That patent only covers emulators that know about specific programs and have special built-in tweaks used when those programs are running. If you just do a straightforward emulation of the whole machine, there's no problem.
I noticed that too. That's very bad. It might be a good idea to hold off on using these drives until MTBF data becomes available. You don't want to fill up a server farm with these things and then discover they have a 2000 hour MTBF.
Actually, the motorcycle entry (the Blue Team, from U. C. Berkeley) is the most innovative vehicle there. Not only does it self-balance, but the computer hardware is quite impressive, with FPGAs used for vision.
It's not going to win, and they don't expect it to. It's a technology demonstration. I've met the people doing that project, and they have a coherent vision. They view the Grand Challenge as a beginning, not an end.
I wonder if the CMU vehicle really steered itself.
I suspect that it did. They have the technology. But I wonder about some of the entries that totally blew it on Tuesday, then did much better on Wednesday.
The original QID plans included a moving car-sized
obstacle to be avoided, but no such obstacle was present at the real QID.
The Ohio State monster truck rammed a mini-van (picture) on Tuesday. On Wednesday, it was stopped before running down a course obstacle. And DARPA is letting them attempt the actual event?
The QID was pathetic. We spent two days watching vehicles move around at 1MPH and hit big, obvious obstacles. No way can most of those vehicles operate effectively offroad.
The big design mistakes seem to be these:
Using a laser rangefinder aimed horizontally forward as primary obstacle detection. That doesn't work reliably on either dark or smooth objects. The black mini-van was both.
Using fixed line scanners. If you miss a data point, you're stuck. There's no way to take a second look.
Overreliance on vision. Computer vision in unstructured situations has a very poor track record.
Only CMU is doing well. It's not the money, by the way. Their actual cash outlays are only about $300K to date. It's the body count and the fear. They have about fifty people on the project, a slavedriver boss, and the full backing of CMU.
CMU has to do well; most of the Robotics Institute funding over the last three decades is from DARPA, and DARPA can turn that money off at any time.
Today's buyback announcement boosted the price of SCO stock for about an hour, between 0930 and 1030 EST. But it didn't help. Thursday's high is still below Wednesday's high, which was below Tuesday's high, which was below Monday's high. The stock is down 50% since December. and about 20% this week.
Note, that in typical SCO style, they didn't do a stock buyback, they announced a stock buyback.
It's that [Linux] reacts to marketing pressure to make design decisions favoring running a few processes faster but then finds itself forced first to layer in backward compatibility and then to engage in a patch-and-kludge upgrade process until the code becomes so bloated, slow and unreliable that wholesale replacement is again called for."
That's a good description of Linux inter-application communication. Linux is still stuck with a antiquated pre-object model of interprocess communication that's based on pipes, signals, forking, and sockets. The Linux/Unix world has never been able to come up with a good answer to COM/DCOM/Active-X. CORBA never caught on. The window managers and OpenOffice have totally different approaches to inter-application communication. In typical Linux fashion, there's an attempt to hack a "gateway" between the two,
rather than standardize.
Because of this Mess Underneath, most interprocess communication is done by adding a bloated layer on top, usually at the language level. This leads to hacks like Java RMI, or the Mozilla "platform".
Cut and paste sucks because the infrastructure needed to do it right is missing.
DARPA is seeking to promote innovative technical approaches that will enable the autonomous operation of unmanned ground combat vehicles. In the future, such combat vehicles will operate over varied terrain without the benefit of road signs, pre-programmed routes, etc. Autonomous vehicles must navigate from point to point in an intelligent manner so as to avoid or accommodate obstacles and other impediments to the completion of their missions.
To insure that teams didn't pre-plan, there were these provisions in the original rules:
The Route Definition Data File (RDDF) will be given to all Participants approximately two hours prior to the first Departure Signal at a pre-Challenge brief.
Only commercially available data (maps, images, other cartographic products) may be downloaded to the autonomous or safety vehicles prior to the challenge. Use of GPS is acceptable.
Then, when the general route leaked from the Bureau of Land Management, preplanning got completely out of hand. Now teams could predrive much of the route or overfly it. And they did. Two teams had the route laser-scanned from aircraft. This produced a very detailed topo map, with an elevation point every 25cm or so, along with equally detailed aerial photos.
On top of this, DARPA increased the number of waypoints from 1000 or so to 5000 or so.
At this point, it started to look like a breadcrumb-following exercise.
CMU will manually plan the exact route in the two hours before the race with a team of people at workstations in a big trailer. So even the planning is mostly manual. Their vehicle really does have some autonomous navigation capability, but it only uses it if the route doesn't match the mapped path.
So that's the history. From true autonomy to connect-the-dots.
It's Office that you lose on. Microsoft makes most of their money on Office. Arguably, Microsoft is a company that sells Office; everything else exists to sell Office.
We've had trouble with that. The transmission in the Polaris Ranger won't reliably go into a gear selected by position only. There's too much slop in the linkage. We need to open up the transmission and put a sensor inside it.
We (Team Overbot) dropped out over a month ago. We couldn't deliver a safe vehicle in time. Two of us are flying down tomorrow to watch.
At least two other teams have formally dropped out, and we expect some no-shows.
CMU is the favorite. Fifty people, $3.4 million spent to date, direct support from aerospace companies, and a team leader who expects people to work all night, day after day. (Read the article in the current Scientific American.) But their technology is rather disappointing. The whole route is preplanned by hand, using a bunch of people at workstations in a big trailer with maps obtained by overflying the route with LIDAR-equipped reconnaissance aircraft. It's not very autonomous. They found a loophole in the rules and exploited it very effectively. There's no breakthrough there.
Anthony Lewandosky, with his self-balancing motorcycle, has the most innovative technology.
We've met him, and are impressed.
Palos Verdes High School has a viable entry, using a Honda Acura. We've loaned them some hardware. They've had autonomous driving working for months. They started by having handicapped driving control actuators put into a car, which simplified their mechanical problems. They debugged using a golf cart. Very nice work.
Caltech tried to qualify today, but their vehicle made an unexpected turn and bumped into something. They get a second chance on Wednesday.
Most likely, no one will finish. Nobody has really done enough field testing yet.
See Clarification - CA Signs Licenses.. This Dow Jones story is very negative on SCO, and it appears on most stock-related sites, although it's not up on Bloomberg yet.
SCOX is down 2% today, reaching a new low for 2004.
The stock has been in a screaming dive since December, dropping from 19 to 11.
I once came across an official "U.S. Navy Calculus" textbook. This was written for use during WWII, when there was an urgent need for engineers. It was utterly practical. Integration methods included the "tables method", looking up the appropriate integral in a table of integrals.
After the war, the theory people took over again, of course.
I have a suspicion that Apple is planting these stories. They all mention Apple's iPod prominently, although personal audio players with headphones have been around for twenty years now.
Through the analysis of Walkman use I propose a re-evaluation of the significance of the auditory in everyday experience. I argue that the role of sound has been largely ignored in the literature on media and everyday life resulting in systematic distortions of the meanings attached to much everyday behaviour. Sound as opposed to vision becomes the site of investigation of everyday life in this
article. In focusing thus, I draw upon a range of neglected texts in order to provide a dialectical account of auditory and technologically mediated experience that avoids reductive and dichotomous categories of explanation. I propose a new evaluation of the relational nature of auditory experience whereby users manage their cognition, interpersonal behaviour and social space. The Walkman is perceived as a tool whereby users manage space, time and the boundaries around the self.
So why is the iPod being mentioned so prominently? Because Bull has articles in the New York Times and the Wired catalog. Both articles mention the iPod, but not the Walkman. Both periodicals run iPod ads.
One of the more amusing controversies was over the X-Files "First Person Shooter", which features a VR-type game with a character called "Jade Blue Afterglow". This annoyed Jade-Blue Eclipse, who's a dancer in San Francisco. Jade never collected from Fox, although she tried.
(If you ever get a chance to see Jade in performance, do so. She's a small Asian woman who's very buffed and does performance pieces that show off her strength.)
"Up to" probably means "when nobody else is using the network". How many people share spectrum with this thing? This is, really, a desktop 3G cellular modem.
Also note that data transfer costs $10/gigabyte after the first 20GB (down) or 5GB (up) in a month.
Personally, I think it should be considered false advertising to advertise "up to" anything. Vendors should have to provide a guaranteed minimum.
Missed opportunity for open source
on
Real's Reality
·
· Score: 2, Insightful
So, considering how badly WMP and Real both suck, why haven't the open source players achieved more market share?
Well, let's look at Freeamp. First of all, it's now called "Zinf", continuing the tradition of stupid names for open source programs. Second, when we go to the Zinf home page, we have to click on "Download", one of a number of options (including "SF", which developers know as SourceForge but users do not.)
On the "Download" page, the first option, in typical user-clueless style, downloads the Linux source distribution. You also have to download and build the "MusicBrainz 2.0 client library",
whatever that is. "RPMs will be available soon for RedHat 9". No date is given.
Further down, there's a Windows version, but it's three revs behind. But at least there's an installer and a binary.
If you want to build the thing, there are obscure instructions. ("You'll need perl and NASM in order to compile the latest MP3 decoder assembly optimizations. If you don't have NASM, you can still compile successfully, but you'll only be able to use some of the older optimizations written in gas.")
The Windows version is built with MSVC 5, circa 1997. Builds require some workarounds. ("NOTE: In order for the build to succeed you will need to install the SGI STL. ")
Now consider a typical Windows user. Will they be able to figure out what they're supposed to do?
Or worse, someone who bought a Linux machine at WalMart and wants to run Freeamp, er, Zinf. Will they succeed building this on Thiz Linux? What do you think?
That database should be very useful for finding bad doctors. You can find doctors who have multiple malpractice claims against them.
A few states have databases of disciplinary actions against doctors, but that only shows up the really awful ones. With this database, you should able to find the ones that are merely mediocre, and avoid them.
That's why the Direct Marketing Association lobbied so hard for the CAN-SPAM act. A California law that did just that was going into effect on January 1, 2004. That had the spam industry really scared.
Apple already had A/UX, which ran the MacOS on top of a UNIX kernel, for servers. A/UX was rather slow (4x slower than System 7 for compiles on the same machine), and had some annoying incompatibilities in the filename area, but Apple did have a UNIX-based server product long before NeXT.
The Lisa was a nice little machine. It was just too early. It had a hard drive and a protected mode OS, which wasn't bad for early 1983. You could actually get work done on a Lisa. It just cost far too much.
The original Macintosh (128K, one floppy, and no hard drive) wasn't very useful. You spent most of your time looking at the watch icon and changing floppies. Not until Macs with hard drives came out was it good for much.
And that took years. Apple even fought a company that managed to put a third-party hard drive into original Macs.
Technically, the big problem with the Lisa was that Motorola was years late with the MMU chip for the M68000. The Lisa had an MMU that Apple put together out of register-level parts. This ran up the parts count and the cost. Worse, the M68000 didn't do instruction resumption after page faults correctly. So code for a M68000 with an MMU had to avoid all instructions that could cause page faults after they'd already changed the machine state. This meant avoiding the use of increment bits to increment index registers. If a load with increment page-faulted, the increment would be done twice. So the compiler had to generate code which incremented the index register in a separate operation. This produced code bloat and a slowdown.
No, Copeland did make it to the point that early developer releases went out to selected developers. There's even a book about it, called "MacOS 8".
Copeland didn't fail for technical reasons. The problem was that Microsoft refused to convert Microsoft Word to run on it. That's what held up the transition to a new OS. Until Jobs did the big suck-up deal with Microsoft, Apple was stuck.
The whole Next thing was to justify paying $400M to bail out Jobs. Supposedly, the NeXT acquisition was to provide a new MacOS within a year. It took much longer.
Slashdot Mirror
That patent only covers emulators that know about specific programs and have special built-in tweaks used when those programs are running. If you just do a straightforward emulation of the whole machine, there's no problem.
I noticed that too. That's very bad. It might be a good idea to hold off on using these drives until MTBF data becomes available. You don't want to fill up a server farm with these things and then discover they have a 2000 hour MTBF.
It's not going to win, and they don't expect it to. It's a technology demonstration. I've met the people doing that project, and they have a coherent vision. They view the Grand Challenge as a beginning, not an end.
I suspect that it did. They have the technology. But I wonder about some of the entries that totally blew it on Tuesday, then did much better on Wednesday.
The original QID plans included a moving car-sized obstacle to be avoided, but no such obstacle was present at the real QID.
The QID was pathetic. We spent two days watching vehicles move around at 1MPH and hit big, obvious obstacles. No way can most of those vehicles operate effectively offroad.
The big design mistakes seem to be these:
Only CMU is doing well. It's not the money, by the way. Their actual cash outlays are only about $300K to date. It's the body count and the fear. They have about fifty people on the project, a slavedriver boss, and the full backing of CMU. CMU has to do well; most of the Robotics Institute funding over the last three decades is from DARPA, and DARPA can turn that money off at any time.
John Nagle
Team Overbot
The stock hasn't had a real up week since December of 2003.
Note, that in typical SCO style, they didn't do a stock buyback, they announced a stock buyback.
That's a good description of Linux inter-application communication. Linux is still stuck with a antiquated pre-object model of interprocess communication that's based on pipes, signals, forking, and sockets. The Linux/Unix world has never been able to come up with a good answer to COM/DCOM/Active-X. CORBA never caught on. The window managers and OpenOffice have totally different approaches to inter-application communication. In typical Linux fashion, there's an attempt to hack a "gateway" between the two, rather than standardize.
Because of this Mess Underneath, most interprocess communication is done by adding a bloated layer on top, usually at the language level. This leads to hacks like Java RMI, or the Mozilla "platform".
Cut and paste sucks because the infrastructure needed to do it right is missing.
I will pay $100 for the name and address of the person responsible for that posting.
John Nagle
nagle@overbot.com
To insure that teams didn't pre-plan, there were these provisions in the original rules:
-
The Route Definition Data File (RDDF) will be given to all Participants approximately two hours prior to the first Departure Signal at a pre-Challenge brief.
-
Only commercially available data (maps, images, other cartographic products) may be downloaded to the autonomous or safety vehicles prior to the challenge. Use of GPS is acceptable.
Under the original rules, CMU's approach was prohibited. But that last provision was dropped by DARPA and does not appear in the final rules. (Here's our archive of all the versions of the rules..)Then, when the general route leaked from the Bureau of Land Management, preplanning got completely out of hand. Now teams could predrive much of the route or overfly it. And they did. Two teams had the route laser-scanned from aircraft. This produced a very detailed topo map, with an elevation point every 25cm or so, along with equally detailed aerial photos.
On top of this, DARPA increased the number of waypoints from 1000 or so to 5000 or so.
At this point, it started to look like a breadcrumb-following exercise.
CMU will manually plan the exact route in the two hours before the race with a team of people at workstations in a big trailer. So even the planning is mostly manual. Their vehicle really does have some autonomous navigation capability, but it only uses it if the route doesn't match the mapped path.
So that's the history. From true autonomy to connect-the-dots.
It's Office that you lose on. Microsoft makes most of their money on Office. Arguably, Microsoft is a company that sells Office; everything else exists to sell Office.
We've had trouble with that. The transmission in the Polaris Ranger won't reliably go into a gear selected by position only. There's too much slop in the linkage. We need to open up the transmission and put a sensor inside it.
At least two other teams have formally dropped out, and we expect some no-shows.
CMU is the favorite. Fifty people, $3.4 million spent to date, direct support from aerospace companies, and a team leader who expects people to work all night, day after day. (Read the article in the current Scientific American.) But their technology is rather disappointing. The whole route is preplanned by hand, using a bunch of people at workstations in a big trailer with maps obtained by overflying the route with LIDAR-equipped reconnaissance aircraft. It's not very autonomous. They found a loophole in the rules and exploited it very effectively. There's no breakthrough there.
Anthony Lewandosky, with his self-balancing motorcycle, has the most innovative technology. We've met him, and are impressed.
Palos Verdes High School has a viable entry, using a Honda Acura. We've loaned them some hardware. They've had autonomous driving working for months. They started by having handicapped driving control actuators put into a car, which simplified their mechanical problems. They debugged using a golf cart. Very nice work.
Caltech tried to qualify today, but their vehicle made an unexpected turn and bumped into something. They get a second chance on Wednesday.
Most likely, no one will finish. Nobody has really done enough field testing yet.
John Nagle
Yes, mod parent down. I'm the author of the original article.
SCOX is down 2% today, reaching a new low for 2004. The stock has been in a screaming dive since December, dropping from 19 to 11.
After the war, the theory people took over again, of course.
In fact, Dr. Bull's published papers are more Walkman-oriented:
-
Through the analysis of Walkman use I propose a re-evaluation of the significance of the auditory in everyday experience. I argue that the role of sound has been largely ignored in the literature on media and everyday life resulting in systematic distortions of the meanings attached to much everyday behaviour. Sound as opposed to vision becomes the site of investigation of everyday life in this
article. In focusing thus, I draw upon a range of neglected texts in order to provide a dialectical account of auditory and technologically mediated experience that avoids reductive and dichotomous categories of explanation. I propose a new evaluation of the relational nature of auditory experience whereby users manage their cognition, interpersonal behaviour and social space. The Walkman is perceived as a tool whereby users manage space, time and the boundaries around the self.
But nobody actually reads the Journal of New Media and Society. (Subscriptions are $809 per year.) Nor do they read his book "Sounding Out the City: Personal Stereos and the Management of Everyday Life" (Amazon sales rank: 613,313.)So why is the iPod being mentioned so prominently? Because Bull has articles in the New York Times and the Wired catalog. Both articles mention the iPod, but not the Walkman. Both periodicals run iPod ads.
Any questions?
(If you ever get a chance to see Jade in performance, do so. She's a small Asian woman who's very buffed and does performance pieces that show off her strength.)
Also note that data transfer costs $10/gigabyte after the first 20GB (down) or 5GB (up) in a month.
Personally, I think it should be considered false advertising to advertise "up to" anything. Vendors should have to provide a guaranteed minimum.
Well, let's look at Freeamp. First of all, it's now called "Zinf", continuing the tradition of stupid names for open source programs. Second, when we go to the Zinf home page, we have to click on "Download", one of a number of options (including "SF", which developers know as SourceForge but users do not.) On the "Download" page, the first option, in typical user-clueless style, downloads the Linux source distribution. You also have to download and build the "MusicBrainz 2.0 client library", whatever that is. "RPMs will be available soon for RedHat 9". No date is given.
Further down, there's a Windows version, but it's three revs behind. But at least there's an installer and a binary.
If you want to build the thing, there are obscure instructions. ("You'll need perl and NASM in order to compile the latest MP3 decoder assembly optimizations. If you don't have NASM, you can still compile successfully, but you'll only be able to use some of the older optimizations written in gas.")
The Windows version is built with MSVC 5, circa 1997. Builds require some workarounds. ("NOTE: In order for the build to succeed you will need to install the SGI STL. ")
Now consider a typical Windows user. Will they be able to figure out what they're supposed to do?
Or worse, someone who bought a Linux machine at WalMart and wants to run Freeamp, er, Zinf. Will they succeed building this on Thiz Linux? What do you think?
A few states have databases of disciplinary actions against doctors, but that only shows up the really awful ones. With this database, you should able to find the ones that are merely mediocre, and avoid them.
That's why the Direct Marketing Association lobbied so hard for the CAN-SPAM act. A California law that did just that was going into effect on January 1, 2004. That had the spam industry really scared.
Apple already had A/UX, which ran the MacOS on top of a UNIX kernel, for servers. A/UX was rather slow (4x slower than System 7 for compiles on the same machine), and had some annoying incompatibilities in the filename area, but Apple did have a UNIX-based server product long before NeXT.
The original Macintosh (128K, one floppy, and no hard drive) wasn't very useful. You spent most of your time looking at the watch icon and changing floppies. Not until Macs with hard drives came out was it good for much. And that took years. Apple even fought a company that managed to put a third-party hard drive into original Macs.
Technically, the big problem with the Lisa was that Motorola was years late with the MMU chip for the M68000. The Lisa had an MMU that Apple put together out of register-level parts. This ran up the parts count and the cost. Worse, the M68000 didn't do instruction resumption after page faults correctly. So code for a M68000 with an MMU had to avoid all instructions that could cause page faults after they'd already changed the machine state. This meant avoiding the use of increment bits to increment index registers. If a load with increment page-faulted, the increment would be done twice. So the compiler had to generate code which incremented the index register in a separate operation. This produced code bloat and a slowdown.
Copeland didn't fail for technical reasons. The problem was that Microsoft refused to convert Microsoft Word to run on it. That's what held up the transition to a new OS. Until Jobs did the big suck-up deal with Microsoft, Apple was stuck.
The whole Next thing was to justify paying $400M to bail out Jobs. Supposedly, the NeXT acquisition was to provide a new MacOS within a year. It took much longer.