future-mecha models made Jar Jar Binks look like something from a lost da Vinci sketch. My opinion of the film improved immeasurably when someone clued me in that the "aliens" were descendents of the mecha.
My understanding from surfing RadioShack Sucks is that their salesdroids would actually be penalized financially, or even fired altogether, for failing to obtain some arbitrary percentage of customer names and addresses. Seems like the quota was something on the order of 80-90% "compliance."
Between local stores like Active Electronics, the utterly-amazing variety of electronic parts on eBay and topnotch mail-order houses like Digi-Key, Jameco, and Mouser, it's pretty darned rare for me to set foot in a RatShit store these days. Their 1/4-watt resistor assortments are still a killer deal, though.
Interesting info. So the van outside watches for leakage of the amplified 90-150 kHz signal from the baseband circuitry, huh? Hard to believe there's enough radiation to pick up... you need a fairly-efficient antenna to 'broadcast' signals at that frequency, not just stray bits and pieces of wire. TEMPEST interception works because deflection yokes make nice loop antennas, but I wouldn't think this carrier would make it through to the deflection circuitry (at least not without causing horrible artifacts in the picture).
But those signals are very weak, especially when going 70 mph. It doesn't make sense to me; seems as though it wouldn't be worth the trouble.
Couple of points in their favor, though: they know exactly where to listen, down to the nearest 1 kHz or so given the stability of modern synthesized LOs. That means they can use a very narrow filter to detect the desired signal, and overall S/N ratio improves with 10 * log(bandwidth). Compare the 200 kHz-wide IF an ordinary FM radio uses to the 5-kHz detection bandwidth a billboard receiver might use. That's like picking up an extra 16 dB of signal strength for free.
Also, the LO signal isn't necessarily that weak to begin with... probably on the order of a milliwatt at the point where it's injected into the mixer. Many of the cheaper FM "wireless broadcasters" sold by outfits like Radio Shack put out less power than that.
I imagine that the power at the antenna might be as high as -60 dBm or greater, which is workable, but that's a wild guess. I'll have to drag a spectrum analyzer into the garage tonight and see what kind of radiated field strength I can pick up from my cars' antennas.
I don't follow your point about "interference produced by the amplifier (in the) car's stereo." That's an audio amp.
have you actually tried to detect that signal outside a vehicle? ok now do it to a MOVING TARGET.
I could tell you, but then I'd have to... aw, never mind.
and what you are talking about is not the case.. the BBC transmits a subcarrier with a tone on it that is easily detectable. It's the same detection scheme used by american cable TV companies to snif out people stealing cable tv. It's a simple device and putting the subcarrier there makes it air tight in court.. trying to say that "we detected what channel your tv is tuned to doesnt work in court... saying we detected our special signal we transmit to catch them.... does.
It's likely that the technique used in the UK has changed over the years, but it's also worth pointing out that what you're describing makes no sense whatsoever.:) If the BBC is transmitting the 'subcarrier', what does a TV set do to the 'subcarrier' to make it detectable from outside the house? If you're talking about a TV receiving a signal off the air (as opposed to through a CATV feed that can be TDR'ed to detect connections), the only way you can detect it is by listening for either an LO signal or a (Tempest-style) deflection signal. Can you nail down some more details about the subcarrier-detection scheme you're referring to?
The reason why LO signal leakage absolutely does occur in any practical radio receiver is that there's no such thing as a perfectly-directional RF amplifier.
The local oscillator signal is generated and mixed with the incoming signal at a very early stage, relativley close to the antenna. A long way before the limiter/discriminator stage, in other words, and in a portion of the receiver that's architecturally identical regardless of the mode being demodulated. Different IF frequencies are used by FM, AM, and TV broadcast receivers, but the front-end topology is the same: one or more RF amplifier stages feed the incoming signal into the first mixer, which is also fed by the first local oscillator. This signal chain is not perfectly unidirectional.
RF amplifiers are usually characterized by a handful of key parameters -- their noise figure, their ability to operate properly in the presence of strong signals as well as weak ones, and their forward gain (also known as 'S21'). But any real RF amplifier will also have a reverse gain parameter ('S12') describing the attenuation a signal applied to its output undergoes on the way back through to the input port. With the untuned, resitively-matched gain blocks popular in RF work these days, the S12 parameter is often just barely better than S21. The same is true of the mixer itself; it has a decidedly-finite port isolation spec that describes how much of the signal at the LO port will leak back through the RF port. The cheap unbalanced mixers used in consumer-grade receivers aren't exactly state-of-the-art in this department.
The bottom line is that significant, detectable LO leakage DOES take place through the antenna regardless of how well-shielded the rest of the receiver is. It's entirely believable that a stationary billboard receiver could be designed and optimized to look for LO leakage from passing cars. Remember that even the cheapest radios today use highly-stable phase-locked LOs, so the required signal-detection bandwidth at the billboard is very small indeed.
Military receivers often have two or even three RF stages preceding the mixer, not to maximize front-end gain but rather to cut down on LO leakage to make the receiver (and its user) harder for the enemy to direction-find. This practice dates back at least to the BC-series receivers used in WWII. Not much new under the sun here.
Slashdot Mirror
Why, exactly, did the EFF choose not to appeal the 2nd Circuit Court decision against 2600 Magazine in the DeCSS-linking case?
Nowadays, readers assume that "golden boys and girls" are basically "princes and princesses," in the metaphorical sense...
:)
:) Humans are susceptible to a broader range of "effects" than machines, which is why natural languages seem so diverse and ambiguous compared to machine-oriented languages. But I still say the differences are more by degree than by design.
Interesting stuff! I've actually never read or seen the play, but I'm familiar with that passage (it was set to music by Loreena McKennitt at one point, and I could be described as a drooling fanboy of hers.) I certainly had no idea there was any ambiguity in those lines.
(Parenthetically, it'd be interesting to know the meaning behind another line: "The sceptre, learning, physic must all follow this, and come to dust". A simple equation of kings, sages, and men of medicine? If so, why the weird verb tense in "learning"? Just to pass on my confusion, I stuck it on a plaque in the middle of a dungeon in Eye of the Beholder III, but nobody ever recognized it or commented on it as far as I could tell.
Anyway, you have a point, but it may actually serve to back up my own in this case. Regardless of the literal meaning behind the terms he used, it's obvious that Shakespeare was saying that we're all equal in the eyes of Death. He may have found a clever way to express the same idea in two different ways recognizable as such by contemporary audiences who knew about the dandelion business, but the effect is the same: the communication (not just the expression, but the transmission and reception) of the idea that death is equally inevitable to all.
Subtlety is usually wasted on machines, so I'm not sure I can think of a good use for double entendre on the programming side of things. But I suppose when you're writing for people it can make the difference between being dismissed as a wannabee and being admired for your cleverness for the next 400 years.
Unfortunately, natural languages have almost nothing in common with computer languages. Computer languages are for the most part 1:1 codes - the same command means the same thing in whatever context it appears in a particular language. Natural languages are not codes; an idiom means different things in different contexts. That's part of the problem comparing the two.
Not that it has anything to do with grade inflation, but that's an overbroad assertion on your part. You're focusing on details of meaning and context when you contrast programming languages with human languages... but neither good programmers nor good writers are concerned with meaning, when you get right down to it. They're concerned with effect. How can I cause my target machine to process the most information in the fewest machine cycles? There's always an endless variety of semantically-interchangeable expressions available to the programmer, no less than to the writer. But they're not all equally efficient. Practitioners of both professions are judged by the efficiency of the expressions they choose.
Whether the language is C++ or English, you're issuing instructions to one or more processors: Think this way. There are compelling parallels between the mechanisms of natural languages and computer languages. When Strunk & White tells us to "Keep related words together," is this any different from what an x86 assembly programmer does to optimize register usage? When we're told to "Avoid fancy words," it's hard not to recall Abrash's admonition against microcoded instructions -- those that seem erudite to the novice but consume far more machine cycles on modern processors than the prosaic RISC-like instruction sequences they replace. Sending the reader to the dictionary to look up one of Strunk's famous "twenty-dollar words" is no less a crime against efficiency than an unnecessary L2 cache miss on an Athlon XP!
I know plenty of humanists who would stupidly assume that programming doesn't require any brains; after all, "it's just writing down instructions for machines. What's so hard about that."
Be careful that you don't make a similar mistake by confusing the meaning of individual terms used by an author (or a programmer) with the effect she's trying to achieve with her text. If context-dependent idioms are in fact exclusive to natural languages, it's only because the designer of an artificial language -- whether intended for use by humans or not -- would probably be loath to include them. There may be a 1:1 mapping of terms to meanings in a typical programming language, but that fact alone does nothing to constrain the programmer's expression space vis-a-vis the writer's.
It's really the same job, except that the programmer is blessed with an entirely-apolitical audience.
No, I'm pretty sure they just licensed the franchise from those wacky Swiss.
That arguments would be more convincing if Big Pharma companies weren't posting such huge profits.
Your argument would be more convincing if Big Pharma companies, primarily in the US, weren't the ones developing all the new drugs.
The last drug of any significance to come out of a socialist-paradise state was probably LSD.
CRTs are still better than most of the crappy LCDs sold at CompUSA, but they have been soundly thrashed by the higher-end LCDs -- the kind that sell for around 3X the price per square inch of a good CRT display.
What Sony's doing here is acknowledging that customers who are after a high-quality display are probably also looking for a large display. There's no point making an expensive, high-quality, small CRT monitor anymore, unless you're selling video reference monitors (an entirely different market).
... although I don't tend to agree with Ed's assertion that 802.11's core technology is some sort of deep voodoo that could never have come out of anyone's ham shack.
Nowadays, the 'interesting' part of the RF world is UWB (ultra-wideband wireless) technology. UWB occupies the embryonic-technology niche that spread-spectrum occupied 20 years ago. And if I can just scrape together whatever Picosecond Pulse Labs wants for their new 100-GHz samplers and the time to fool around with them, well... there ya go.
(begin quote of parent message from Ed Hare)
> But, no, instead what I see here is the ham
> radio organizations trying (hopelessly, I might > add) to kill or restrict one of the best things > to happen to public communications since
> Marconi flew a kite.
ARRL is not trying to kill or restrict the best thing to happen since Marconi, it is trying to use the 802.11b technology within amateur radio. Although I am sure that amateurs will make some progress with pushing the state of the art, the concept that the 802.11b standard could have been developed by a handful of experimenters is ludicrous. It took an industry millions and millions of dollars to make it happen.
802.11 is really not even very good spread spectrum. One of the future goals of the HSMM group is to develop ways that amateurs can experiment with non-standard spread spectrum and possibly make some real improvements along the way. Please don't feel that these amateurs see the Part 15 users as the enemy, because they don't. Some are even professionals in the field.
Just as there is overlap in our spectrum, there is also a natural separation, and the goals and intent of unlicensed operation are not necessarily the same as Part 97. Those that want to build a network so that anyone in their community can access xxx.com should do so under Part 97. Those that want to experiment with protocols and equipment and antennas and amplification should do so under Part 97, with the clear understanding that the result can only be used by licensed amateurs and the prohibition against business use and the content of xxx.com are clearly spelled out in Part 97. I expect that, just as is happening now, no small number of the Part 15 users will continue to be hams, and those who want to do more experimentation than allowed under Part 15 will become licensed under Part 97. The HSMM group is sure hoping that the latter happens.
Those who are interested in amateur radio and its potential can vistit the ARRL Web site at http://www.arrl.org.
73, Ed Hare, W1RFI ARRL Lab
(end quote)
I see some of your points, I guess, but I'm a little unclear as to how "This does not allow for any content provider to enter the market without an affiliation." Any site would be free to set up its own Salon-like subscription model, or continue begging on the streetcorner for PayPal/Amazon donations just like many do now. Who is doing the "allowing" that you're referring to here?
How else do you suggest that small sites overcome the obscurity and inertia that limits their financial models these days? And how am I supposed to support my 10 favorite sites while spending less than $500 a year? Those aren't rhetorical questions -- I'd really be curious if there are any answers besides syndication.
I don't see how the RIAA is "not a syndicate." Which definition are you using?
Well, you and some other people have raised a (very valid) point in this thread: syndicates that start out representing artists often don't end up that way.
I don't think many genuine artists (as opposed to synth-pop replicants) would argue with a straight face that the RIAA actually represents them. Today, the RIAA is purely a legislative mouthpiece for the music industry's corporate interests, one that began as an innocuous-sounding standards organization (ever heard of the "RIAA equalization curve" for phonograph records?).
In the case of a Web content syndicate, such an organization's reach would (ideally) be contractually constrained from the very first click. What I'd like to see is simply a way for ISPs to put a list of check boxes on their member pages, with labels like Slashdot, Kuro5hin, Salon, BoingBoing, Joe's Lame-Ass Weblog, whatever. As a subscriber, I check the boxes for the sites I want to read. A "total subscription amount" field above the boxes would go up by $10.00 for a year of Salon (which employs professional journalists and editors), $2.50 for a year of Slashdot (which does not), and $0.25 for BoingBoing (which is primarily a hobbyist weblog with no significant costs or original content to pay for). The final amount would be added to the ISP bill I already pay. Most of the revenue would then be divided proportionally among the content providers, with the ISP keeping the rest (and competing with other ISPs to offer me the most content for my money).
Similarly, when Joe and his crappy weblog join the syndicate, he's asked "What do you want to be paid for your content?" If Joe says "I'd like to be paid $0.05 per viewer-year," that's what the syndicate charges me to subscribe. If he says "I'd like to be paid $50.00 per viewer-year," that's fine too, from the syndicate's point of view. It doesn't cost anything for them to add Joe's check box to their web page, after all. In both cases, the users are the ones who decide what sites are worth paying for based on their operators' asking prices. That's already a pretty big departure from the way most big, bad, evil syndicates work, right?
At the end of the day, the big, bad, evil syndicate should really be nothing more than a clearinghouse that keeps the good sites like kuro5hin.org from having to beg on the street corner for financial patronage. Right now -- excluding paid text ads -- K5's business model relies on selling a small number of relatively-expensive memberships to a (necessarily) small number of readers. That sucks. It would be much better if K5 could lower their subscription prices by 10X and pick up 20X the subscriber base as a result. But not only is that too expensive due to financial-processing overhead, it's just too much of a pain in the neck for large numbers of users to bother with. A useful syndication model would (1) make life easy for both consumers and providers; (2) result in revenue streams that are much finer-grained and more intimately tied to the value (or lack thereof) of the content being provided; and (3) stick to its core mission rather than trying to become a power-drunk cartel like the RIAA and MPAA.
Maybe the latter course is inevitable, assuming the syndication concept succeeds spectacularly enough. But we need to at least try. Currently, Web content providers and have no mojo to abuse in the first place, which is no better for us all in the long run than the appearance and domination of the next RIAA-like organization. Either way, we, the consumers of content, risk losing out on some good stuff.
No, not like the RIAA. The RIAA is a cartel of content companies, not a syndicate that serves content creators.
... is the right way to address this, not micropayments (which will never be economically viable without a syndicate-style clearinghouse that insulates the participants from contact with actual financial institutions).
Paying $50/year to subscribe to a site sounds like a lot of money, because it is. But if I could pay $100/year for ad-free access to all of my favorite sites on an a la carte basis, it'd sound like a bargain. That's where commercial Web content will have to go eventually. I can't imagine any alternatives that will meet the needs of both consumers and site operators in the long term.
It'd be nice if one or more of the major ISPs would offer a pilot program along these lines. Not necessarily MSN or TW/AOL; even someone like Speakeasy could make a credible effort at syndicating content for their members, IMHO.
1. Make nitroglycerin from a recipe written by someone who can't spell the names of the chemicals ("sulferic acid") required.
2. Light fuse.
3. Get away.
4. (Your heirs) profit!
But that line of reasoning often turns into a psychological crutch for chronic whiners. How many posts on Slashdot read something like, "Dammit, I know Logo, BASIC, Pascal, VB, FORTRAN, assembly, Java, C++, and C#... and I still got laid off!" Sure, but how good were you at solving problems? Should an auto shop manager be impressed when a job applicant claims to have worked on Pintos, Novas, Malibus, Mustangs, Explorers, Cavaliers, and Excursions? How many of those cars drove away from the applicant's garage bay with their lugnuts cross-threaded?
Quality software engineering is more than a resume full of hip languages and buzzwords from the Gamma book. The best software engineering is usually done by people who got into the business because computers seemed like a really powerful and enjoyable way to solve engineering or (in the games biz) aesthetic problems. Those folks -- not the language lawyers, tool fetishists, and epicene gnomes of Unix who still have their home page set to schwab.com -- are the ones who tend to have the best answers to the question, "OK, why should I hire you?"
Which liberal are you funding?
From the looks of my 1040 this year, all of them.
I would recommend against using widget sets with OpenGL as a graphics layer unless you really need it: OpenGL is less than ideal for that purpose.
Just curious: why do you say that?
If your aircraft transceiver is running on 2.4 GHz, you've got bigger problems than renegade WiFi networks.
Good point, if that's actually when the artifacts first showed up. (I actually never played through any of the Ultimas after V, so what I wrote shouldn't be taken as canon.)
Actually the 'O' was just a stylized sphere interposed between the cube ('E') and pyramid ('A') symbols that comprised the EA logo. It made for a memorable if misleading trademark.
:)
The cube, sphere, and pyramid objects in Ultima VIII were originally supposed to be evil artifacts in Ultima VI, if I remember correctly. Richard Garriott had a bit of a grudge against Trip Hawkins ("Pirt Snikwah") back in those days, as Origin was less than thrilled with their treatment as an EA affiliated label in the late Eighties. Richard had painstakingly planned a whole boatload of EA-as-the-embodiment-of-ultimate-evil allusions for the Ultimas beyond V, but EA skillfully dodged his wrath by offering to buy the company.
I'm not involved with that particular product and can't rattle any stats off the top of my head, but I remember being shocked to hear how much download traffic our site sees per month. The donation button shows up only when you're downloading a freebie like the video toolset that's of primary interest to (non-paying) end users.
I made the same mistake ("aliens? WTF?!?") the first time I saw AI.
If there's a time on every project schedule to shoot the engineer, and a time on every movie schedule to shoot the director, there was DEFINITELY an overlooked time on AI at which the VFX lead should have been shot. Those future-mecha models made Jar Jar Binks look like something from a lost da Vinci sketch. My opinion of the film improved immeasurably when someone clued me in that the "aliens" were descendents of the mecha.
My understanding from surfing RadioShack Sucks is that their salesdroids would actually be penalized financially, or even fired altogether, for failing to obtain some arbitrary percentage of customer names and addresses. Seems like the quota was something on the order of 80-90% "compliance."
Between local stores like Active Electronics, the utterly-amazing variety of electronic parts on eBay and topnotch mail-order houses like Digi-Key, Jameco, and Mouser, it's pretty darned rare for me to set foot in a RatShit store these days. Their 1/4-watt resistor assortments are still a killer deal, though.
Interesting info. So the van outside watches for leakage of the amplified 90-150 kHz signal from the baseband circuitry, huh? Hard to believe there's enough radiation to pick up... you need a fairly-efficient antenna to 'broadcast' signals at that frequency, not just stray bits and pieces of wire. TEMPEST interception works because deflection yokes make nice loop antennas, but I wouldn't think this carrier would make it through to the deflection circuitry (at least not without causing horrible artifacts in the picture).
I'll have to do some Googling on this one.
But those signals are very weak, especially when going 70 mph. It doesn't make sense to me; seems as though it wouldn't be worth the trouble.
Couple of points in their favor, though: they know exactly where to listen, down to the nearest 1 kHz or so given the stability of modern synthesized LOs. That means they can use a very narrow filter to detect the desired signal, and overall S/N ratio improves with 10 * log(bandwidth). Compare the 200 kHz-wide IF an ordinary FM radio uses to the 5-kHz detection bandwidth a billboard receiver might use. That's like picking up an extra 16 dB of signal strength for free.
Also, the LO signal isn't necessarily that weak to begin with... probably on the order of a milliwatt at the point where it's injected into the mixer. Many of the cheaper FM "wireless broadcasters" sold by outfits like Radio Shack put out less power than that.
I imagine that the power at the antenna might be as high as -60 dBm or greater, which is workable, but that's a wild guess. I'll have to drag a spectrum analyzer into the garage tonight and see what kind of radiated field strength I can pick up from my cars' antennas.
I don't follow your point about "interference produced by the amplifier (in the) car's stereo." That's an audio amp.
have you actually tried to detect that signal outside a vehicle? ok now do it to a MOVING TARGET.
:) If the BBC is transmitting the 'subcarrier', what does a TV set do to the 'subcarrier' to make it detectable from outside the house? If you're talking about a TV receiving a signal off the air (as opposed to through a CATV feed that can be TDR'ed to detect connections), the only way you can detect it is by listening for either an LO signal or a (Tempest-style) deflection signal. Can you nail down some more details about the subcarrier-detection scheme you're referring to?
I could tell you, but then I'd have to... aw, never mind.
and what you are talking about is not the case.. the BBC transmits a subcarrier with a tone on it that is easily detectable. It's the same detection scheme used by american cable TV companies to snif out people stealing cable tv. It's a simple device and putting the subcarrier there makes it air tight in court.. trying to say that "we detected what channel your tv is tuned to doesnt work in court... saying we detected our special signal we transmit to catch them.... does.
It's likely that the technique used in the UK has changed over the years, but it's also worth pointing out that what you're describing makes no sense whatsoever.
The reason why LO signal leakage absolutely does occur in any practical radio receiver is that there's no such thing as a perfectly-directional RF amplifier.
The local oscillator signal is generated and mixed with the incoming signal at a very early stage, relativley close to the antenna. A long way before the limiter/discriminator stage, in other words, and in a portion of the receiver that's architecturally identical regardless of the mode being demodulated. Different IF frequencies are used by FM, AM, and TV broadcast receivers, but the front-end topology is the same: one or more RF amplifier stages feed the incoming signal into the first mixer, which is also fed by the first local oscillator. This signal chain is not perfectly unidirectional.
RF amplifiers are usually characterized by a handful of key parameters -- their noise figure, their ability to operate properly in the presence of strong signals as well as weak ones, and their forward gain (also known as 'S21'). But any real RF amplifier will also have a reverse gain parameter ('S12') describing the attenuation a signal applied to its output undergoes on the way back through to the input port. With the untuned, resitively-matched gain blocks popular in RF work these days, the S12 parameter is often just barely better than S21. The same is true of the mixer itself; it has a decidedly-finite port isolation spec that describes how much of the signal at the LO port will leak back through the RF port. The cheap unbalanced mixers used in consumer-grade receivers aren't exactly state-of-the-art in this department.
The bottom line is that significant, detectable LO leakage DOES take place through the antenna regardless of how well-shielded the rest of the receiver is. It's entirely believable that a stationary billboard receiver could be designed and optimized to look for LO leakage from passing cars. Remember that even the cheapest radios today use highly-stable phase-locked LOs, so the required signal-detection bandwidth at the billboard is very small indeed.
Military receivers often have two or even three RF stages preceding the mixer, not to maximize front-end gain but rather to cut down on LO leakage to make the receiver (and its user) harder for the enemy to direction-find. This practice dates back at least to the BC-series receivers used in WWII. Not much new under the sun here.