I remember hearing how back in the day he had a Mac Plus which had blown its analog board, so he couldn't get online. (As if there was much time for that while doing the weekly TV series grind.) Ah, the good old Mac analog board.
It wasn't the power supply that blew, like most people said, it was actually part of the video circuitry. And usually it was a 25 volt non-polarized electrolytic capacitor that went, in a spew of orange smoke. When it happened to mine, I took it to the local Mac fixit guy, and he replaced it with a 35 or 50 or so volt Mylar non-electrolytic capacitor. I guess Apple didn't learn its lesson, because AirPort base stations have had a problem with marginally rated capacitors, too.
Absolutely! After all, once it was decided that the CIA couldn't hire operatives with shady backgrounds, Bert was the only remaining hope of the free world. And now, thanks to a careless photo montage, his cover has been blown.
Bert now only has one chance to survive (make your time): Ernbo! (insert picture of Ernie with M-16 and headband here)
(They would send Mr. T, but it seems he's gotten wise to all the A-Team tricks to get him onto airplanes.)
I remember that one from a previous job. The thing about it was that we normally never noticed it, but there was one cash register PC where the floppy drive would stop working once NATAS took hold. It wasn't used very often to test code, but when one of those stopped reading floppies, it was time to go around and run a virus checker on hard drives and stacks of floppies. Also nasty was that it would infect an executable file of some sort in our software which had a file extension that the virus scanner's "quick" mode didn't scan! After about four years (yes, years), I think we finally got rid of it.
even if you have EVERY patch installed on your MS IIS servers, you still get slammed by random IP's from MS servers that weren't patched, thereby bringing "your internet" to a slow crawl (bandwidth/data rate dependent, of course).
First of all, if you have every patch installed and don't get infected, then you're helping by not contributing to the problem. And you're keeping internal machines (which might not be accessible from the general Internet due to firewalling or NAT) from getting infected and making your own problem worse.
Second, a lot of the problems people are encountering are due to ARP storms as unused IP addresses are scanned. The solution is to have a proper mix of routing and switching, and not to simply pipe a LAN in one branch office to every other branch office by using VLAN trunking. You have to break up the ARP broadcast domains to keep the effect under control (which it seems most cable modem companies are not doing). VLAN switching isn't a substitute for proper routing.
Also in this Register article. Encryption just makes secrecy as easy and convienent. If your suspect is willing to go to the extra trouble of avoiding high-tech communications entirely, all the crypto restrictions in the world won't help.
Basically there are TWO places in a DVD player that contain zone information. There's a 8-bit bitmap in the MPEG stream that says which regions it's playable on (enforced by the player), and there's also an region number register in one of the GUI registers of the playback control software which can be read by the DVD's scripting (allowing enforcement by the DVD disc).
The bitmap can easily be ignored, but it's hard to put six or eight integers simultaneously into a single register without resorting to quantum computing, which is still a few years off. So-called "region zero" players have the bitmap bypassed, but "region switchable" players let the user choose (and maybe automatically try based on the bitmap) which region number will be presented to the DVD's scripting code.
Region switchable players also let you take advantage of discs which enable extra features (like subtitles in Asian languages) based on the player's region number.
Anyhow, I don't expect to be getting this disc because I don't f'ing care about Episode I. I know it's heresy, but I never saw it in the theatres, and aside from Wierd Al Yankovic's summary of the movie, I have only the vaguest idea of what is contained within. I do know that I would hate Jar Jar, though.
I also doubt I'll be getting the Episode 4-6 DVDs, because they'll likely only be available in the "Ministry of Information approved" edition. I've got the widescreen stereo release and the widescreen THX release on LD, without the "Han Solo didn't really shoot first" crap. I'll live with the disc change to watch the end of the movies on side 3, if that's what it takes to avoid the Politically Correct version.
Sendmail does have anti-spam features. If you configure it properly. By building the sendmail.cf file using the M4 macros.
Using (gag, choke) Linuxconf does not count as "configuring it properly", because Linuxconf merely glues together random snippets of a sendmail.cf file of questionable vintage.
I'm sure glad I have DSL. Considering that the cable modem people still haven't learned from Code Red, and still have ARP broadcast domains that spam^Hn the entire United States, cable modem service is gonna suck for another two weeks or so until this new mess is cleaned up.
Interesting that this is on their list. Why? Because Rush Limbaugh uses that tune (the instrumental bits only) as his semi-official theme song. Clear Channel now owns Premiere Networks, which is the syndicator of Rush Limbaugh, and they make every effort to keep it exclusively on Clear Channel stations whenever possible. I mean, I know the lyrics (and title) are why this is on the list, but isn't it ironic (oops, another song from the list!) that Clear Channel stations everywhere are broadcasting part of one of these songs five days a week?
I believe he is referring to ATX. I doubt that VME equipment would go for much less than what Apple is charging. I notice a lack of pricing on that web page.
Is it just market forces that keep Asus, Tyan, and ABit from producing a PPC MB? I suppose a standard BIOS is lacking (other than Apples)... surely someone could come up with a non-Apple hardware solution, though.
Apple now uses the Forth-based OpenBoot or whatever they call it, which is based on what Sun uses to boot their systems. The MacOS ROM image is loaded from a disk file during the boot process of OS 9.x. I doubt BIOS issues are keeping people from making PPC motherboards, other than maybe the lack of a vendor (like Award or Phoenix) for a standard BIOS. But I'm pretty sure the spec is sufficient to allow a third-party BIOS to be written.
One unique thing about the PowerPC architecture is that it has eight sets of condition code flags. So you can pipeline condition codes just like you can pipeline the general purpose registers. By the time you need to decide the branch, the condition can already be known, resulting in a zero-cycle branch. Meanwhile, another condition or three can be computed without disturbing the one you've got pipelined.
Before people start throwing stuff at the Red Cross that they can't handle, based on instructions in an article from a third-party, check with the Red Cross web first. Keep in mind that giving them stuff (other than commodities like blankets, etc.) requires them to divert resources to processing the stuff. As this comment says:
It looks to me like if you have a container-load of blankets or something, or want to just give them money with your credit card, they
would be happy to have it. But wouldn't just mailing them one monitor (with desktop???) or one PC just add to the logistical load?
I like the idea of sending them some hardware, but follow their instructions, not some third party.
If you can find something on the Red Cross's site saying they want this equipment, send it. Otherwise, only send what they have asked for.
Thrice Upon a Time has a plot thread about an experiment that generates tiny black holes that don't show themselves as growing until a few months pass (with wierd unexplainable holes in things until people figure it out). But don't worry, their trusty DEC PDP-21 will help fix things!
The problem isn't getting an IPv6 node number. There's already a pre-defined IPv6 number range for IPv4 addresses. The problem is that there need to be IPv6 routing protocols for routers, and backbones that use them.
You're thinking about this completely wrong. What was it that made TCP/IP the 800 pound gorilla standard in the first place? The US Government, especially the military, standardized on it. What we need is to get the US Government to start requiring IPv6 in contracts.
And this is the Slashdot article about this. If it weren't for the linked article, this topic would be totally redundant. Except that it contains nothing that isn't in the Scientific American article, so not only is this typical Slashdot regurgitation of old topics, but the linked article is, too.
Re:Linux will be just as bad...discuss....
on
Windows in 2020
·
· Score: 2
Ah yes, your lovely Cobalt Qube, where upgrading the OS with anything other the official Cobalt patches violates the warranty? What an excellent way to ensure homogenity!
Dammit, brain fart. It's the RaQ that has this warranty policy. I have no idea whether the Qube has this.
Re:Is Windows security full of holes?
on
Windows in 2020
·
· Score: 2
[28 Jul 2001] DSA-067 apache - Remote exploit
"Remote exploit"? That makes it sound like someone can 0wn your machine. When in actuality, this particular "exploit" merely allowed people to bypass an index.html file and see a raw directory listing.
[10 Aug 2001] DSA-071 fetchmail - memory corruption
They may call it a "security alert", but "memory corruption" sounds more like a simple bug to me.
Re:Linux will be just as bad...discuss....
on
Windows in 2020
·
· Score: 2
Posit: once Linux reaches a certain saturation, it will suffer the same security issues as Microsoft does.
I believe you are making an invalid assumption that Microsoft == Windows OS. The reason Code Red took off so quickly is that the software with the real problem wasn't Windows, it was IIS. Windows does have a bit of variety (I've heard the original patch would crash IIS on NT4 when Code Red came around again), but IIS is a lot more homogenous. In this case it should be compared to Apache. I don't know when the last remote root exploit was found in Apache, but I'm sure it was found through source code inspection. I seem to recall hearing that the default.ida exploit used by Code Red was found by someone disassembling (!) IIS to look for vulnerabilites.
You're also making the invalid assumption that Linux == Linux. The reason that Microsoft OS'es, and IIS in particular, are so homogenous is that they are installed from CD-ROMs that are produced in the millions! The same can be said for Red Hat, which, not coincidentally, is the distribution of choice for people running honeypots. Just do a full install of RH 6.0 and wait for the 'sploits to come flooding in! On the other hand, a Slackware install with half of the daemons recompiled from freshly downloaded source, and a decent ipchains firewall is much harder to break into. It's not Linux becoming dominant that will be a problem, it's a single distro like RedHat (aka DeadRat) becoming popular among people who don't know what they're doing that is the problem. And right now, 4 out of 5 people who don't know what they're doing choose Microsoft.
Once Linux systems are consumer devices (like my lovely Cobalt Qube) - and there is every good reason for them to become so - then no amount of open source hacking, patches, multiple eyeballs and bugtraqery will stop these systems from being compromised once a hole is found and made public.
Ah yes, your lovely Cobalt Qube, where upgrading the OS with anything other the official Cobalt patches violates the warranty? What an excellent way to ensure homogenity!
Speaking as someone with experience unlike the parent of this message
Please don't use such non-specific references as "the parent of this message". There may be one true parent message, but it may be difficult to find without browsing at lower thresholds, especially if it gets modded down. And it becomes almost impossible to determine after the thread is archived.
Anyhow, I didn't say Mozilla wasn't stable. I said NS 4.x was so unstable that it was meaningless to say Mozilla was more stable. Most of the Mozilla builds since around 0.8 or so have been very stable for me under W2K at work. (And I have a 90+ day uptime on that W2K box, too.) The only problem I have had was right after I installed 0.9.3 it crashed twice. So I uninstalled it and re-installed with talkback and it hasn't crashed since then. Note that I do NOT install Mail/News/Chat because I don't believe in using web browsers for anything but web browsing. (And I wish I had the option to disable the freaking HTML editor too!)
To the person who said "turn off everything including images and it's great!", I say at that point you might as well be running Lynx. To the other person who says NS 4.x is stable for him, I suppose he might still be a modem user. I gave up NS 4.x when I got DSL and could crash it every two minutes. And to the person who says "just learn how not to annoy Windows 98", I say get a real operating system!
But I wish they hadn't broken auto-completion of URLs. (in 0.9.2, I think) Sure, the menu pops up, but it doesn't complete the URL that you're typing into.
Re:Code Red is trying to eat me!
on
Code Red III
·
· Score: 2
If your router is a Cisco and hasn't had a firmware upgrade in the past six months or so, it may have a crashing bug that can be triggered by Code Red's exploit. But if you have _any_ service, this is not the problem, and the guy who called you was a total idiot.
Re:It's not like they haven't announced the patch
on
Code Red III
·
· Score: 2
There's also the subtle difference that flaws in Microsoft products don't kill people.
Yet.
The US Navy is giving it a good try, though.
Mozilla is the cure
on
Code Red III
·
· Score: 1, Offtopic
Download the latest Mozilla and read the version notes for 0.9.2 to see how to fix your setup file to kill pop-up ads. I've been pop-under free for a couple of months now.
As usual, the cure is to ditch the SmallLimp crapware and replace it with mature open source code.
Slashdot Mirror
It wasn't the power supply that blew, like most people said, it was actually part of the video circuitry. And usually it was a 25 volt non-polarized electrolytic capacitor that went, in a spew of orange smoke. When it happened to mine, I took it to the local Mac fixit guy, and he replaced it with a 35 or 50 or so volt Mylar non-electrolytic capacitor. I guess Apple didn't learn its lesson, because AirPort base stations have had a problem with marginally rated capacitors, too.
Bert now only has one chance to survive (make your time): Ernbo! (insert picture of Ernie with M-16 and headband here)
(They would send Mr. T, but it seems he's gotten wise to all the A-Team tricks to get him onto airplanes.)
Well, I'll admit I run SCSI or Firewire when I can, but ATA is still a pretty decent hard drive interface. At least it's cheap.
I remember that one from a previous job. The thing about it was that we normally never noticed it, but there was one cash register PC where the floppy drive would stop working once NATAS took hold. It wasn't used very often to test code, but when one of those stopped reading floppies, it was time to go around and run a virus checker on hard drives and stacks of floppies. Also nasty was that it would infect an executable file of some sort in our software which had a file extension that the virus scanner's "quick" mode didn't scan! After about four years (yes, years), I think we finally got rid of it.
First of all, if you have every patch installed and don't get infected, then you're helping by not contributing to the problem. And you're keeping internal machines (which might not be accessible from the general Internet due to firewalling or NAT) from getting infected and making your own problem worse.
Second, a lot of the problems people are encountering are due to ARP storms as unused IP addresses are scanned. The solution is to have a proper mix of routing and switching, and not to simply pipe a LAN in one branch office to every other branch office by using VLAN trunking. You have to break up the ARP broadcast domains to keep the effect under control (which it seems most cable modem companies are not doing). VLAN switching isn't a substitute for proper routing.
Also in this Register article. Encryption just makes secrecy as easy and convienent. If your suspect is willing to go to the extra trouble of avoiding high-tech communications entirely, all the crypto restrictions in the world won't help.
The bitmap can easily be ignored, but it's hard to put six or eight integers simultaneously into a single register without resorting to quantum computing, which is still a few years off. So-called "region zero" players have the bitmap bypassed, but "region switchable" players let the user choose (and maybe automatically try based on the bitmap) which region number will be presented to the DVD's scripting code.
Region switchable players also let you take advantage of discs which enable extra features (like subtitles in Asian languages) based on the player's region number.
Anyhow, I don't expect to be getting this disc because I don't f'ing care about Episode I. I know it's heresy, but I never saw it in the theatres, and aside from Wierd Al Yankovic's summary of the movie, I have only the vaguest idea of what is contained within. I do know that I would hate Jar Jar, though.
I also doubt I'll be getting the Episode 4-6 DVDs, because they'll likely only be available in the "Ministry of Information approved" edition. I've got the widescreen stereo release and the widescreen THX release on LD, without the "Han Solo didn't really shoot first" crap. I'll live with the disc change to watch the end of the movies on side 3, if that's what it takes to avoid the Politically Correct version.
...and Little Brick Out.
Using (gag, choke) Linuxconf does not count as "configuring it properly", because Linuxconf merely glues together random snippets of a sendmail.cf file of questionable vintage.
I'm sure glad I have DSL. Considering that the cable modem people still haven't learned from Code Red, and still have ARP broadcast domains that spam^Hn the entire United States, cable modem service is gonna suck for another two weeks or so until this new mess is cleaned up.
Interesting that this is on their list. Why? Because Rush Limbaugh uses that tune (the instrumental bits only) as his semi-official theme song. Clear Channel now owns Premiere Networks, which is the syndicator of Rush Limbaugh, and they make every effort to keep it exclusively on Clear Channel stations whenever possible. I mean, I know the lyrics (and title) are why this is on the list, but isn't it ironic (oops, another song from the list!) that Clear Channel stations everywhere are broadcasting part of one of these songs five days a week?
I believe he is referring to ATX. I doubt that VME equipment would go for much less than what Apple is charging. I notice a lack of pricing on that web page.
Is it just market forces that keep Asus, Tyan, and ABit from producing a PPC MB? I suppose a standard BIOS is lacking (other than Apples)... surely someone could come up with a non-Apple hardware solution, though.
Apple now uses the Forth-based OpenBoot or whatever they call it, which is based on what Sun uses to boot their systems. The MacOS ROM image is loaded from a disk file during the boot process of OS 9.x. I doubt BIOS issues are keeping people from making PPC motherboards, other than maybe the lack of a vendor (like Award or Phoenix) for a standard BIOS. But I'm pretty sure the spec is sufficient to allow a third-party BIOS to be written.
One unique thing about the PowerPC architecture is that it has eight sets of condition code flags. So you can pipeline condition codes just like you can pipeline the general purpose registers. By the time you need to decide the branch, the condition can already be known, resulting in a zero-cycle branch. Meanwhile, another condition or three can be computed without disturbing the one you've got pipelined.
Thrice Upon a Time has a plot thread about an experiment that generates tiny black holes that don't show themselves as growing until a few months pass (with wierd unexplainable holes in things until people figure it out). But don't worry, their trusty DEC PDP-21 will help fix things!
We didn't. Ever hear of CIDR?
So what's this NAP you're babbling about?
You're thinking about this completely wrong. What was it that made TCP/IP the 800 pound gorilla standard in the first place? The US Government, especially the military, standardized on it. What we need is to get the US Government to start requiring IPv6 in contracts.
What does it mean? It means that pricewatch.com is your friend.
And this is the Slashdot article about this. If it weren't for the linked article, this topic would be totally redundant. Except that it contains nothing that isn't in the Scientific American article, so not only is this typical Slashdot regurgitation of old topics, but the linked article is, too.
Dammit, brain fart. It's the RaQ that has this warranty policy. I have no idea whether the Qube has this.
"Remote exploit"? That makes it sound like someone can 0wn your machine. When in actuality, this particular "exploit" merely allowed people to bypass an index.html file and see a raw directory listing.
[10 Aug 2001] DSA-071 fetchmail - memory corruption
They may call it a "security alert", but "memory corruption" sounds more like a simple bug to me.
I believe you are making an invalid assumption that Microsoft == Windows OS. The reason Code Red took off so quickly is that the software with the real problem wasn't Windows, it was IIS. Windows does have a bit of variety (I've heard the original patch would crash IIS on NT4 when Code Red came around again), but IIS is a lot more homogenous. In this case it should be compared to Apache. I don't know when the last remote root exploit was found in Apache, but I'm sure it was found through source code inspection. I seem to recall hearing that the default.ida exploit used by Code Red was found by someone disassembling (!) IIS to look for vulnerabilites.
You're also making the invalid assumption that Linux == Linux. The reason that Microsoft OS'es, and IIS in particular, are so homogenous is that they are installed from CD-ROMs that are produced in the millions! The same can be said for Red Hat, which, not coincidentally, is the distribution of choice for people running honeypots. Just do a full install of RH 6.0 and wait for the 'sploits to come flooding in! On the other hand, a Slackware install with half of the daemons recompiled from freshly downloaded source, and a decent ipchains firewall is much harder to break into. It's not Linux becoming dominant that will be a problem, it's a single distro like RedHat (aka DeadRat) becoming popular among people who don't know what they're doing that is the problem. And right now, 4 out of 5 people who don't know what they're doing choose Microsoft.
Once Linux systems are consumer devices (like my lovely Cobalt Qube) - and there is every good reason for them to become so - then no amount of open source hacking, patches, multiple eyeballs and bugtraqery will stop these systems from being compromised once a hole is found and made public.
Ah yes, your lovely Cobalt Qube, where upgrading the OS with anything other the official Cobalt patches violates the warranty? What an excellent way to ensure homogenity!
Please don't use such non-specific references as "the parent of this message". There may be one true parent message, but it may be difficult to find without browsing at lower thresholds, especially if it gets modded down. And it becomes almost impossible to determine after the thread is archived.
Anyhow, I didn't say Mozilla wasn't stable. I said NS 4.x was so unstable that it was meaningless to say Mozilla was more stable. Most of the Mozilla builds since around 0.8 or so have been very stable for me under W2K at work. (And I have a 90+ day uptime on that W2K box, too.) The only problem I have had was right after I installed 0.9.3 it crashed twice. So I uninstalled it and re-installed with talkback and it hasn't crashed since then. Note that I do NOT install Mail/News/Chat because I don't believe in using web browsers for anything but web browsing. (And I wish I had the option to disable the freaking HTML editor too!)
To the person who said "turn off everything including images and it's great!", I say at that point you might as well be running Lynx. To the other person who says NS 4.x is stable for him, I suppose he might still be a modem user. I gave up NS 4.x when I got DSL and could crash it every two minutes. And to the person who says "just learn how not to annoy Windows 98", I say get a real operating system!
But I wish they hadn't broken auto-completion of URLs. (in 0.9.2, I think) Sure, the menu pops up, but it doesn't complete the URL that you're typing into.
If your router is a Cisco and hasn't had a firmware upgrade in the past six months or so, it may have a crashing bug that can be triggered by Code Red's exploit. But if you have _any_ service, this is not the problem, and the guy who called you was a total idiot.
Yet.
The US Navy is giving it a good try, though.
As usual, the cure is to ditch the SmallLimp crapware and replace it with mature open source code.