Patriot Act demands are supposed to be handled at the FBI Director level. But the FBI has delegated authority to issue them all the way down to the Special Agent in Charge at each FBI office.
The FBI wants the name of the ISP involved kept secret. But from this deposition it's very tiny. "I am the President and sole employee of (blacked out)".
He's currently subject to a gag order, and the ACLU is trying to get that lifted.
The ACLU recently moved for summary judgement in this case, because there are no factual issues in dispute. This will need to be resolved on appeal, as a constitutional issue.
3COM, first with TCP/IP
on
The 3Com Saga
·
· Score: 4, Informative
3COM was the first company to support TCP/IP in a commercial product. 3COM's UNET, released in 1980, was a TCP/IP stack for PDP-11 machines. But they dropped TCP/IP in favor of their own private protocol set for their PC LAN.
That sold for a few years, then tanked.
I did considerable work on that product while at Ford Aerospace. Basically, I had to overhaul TCP, and wrote ICMP and UDP from scratch. We used this internally within Ford, but couldn't sell it or give it away, since UNET was proprietary.
Bill Joy's TCP implementation in BSD came years later. But because he was funded to give it away, it became popular, even though it sucked until the second release of 4.3BSD.
ICANN holds meetings as if it were some big-time international organization. It's Esther Dyson's "queen of the Internet" arrogance. They need to downgrade a bit. Here's their July meeting announcement:
You'll be guaranteed a tan this summer if you come to ICANN Kuala Lumpur which takes place from 19-23 July.
Be a part of the meeting and enjoy the beauty and hospitality of Kuala Lumpur where the Malaysian Communications and Multimedia Commission (MCMC), will be your host.
The meeting will take place in the award winning Shangri-La Hotel, Kuala Lumpur. Centrally located in KL's Golden Triangle the meeting venue is centrally located to allow guests easy access to the best of Kuala Lumpur's sights and sounds.
Set against the vibrant backdrop of KL, ICANN Kuala Lumpur will also be an opportunity for you to meet, network and interact with the Malaysian communications and multimedia community and industry.
So book your berth to ICANN Kuala Lumpur by registering early....
There's always lots to see and do in Kuala Lumpur. KL really is the city that never sleeps. From late night latte to night clubs, from some of the world's best scuba diving spots to tropical jungle getaways and some of Asia's best golf courses, you'll find yourself spoilt for choice.
They're all like that. The last four were in Rio, Montreal, Carthage, and Rome.
The RFID tag probably reports the SKU, and maybe the UPC code. UPC codes are generally available, and you can get the SKUs for products of interest by walking through the store. Finding out the top 20 or 100 SKUs a store sells is very valuable info to a competitor.
I heard you on the wireless back in Fifty Two
Lying awake intent at tuning in on you
If I was young it didn't stop you coming through
Oh-a oh
They took the credit for your second symphony
Rewritten by machine and new technology,
and now I understand the problems you can see
Oh-a oh
I met your children
Oh-a oh
What did you tell them?
Video killed the radio star
Video killed the radio star
Pictures came and broke your heart
Oh-a-a-a oh
And now we meet in an abandoned studio
We hear the playback and it seems so long ago
And you remember the jingles used to go
Crosby has a face made for radio. That doesn't make a star any more.
Most of what he's talking about was in Interlisp, the first Really Big Integrated Programming Environment. Integrated debugging. EMACS. Program storage as "workspaces". Extensibility. Intelligent assistants ("DWIM", or "Do What I Mean", a set of heuristics for correcting Warren Titelbaum's most common typing errors.)
The ultimate expression of this was realized with the Symbolics LISP machine. Everything was in LISP. Everything was hackable. The MIT Space Cadet keyboard, with six shift keys (Shift, Ctrl, Meta, Super, Hyper, and Top). All 2^16 keycodes could be bound to any EMACS function.
I've used both. They sucked. Partly because they didn't work very well, but mostly because all that flexibilty and programmability had negative value.
Language and UI design are hard. Evading the problem by making everything changeable does not fix the problem.
His point about XML being another way to put LISP S-expressions into textual form is well taken, though. They're both trees. The problem with LISP is that while the data structures are valuable, the programming notation really is a pain.
LISP works well as a web development environment. Viamall, which later became Yahoo Store, was written in LISP. That was one of the first web applications that really did something elaborate on the server. You could create web pages on the server from a web browser. And the overhead was lower than with XML, where you're forever re-parsing text strings into trees.
The Groom Lake complex was originally built to support the SR-71 test program in the early 1960s, back when that was a big secret. The SR-71, remember, was way ahead of its time. Intercontinental cruise at mach 3, in the early 1960s. As the original poster pointed out, the Groom Lake/SR-71 story was written up in "Skunk Works"
If you want to see an SR-71 up close, the Boeing Museum of Flight has one. They even have a Tagboard drone. They also have the technical documents for the aircraft, and you can read them if you make an appointment.
Sadly, the Skunk Works is no more. The hangars in Burbank are abandoned, as a result of the Lockheed/Martin merger.
Recon aircraft today are less dramatic. We don't need to build something to overfly Russia or Hanoi. What we need is something that can overfly Afghanistan and find bin Laden. That's more of a sensor problem than an airframe problem.
Still, the Air Force has a huge "black budget". It's not at all clear what they're hiding from whom.
The mPhase press release sounds bogus. What seems to have happened is that they licensed a technology for manipulating very tiny drops of fluids from Lucent and then hyped it into a "breakthrough".
There's no indication that they've ever developed a prototype battery at all. It's not clear this approach leads to better battery densities. There's no mention of what battery chemistry they have in mind. They don't even indicate whether this is for rechargeable or primary batteries.
It't not clear that this approach will even get a better surface to area ratio than existing approaches. Or that making a battery in a wafer fab would be cost effective. No way are those guys on track for a product in 2005.
Battery hype has been around for a century. If you've followed the electric car industry, you're familiar with the frustrations of listening to new battery technology claimants. A basic problem is that more powerful battery technologies tend to require more reactive materials, ones further from the center of the electromotive scale. Lithium has been made to work, but it took a long time and a few laptop fires. Sodium-sulfur batteries seem to be too dangerous. There are some workable chemistries, like silver-cadmium, that require overly expensive materials. Thus, there are some high-power battery technologies which have been successfully demonstrated but aren't going mainstream. The mPhase people aren't even at that point.
This is a consistent problem with Piquepaille's blog. He comes across some overhyped press release and writes it up as a "technology trend".
He seems to want to be the next George Gilder, who you may remember as a pundit from the days of dot-com hype.
It's now official. This beast QNX.
(Someone else pointed this out, but didn't provide a link.)
There are many QNX machines around, but most of them are "faceless". Railroad switchyard control. Nuclear power plants. Avionics. And now, big routers.
But it's not getting cheaper
on
Shrek 2 How-To
·
· Score: 3, Interesting
The big headache with all this is that the technology isn't making animated features cheaper.
The project headcounts are still huge.
A few years ago, when I was peripherally involved with the effects industry, everybody was looking forward to the coming era of low-budget effects. "Reboot" and "Starship Troopers" (the TV show, not the movie) seemed to herald the beginning of a new era of feature films at TV production prices.
Didn't happen. The first problem was with live-action directors who didn't understand the inflexibility and costs of CG. As one art director with experience from the pre-computer era put it, "now you can make changes until you run out of money".
Then came the "no limits" problem - "Let's have a drive-through of ancient Rome". Speilberg started it with Jurassic Park, and now everybody expects it in every film. Minor directors plan shots DeMille would have envied. And somewhere, a modelling department has a hundred people busy for months, often for less than a minute of screen time.
The result has been $100M animation budgets. Even "Sky Captain", which was supposed to be a low-budget effects movie, is headed towards that figure. (The production team screwed up, and now ILM is bailing them out. ILM makes a sizable fraction of their money bailing out the botched productions of others.)
It's not about compute power. It's a labor cost issue. It still takes too many bodies to do this stuff.
I wouldn't worry about it. Look how much effort SCO has put into finding infringements, how unsuccessful they've been, and how much trouble they're in now.
Once the SCO case is over, nobody is going to challenge Linux for a long time.
Meanwhile, SCOX is down to 4.74 today. Volume is about a third of the 3-month average; they're falling off the investment radar. IBM's latest set of legal moves put SCO in worst shape than they've been since the litigation started. SCO has an earnings call and webcast on June 2. Tune in and hear Darl try to talk his way out of this one.
General Electric had an actual fusion demonstration at the 1964 World's Fair. Less energy came out than went in, of course.
Forty years later, there's still no useful fusion power technology.
The US Department of Energy is terminating all work on fusion effective September 30, 2004.
That's probably a good thing; it will free up activities in the EU and Japan from US interference.
One advantage of MIDI ringtones is that they don't involve the RIAA. You may need BMI/ASCAP clearance, but you're not reproducing a recording, you're generating a performance. That's far cheaper and there's a compulsory license.
That's a complement to Moore. Leni Riefenstahl is considered the greatest political film producer of all time. Her "Triumph of the Will" won awards in its day, even in France.
What's so striking about "Triumph of the Will" is that there's so little talking. The first words are heard more than twenty minutes into the film.
The early parts of the film show Hitler's aircraft flying over the city where the big rally will be held, and the preparations for the rally. That's all. Yet the imagery and editing are very powerful.
Most propaganda films consist of endless blithering. Compare the US's "Why We Fight" films, which are illustrated lectures.
When she shows someone giving a speech, she cuts them down to four or five sentences. But sequences with movement run in real time out to the end, giving a feeling of unstoppable power.
Riefenstahl is the inventor of modern sports photography, with trackside cameras, slow motion, high angle shots, and background interviews. She also introduced shooting day-for-night, which required the development of special film stock.
Leni died last year, at the age of 101. She released her last film at the age of 100.
As an obituary put it, "Next time you see that footage of President Bush in his flight suit, landing on that aircraft carrier lit by the rented stadium lights, remember that Karl Rove and his ilk owe no small debt to Riefenstahl."
Cell phones need much better bass response for this to work well. A bass driver that works between the front and back of the case would help. Plus, you could use the same device for vibrate mode.
For starters, I'm reading this discussion on a computer running a microkernel. This machine is running QNX 6.2 on a Shuttle 1.5GHz AMD desktop box. The browser is Mozilla 1.6, running under the QNX Photon GUI. It runs about as well as the same version of Mozilla on a comparable Windows machine. Even the same Mozilla bugs show up.
The file systems and networking are user programs.
You can add new file systems; there's one that mounts.zip files, there's NFS, and there's Samba.
In Linux terms, visualize a system where there's the/proc file system for inter-program communication, and everything works through that mechanism.
The drivers really are outside the OS. I've written a FireWire camera driver for QNX, and it's a user program. It's privileged in that it does map some real memory shared by the device, and it can talk to the device directly, so it could potentially cause a crash by making the device write someplace it shouldn't. (That's really a weakness in the PC's I/O architecture; there's no MMU between devices
and memory, for historical reasons dating back to the original IBM PC.)
Debugging a driver is like debugging a normal program. You can even run a driver under a debugger. You can kill a driver while it's running, and it's no big deal. (If you have real memory mapped, it's not recovered until the next boot, so I had to restart my machine about once a week while doing driver development.) Mainframe people have been doing this since the 1960s, but it's rare on PCs.
The basic penalty for using a microkernel is one extra copy and context switch for every file system operation. If your system is doing anything besides I/O, you'll probably never notice. If you're running a web server that serves mostly plain pages (little Perl, Java, PHP, etc.),
you'd probably notice the overhead.
So why are microkernels so rare? They're hard to write well. You can't just hack them together like a UNIX clone. There are some tough design problems to be solved. If those are botched, message passing performance will be terrible.
Message passing and CPU scheduling need to work together. This forces certain design decisions in the scheduler. It's also why adding message passing to an existing system tends not to work well. The Hurd crowd has been thrashing on this issue for a decade. I would have loved to see something as good as QNX from the Hurd people.
But it didn't happen.
Mach didn't really work out as a microkernel. Mach started from 4.3BSD (considered bloated in its day), and versions of Mach below 3 had 4.3BSD in the kernel. MacOS X is not a microkernel system; the BSD stuff is in the kernel. Basically, retrofitting a microkernel architecture to an existing UNIX kernel didn't work.
What you do get from a microkernel like QNX is predictablity. The kernel changes very little and is very reliable. Good microkernels, like QNX and IBM's VM, settle down into versions that almost never change and have very long MTBFs. This brings down total cost of ownership.
This is not the end, but it is the beginning of the end. SCO has stalled for some months, but eventually, the courts won't put up with any more stalling. IBM is now saying that "eventually" has come, and asking the court to rule accordingly.
This is unlkely to result in outright dismissal.
But IBM is also moving, alternatively, for "foreclosure of factual defenses pursuant to discovery rule
37(b)(2)," That's the real "put up or shut up" motion. IBM is saying that SCO has had enough opportunities, after two court orders, to come up with specific evidence of copyright infringement.
So SCO should now be barred from submitting any new evidence of infringement in future.
They'll probably get that. They cite all the embarassing background material, including press statements by SCO and an.MP3 of Darl mouthing off. IBM uses the words "fear, uncertainty and doubt" twice in its latest court filings. But what really matters is that two tightly focused discovery orders have failed to elicit evidence of copyright infringement from SCO. That's so unusual that a ruling by the judge that SCO has had its chance isn't unreasonable.
It introduced the concept of modern phone tech support. Whenever someone has a problem, they open their medicine cabinet and look into a lens, and a concerned voice says "what's wrong?". Later, you see the call center, where someone listens to the complaints and pushes a button to send out the appropriate canned response.
The chase scene's ending (this is a spoiler) is
unique. The chase ends because the people chasing the escaped guy have reached their budget limit and chasing him is no longer cost-effective. So they let him go.
Lucas went on to do some more commercially successful work later, but THX-1138 represents his most original thinking.
Slashdot Mirror
Those terms make Microsoft look good. Does it self-destruct if you don't send in more money each year?
Go with OpenOffice. Less hassle.
A few key points:
I did considerable work on that product while at Ford Aerospace. Basically, I had to overhaul TCP, and wrote ICMP and UDP from scratch. We used this internally within Ford, but couldn't sell it or give it away, since UNET was proprietary.
Bill Joy's TCP implementation in BSD came years later. But because he was funded to give it away, it became popular, even though it sucked until the second release of 4.3BSD.
Stanford has had air-bearing robots to simulate space operations for over a decade. Theirs, though, carry an air tank and work against a flat granite slab.
Be a part of the meeting and enjoy the beauty and hospitality of Kuala Lumpur where the Malaysian Communications and Multimedia Commission (MCMC), will be your host. The meeting will take place in the award winning Shangri-La Hotel, Kuala Lumpur. Centrally located in KL's Golden Triangle the meeting venue is centrally located to allow guests easy access to the best of Kuala Lumpur's sights and sounds.
Set against the vibrant backdrop of KL, ICANN Kuala Lumpur will also be an opportunity for you to meet, network and interact with the Malaysian communications and multimedia community and industry.
So book your berth to ICANN Kuala Lumpur by registering early. ...
There's always lots to see and do in Kuala Lumpur. KL really is the city that never sleeps. From late night latte to night clubs, from some of the world's best scuba diving spots to tropical jungle getaways and some of Asia's best golf courses, you'll find yourself spoilt for choice.
They're all like that. The last four were in Rio, Montreal, Carthage, and Rome.
The RFID tag probably reports the SKU, and maybe the UPC code. UPC codes are generally available, and you can get the SKUs for products of interest by walking through the store. Finding out the top 20 or 100 SKUs a store sells is very valuable info to a competitor.
Present examples like that when talking to retailers. They value their own "privacy". Mall operators hate it when you take pictures of store displays.
Lying awake intent at tuning in on you
If I was young it didn't stop you coming through
Oh-a oh
They took the credit for your second symphony ,
Rewritten by machine and new technology
and now I understand the problems you can see
Oh-a oh
I met your children
Oh-a oh
What did you tell them?
Video killed the radio star
Video killed the radio star
Pictures came and broke your heart
Oh-a-a-a oh
And now we meet in an abandoned studio
We hear the playback and it seems so long ago
And you remember the jingles used to go
Crosby has a face made for radio. That doesn't make a star any more.
The ultimate expression of this was realized with the Symbolics LISP machine. Everything was in LISP. Everything was hackable. The MIT Space Cadet keyboard, with six shift keys (Shift, Ctrl, Meta, Super, Hyper, and Top). All 2^16 keycodes could be bound to any EMACS function.
I've used both. They sucked. Partly because they didn't work very well, but mostly because all that flexibilty and programmability had negative value. Language and UI design are hard. Evading the problem by making everything changeable does not fix the problem.
His point about XML being another way to put LISP S-expressions into textual form is well taken, though. They're both trees. The problem with LISP is that while the data structures are valuable, the programming notation really is a pain.
LISP works well as a web development environment. Viamall, which later became Yahoo Store, was written in LISP. That was one of the first web applications that really did something elaborate on the server. You could create web pages on the server from a web browser. And the overhead was lower than with XML, where you're forever re-parsing text strings into trees.
If you want to see an SR-71 up close, the Boeing Museum of Flight has one. They even have a Tagboard drone. They also have the technical documents for the aircraft, and you can read them if you make an appointment.
Sadly, the Skunk Works is no more. The hangars in Burbank are abandoned, as a result of the Lockheed/Martin merger.
Recon aircraft today are less dramatic. We don't need to build something to overfly Russia or Hanoi. What we need is something that can overfly Afghanistan and find bin Laden. That's more of a sensor problem than an airframe problem.
Still, the Air Force has a huge "black budget". It's not at all clear what they're hiding from whom.
Battery hype has been around for a century. If you've followed the electric car industry, you're familiar with the frustrations of listening to new battery technology claimants. A basic problem is that more powerful battery technologies tend to require more reactive materials, ones further from the center of the electromotive scale. Lithium has been made to work, but it took a long time and a few laptop fires. Sodium-sulfur batteries seem to be too dangerous. There are some workable chemistries, like silver-cadmium, that require overly expensive materials. Thus, there are some high-power battery technologies which have been successfully demonstrated but aren't going mainstream. The mPhase people aren't even at that point.
This is a consistent problem with Piquepaille's blog. He comes across some overhyped press release and writes it up as a "technology trend". He seems to want to be the next George Gilder, who you may remember as a pundit from the days of dot-com hype.
There are many QNX machines around, but most of them are "faceless". Railroad switchyard control. Nuclear power plants. Avionics. And now, big routers.
A few years ago, when I was peripherally involved with the effects industry, everybody was looking forward to the coming era of low-budget effects. "Reboot" and "Starship Troopers" (the TV show, not the movie) seemed to herald the beginning of a new era of feature films at TV production prices.
Didn't happen. The first problem was with live-action directors who didn't understand the inflexibility and costs of CG. As one art director with experience from the pre-computer era put it, "now you can make changes until you run out of money".
Then came the "no limits" problem - "Let's have a drive-through of ancient Rome". Speilberg started it with Jurassic Park, and now everybody expects it in every film. Minor directors plan shots DeMille would have envied. And somewhere, a modelling department has a hundred people busy for months, often for less than a minute of screen time.
The result has been $100M animation budgets. Even "Sky Captain", which was supposed to be a low-budget effects movie, is headed towards that figure. (The production team screwed up, and now ILM is bailing them out. ILM makes a sizable fraction of their money bailing out the botched productions of others.)
It's not about compute power. It's a labor cost issue. It still takes too many bodies to do this stuff.
Meanwhile, SCOX is down to 4.74 today. Volume is about a third of the 3-month average; they're falling off the investment radar. IBM's latest set of legal moves put SCO in worst shape than they've been since the litigation started. SCO has an earnings call and webcast on June 2. Tune in and hear Darl try to talk his way out of this one.
Forty years later, there's still no useful fusion power technology.
The US Department of Energy is terminating all work on fusion effective September 30, 2004. That's probably a good thing; it will free up activities in the EU and Japan from US interference.
One advantage of MIDI ringtones is that they don't involve the RIAA. You may need BMI/ASCAP clearance, but you're not reproducing a recording, you're generating a performance. That's far cheaper and there's a compulsory license.
What's so striking about "Triumph of the Will" is that there's so little talking. The first words are heard more than twenty minutes into the film. The early parts of the film show Hitler's aircraft flying over the city where the big rally will be held, and the preparations for the rally. That's all. Yet the imagery and editing are very powerful. Most propaganda films consist of endless blithering. Compare the US's "Why We Fight" films, which are illustrated lectures.
When she shows someone giving a speech, she cuts them down to four or five sentences. But sequences with movement run in real time out to the end, giving a feeling of unstoppable power.
Riefenstahl is the inventor of modern sports photography, with trackside cameras, slow motion, high angle shots, and background interviews. She also introduced shooting day-for-night, which required the development of special film stock.
Leni died last year, at the age of 101. She released her last film at the age of 100. As an obituary put it, "Next time you see that footage of President Bush in his flight suit, landing on that aircraft carrier lit by the rented stadium lights, remember that Karl Rove and his ilk owe no small debt to Riefenstahl."
This is a job made for outsourcing to India.
NASA has a great PR operation. They should realize that's their core business area and dump the space operations.
Cell phones need much better bass response for this to work well. A bass driver that works between the front and back of the case would help. Plus, you could use the same device for vibrate mode.
It's like sending John Glenn back into orbit. Oh, we did that?
The file systems and networking are user programs. You can add new file systems; there's one that mounts .zip files, there's NFS, and there's Samba.
In Linux terms, visualize a system where there's the /proc file system for inter-program communication, and everything works through that mechanism.
The drivers really are outside the OS. I've written a FireWire camera driver for QNX, and it's a user program. It's privileged in that it does map some real memory shared by the device, and it can talk to the device directly, so it could potentially cause a crash by making the device write someplace it shouldn't. (That's really a weakness in the PC's I/O architecture; there's no MMU between devices and memory, for historical reasons dating back to the original IBM PC.)
Debugging a driver is like debugging a normal program. You can even run a driver under a debugger. You can kill a driver while it's running, and it's no big deal. (If you have real memory mapped, it's not recovered until the next boot, so I had to restart my machine about once a week while doing driver development.) Mainframe people have been doing this since the 1960s, but it's rare on PCs.
The basic penalty for using a microkernel is one extra copy and context switch for every file system operation. If your system is doing anything besides I/O, you'll probably never notice. If you're running a web server that serves mostly plain pages (little Perl, Java, PHP, etc.), you'd probably notice the overhead.
So why are microkernels so rare? They're hard to write well. You can't just hack them together like a UNIX clone. There are some tough design problems to be solved. If those are botched, message passing performance will be terrible. Message passing and CPU scheduling need to work together. This forces certain design decisions in the scheduler. It's also why adding message passing to an existing system tends not to work well. The Hurd crowd has been thrashing on this issue for a decade. I would have loved to see something as good as QNX from the Hurd people. But it didn't happen.
Mach didn't really work out as a microkernel. Mach started from 4.3BSD (considered bloated in its day), and versions of Mach below 3 had 4.3BSD in the kernel. MacOS X is not a microkernel system; the BSD stuff is in the kernel. Basically, retrofitting a microkernel architecture to an existing UNIX kernel didn't work.
What you do get from a microkernel like QNX is predictablity. The kernel changes very little and is very reliable. Good microkernels, like QNX and IBM's VM, settle down into versions that almost never change and have very long MTBFs. This brings down total cost of ownership.
This is unlkely to result in outright dismissal. But IBM is also moving, alternatively, for "foreclosure of factual defenses pursuant to discovery rule 37(b)(2)," That's the real "put up or shut up" motion. IBM is saying that SCO has had enough opportunities, after two court orders, to come up with specific evidence of copyright infringement. So SCO should now be barred from submitting any new evidence of infringement in future.
They'll probably get that. They cite all the embarassing background material, including press statements by SCO and an .MP3 of Darl mouthing off. IBM uses the words "fear, uncertainty and doubt" twice in its latest court filings. But what really matters is that two tightly focused discovery orders have failed to elicit evidence of copyright infringement from SCO. That's so unusual that a ruling by the judge that SCO has had its chance isn't unreasonable.
It introduced the concept of modern phone tech support. Whenever someone has a problem, they open their medicine cabinet and look into a lens, and a concerned voice says "what's wrong?". Later, you see the call center, where someone listens to the complaints and pushes a button to send out the appropriate canned response.
The chase scene's ending (this is a spoiler) is unique. The chase ends because the people chasing the escaped guy have reached their budget limit and chasing him is no longer cost-effective. So they let him go.
Lucas went on to do some more commercially successful work later, but THX-1138 represents his most original thinking.