Well, what do you expect? Bush's appointments have been terrible across the board. His original economic advisers were mostly from Enron. His energy advisers were Cheney's buddies from the oil industry. His head of FEMA, well, we know about that bozo. In all those areas, the government is doing a poor job. Why should cybersecurity be doing better?
It's a real problem. The President's key job is is appoint the top people in the federal government, about 3000 of them. That determines how well the Government works. This is not one of the better administrations in that area.
It's not a party thing. Some presidents do a good job in this area, and some don't. Eisenhower was very good, Kennedy was good at it, Johnson was OK, Nixon was terrible, Ford wasn't around long enough to matter, Carter was mediocre, Reagan was spotty, Bush I was OK, Clinton was OK, and Bush II is terrible.
Internal hard drives didn't come 'til, I want to say Mac II? Was there one for the SE?
There was a Mac SE FDHD, but that wasn't until 1989. The Macintosh Plus, released in 1986, was the first Mac with a SCSI port, and Apple sold a matching hard drive. That was the first Mac with a supported hard drive. There had been previous third-party attempts to add a hard drive, but they either required internal mods or worked, slowly, through the printer port.
Multifinder originally was an add-on for System 4. It came standard with system 6 but wasn't installed by default. Not until System 7 was it standard. Many early apps wouldn't run under MultiFinder, and the transition generally required application upgrades. Internally, it was a horror, since it began as an external add-on above a single-thread OS, not as a scheduler installed in the core system. Developers used to refer to the Mess Inside.
It's worth remembering that the original Macintosh was a flop. The attempt to cost-reduce the Lisa resulted in a machine too weak to do much of anything. Remember the original specs: 128K, no hard drive, one floppy. Ever use one? Ever actually try to get work done on one? You had to fit the OS, the app, and your documents on one floppy. Or you could get an external floppy, which made the thing marginally useable. It was cute, but not productive.
The lack of a hard drive was the killer. By the time the Mac came out, IBM PCs had a hard drive, so Apple was playing catch-up. Apple had tried building hard drives (the LisaFile), but they were slow and crashed frequently. But at least the Lisa had a hard drive.
Third parties added a 10MB hard drive to the Mac in early 1985, which brought performance up to an acceptable level. Some people say that third-party hard drives saved the Mac. But Apple fought them tooth and nail. Apple finally came out with a 20MB external hard drive for the Mac in 1986. This was very late; IBM PCs had been shipping with hard drives for five years.
Sales for the Mac were well below expectations. Apple had been outselling IBM in the Apple II era. (Yes, Apple was once #1 in personal computers.) In the Mac era, Apple's market share dropped well below that of IBM.
What really saved the Mac was the LaserWriter, which launched the "desktop publishing" era. But that required a "Fat Mac" with a hard drive and 512K.
By then, the Mac had reached parity with the Lisa specs, except that the Lisa had an MMU and the Mac didn't. The Lisa also had a real operating system, with protected mode processes; the Mac had "co-operative multitasking" in a single address space, which was basically a DOS-like system with hacks to handle multiple psuedo-threads.
The MMU issue was actually Motorola's fault. The 68000 couldn't do page faults right, and Motorola's first MMU, the Motorola 68451, was a terrible design. The Lisa had an Apple-built MMU made out of register-level parts, which pushed the price up.
Apple might have been more successful if they'd just stayed with the Lisa and brought the cost down as the parts cost decreased. They would have had to push Motorola to fix the MMU problem, but as the biggest 68000 customer, they could have.
Here's a better idea, and one that's quite buildable -
the two piece jewelry phone. One piece is an earring; that's the speaker end. The other is a necklace or choker; that has the microphone. Voice controlled, using something like Wildfire's semi-intelligent interface.
No ring tones; a subtle whisper tells the user who's calling, and a head-toss quietly diverts the call to voicemail. Finally, an elegant phone.
Under the rules, slow vehicles might be stopped overnight and restarted in the morning. That's what "October 9th operations possible" probably means. TerraMax is still chugging away, at 77 miles. Grey Team, at 120 miles, is coming up on the tough mountain pass with the steep dropoff. Neither will make it under 10 hours, but it's worth letting them finish.
There's someone in a pickup following each 'bot, carrying an emergency stop transmitter. Some of those roads you don't want to drive at night.
The rules don't allow for a tie-breaker. First 'bot to finish in under 10 hours wins.
It's not clear why the displayed timing information is so bad. Each 'bot has a DARPA-supplied box which handles emergency stop, tracking, and timing. The units have two radio links (1W at 900MHz for emergency stop, and an Iridium satellite uplink for tracking), plus onboard logging of all events. When the 'bot gets a PAUSE command, the clock stops running at the moment the RUN signal to the bot's onboard electronics goes low; on remote restart, it continues. That timing info goes out over the radio link, and the E-stop and tracking systems can read it remotely. That's the official timing. But that timing info doesn't seem to be what's driving the "grandchallenge.org" web site, because the diplayed time keeps increasing on the bots that finished hours ago.
Stanford is about half an hour ahead on time, so they win. (The numbers on the DARPA tracking site are wrong; the clocks are still running for the teams that have finished.) Insight Racing, TerraMax, and the Grey team are still running, but they're not even half way and can't finish within ten hours.
Stanford's vehicle is through the last hard section and out of the mountains. It's all straight, flat power line road to the finish now. Four miles to go.
Ten miles to go, three entries (Stanford and both CMU vehicles) are close together, with Stanford well in the lead. Now approaching the section with the one-lane dirt road with the 1000 foot drop on one side. Is there a live video feed of this area anywhere?
As a team leader of one of the teams eliminated at the NQE, I didn't see any visible favoritism by the DARPA staff. The teams that went to Primm are the teams that should have gone.
Funding is more of an issue. Teams were supposed to have no Government funding whatsoever, either direct or indirect. Yet MITRE had a team, and they're a quasi-governmental agency.
CMU has received DARPA robotics contracts for years, as has Stanford. Red Whittaker of the CMU team is still the principal investigator on a NASA grant (#NAG5-12890) until February 2006. Stanford used software developed under DoD contract, although anyone can download it and they asked DARPA for permission.
It's more of a revolving-door issue than direct diversion of Government funds.
But the real incentive for the big university teams was fear. If Joe's Auto Parts fielded a better robot than some university getting $20 million a year in robotics funding from DARPA, DARPA might well pull the plug on the school. CMU faced that prospect; originally, they weren't going to enter the Grand Challenge at all. The whole Grand Challenge was created because of unhappiness at DARPA with the rate of progress in mobile robotics. DARPA has been pouring robotics money into CMU and Stanford for thirty years, without getting much back. The head of DARPA, Dr. Tony Tether, decided that it was time to do something about that. It worked.
"You know what makes rockets fly? Funding." - The Right Stuff.
What's really making this go is not new technology, but money. Most of the designs are quite straightforward. But nobody in the US has ever spent money at this rate in robotics research. CMU spent $3 million last year, and this year the total costs of their efforts were much higher. The major teams have direct engineering support, including on-site people, from major auto and aerospace companies. Huge field test and support operations have made it possible to pound existing technology into working shape by sheer debugging effort.
Most entries have a bunch of LIDAR line scanners as their primary sensors. Some also have vision systems, but usually those systems don't do much. At best, they're just depth from stereo or road following. Nobody has image understanding. But with enough hammering, those technologies can be made to do the job, more or less.
Anyway, it was great fun being part of it. But we won't do it again. We'll leave that to the organizations that get Government funding. For them, it's a marketing expense.
I thought that the new features on the upgraded satellites included additional carriers (L2-C/A and L5) for civil use, which would allow for measurements of frequency-dependent ionospheric delay.
Around 2010-2013 that should become possible, as enough new sats are launched.
I used to dream when I was younger of "cheap cheap energy" and how much society would be affected.
And you got to see it. Suburbs, interstates, SUVs, detached houses for the masses, low-cost air travel, aluminum cans, plastics, all made possible by cheap oil. And now that's over.
The next 50 years may look more like the first half of the 20th century. Big apartment buildings, living near work, trains, glass bottles, flying as a luxury. All the little low-power gadgets we have now will be around, but US-type urban sprawl just won't be affordable any more.
In this new release, the installer is in Java. And apparently only some versions of Java work. Guess whose.
Each new version of StarOffice seems to have more dependencies on Sun's Java. This is not good for OpenOffice.
It's not Java, per se, that's the problem. It's the dependency of open source software on closed source software, the evil that Stallman always warns about. You don't want someone to be in a position where they can cut off your air supply.
We've recently been struggling with a Novatel GPS that receives Omnistar High Precision corrections, and supposedly provides 15cm accuracy. The problem is that it needs to see at least five GPS satellites for Omnistar HP to work. Regular GPS requires only four, but the ionospheric corrections for Omnistar require some redundancy. Five sats are the minimum; six are better.
Unless you're in a very flat area, in the air, or on an ocean, you won't see five or six sats 100% of the time. 70-80% is more like it. If one of the sats is down (which happens; PRN #5, plane B, slot 4, wss down for 8 days recently), the outages are longer.
GPS uses six rings of four satellites each, with all rings in polar orbit. The four satellites in each ring are 90 degrees apart. So, when a satellite in a ring is near the zenith, it's usually the only one visible in that ring. The original design called for more satellites per ring; with six per ring, you'd always have at least two satellites visible per ring, as long as you could see to within 30 degrees of the horizon. But there was a budget cut in the early days of GPS.
What he's really describing is Microsoft's "zero administration PC" concept from 1997, where Central Control in the IT department controls everything on the machine. The user interface concept is borrowed from Go Computer's tablet machine from 1994. The I-opener, circa 1999, had many of the same concepts.
All were failures.
The key line in his article is this:
Because nearly every Appliance will generate a $200-600 per year DSL or cable modem revenue stream for the carrier...
This sounds like a fantasy written to entertain telco executives.
What's more likely to happen is the return of the "Internet appliance", but at a really low price point comparable to a DVD player.
Now that was dumb. Polycarbonates are strong, but not hard. The eyeglass lens industry solved this problem years ago. Even the bus window industry has solved this problem. Optical polycarbonate surfaces are routinely hard-coated, and an anti-glare coating is often added at the same time.
The cool solution, which Apple probably now has to use to get their reputation back, is sapphire. That's what scratch-resistant high-end watches use. Put an 0.15mm sapphire layer on top of the polycarbonate, and you can dump the thing in with your keys without worrying. It's not that expensive for a phone or music player sized screen. Some of Nokia's high-end phones have a sapphire screen.
Of course, doing it right might cut into those 40% profit margins at Apple.
"We have to keep thinking how we are going to monetize our product for our shareholders. We are the arms supplier in the device wars between Samsung, Sony, Apple, and others."
- Bronfman, Warner Music Group CEO
Slashdot Mirror
Developers, you don't get to check in code until you've watched the video of users struggling with your program. OK?
Could be worse. There are winding machines for self-winding watches. Little powered turntables. Really.
It's a real problem. The President's key job is is appoint the top people in the federal government, about 3000 of them. That determines how well the Government works. This is not one of the better administrations in that area.
It's not a party thing. Some presidents do a good job in this area, and some don't. Eisenhower was very good, Kennedy was good at it, Johnson was OK, Nixon was terrible, Ford wasn't around long enough to matter, Carter was mediocre, Reagan was spotty, Bush I was OK, Clinton was OK, and Bush II is terrible.
There was a Mac SE FDHD, but that wasn't until 1989. The Macintosh Plus, released in 1986, was the first Mac with a SCSI port, and Apple sold a matching hard drive. That was the first Mac with a supported hard drive. There had been previous third-party attempts to add a hard drive, but they either required internal mods or worked, slowly, through the printer port.
Multifinder originally was an add-on for System 4. It came standard with system 6 but wasn't installed by default. Not until System 7 was it standard. Many early apps wouldn't run under MultiFinder, and the transition generally required application upgrades. Internally, it was a horror, since it began as an external add-on above a single-thread OS, not as a scheduler installed in the core system. Developers used to refer to the Mess Inside.
Schwarzenegger's "True Lies" has all of A through E. Any questions?
The lack of a hard drive was the killer. By the time the Mac came out, IBM PCs had a hard drive, so Apple was playing catch-up. Apple had tried building hard drives (the LisaFile), but they were slow and crashed frequently. But at least the Lisa had a hard drive. Third parties added a 10MB hard drive to the Mac in early 1985, which brought performance up to an acceptable level. Some people say that third-party hard drives saved the Mac. But Apple fought them tooth and nail. Apple finally came out with a 20MB external hard drive for the Mac in 1986. This was very late; IBM PCs had been shipping with hard drives for five years.
Sales for the Mac were well below expectations. Apple had been outselling IBM in the Apple II era. (Yes, Apple was once #1 in personal computers.) In the Mac era, Apple's market share dropped well below that of IBM.
What really saved the Mac was the LaserWriter, which launched the "desktop publishing" era. But that required a "Fat Mac" with a hard drive and 512K. By then, the Mac had reached parity with the Lisa specs, except that the Lisa had an MMU and the Mac didn't. The Lisa also had a real operating system, with protected mode processes; the Mac had "co-operative multitasking" in a single address space, which was basically a DOS-like system with hacks to handle multiple psuedo-threads.
The MMU issue was actually Motorola's fault. The 68000 couldn't do page faults right, and Motorola's first MMU, the Motorola 68451, was a terrible design. The Lisa had an Apple-built MMU made out of register-level parts, which pushed the price up.
Apple might have been more successful if they'd just stayed with the Lisa and brought the cost down as the parts cost decreased. They would have had to push Motorola to fix the MMU problem, but as the biggest 68000 customer, they could have.
Here's a better idea, and one that's quite buildable - the two piece jewelry phone. One piece is an earring; that's the speaker end. The other is a necklace or choker; that has the microphone. Voice controlled, using something like Wildfire's semi-intelligent interface. No ring tones; a subtle whisper tells the user who's calling, and a head-toss quietly diverts the call to voicemail. Finally, an elegant phone.
This is impressive. Four successful finishers in under 10 hours.
There's someone in a pickup following each 'bot, carrying an emergency stop transmitter. Some of those roads you don't want to drive at night.
The rules don't allow for a tie-breaker. First 'bot to finish in under 10 hours wins.
It's not clear why the displayed timing information is so bad. Each 'bot has a DARPA-supplied box which handles emergency stop, tracking, and timing. The units have two radio links (1W at 900MHz for emergency stop, and an Iridium satellite uplink for tracking), plus onboard logging of all events. When the 'bot gets a PAUSE command, the clock stops running at the moment the RUN signal to the bot's onboard electronics goes low; on remote restart, it continues. That timing info goes out over the radio link, and the E-stop and tracking systems can read it remotely. That's the official timing. But that timing info doesn't seem to be what's driving the "grandchallenge.org" web site, because the diplayed time keeps increasing on the bots that finished hours ago.
So it's over.
Autonomous vehicles will never be a joke again.
The two CMU vehicles have made it through the last tough parts, and they should finish in ten to fifteen minutes.
Four other teams are still running, but are too slow. None are halfway yet.
Stanford's vehicle is through the last hard section and out of the mountains.
It's all straight, flat power line road to the finish now. Four miles to go.
Ten miles to go, three entries (Stanford and both CMU vehicles) are close together, with Stanford well in the lead. Now approaching the section with the one-lane dirt road with the 1000 foot drop on one side. Is there a live video feed of this area anywhere?
Funding is more of an issue. Teams were supposed to have no Government funding whatsoever, either direct or indirect. Yet MITRE had a team, and they're a quasi-governmental agency. CMU has received DARPA robotics contracts for years, as has Stanford. Red Whittaker of the CMU team is still the principal investigator on a NASA grant (#NAG5-12890) until February 2006. Stanford used software developed under DoD contract, although anyone can download it and they asked DARPA for permission. It's more of a revolving-door issue than direct diversion of Government funds.
But the real incentive for the big university teams was fear. If Joe's Auto Parts fielded a better robot than some university getting $20 million a year in robotics funding from DARPA, DARPA might well pull the plug on the school. CMU faced that prospect; originally, they weren't going to enter the Grand Challenge at all. The whole Grand Challenge was created because of unhappiness at DARPA with the rate of progress in mobile robotics. DARPA has been pouring robotics money into CMU and Stanford for thirty years, without getting much back. The head of DARPA, Dr. Tony Tether, decided that it was time to do something about that. It worked.
If Level 3 doesn't want to peer with Cogent, that's fine. If they export incorrect routing information to gain a business advantage, that's fraud.
What's really making this go is not new technology, but money. Most of the designs are quite straightforward. But nobody in the US has ever spent money at this rate in robotics research. CMU spent $3 million last year, and this year the total costs of their efforts were much higher. The major teams have direct engineering support, including on-site people, from major auto and aerospace companies. Huge field test and support operations have made it possible to pound existing technology into working shape by sheer debugging effort.
Most entries have a bunch of LIDAR line scanners as their primary sensors. Some also have vision systems, but usually those systems don't do much. At best, they're just depth from stereo or road following. Nobody has image understanding. But with enough hammering, those technologies can be made to do the job, more or less.
Anyway, it was great fun being part of it. But we won't do it again. We'll leave that to the organizations that get Government funding. For them, it's a marketing expense.
John Nagle
Team Overbot.
Around 2010-2013 that should become possible, as enough new sats are launched.
And you got to see it. Suburbs, interstates, SUVs, detached houses for the masses, low-cost air travel, aluminum cans, plastics, all made possible by cheap oil. And now that's over.
The next 50 years may look more like the first half of the 20th century. Big apartment buildings, living near work, trains, glass bottles, flying as a luxury. All the little low-power gadgets we have now will be around, but US-type urban sprawl just won't be affordable any more.
Each new version of StarOffice seems to have more dependencies on Sun's Java. This is not good for OpenOffice.
It's not Java, per se, that's the problem. It's the dependency of open source software on closed source software, the evil that Stallman always warns about. You don't want someone to be in a position where they can cut off your air supply.
Unless you're in a very flat area, in the air, or on an ocean, you won't see five or six sats 100% of the time. 70-80% is more like it. If one of the sats is down (which happens; PRN #5, plane B, slot 4, wss down for 8 days recently), the outages are longer.
GPS uses six rings of four satellites each, with all rings in polar orbit. The four satellites in each ring are 90 degrees apart. So, when a satellite in a ring is near the zenith, it's usually the only one visible in that ring. The original design called for more satellites per ring; with six per ring, you'd always have at least two satellites visible per ring, as long as you could see to within 30 degrees of the horizon. But there was a budget cut in the early days of GPS.
Misery is a legacy system written in Perl. COBOL is reasonably readable.
All were failures.
The key line in his article is this: Because nearly every Appliance will generate a $200-600 per year DSL or cable modem revenue stream for the carrier ...
This sounds like a fantasy written to entertain telco executives.
What's more likely to happen is the return of the "Internet appliance", but at a really low price point comparable to a DVD player.
The cool solution, which Apple probably now has to use to get their reputation back, is sapphire. That's what scratch-resistant high-end watches use. Put an 0.15mm sapphire layer on top of the polycarbonate, and you can dump the thing in with your keys without worrying. It's not that expensive for a phone or music player sized screen. Some of Nokia's high-end phones have a sapphire screen.
Of course, doing it right might cut into those 40% profit margins at Apple.
With one show listed per page, and surrounded by ads, this is a click troll. Somewhere, somebody is getting pay per click payments for this turkey.
"We have to keep thinking how we are going to monetize our product for our shareholders. We are the arms supplier in the device wars between Samsung, Sony, Apple, and others." - Bronfman, Warner Music Group CEO