Last time I looked, modern motherboards have up to six layers of printed circuits. That's the whole point. You can lay down non-conductive plastic, then some wiring, and repeat the process five times for a six layer board. If the traces aren't as fine as from photolithography, you can compensate by using more layers. You can build risers at the same time, creating a monolithic block of circuits equivalent to an entire PC. (Admittedly, getting the ICs mounted in the center may be trickier, but not insurmountable.)
The issue I have (and it's a minor point) is that, based on the melting point of the metal they're using, they're basically using solder for the wiring instead of copper.
New features include, for example, heads that can be changed for different kinds of plastic. A head that deposits low melting-point metal is in development, he says. The metal melts at a lower temperature than that at which plastic melts, which means the metal can be put inside plastic, says Olliver. "That means, in theory, we could build structures like motors." Of course, the main part of a motor usually consists of really long wires wrapped into coils. I'm not sure how well a non-wrapped version would work, but yes, in theory it's possible. More feasible would be building a jig to help me wind my own motor coils.
Also, it sounds like it would be trivial to build a PC "board". It wouldn't have to be flat, and you wouldn't need to etch it. You could have places on your device to surface attach ICs.
Real scientists should shun engineers who warn about. This guy has a completely unverifiable model and feeds garbage information into it. He's trying to predict the likelihood of deterrence failing. But it's never failed, so he has no data to go off of. Not only has it never failed, when we think deterrence has been close to failing, we have no way of knowing how close. By that logic, on the morning January 28, 1986, NASA's management was right to ignore the engineers warning that the Space Shuttle Challenger might explode. Those guys also had an unverifiable model: A shuttle had never failed, so they had no data to go off of. Not only had it never failed, they had no way of knowing how close it had ever come to failing. http://en.wikipedia.org/wiki/Space_Shuttle_Challenger_disaster
I got inducted as a Tau Bate in college, when I was studying EE. A lifetime subscription to 'The Bent' was one of the two biggest benefits (the other being the special ring that gets you free sodas from vending machines).
The Bent usually has great cover articles. Sometimes you get an article of the multi-cnetury history of global position determination, sometimes an explaination of the LIGO project. On rare occasion, you get a Libertarian discussing how things could be changed for the better in modern America; I generally agree with the articles, but realisticly they have as much chance of implementation as Ron Paul has of getting elected. Getting an article into The Bent is a great way to get it in front of some of the brightest people in the world, so I can see why Hellman used it.
Way back when, SGI sent out a CD-ROM that contained a product catalog as a collection of HTML files. The CD also included all of the binaries for Mozilla 0.9 listed at http://www.mcom.com/archives/. For several years thereafter (until browsers became a standard part of software distributions) I kept that CD close at hand; whenever I had to work on a particular workstation or PC, I used it to install a browser (and usually then bootstrap a more recent version).
If this happens, I think it will be confined to the motorways and dual carriageways. I can't see it being rolled out to single carriageway roads. Can someone point me to a good American-British version of Bablefish? I tried reading the referenced Wikipedia articles but discovered that they were written in a foreign language.
I use a laptop with a second LCD monitor. There's also an external keyboard and mouse and a USB hard drive. I leave my laptop on at night so I can remote-desktop into it, so I'm not really happy about putting it in a drawer (no ventilation), plus I don't like the idea of having to 'unharness' everything every time I want to put it away. Two monitors, keyboard and mouse, an extra drive and you don't like to unharness it. Why do you even have a laptop? You need to get an honest-to-gawd beige box.
I don't trust cable locks. Besides, cable locks won't help me secure my the USB drive and other electronics that might wander off. I don't know why you don't trust cable locks. Have you tried to steal something that's attached to one? You'll need a big pair of bolt cutters or thermite or someting equally obvious to defeat it. OTOH, in my cubicle I have a USB hub in a locked drawer, with a cable that's permenently plugged into my docking station. Anything too small for a cable lock goes into the drawer.
How do you pronounce "regex"? I see four possibilities: 1) "regh-ex" (hard 'g', like 'ghost') 2) "rej-ex" (soft 'g', like 'gerbil') 3) "re-gex" (hard 'g') 4) "re-jex" (soft 'g')
I use the first one, since those are the two initial syllables of 'regular' and 'expression', but I can see arguments for the others.
So he wants everyone, especially P2P users, to voluntarily update their TCP stack? Why in the world would a P2P user do that, when they know that (a) older stacks would be supported forever, and (b) a new stack would slow down their transfer rates? I'm sure that if Microsoft pushed an update, it would handle more than half of the P2P community. Over time, when the successor to Vista arrived, you wouldn't have an older stack to fall back upon. Linux might be a bit harder, since old stacks could still float around forever, but there's nothing today that's stopping anyone today from running a stack that has the Jacobson code disabled.
Instead of throttling based on per-host, though, I'd do it per process or process group. Right now, every P2P app that I've seen has a way to schedule bandwidth limits so that you can sure of full speed during the day while downloading at full speed when you're asleep or at work. The problem is, a lot of times you don't conform to your schedule. Then either your websurfing is slow (because you're home from work or just can't sleep) or you aren't downloading as fast as possible (while out you're out drinking with your friends). Per-process throttles would allow the P2P apps to get out of the way as soon as you request a web-page, yet build back up to full speed as soon as you step away from your computer for any reason. That's a stack that i think a lot of people would willingly use.
Simply by opening up 10 to 100 TCP streams, P2P applications can grab 10 to 100 times more bandwidth than a traditional single-stream application under a congested Internet link. [...] The other major loophole in Jacobson's algorithm is the persistence advantage of P2P applications where P2P applications can get another order of magnitude advantage by continuously using the network 24×7. I agree with the first point, but not with the second. One of the whole points of having a computer is that it can do things unattended. Fortunately, the proposal seems to only fix the first issue.
I'd think that a simple fix to Jacobson's algorithm could help a lot. Instead of resetting the transmission rate on just one connection on a dropped packet, reset all of them. This would have no effect on anyone using a single stream, and would eliminate problems with the source of the congestion is nearby. Variations on this theme would included resetting all connections for a single process or process group, which would throttle my P2P without affecting my browser. This alone would be more than enough incentive for me to adopt the patch: instead of having to schedule different bandwidth limits during the day, I could just let everything flow at full speed 24x7. And by putting the patch into the kernel, you'd have less to worry about individual applications and/or users deciding to adopt it.
I hope that that statue of limitations in in effect now, but in case it isn't I'll fuzz a few of the facts. A few years back, I was working for a state office that had a disaster recovery aggreement with the department that handles driver's licenses. So, I was alone in their computer room, and there was a terminal logged into the driver's license database. I did a search of my name, and sure enough there were my records. Then I did searches of several other people, including the governor. At the time, the records included your SSN, but this was before anyone had heard of identity theft so I didn't think anything of it. I didn't take any notes of anything I saw, and cleared the screen before anyone got back. I don't think any investigation was done; at least no one contacted me wanting to know why my records might have been the first ones searched.
Not that that's a bad thing. I don't see why you can't attach it to anything that's a USB master, say a Linux smartphone, or use it as a dumb terminal for a desktop system.
Honestly I think its kind of a cool idea, but the sad part is I don't really see how this could be done on a software level... Why not just do a 'traceroute' to all of the seeds as you discover them, and penalize the ones that are more hops away?
Coincidentally, I wrote about this just last month:
[...] 4) Send 12-25 people, aged 65+, with enough supplies to spend the rest of their lives on Mars. [...] Here's the entire discussion: http://slashdot.org/comments.pl?sid=453282&cid=22424818
Trusting her to log off unsupervised? Pretty obvious one, there, sparky. No, it's not. Megan was still alive when her mother returned home and discovered that Megan hadn't done as she was told. She only killed herself after being sent to her room. And remember, Megan had been closely supervised for months. At some point, her parents had to start trusting her again. The alternative would be to keep her off the Internet until she turned 18. There probably are some people for whom that would be appropriate, but only in hindsight does this seem like such a case.
They trusted a 13 year old girl to do what she was told when she'd previously shown she was untrustworthy in that regard. Bad parenting. Megan had acted out before, her parents clamped down, and they were just starting to give her a bit more freedom again. Not a lot, just a little; the only computer in the house was in the family room and Megan was only allowed to use it with a perent sitting next to her. On Megan's last day, Mom was taking the youngest child to the orthodontist. Mom told Megan to log off, Megan said she would after one more message, and Mom decided to trust her to do so. When Mom got home and discovered that Megan hadn't done as she had said she would, Mom punished her by sending here to her bedroom. Ten minutes later, Megan had hung herself.
So, where's the bad parenting? Should Mom have not punished Megan for breaking the rules? Should they have waited another year or two before letting Megan online again?
They raised a kid that could be bullied into committing suicide at the age of 13. There's something really wrong with that. Yes, but is it the parent's fault? Megan apparently had ADHD. I'm definitely don't believe that every diagnosis of ADHD is correct, but in this case there might have been something wrong with Megan in a genetic sense, not just due to how her parent's raised her. Megan had acted out before, her parents clamped down, and they were just starting to give her a bit more freedom again. Not a lot, just a little; the only computer in the house was in the family room and Megan was only allowed to use it with a perent sitting next to her. On Megan's last day, Mom was taking the youngest child to the orthodontist. Mom told Megan to log off, Megan said she would after one more message, and Mom decided to trust her to do so. When Mom got home and discovered that Megan hadn't done as she had said she would, Mom sent her to her bedroom. Ten minutes later, Megan had hung herself.
if my parents said 'no', they said so for a reason and they meant 'no'. And that reason never changed? They never said 'no' because they were busy with something else at the moment? They never said 'no' because they thought you weren't mature enough, and you decided to prove them wrong? Your parents never just had a bad day, and changed their minds later? The DaVinci Code aside, I didn't think that Jesus had any kids.
Kids don't gain maturity unless they're allowed to leave the nest. You start with small steps, then allow larger excursions as the kids show that they can handle it. Mom and Dad *were* monitoring Megan's online access; they had shut Megan down when she secretly got a MySpace account earlier, and were just starting to let her back on-line with supervision. Unfortunately, there was a "perfect storm" of bad breaks. According to http://stcharlesjournal.stltoday.com/stevepokin/2007_12_01_archive.html#1018547311670903338>:
Tina Meier had left Megan alone on the MySpace account because she had to take her younger daughter to an orthodontist appointment. Megan had promised to sign off as soon as she had finished writing a message. She didn't and things got worse. From everything I've seen, the Meier parents did everything that everyone here has been suggesting, but they lost their daughter anyway.
Is this the sort of close parental supervision you're thinking of?
Tina Meier was wary of the cyber-world of MySpace and its 70 million users. People are not always who they say they are. Tina knew firsthand. Megan and the girl down the block, the former friend, once had created a fake MySpace account, using the photo of a good-looking girl as a way to talk to boys online, Tina says. When Tina found out, she ended Megan's access. [...] As Megan's 14th birthday approached, she pleaded for her mom to give her another chance on MySpace, and Tina relented. She told Megan she would be all over this account, monitoring it. Megan didn't always make good choices because of her ADD, Tina says. And this time, Megan's page would be set to private and only Mom and Dad would have the password. http://stcharlesjournal.stltoday.com/articles/2007/11/10/news/sj2tn20071110-1111stc_pokin_1.ii1.txt
What would you have done differently? Not allowed Megan back on-line? That's an easy idea in retrospect, but growing up did you ever bug your parents over and over about something until they decided to let you do it?
The problem with your switchless proposal is that you presume two ports of network connectivity per node and that all can be dedicated to inter-node communication. It's then hardly useful as there is no way to uplink it to transfer meaningful work/results in and out. I guess your answer would be to slap an ethernet card, but at three nodes, why bother, particularly if you are advocating on board NICs as the interconnect. You're certaining devoting a lot of energy to discrediting a spur-of-the-moment design (you did notice that the post was titled "My next Christmas wish", didn't you?), however it was based upon typical Beowulf clusters. For example, Microwulf (http://www.calvin.edu/~adams/research/microwulf/design/) has just four nodes and uses $40 T-base/1000 for internal communications and a single $15 T-base/100 card installed on one of the ndoes to talk to the outside world. A year ago, the prototypical Microwulf provided 4 dual-core clusters for under $2,500; I was suggesting a similar design supporting 27 cores at a hopefully comparable cost.
Switchs may be cheap, but they ain't free: they cost money, they draw power, and they occupy space. Infiniband doesn't use cross-over cables, so there's no extra cost there; you just need to reconfigure one end of the connection in the software. It's always better to configure your nodes as similarly as possible, so if I were actually going to use Infiniband as an interconnect, I'd probably enable the subnet manager on the first port on all nodes, and disable it on the other port. This would allow easy expansion to a larger loop topology if my hypothtical budget ever allowed for the purchase of more of my fictitious three-socket mother boards.
It has no relevance to a cluster, where it's hard to imagine *any* node interconnect that would care about having 3 vs. 4 nodes in it. Go back and read my post. Node interconnects can be standard switches, where communications between one pair of nodes can delay all of the other nodes, or you can use a cross-bar, where distinct pairs can communicate, but one code can only talk to one other system at a time. Both require additional hardware external to the nodes, which costs money and power.
Or you can use direct connections, wherein a node can always talk to the other end of a wire with no delay. If your mobo has two Ethernet ports, then each node can talk to two others simultaneously, while if it has four then each can talk to three, etc. Unfortunately, each port costs you both money and real estate, so you can't continue the process indefinitely. The mobo that I was describing has two ports, which in a full mesh means that you wind up with a three node cluster.
Yeah, if you have an infinite budget, then by all means buy as many nodes as you can fit into your closet. Those of us in the real world, unfortunately, have limits to our purchasing power, so we need to find the best cost/performance ratio that we can.
IIRC, somebody designed and sells a three socket mobo where all the data paths are also equal. (Ah, here it is: http://hardware.slashdot.org/hardware/07/08/13/1749213.shtml, a three socket Opteron machine with two PCIe slots and two Infiniband 4x ports.) I'd like to see a version for the Phenom 3-core CPUs; even better would be building some sort of Beowulf cluster using three of them, each using a pair of cross-over cables for the interconnects. That would give you one sweet 27-way cluster.
First, Japan is working on using robots to assist an aging population, so the "nurses" don't have to be that young. Also, I assume that there would be follow-up missions delivering more colonists, so the original mission's final years wouldn't be too lonely; if not, we could still return the last two or three people a lot cheaper than a few dozen.
Slashdot Mirror
The issue I have (and it's a minor point) is that, based on the melting point of the metal they're using, they're basically using solder for the wiring instead of copper.
Also, it sounds like it would be trivial to build a PC "board". It wouldn't have to be flat, and you wouldn't need to etch it. You could have places on your device to surface attach ICs.
I got inducted as a Tau Bate in college, when I was studying EE. A lifetime subscription to 'The Bent' was one of the two biggest benefits (the other being the special ring that gets you free sodas from vending machines).
The Bent usually has great cover articles. Sometimes you get an article of the multi-cnetury history of global position determination, sometimes an explaination of the LIGO project. On rare occasion, you get a Libertarian discussing how things could be changed for the better in modern America; I generally agree with the articles, but realisticly they have as much chance of implementation as Ron Paul has of getting elected. Getting an article into The Bent is a great way to get it in front of some of the brightest people in the world, so I can see why Hellman used it.
Way back when, SGI sent out a CD-ROM that contained a product catalog as a collection of HTML files. The CD also included all of the binaries for Mozilla 0.9 listed at http://www.mcom.com/archives/. For several years thereafter (until browsers became a standard part of software distributions) I kept that CD close at hand; whenever I had to work on a particular workstation or PC, I used it to install a browser (and usually then bootstrap a more recent version).
How do you pronounce "regex"? I see four possibilities:
1) "regh-ex" (hard 'g', like 'ghost')
2) "rej-ex" (soft 'g', like 'gerbil')
3) "re-gex" (hard 'g')
4) "re-jex" (soft 'g')
I use the first one, since those are the two initial syllables of 'regular' and 'expression', but I can see arguments for the others.
Instead of throttling based on per-host, though, I'd do it per process or process group. Right now, every P2P app that I've seen has a way to schedule bandwidth limits so that you can sure of full speed during the day while downloading at full speed when you're asleep or at work. The problem is, a lot of times you don't conform to your schedule. Then either your websurfing is slow (because you're home from work or just can't sleep) or you aren't downloading as fast as possible (while out you're out drinking with your friends). Per-process throttles would allow the P2P apps to get out of the way as soon as you request a web-page, yet build back up to full speed as soon as you step away from your computer for any reason. That's a stack that i think a lot of people would willingly use.
I'd think that a simple fix to Jacobson's algorithm could help a lot. Instead of resetting the transmission rate on just one connection on a dropped packet, reset all of them. This would have no effect on anyone using a single stream, and would eliminate problems with the source of the congestion is nearby. Variations on this theme would included resetting all connections for a single process or process group, which would throttle my P2P without affecting my browser. This alone would be more than enough incentive for me to adopt the patch: instead of having to schedule different bandwidth limits during the day, I could just let everything flow at full speed 24x7. And by putting the patch into the kernel, you'd have less to worry about individual applications and/or users deciding to adopt it.
I hope that that statue of limitations in in effect now, but in case it isn't I'll fuzz a few of the facts. A few years back, I was working for a state office that had a disaster recovery aggreement with the department that handles driver's licenses. So, I was alone in their computer room, and there was a terminal logged into the driver's license database. I did a search of my name, and sure enough there were my records. Then I did searches of several other people, including the governor. At the time, the records included your SSN, but this was before anyone had heard of identity theft so I didn't think anything of it. I didn't take any notes of anything I saw, and cleared the screen before anyone got back. I don't think any investigation was done; at least no one contacted me wanting to know why my records might have been the first ones searched.
Not that that's a bad thing. I don't see why you can't attach it to anything that's a USB master, say a Linux smartphone, or use it as a dumb terminal for a desktop system.
Star's next court date is March 21st. Her attorney has moved to have all charges dismissed.
http://www-tech.mit.edu/V128/N1/simpson.html
4) Send 12-25 people, aged 65+, with enough supplies to spend the rest of their lives on Mars.
[...] Here's the entire discussion: http://slashdot.org/comments.pl?sid=453282&cid=22424818
So, where's the bad parenting? Should Mom have not punished Megan for breaking the rules? Should they have waited another year or two before letting Megan online again?
So, where's the bad parenting?
What would you have done differently? Not allowed Megan back on-line? That's an easy idea in retrospect, but growing up did you ever bug your parents over and over about something until they decided to let you do it?
Switchs may be cheap, but they ain't free: they cost money, they draw power, and they occupy space. Infiniband doesn't use cross-over cables, so there's no extra cost there; you just need to reconfigure one end of the connection in the software. It's always better to configure your nodes as similarly as possible, so if I were actually going to use Infiniband as an interconnect, I'd probably enable the subnet manager on the first port on all nodes, and disable it on the other port. This would allow easy expansion to a larger loop topology if my hypothtical budget ever allowed for the purchase of more of my fictitious three-socket mother boards.
Or you can use direct connections, wherein a node can always talk to the other end of a wire with no delay. If your mobo has two Ethernet ports, then each node can talk to two others simultaneously, while if it has four then each can talk to three, etc. Unfortunately, each port costs you both money and real estate, so you can't continue the process indefinitely. The mobo that I was describing has two ports, which in a full mesh means that you wind up with a three node cluster.
Yeah, if you have an infinite budget, then by all means buy as many nodes as you can fit into your closet. Those of us in the real world, unfortunately, have limits to our purchasing power, so we need to find the best cost/performance ratio that we can.
IIRC, somebody designed and sells a three socket mobo where all the data paths are also equal. (Ah, here it is: http://hardware.slashdot.org/hardware/07/08/13/1749213.shtml, a three socket Opteron machine with two PCIe slots and two Infiniband 4x ports.) I'd like to see a version for the Phenom 3-core CPUs; even better would be building some sort of Beowulf cluster using three of them, each using a pair of cross-over cables for the interconnects. That would give you one sweet 27-way cluster.
First, Japan is working on using robots to assist an aging population, so the "nurses" don't have to be that young. Also, I assume that there would be follow-up missions delivering more colonists, so the original mission's final years wouldn't be too lonely; if not, we could still return the last two or three people a lot cheaper than a few dozen.