The reason SCO was able to do that is that while the system is designed to thwart foot-dragging by defendants, it doesn't do as much to prevent foot-dragging plaintiffs.
Right. But in bankruptcy court, that's reversed. The debtor isn't in charge. The court, the U.S. Trustee, and the creditor's committee are. Notice how, since SCO declared bankruptcy, their wierd legal moves have been shot down fast. SCO's "emergency motion" for a quick sale - deferred by judge without even asking SCO.
Quick asset sale to York under wierd conditions - withdrawn once the creditors objected. Novell lawsuit - unstayed.
Judge Kimball's calendar has an open day on December 11.
The SCO cases could drag on for years, or they could end next month. It's hard to say.
Groklaw just covering SCO is enough for now.
As SCO management is discovering, they have far less control in bankruptcy than they could before. They were able to control the pace of the pre-bankruptcy cases because they were the plaintiff in most of them. It's different in bankruptcy. The bankruptcy judge, the U.S. Trustee, and the creditors are in charge. The bankruptcy judge keeps shooting down SCO management's wierd legal ideas.
Yahoo Finance says it all: "SCOX is deficient and bankrupt."
Well, there's IntelliCAD 2000.
This was free; it's a Windows-based clone of AutoCAD 98. It's very like 2D AutoCAD. There are later, non-free versions. There's a consortium which maintains the base engine, and various products built on top of it.
IntelliCAD 2000 is amusing, because it replicates some of the features of AutoCAD from the DOS era, like having to explicitly refresh the drawing.
Wierd. First, it's not a "nuclear battery". Those have been around since the 1950s, and they typically have quite modest power output, from a few watts to a few hundred watts. They're just some radioactive material decaying at its normal rate; they don't use a chain reaction. If this thing is supposed to produce 27MW, it has to be a real nuclear reactor.
And it is. Here's the patent application, out of Los Alamos National Laboratory. The basic idea is this: "This present invention achieves control by utilizing the properties of a fissile metal hydride as a self-contained nuclear fuel and neutron energy moderator. If the physical size, fissile metal content and enrichment are appropriately selected, the metal will absorb ambient hydrogen, which moderates the neutron energies so that nuclear fission criticality is achieved. The temperature will then be increased by the fission reactions until the dissociation pressure of the hydrogen for that temperature is greater than the ambient pressure of the hydrogen, at which point the hydrogen dissociates from the hydride and the source becomes sub-critical." So that's the way it self-regulates. It's supposed to operate at a constant temperature; if you remove heat with a working fluid, it produces more heat; if you don't, it stabilizes at its normal operating temperature. It's a uranium reactor, using 5% enriched uranium. Runs at 350C to 800C. Uses heat pipes to get the heat out to a working fluid, probably water, used to make steam and drive a turbine.
It's not clear if this is a workable design. There's no prototype. But it's at least plausible. It's not a totally new idea; the TRIGA reactors are self-regulating in a somewhat similar fashion.
Their definition of "consistently produces power" is 79% to 92% uptime. This figure is based on the uptime for a typical single coal-fired generation unit. But they're using those numbers for a whole collection of widely distributed wind farms. That's not an appropriate comparison.
They have some moderately encouraging numbers for a set of 19 wind farms spread across a thousand kilometers, from New Mexico to Kansas. But look at Figure 3.
92% of the time, at least a quarter of average output is available. The output reliably available 99+% of the time is near zero.
What this paper actually demonstrates is that "baseload wind" isn't going to consistently provide power, even with a big grid. You need peaking plants or energy storage.
Right. The Copenhagen interpretation is a hack developed because the many-worlds theory is emotionally unsatisfying to some. But it leads to paradoxes like this one. Many-worlds is consistent with observation (Hawking once said it's "trivially true"), but the notion that the universe forks on every quantum-level event bothers some people. It seems uneconomical, or theologically objectionable, or something.
There are "open redirectors" on many major sites, including Google, AOL, eBay, and Microsoft Live. (Yahoo plugged their hole by giving their open redirector its own, easily blockable, domain.) We mentioned this on Slashdot a few days ago, and someone immediately followed up by using the Google exploit to get through Slashdot's filters.
These open redirectors are regularly exploited by phishing scams. People report them to PhishTank, and over at SiteTruth, we tie them back to the domain responsible and fix blame. PhishTank is too nice about this. They just blacklist the phishing URL. That stopped working a few months back, when phishers started generating random URLs and subdomains for each e-mail. We down-rate the whole base domain.
It's time to take a hard line on this. The Internet used to tolerate open mail relays, which were a nice feature until spammers started exploiting them. Now they're routinely blocked. Open redirectors now need similar treatment.
It would have been better if they tied the map data to real estate ownership records. Much of that data is available in machine-readable form. It would be cool, and useful, to zoom in and see the property lines. Displaying the ownership information would be even better. It's a public record, after all.
Or if they recognized house numbers in the imagery taken by the StreetView truck.
Here's yet another redirection exploit on Google, reported in a Symantec security bulletin. This one exploits redirection in the "I'm Feeling Lucky" feature.
You should look the open source BlackBox CP compiler... their garbage collector works roughly that way (or better i think)
Sadly, no. Component Pascal does not have deterministic destructors. So you can't use destructors to release resources at the appropriate time, and you have all the problems mentioned above with destructors being called at a random time. See section 10.4: "The FINALIZE procedure can be implemented for any pointer type. The method is called at some
unspecified time after an object of that type (or a base type of it) has become unreachable via other
pointers (not globally anchored anymore) and before the object is deallocated."
Stock spam of lube additive treated as terrorism
on
10 Great Snake-Oil Gadgets
·
· Score: 4, Interesting
A few years ago, I received many stock spams for "XLPI.PK", or Xcel Plus, which sells fuel and lubricant additives.
Such additives are referred to in the automotive industry as "mouse milk"; they usually don't do much, and may make things worse. That whole category of products is mostly bogus.
A few weeks later, I got a call from an anti-terrorism investigator at NCIS. Someone at the FAA had looked at the letter and the web site. They apparently didn't like what they saw, and referred the matter for investigation of the use of unapproved lubricants in military equipment. That comes under the "sabotaging the war effort" laws, which brings in military investigators.
I'm not sure what happened thereafter, but the spamming stopped and "XLPI.PK" is now trading at $0.001.
Note that it fools Slashdot, and most link scanners in spam filters, into accepting the URL as leading to "google.com". But, in fact, it redirects to the "malware-scan.com" hostile site, which will try to install an Active-X control.
We've been finding attacks like this up with SiteTruth, by using PhishTank information to down-rate sites that have open redirectors. We've found open redirectors on Google and AOL. They're actively being exploited.
So we're currently down-rating Google,
and AOL.. It may seem drastic to downrate an entire major site because they have a few "minor" exploits. PhishTank itself only blacklists specific hostile URLs. But that's no longer enough. Most modern phishing attacks use a unique URL, and often a unique subdomain, for each user attacked. SiteTruth thus takes a harder line. If a domain hosts something one of the data sources says is an attack, it downrates the whole domain automatically.
It's within the power of the site operator to close such security holes. We encourage them to do so.
It's not that hard with the right tools.
on
Star Trek Home Theater
·
· Score: 3, Interesting
The ceiling wouldn't be that hard to do. Use a CNC router to mill a clay mould, then vacuum-form plastic sheet over it. TechShop in Silicon Valley has all the gear for that, and there are shops that do large-area vacuum forming. Up to 6' x 11' vacuum forming of single pieces is commercially available.
Much of the "future" that comes from Hollywood is made by vacuum forming. It's cheap.
There's actually an accepted safe way to do memory management - reference counts and weak references. That's what both Perl and Python have settled on, and it's worth noting that programmers in those languages seldom have serious memory management problems. In C and C++, one has to obsess on memory management issues, and even in Java and C#, which are garbage collected, it takes more attention than it should.
Reference counts have the advantage of repeatability - deletion will occur at predictable times. This allows the use of destructors. You can safely use destructors
to manage other assets, like windows, open files, network connections, and such.
Destructors in systems with garbage collection make for an unhappy marriage. Calling a destructor or finalizer from the garbage collector is essentially equivalent to calling it at some random time from another thread. So race conditions are possible. Check out Microsoft's "managed C++" for an attempt to get all the cases for this right. It's not pretty.
The classic complaint about reference counts is "what about cycles"? There's a simple answer - cycles, that is, loops of strong pointers, are errors. This isn't a severe restriction; it just requires some data structure design. With trees, for example, links towards the leaves are strong pointers, and links towards the root are weak. (I've revised Python's BeautifulSoup HTML parser to work that way; "down" and "forward" links are strong, while "up" and "backwards" links are weak. It took about 20 lines of code and eliminated annoying problems in programs dealing with HTML trees.)
If you really need a symmetrical circular list, which might happen in, say, a window library with many links between widgets, there's a simple solution. Have all the objects owned by some collection, then use weak pointers between them. When the collection is dropped, all the bits and pieces go away, in a well defined order.
In Python, you can turn off garbage collection while leaving reference counting active, then list any orphaned cycles at program end for debugging purposes. This is a practical way to program without leaks or garbage collection. It's generally easy to find cycles, because cycles are created by data structure design, not by bugs. So if a program has cycles, it will probably have them every time, and thus they can be found early in debugging. With better language support for debugging, cycles could be caught at the moment of creation, which would make it easy to eliminate them.
Now if we could get this into a hard-compiled language, we'd have the problem solved. Repeated attempts to bolt reference counting onto C++ via templates have resulted in fragile systems. The fundamental problem is that C++ still requires access to raw pointers to get anything done, and this puts a hole in the protection provided by the reference counting system. It takes language support to make this work right.
Grand Central Station had an extensive 120V DC system until an overdue rewiring in the 1990s. Grand Central had several independent power systems - 600VDC subway traction power, 25HZ 12KV Pennsylvania Railroad traction power, original 120VDC Edison utility power, 60HZ AC utility power at the usual voltages, and a sizable in-house steam plant, all fed from different sources. Old equipment was running off of one or another of these sources, and had to be replaced. After the rewiring, everything that didn't run on rails was powered from commercial 60Hz power.
Grand Central provided power and steam for several nearby buildings, and during the cutover, some businesses were discovered to be mooching off Grand Central's power.
Later elevators still used 600VDC but used a dynamotor
What you're hearing is not a dynamotor, but something called a Ward Leonard drive. It's a fixed-speed motor driving a generator, but its purpose is speed control. The field current of the generator, which is small, is adjusted to control the larger output of the generator. The variable output of the generator then drives the elevator motor. The Ward Leonard drive is thus a big power amplifier. Until power semiconductors got big enough, which wasn't really until the 1980s, this was the most effective way to smoothly speed-control large motors.
A dynamotor has a common field for the input and output sides, but a Ward Leonard drive does not.
Wierd Stuff Warehouse in Mountain View, CA offers free electronics recycling. If it works, they'll put it up for sale; if not, they'll scrap it properly.
Good place to get CRT monitors cheap, if you want one.
Well, what do you expect? Read Bronfman's entry on Wikipedia. He was the heir to Seagram's Liquor. His whole life has been carried along by family connections. Highlights from Wikipedia:
"He was particularly active in school theatre, an interest his parents supported by donating to construct The Ann and Edgar Bronfman Theatre during a 1967 expansion at The Collegiate School, the prestigious private school in Manhattan which Edgar Jr. attended."
"The summer before his final year of high school, in 1972, he was a credited producer on the film, The Blockhouse. Despite his inexperience, Bronfman's involvement was accepted because of his connections and access to financing."
"By 1994 he became the Chief Executive Officer (of Seagrams), where he began a move away from the traditional liquor business and into entertainment.
The first step in this diversification was the widely criticized sale of Seagram's stake in DuPont."
"Bronfman, Jr., then led Seagram into a disastrous all-stock acquisition by French conglomerate Vivendi in 2000."
"Seagram's for all intents and purposes ceased to exist."
"On February 27, 2004, Bronfman finalized the acquisition of Warner Music Group and he has served as Chairman and CEO of the music company since that time."
He didn't build up Warner Music, or move up within the company, or come to it from success elsewhere. He bought the thing with inherited money, after a long career as a failed executive.
That reads like a SOP for a well-funded maximum security prison. It's rather labor-intensive; a US prison wouldn't be that heavily staffed. It's amusing that "punishment food" is MREs, which is what our soldiers eat. But that's not a big deal.
The terms are incredibly permissive in one area - religion. Considerable efforts are made to accommodate Islamic worship. The guards are required to handle a Koran in very specific ways. Prayer mats are provided. Even honey and dates are supplied for Ramadan.
When softening up prisoners for interrogation, the US military might do better to provide inmates with lots of American movies and music, but less religious support. Islamic fundamentalism is instilled by emphasis on Islam to the exclusion of all else, and the Camp Delta procedures reinforce that. If prisoners want a Koran, they should get a paperback copy, maybe a Xerox. Let them watch Baywatch reruns, and schedule the good parts to conflict with their prayer schedules. Have different prisoners doing different things at different times, to discourage synchronized prayer. The official attitude should be "if you want to pray, we're not going to stop you. Whatever".
Most large refrigeration compressors have "short-cycling protection". The compressor motor is overloaded during startup, and needs time to cool. So there's a timer that limits the time between two compressor starts. 4 minutes is a typical delay for a large unit. If you don't have this delay, compressor motors burn out.
Some fancy short-cycling protection timers have backup power, so the the "start to start" time is measured even through power failures. But that's rare. Here's a typical short-cycling timer. For the ones that don't, like that one, a power failure restarts the timer, so you have to wait out the timer after a power glitch.
The timers with backup power, or even the old style ones with a motor and cam-operated switch, allow a quick restart after a power failure if the compressor was already running. Once. If there's a second power failure, the compressor has to wait out the time delay.
So it's important to ensure that a data center's chillers have time delay units that measure true start-to-start time, or you take a cooling outage of several minutes on any short power drop. And, after a power failure and transfer to emergency generators, don't go back to commercial power until enough time has elapsed for the short-cycling protection timers to time out. This last appears to be where Rackspace failed.
Dealing with sequential power failures is tough. That's what took down that big data center in SF a few months ago.
Skip the ad-laden overloaded blogodreck site and go directly to StupidFilter. The concept is straightforward - they're training a naive Bayesian classifier, like a spam filter, on a set of text excerpts rated by humans. You can look at random samples from the training set for amusement.
Wikipedia already has some 'bots that do somewhat similar things, looking for totally bogus edits and reverting them. Yahoo's "commercial intent" filter also does something like that, to separate commercial and non-commercial sites. We considered something like that for SiteTruth, where we need to distinguish non-commercial sites so we don't rate them by business criteria.
This approach to filtering will probably need domain-dependent filters. A political site, a social site, a sports site, and a game site all need different training sets. I'd go for a two-stage classifier, one that divided sites into about ten to twenty major categories, and then a stupidity filter trained for each of those categories.
Applying such a filter at blog posting time should be interesting.
And the characters in these books, and plays, and so on, and in real life, I might add, spend hours bemoaning the fact that they can't communicate. I feel that if a person can't communicate the very least he can do is to shut up. - Tom Lehrer.
The US military has had something like this for a while. It's a Jeep-sized thing used by special ops types. Windshield and roll bar can be folded down for stacking, so the things can be stacked two-high in a C-130. With loading ramps, you can drive one onto the top of another one.
Slashdot Mirror
The reason SCO was able to do that is that while the system is designed to thwart foot-dragging by defendants, it doesn't do as much to prevent foot-dragging plaintiffs.
Right. But in bankruptcy court, that's reversed. The debtor isn't in charge. The court, the U.S. Trustee, and the creditor's committee are. Notice how, since SCO declared bankruptcy, their wierd legal moves have been shot down fast. SCO's "emergency motion" for a quick sale - deferred by judge without even asking SCO. Quick asset sale to York under wierd conditions - withdrawn once the creditors objected. Novell lawsuit - unstayed.
Judge Kimball's calendar has an open day on December 11.
The SCO cases could drag on for years, or they could end next month. It's hard to say. Groklaw just covering SCO is enough for now.
As SCO management is discovering, they have far less control in bankruptcy than they could before. They were able to control the pace of the pre-bankruptcy cases because they were the plaintiff in most of them. It's different in bankruptcy. The bankruptcy judge, the U.S. Trustee, and the creditors are in charge. The bankruptcy judge keeps shooting down SCO management's wierd legal ideas.
Yahoo Finance says it all: "SCOX is deficient and bankrupt."
Well, there's IntelliCAD 2000. This was free; it's a Windows-based clone of AutoCAD 98. It's very like 2D AutoCAD. There are later, non-free versions. There's a consortium which maintains the base engine, and various products built on top of it.
IntelliCAD 2000 is amusing, because it replicates some of the features of AutoCAD from the DOS era, like having to explicitly refresh the drawing.
Wierd. First, it's not a "nuclear battery". Those have been around since the 1950s, and they typically have quite modest power output, from a few watts to a few hundred watts. They're just some radioactive material decaying at its normal rate; they don't use a chain reaction. If this thing is supposed to produce 27MW, it has to be a real nuclear reactor.
And it is. Here's the patent application, out of Los Alamos National Laboratory. The basic idea is this: "This present invention achieves control by utilizing the properties of a fissile metal hydride as a self-contained nuclear fuel and neutron energy moderator. If the physical size, fissile metal content and enrichment are appropriately selected, the metal will absorb ambient hydrogen, which moderates the neutron energies so that nuclear fission criticality is achieved. The temperature will then be increased by the fission reactions until the dissociation pressure of the hydrogen for that temperature is greater than the ambient pressure of the hydrogen, at which point the hydrogen dissociates from the hydride and the source becomes sub-critical." So that's the way it self-regulates. It's supposed to operate at a constant temperature; if you remove heat with a working fluid, it produces more heat; if you don't, it stabilizes at its normal operating temperature. It's a uranium reactor, using 5% enriched uranium. Runs at 350C to 800C. Uses heat pipes to get the heat out to a working fluid, probably water, used to make steam and drive a turbine.
It's not clear if this is a workable design. There's no prototype. But it's at least plausible. It's not a totally new idea; the TRIGA reactors are self-regulating in a somewhat similar fashion.
The "Los Alamos Study Group" that made critical comments has nothing to do with Los Alamos National Laboratories. Their director "worked as a transportation planner, natural foods manufacturing entrepreneur, high school teacher, hazardous waste investigator, and contaminant hydrologist."
It's Roland the Plogger again, trying to drive traffic to his blog. It's not like he actually understands what he posts.
Here's the actual paper, Supplying Baseload Power and Reducing Transmission Requirements by Interconnecting Wind Farms. The authors have been crunching on wind speed data to try to figure out if a widespread enough set of wind farms would statistically be able to consistently produce power.
Their definition of "consistently produces power" is 79% to 92% uptime. This figure is based on the uptime for a typical single coal-fired generation unit. But they're using those numbers for a whole collection of widely distributed wind farms. That's not an appropriate comparison.
They have some moderately encouraging numbers for a set of 19 wind farms spread across a thousand kilometers, from New Mexico to Kansas. But look at Figure 3. 92% of the time, at least a quarter of average output is available. The output reliably available 99+% of the time is near zero.
What this paper actually demonstrates is that "baseload wind" isn't going to consistently provide power, even with a big grid. You need peaking plants or energy storage.
Right. The Copenhagen interpretation is a hack developed because the many-worlds theory is emotionally unsatisfying to some. But it leads to paradoxes like this one. Many-worlds is consistent with observation (Hawking once said it's "trivially true"), but the notion that the universe forks on every quantum-level event bothers some people. It seems uneconomical, or theologically objectionable, or something.
There are "open redirectors" on many major sites, including Google, AOL, eBay, and Microsoft Live. (Yahoo plugged their hole by giving their open redirector its own, easily blockable, domain.) We mentioned this on Slashdot a few days ago, and someone immediately followed up by using the Google exploit to get through Slashdot's filters.
These open redirectors are regularly exploited by phishing scams. People report them to PhishTank, and over at SiteTruth, we tie them back to the domain responsible and fix blame. PhishTank is too nice about this. They just blacklist the phishing URL. That stopped working a few months back, when phishers started generating random URLs and subdomains for each e-mail. We down-rate the whole base domain.
It's time to take a hard line on this. The Internet used to tolerate open mail relays, which were a nice feature until spammers started exploiting them. Now they're routinely blocked. Open redirectors now need similar treatment.
Beyond simple URL redirectors are exploits of JavaScript redirectors. Efforts are underway to detect and block those.
It would have been better if they tied the map data to real estate ownership records. Much of that data is available in machine-readable form. It would be cool, and useful, to zoom in and see the property lines. Displaying the ownership information would be even better. It's a public record, after all.
Or if they recognized house numbers in the imagery taken by the StreetView truck.
Here's yet another redirection exploit on Google, reported in a Symantec security bulletin. This one exploits redirection in the "I'm Feeling Lucky" feature.
You should look the open source BlackBox CP compiler... their garbage collector works roughly that way (or better i think)
Sadly, no. Component Pascal does not have deterministic destructors. So you can't use destructors to release resources at the appropriate time, and you have all the problems mentioned above with destructors being called at a random time. See section 10.4: "The FINALIZE procedure can be implemented for any pointer type. The method is called at some unspecified time after an object of that type (or a base type of it) has become unreachable via other pointers (not globally anchored anymore) and before the object is deallocated."
A few years ago, I received many stock spams for "XLPI.PK", or Xcel Plus, which sells fuel and lubricant additives. Such additives are referred to in the automotive industry as "mouse milk"; they usually don't do much, and may make things worse. That whole category of products is mostly bogus.
Back then, their web site contained endorsements from the FAA and the US Army. The web site reproduced a a letter of endorsement appearing to be from an FAA representative. I thought this was a bit strange, so I sent off a note to the regional FAA office asking if it was legitimate.
A few weeks later, I got a call from an anti-terrorism investigator at NCIS. Someone at the FAA had looked at the letter and the web site. They apparently didn't like what they saw, and referred the matter for investigation of the use of unapproved lubricants in military equipment. That comes under the "sabotaging the war effort" laws, which brings in military investigators.
I'm not sure what happened thereafter, but the spamming stopped and "XLPI.PK" is now trading at $0.001.
There's a related hole in Google Maps, an "open redirector", that allows this exploit. Here's an example:
Caution - hostile URL Close the page displayed; don't click on anything on it. .
Note that it fools Slashdot, and most link scanners in spam filters, into accepting the URL as leading to "google.com". But, in fact, it redirects to the "malware-scan.com" hostile site, which will try to install an Active-X control.
We've been finding attacks like this up with SiteTruth, by using PhishTank information to down-rate sites that have open redirectors. We've found open redirectors on Google and AOL. They're actively being exploited.
So we're currently down-rating Google, and AOL.. It may seem drastic to downrate an entire major site because they have a few "minor" exploits. PhishTank itself only blacklists specific hostile URLs. But that's no longer enough. Most modern phishing attacks use a unique URL, and often a unique subdomain, for each user attacked. SiteTruth thus takes a harder line. If a domain hosts something one of the data sources says is an attack, it downrates the whole domain automatically.
It's within the power of the site operator to close such security holes. We encourage them to do so.
The ceiling wouldn't be that hard to do. Use a CNC router to mill a clay mould, then vacuum-form plastic sheet over it. TechShop in Silicon Valley has all the gear for that, and there are shops that do large-area vacuum forming. Up to 6' x 11' vacuum forming of single pieces is commercially available.
Much of the "future" that comes from Hollywood is made by vacuum forming. It's cheap.
There's actually an accepted safe way to do memory management - reference counts and weak references. That's what both Perl and Python have settled on, and it's worth noting that programmers in those languages seldom have serious memory management problems. In C and C++, one has to obsess on memory management issues, and even in Java and C#, which are garbage collected, it takes more attention than it should.
Reference counts have the advantage of repeatability - deletion will occur at predictable times. This allows the use of destructors. You can safely use destructors to manage other assets, like windows, open files, network connections, and such.
Destructors in systems with garbage collection make for an unhappy marriage. Calling a destructor or finalizer from the garbage collector is essentially equivalent to calling it at some random time from another thread. So race conditions are possible. Check out Microsoft's "managed C++" for an attempt to get all the cases for this right. It's not pretty.
The classic complaint about reference counts is "what about cycles"? There's a simple answer - cycles, that is, loops of strong pointers, are errors. This isn't a severe restriction; it just requires some data structure design. With trees, for example, links towards the leaves are strong pointers, and links towards the root are weak. (I've revised Python's BeautifulSoup HTML parser to work that way; "down" and "forward" links are strong, while "up" and "backwards" links are weak. It took about 20 lines of code and eliminated annoying problems in programs dealing with HTML trees.)
If you really need a symmetrical circular list, which might happen in, say, a window library with many links between widgets, there's a simple solution. Have all the objects owned by some collection, then use weak pointers between them. When the collection is dropped, all the bits and pieces go away, in a well defined order.
In Python, you can turn off garbage collection while leaving reference counting active, then list any orphaned cycles at program end for debugging purposes. This is a practical way to program without leaks or garbage collection. It's generally easy to find cycles, because cycles are created by data structure design, not by bugs. So if a program has cycles, it will probably have them every time, and thus they can be found early in debugging. With better language support for debugging, cycles could be caught at the moment of creation, which would make it easy to eliminate them.
Now if we could get this into a hard-compiled language, we'd have the problem solved. Repeated attempts to bolt reference counting onto C++ via templates have resulted in fragile systems. The fundamental problem is that C++ still requires access to raw pointers to get anything done, and this puts a hole in the protection provided by the reference counting system. It takes language support to make this work right.
Grand Central Station had an extensive 120V DC system until an overdue rewiring in the 1990s. Grand Central had several independent power systems - 600VDC subway traction power, 25HZ 12KV Pennsylvania Railroad traction power, original 120VDC Edison utility power, 60HZ AC utility power at the usual voltages, and a sizable in-house steam plant, all fed from different sources. Old equipment was running off of one or another of these sources, and had to be replaced. After the rewiring, everything that didn't run on rails was powered from commercial 60Hz power.
Grand Central provided power and steam for several nearby buildings, and during the cutover, some businesses were discovered to be mooching off Grand Central's power.
Later elevators still used 600VDC but used a dynamotor
What you're hearing is not a dynamotor, but something called a Ward Leonard drive. It's a fixed-speed motor driving a generator, but its purpose is speed control. The field current of the generator, which is small, is adjusted to control the larger output of the generator. The variable output of the generator then drives the elevator motor. The Ward Leonard drive is thus a big power amplifier. Until power semiconductors got big enough, which wasn't really until the 1980s, this was the most effective way to smoothly speed-control large motors.
A dynamotor has a common field for the input and output sides, but a Ward Leonard drive does not.
Incidentally, the Wikipedia article in Ward Leonard drives is bogus. Here's a better reference.
Wierd Stuff Warehouse in Mountain View, CA offers free electronics recycling. If it works, they'll put it up for sale; if not, they'll scrap it properly.
Good place to get CRT monitors cheap, if you want one.
Well, what do you expect? Read Bronfman's entry on Wikipedia. He was the heir to Seagram's Liquor. His whole life has been carried along by family connections. Highlights from Wikipedia:
- "He was particularly active in school theatre, an interest his parents supported by donating to construct The Ann and Edgar Bronfman Theatre during a 1967 expansion at The Collegiate School, the prestigious private school in Manhattan which Edgar Jr. attended."
-
"The summer before his final year of high school, in 1972, he was a credited producer on the film, The Blockhouse. Despite his inexperience, Bronfman's involvement was accepted because of his connections and access to financing."
-
"By 1994 he became the Chief Executive Officer (of Seagrams), where he began a move away from the traditional liquor business and into entertainment.
The first step in this diversification was the widely criticized sale of Seagram's stake in DuPont."
-
"Bronfman, Jr., then led Seagram into a disastrous all-stock acquisition by French conglomerate Vivendi in 2000."
-
"Seagram's for all intents and purposes ceased to exist."
-
"On February 27, 2004, Bronfman finalized the acquisition of Warner Music Group and he has served as Chairman and CEO of the music company since that time."
He didn't build up Warner Music, or move up within the company, or come to it from success elsewhere. He bought the thing with inherited money, after a long career as a failed executive.That reads like a SOP for a well-funded maximum security prison. It's rather labor-intensive; a US prison wouldn't be that heavily staffed. It's amusing that "punishment food" is MREs, which is what our soldiers eat. But that's not a big deal.
The terms are incredibly permissive in one area - religion. Considerable efforts are made to accommodate Islamic worship. The guards are required to handle a Koran in very specific ways. Prayer mats are provided. Even honey and dates are supplied for Ramadan.
When softening up prisoners for interrogation, the US military might do better to provide inmates with lots of American movies and music, but less religious support. Islamic fundamentalism is instilled by emphasis on Islam to the exclusion of all else, and the Camp Delta procedures reinforce that. If prisoners want a Koran, they should get a paperback copy, maybe a Xerox. Let them watch Baywatch reruns, and schedule the good parts to conflict with their prayer schedules. Have different prisoners doing different things at different times, to discourage synchronized prayer. The official attitude should be "if you want to pray, we're not going to stop you. Whatever".
Most large refrigeration compressors have "short-cycling protection". The compressor motor is overloaded during startup, and needs time to cool. So there's a timer that limits the time between two compressor starts. 4 minutes is a typical delay for a large unit. If you don't have this delay, compressor motors burn out.
Some fancy short-cycling protection timers have backup power, so the the "start to start" time is measured even through power failures. But that's rare. Here's a typical short-cycling timer. For the ones that don't, like that one, a power failure restarts the timer, so you have to wait out the timer after a power glitch.
The timers with backup power, or even the old style ones with a motor and cam-operated switch, allow a quick restart after a power failure if the compressor was already running. Once. If there's a second power failure, the compressor has to wait out the time delay.
So it's important to ensure that a data center's chillers have time delay units that measure true start-to-start time, or you take a cooling outage of several minutes on any short power drop. And, after a power failure and transfer to emergency generators, don't go back to commercial power until enough time has elapsed for the short-cycling protection timers to time out. This last appears to be where Rackspace failed.
Dealing with sequential power failures is tough. That's what took down that big data center in SF a few months ago.
Skip the ad-laden overloaded blogodreck site and go directly to StupidFilter. The concept is straightforward - they're training a naive Bayesian classifier, like a spam filter, on a set of text excerpts rated by humans. You can look at random samples from the training set for amusement.
Wikipedia already has some 'bots that do somewhat similar things, looking for totally bogus edits and reverting them. Yahoo's "commercial intent" filter also does something like that, to separate commercial and non-commercial sites. We considered something like that for SiteTruth, where we need to distinguish non-commercial sites so we don't rate them by business criteria.
This approach to filtering will probably need domain-dependent filters. A political site, a social site, a sports site, and a game site all need different training sets. I'd go for a two-stage classifier, one that divided sites into about ten to twenty major categories, and then a stupidity filter trained for each of those categories.
Applying such a filter at blog posting time should be interesting.
And the characters in these books, and plays, and so on, and in real life, I might add, spend hours bemoaning the fact that they can't communicate. I feel that if a person can't communicate the very least he can do is to shut up. - Tom Lehrer.
Is Donald Kerr's house in Google StreetView? What's the link?
Coming up next, a history of the dial tone.
Very nice. The images are still very blurry (resolution 81×61×11), and the detectors, at 37mm, are big, but it's a start.
The US military has had something like this for a while. It's a Jeep-sized thing used by special ops types. Windshield and roll bar can be folded down for stacking, so the things can be stacked two-high in a C-130. With loading ramps, you can drive one onto the top of another one.
But that's a rather specialized application.