The only real downside is that if this (or another) vaccine is effective and reliable, then there's the risk of other STDs becoming more prevelant again as people relax their safe sex practices. That includes unplanned pregnancies. Some people really do need a hypothetical gun to their heads to think about using condoms or monogamy.
Hate to burst your bubble, but most people I know don't use condoms to avoid disease, they use them to avoid pregnancy. Condoms only reduce the transmission of a subset of STDs. Crabs and herpes are just a couple of examples that condoms won't block.
That's essentially what we do...keep machines disconnected unless they need to be. Unfortunately, that adds a bottleneck to our servicing process, which is why I'm looking for a higher-level solution.
iptables (and other filtering software) would only be effective if the bad packets were required to pass through the machine with the filtering on it. The problem is, if two machines are on the same subnet, they can communicate directly with each other, ignoring the machine with the filtering software on it.
One possible solution is to force each machine attached to the network onto its own subnet. The problem is, I don't know of a dhcpd that will put machines on different subnets without a means of identifying which machines should go to which subnet. And then there's the issue of creating alias IP addresses on the filter/gateway machine for each subnet in use. (Otherwise, the "patient" computer won't be able to access the network at all.)
Of course, a decent Cisco switch should be able to do it...but by the time any student at our college knows how to work Cisco, he's already moved on to a four-year institution. Hmm...maybe I should invite GVSU's or Davenport's Computer Club equivalents to participate in our clinic. I've got contacts...
Funny how the timing works out. One of the U3 techs stopped in here, and responded to comments and questions. Interesting answers.. (And yes, I made a fool of myself at the beginning.)
It'll likely remain in systems, just rarely used or updated. Like floppy drives, serial ports parallel ports, the AT keyboard architecture (of which PS/2 keyboards are essentially a clone), and CGA and EGA video modes.
To cut costs, it'll get integrated into the northbridge/southbridge pair of chips. The x86 system is (in)famous for its support and occasional dependancy on legacy systems. Did you know that you can still run MS DOS on most modern computers?
Well, systems are only connected to our network for a few hours at most. Less, if we see traffic that bothers us. Like this last time, two of the machines started scanning all the IPs on the class C subnets adjacent to the subnet we were using. We put a stop to that. The only botnet activity I saw was repeated attempts to connect to the IRC port of a domain name. However, that domain had expired, so the bots couldn't connect.
I'm looking around for a way to prevent machines on our network from talking to each other...putting each one on its own subnet seems like a good idea, but I don't know how to set up Linux dhcp to do it.
At my computer club's PC Clinic, I set up Ethereal on our network gateway computer, to keep track of things. You can easily see this kind of crap going on.
When I was a teenager (and for a couple years after I turned 18), I was the primary phone tech guy for the family business, an ISP. I wasn't the one to receive the worst, call, fortunately...that pain fell to my mother, who was the other phone tech.
Here's a rough transcript of that conversation:
"Cyberspace, this is Darcy"
"Yeah, I'm having problems getting on the Internet."
Could be the modem was no longer capable of handling normal signal voltages.
As for the ASCII art...didn't you use hardware error correction? Granted, it was annoying when the data transfer paused for no apparent reason, but it was usually better than getting a quick demo of the extended ASCII character set and control characters.
BTW...Some old-school BBSes are still around. Try telnetting to grnet.com.
The only times I've seen magic smoke were when capacitors blew. And sometimes, those capacitors weren't really necessary, anyway.
Showering the system helped get all the conductive goo off the board before you applied power.
Water isn't really dangerous to electronics; It's the ions found in unpure water that makes it conductive. I'm still waiting for someone to give a PC a really thorough cleaning job, and demonstrate real water cooling by submerging a system into a cleaned fishtank full of deionized water. (The cleaning job is required to get all the dust and miscellaneous substances off the computer. If they dissolve, the fishtank water will no longer be deionized, and would start conducting again.)
I would bet it had to do with the signal timing on the bus. If one of the cards absolutely had to be seen by the PCI controller before the other, it could be important that that card lay closer on the bus to the PCI controller than the other, and that there be a minimum distance between the two cards.
What exactly does a nubus card do, anyway?
Re:One piece of hardware I'll never understand:
on
Computer Voodoo?
·
· Score: 1
When I was a runt, I had both a SEGA 32X and a Sega Channel adapter. Unfortunately, you couldn't use them both at the same time. I had to remove the 32X whenever I wanted to use Sega Channel, and reinstall it when I wanted to go back to playing Star Wars 32X or Tempo.
Well, I got tired of putting in those little metal plates and stopped bothering with them. Within a couple uses, the 32X was dead as a doornail. I didn't realize until years later how they could be important to the functioning of the device.
(Darn, I miss the 32X. And Sega Channel. Two of the neatest things an 11-year-old could have. Long live Road Rash!)
"Freeze" might be a more apt term for it than you know. Sounds like the fault could have been dependant on the temperature of the CPU/Heatsink and/or motherboard. That could mean anything from a faulty solder connection to a problem with the CPU itself. (You didn't overclock it at one point, did you?)
I could easily see a couple of bad ground pins causing a high enough current flow in the rest of them to cause heat issues on the motherboard. A subsequent cool down could cause some now-soft solder connections to break contact. On the other hand, silicon dies are finicky things...overheating could cause problems internal to a rarely-used section of the CPU that only manifest themselves when heat isn't applied.
A real art major also knows the difference between color displays based on adding and subtracting primary colors. The ones who use computers for their projects actually apply these principles.
I used to have an old pentium (133 I think) that ran well, except that the CD drive would only actually recognize a disk if you tilted the computer at about a 20 to 30 degree angle when the disk was inserted. I never did figure out why this fixed it, luckily I didn't need to use the cd drive very often.
The assembly that latches onto the CD might be loose. Tilting it could do anything from fixing a faulty wire connection to aligning the read laser just right.
I also used to have a cable modem that would drop the connection if you so much as blew on the power cord. I always just figgured that was just some flaky hardware, and eventually got the cable company to replace it.
Bad electrical connection. Either the board was developng a problem at the power connector, or there was a frayed wire in your power cable. I've run into both problems. The first condition happened with the laptop I'm typing this on...tension on the cable caused the soldering connections to break. A friend resoldered it, and now it works fine. I see the second condition all the time...people in the computer lab keep kicking the wiring, causing stress.
Another really aggrevating hardware problem that I never figgured out was an old Sony DVD drive that I had. When you opened the tray, it would about 1 to 2 seconds later automatically close the tray, but when you opened it again it would stay open for about 10 seconds, just long enough to remove or insert a disk.
Sounds like there was something loose inside the drive, or there might have been a bad gear. If something causes the drive to think the tray isn't opening all the way, it will draw the tray back in to avoid damage.
My brother had that problem with WoW on this laptop. Don't think he ever tried opening the CD-ROM drive, though. He's "given" it to me so long as I lose weight. (40 lbs down and counting. Woohoo!)
For a good while, I ran it underclocked to 1.06GHz from the max of 1.79GHz. Then I had a friend service a bad part on it, and found out what the problem was. The inside of the system was caked with dust, with the fans and heat sink being especially bad. He cleaned all that crap out, and now the system runs happily at full speed.
Don't forget those bad solder connections on the CRT (The tube itself, not the monitor as a whole). When those things get hot enough, the color starts to flicker. A good thwack seems to do the trick.
Wanna impress an art major? Tell them which color signal went out. Then fix it. "Everything's magenta? Guess Green went out. Watch this. Don't *whack* you *whack* ever *whack* do *whack* that *whack* again *whack*."
Well, leave out the theatrics if the art major is female...
Half the students that come through my lab don't even know how to use email. And learning how is not, unfortunately, a prerequesite to classes in departments other than Computer Apps. And not even for most Computer Apps classes.
It's a community college. I'd wager at least a third of my student encounters are where the student doesn't even speak English fluently. Ninety eight percent of my student encounters are with students never used a computer before college, or only know enough to use AIM and Myspace.
I wish that untouched two percent of students asked more questions. I really do.
Again, the other possibility might be the way drive letters are in use on the systems at GRCC.
The college I work at has a large number of letter drives assigned. The first partition of the U3 device comes up as the F drive. The G drive is a globally-assigned network drive that supports a large number of applications. If something tries to create the "next" letter drive, it tries to count up from the local drives already created. Since the G drive is in use as a network drive, the operation fails.
We used to have that problem with normal flash drives, as the F drive had been assigned as a network drive. When flash drives became prevalent, they moved the F drive data somewhere else, so it wouldn't have to be disconnected before the flash device was inserted. Now the same thing may be happening with the G drive.
Unfortunately, the same solution isn't as easy with the G drive as with the F drive; there are a much larger number of applications, with a much higher demand, on the G drive than there were with the F drive. Shifting the drive letters up doesn't seem to be possible; some roles on the network already have X, Y and Z drives. (Mostly instructors.)
Drop me an email...if you can help me find a better solution for this, we'd both be better off.
Slashdot Mirror
The only real downside is that if this (or another) vaccine is effective and reliable, then there's the risk of other STDs becoming more prevelant again as people relax their safe sex practices. That includes unplanned pregnancies. Some people really do need a hypothetical gun to their heads to think about using condoms or monogamy.
Hate to burst your bubble, but most people I know don't use condoms to avoid disease, they use them to avoid pregnancy. Condoms only reduce the transmission of a subset of STDs. Crabs and herpes are just a couple of examples that condoms won't block.
Log. It's better than bad...it's good!
That's essentially what we do...keep machines disconnected unless they need to be. Unfortunately, that adds a bottleneck to our servicing process, which is why I'm looking for a higher-level solution.
Sortof.
iptables (and other filtering software) would only be effective if the bad packets were required to pass through the machine with the filtering on it. The problem is, if two machines are on the same subnet, they can communicate directly with each other, ignoring the machine with the filtering software on it.
One possible solution is to force each machine attached to the network onto its own subnet. The problem is, I don't know of a dhcpd that will put machines on different subnets without a means of identifying which machines should go to which subnet. And then there's the issue of creating alias IP addresses on the filter/gateway machine for each subnet in use. (Otherwise, the "patient" computer won't be able to access the network at all.)
Of course, a decent Cisco switch should be able to do it...but by the time any student at our college knows how to work Cisco, he's already moved on to a four-year institution. Hmm...maybe I should invite GVSU's or Davenport's Computer Club equivalents to participate in our clinic. I've got contacts...
Read this thread.
Funny how the timing works out. One of the U3 techs stopped in here, and responded to comments and questions. Interesting answers.. (And yes, I made a fool of myself at the beginning.)
It'll likely remain in systems, just rarely used or updated. Like floppy drives, serial ports parallel ports, the AT keyboard architecture (of which PS/2 keyboards are essentially a clone), and CGA and EGA video modes.
To cut costs, it'll get integrated into the northbridge/southbridge pair of chips. The x86 system is (in)famous for its support and occasional dependancy on legacy systems. Did you know that you can still run MS DOS on most modern computers?
are you sure you can tell what's going on?
Well, systems are only connected to our network for a few hours at most. Less, if we see traffic that bothers us. Like this last time, two of the machines started scanning all the IPs on the class C subnets adjacent to the subnet we were using. We put a stop to that. The only botnet activity I saw was repeated attempts to connect to the IRC port of a domain name. However, that domain had expired, so the bots couldn't connect.
I'm looking around for a way to prevent machines on our network from talking to each other...putting each one on its own subnet seems like a good idea, but I don't know how to set up Linux dhcp to do it.
At my computer club's PC Clinic, I set up Ethereal on our network gateway computer, to keep track of things. You can easily see this kind of crap going on.
...that you submit an Ask Slashdot article?
Make sure to work on your formatting, though. Use the Preview button, and don't use the Submit button until it looks right.
When I was a teenager (and for a couple years after I turned 18), I was the primary phone tech guy for the family business, an ISP. I wasn't the one to receive the worst, call, fortunately...that pain fell to my mother, who was the other phone tech.
Here's a rough transcript of that conversation:
"Cyberspace, this is Darcy"
"Yeah, I'm having problems getting on the Internet."
swish
"OK, what error message does it give you?"
slorsh
"Uh, I don't know."
"Can you try connecting and call me back?"
"Yeah, sure. Let me get out of the tub."
slorsh, swish, dripdribble
Could be the modem was no longer capable of handling normal signal voltages.
As for the ASCII art...didn't you use hardware error correction? Granted, it was annoying when the data transfer paused for no apparent reason, but it was usually better than getting a quick demo of the extended ASCII character set and control characters.
BTW...Some old-school BBSes are still around. Try telnetting to grnet.com.
I had to laugh at myself after encountering NuBus on LowEndMac. Everything I said stands, just remove the term "PCI" from my post.
Usually, the board attached to the CRT tube. Those tubes get hot, and can desolder the boards they're attached to.
The only times I've seen magic smoke were when capacitors blew. And sometimes, those capacitors weren't really necessary, anyway.
Showering the system helped get all the conductive goo off the board before you applied power.
Water isn't really dangerous to electronics; It's the ions found in unpure water that makes it conductive. I'm still waiting for someone to give a PC a really thorough cleaning job, and demonstrate real water cooling by submerging a system into a cleaned fishtank full of deionized water. (The cleaning job is required to get all the dust and miscellaneous substances off the computer. If they dissolve, the fishtank water will no longer be deionized, and would start conducting again.)
Bravo on a good solution.
I would bet it had to do with the signal timing on the bus. If one of the cards absolutely had to be seen by the PCI controller before the other, it could be important that that card lay closer on the bus to the PCI controller than the other, and that there be a minimum distance between the two cards.
What exactly does a nubus card do, anyway?
When I was a runt, I had both a SEGA 32X and a Sega Channel adapter. Unfortunately, you couldn't use them both at the same time. I had to remove the 32X whenever I wanted to use Sega Channel, and reinstall it when I wanted to go back to playing Star Wars 32X or Tempo.
Well, I got tired of putting in those little metal plates and stopped bothering with them. Within a couple uses, the 32X was dead as a doornail. I didn't realize until years later how they could be important to the functioning of the device.
(Darn, I miss the 32X. And Sega Channel. Two of the neatest things an 11-year-old could have. Long live Road Rash!)
"Freeze" might be a more apt term for it than you know. Sounds like the fault could have been dependant on the temperature of the CPU/Heatsink and/or motherboard. That could mean anything from a faulty solder connection to a problem with the CPU itself. (You didn't overclock it at one point, did you?)
I could easily see a couple of bad ground pins causing a high enough current flow in the rest of them to cause heat issues on the motherboard. A subsequent cool down could cause some now-soft solder connections to break contact. On the other hand, silicon dies are finicky things...overheating could cause problems internal to a rarely-used section of the CPU that only manifest themselves when heat isn't applied.
A real art major also knows the difference between color displays based on adding and subtracting primary colors. The ones who use computers for their projects actually apply these principles.
The assembly that latches onto the CD might be loose. Tilting it could do anything from fixing a faulty wire connection to aligning the read laser just right.
Bad electrical connection. Either the board was developng a problem at the power connector, or there was a frayed wire in your power cable. I've run into both problems. The first condition happened with the laptop I'm typing this on...tension on the cable caused the soldering connections to break. A friend resoldered it, and now it works fine. I see the second condition all the time...people in the computer lab keep kicking the wiring, causing stress.
Sounds like there was something loose inside the drive, or there might have been a bad gear. If something causes the drive to think the tray isn't opening all the way, it will draw the tray back in to avoid damage.
Was it a Presario 2100?
My brother had that problem with WoW on this laptop. Don't think he ever tried opening the CD-ROM drive, though. He's "given" it to me so long as I lose weight. (40 lbs down and counting. Woohoo!)
For a good while, I ran it underclocked to 1.06GHz from the max of 1.79GHz. Then I had a friend service a bad part on it, and found out what the problem was. The inside of the system was caked with dust, with the fans and heat sink being especially bad. He cleaned all that crap out, and now the system runs happily at full speed.
I tell students, "There's no easier way to get a computer to behave than to have a tutor look over your shoulder."
Gotta stop doing that, though. A couple students took it to heart, and made work living hell for me this semester.
Don't forget those bad solder connections on the CRT (The tube itself, not the monitor as a whole). When those things get hot enough, the color starts to flicker. A good thwack seems to do the trick.
Wanna impress an art major? Tell them which color signal went out. Then fix it. "Everything's magenta? Guess Green went out. Watch this. Don't *whack* you *whack* ever *whack* do *whack* that *whack* again *whack*."
Well, leave out the theatrics if the art major is female...
Half the students that come through my lab don't even know how to use email. And learning how is not, unfortunately, a prerequesite to classes in departments other than Computer Apps. And not even for most Computer Apps classes.
It's a community college. I'd wager at least a third of my student encounters are where the student doesn't even speak English fluently. Ninety eight percent of my student encounters are with students never used a computer before college, or only know enough to use AIM and Myspace.
I wish that untouched two percent of students asked more questions. I really do.
Gah. Helps if I finish my thought.
Again, the other possibility might be the way drive letters are in use on the systems at GRCC.
The college I work at has a large number of letter drives assigned. The first partition of the U3 device comes up as the F drive. The G drive is a globally-assigned network drive that supports a large number of applications. If something tries to create the "next" letter drive, it tries to count up from the local drives already created. Since the G drive is in use as a network drive, the operation fails.
We used to have that problem with normal flash drives, as the F drive had been assigned as a network drive. When flash drives became prevalent, they moved the F drive data somewhere else, so it wouldn't have to be disconnected before the flash device was inserted. Now the same thing may be happening with the G drive.
Unfortunately, the same solution isn't as easy with the G drive as with the F drive; there are a much larger number of applications, with a much higher demand, on the G drive than there were with the F drive. Shifting the drive letters up doesn't seem to be possible; some roles on the network already have X, Y and Z drives. (Mostly instructors.)
Drop me an email...if you can help me find a better solution for this, we'd both be better off.
No device drivers? Odd. I thought that was what was killing it.
The only other possibility is that the device tries to take two letter drives.