Reported in Oregon, 1986:
Farewell O verse
Along the road
How sad to see
You're out of mode
Burma Shave
05 September 2012
02 July 2012
Maybe you just predicted wrong, Cringe?
In Cringely's latest article he declares that the PC is about to become something unrecognizably different. Because that's what he predicted in 1992 would have happened by now, and anyway that's what happens to technology after about 30 years. That's why the telephone of today is unrecognizably different from the telephone of 1900, the movie from the movies of the 1920s, and television from the televisions of the 1950s.
16 June 2012
05 June 2012
Footprints In The Sand
"I dreamed that I was walking down the beach with the Flying
Spaghetti Monster. And I looked back and saw footprints in the sand. But
sometimes there were two pairs of footprints and sometimes there was
only one. Which was really weird, because His Noodly Appendage doesn't
leave footprints."
18 May 2012
How do you tell if your computer is infected?
My brother recently asked me how to tell if your computer was part of a botnet. Now, I'm not a real Windows jock any more, so I don't know how much help I can be:
I'm not really up on the mechanics of telling if your computer is
compromised from the inside.
I can tell you, though, that it may not be practical to even try.
The problem is that if your computer has been compromised, the attacker can install components that change the behavior of tools running inside the computer. There are specific toolkits, "rootkits", that target antivirus and other security software to hide the presence of the compromising code, even modifying the running security applications so they don't see the files that the malware is hiding in. If you "run a security" and don't see anything, you still can't be sure.
The best way to tell if your computer is part of a botnet is from outside your computer. What I used to do, before virtual machines, was run an external firewall router. Since the firewall isn't inside the computer's trust boundary, it's not going to be compromised by the malware. You can look through your logs on the firewall to see connections to unexpected sites or connections when you're not actually using the computer. These days, you can run Windows in a virtual machine, and get the same result more cheaply.
But really, for most people, you still can't be sure. It could be sleeping. It could be that you're just not familiar with reading log files.
So the best thing to do if you even suspect your computer might be infected is to back up all the data (you're keeping backups, right?) and reinstall from scratch, reinstalling any applications from the original media. Again, if you're running Windows in a virtual machine, you can roll back to the original snapshot. Even if you're not infected, this cleans up cruft and dander and leaves your computer feeling years younger with a lovely silky complexion.
I used to do that every year or so, when I was using Windows at home for more than the occasional videogame. Just because it ended up making my computer faster. Apart from the external backup drive, I kept two drives inside the system and switched which one was primary. The process: unhook the "current" drive, and install Windows over again on the "old", reformatting the drive in the install process. Now the "old" drive is "current", and the current is "old". Hook up the "old" drive long enough to copy the data off it (just the data, no applications), then disconnect it, leaving it powered off until next time. If you put your ear up against it, maybe you can hear the botnet screaming to be let out...
I can tell you, though, that it may not be practical to even try.
The problem is that if your computer has been compromised, the attacker can install components that change the behavior of tools running inside the computer. There are specific toolkits, "rootkits", that target antivirus and other security software to hide the presence of the compromising code, even modifying the running security applications so they don't see the files that the malware is hiding in. If you "run a security" and don't see anything, you still can't be sure.
The best way to tell if your computer is part of a botnet is from outside your computer. What I used to do, before virtual machines, was run an external firewall router. Since the firewall isn't inside the computer's trust boundary, it's not going to be compromised by the malware. You can look through your logs on the firewall to see connections to unexpected sites or connections when you're not actually using the computer. These days, you can run Windows in a virtual machine, and get the same result more cheaply.
But really, for most people, you still can't be sure. It could be sleeping. It could be that you're just not familiar with reading log files.
So the best thing to do if you even suspect your computer might be infected is to back up all the data (you're keeping backups, right?) and reinstall from scratch, reinstalling any applications from the original media. Again, if you're running Windows in a virtual machine, you can roll back to the original snapshot. Even if you're not infected, this cleans up cruft and dander and leaves your computer feeling years younger with a lovely silky complexion.
I used to do that every year or so, when I was using Windows at home for more than the occasional videogame. Just because it ended up making my computer faster. Apart from the external backup drive, I kept two drives inside the system and switched which one was primary. The process: unhook the "current" drive, and install Windows over again on the "old", reformatting the drive in the install process. Now the "old" drive is "current", and the current is "old". Hook up the "old" drive long enough to copy the data off it (just the data, no applications), then disconnect it, leaving it powered off until next time. If you put your ear up against it, maybe you can hear the botnet screaming to be let out...
21 April 2012
Writing code to fit into small places...
Somewhere around here I have the source to Jim Penny's "Freelancin'
Roundtable" chat system, which I asked him for because it was supposed
to be an amazing piece of tight coding... I thought I might learn
something from it. This was in the early '80s, and an 8-user dialup chat
system was an amazing thing. I learned a lot, actually, particularly
about the kind of things that become reasonable when you're up against
the wall and have absolutely no cycles left to spare.
One quirk of Roundtable was that your user ID was 4 hex digits, and your password was a fixed random string that you couldn't change... every now and then he'd mail all the users with a new user ID and password.
It ran on a TRS-80 and communicated with 8 modems through unbuffered UARTs, so it absolutely had to get back to all 8 serial ports 30 times a second, every second, to output or input the next character of text.
Reading data from files or writing it out again was obviously out of the question, and memory was really tight, so he didn't even implement any lookup tables. Instead, he would eyeball the binary code of the program for printable strings. Your user ID was the address of the string. Every time he recompiled the program he had to come up with a new set of user IDs and passwords and send them out to all the users before they could log in again.
One quirk of Roundtable was that your user ID was 4 hex digits, and your password was a fixed random string that you couldn't change... every now and then he'd mail all the users with a new user ID and password.
It ran on a TRS-80 and communicated with 8 modems through unbuffered UARTs, so it absolutely had to get back to all 8 serial ports 30 times a second, every second, to output or input the next character of text.
Reading data from files or writing it out again was obviously out of the question, and memory was really tight, so he didn't even implement any lookup tables. Instead, he would eyeball the binary code of the program for printable strings. Your user ID was the address of the string. Every time he recompiled the program he had to come up with a new set of user IDs and passwords and send them out to all the users before they could log in again.
24 March 2012
The Towers of Utopia
In this 1975 novel Mack Reynolds
equipped every resident of Skyler Deme with a "TV Phone" that was also a
"Universal Credit Card", and was routinely used to query "the
computers" as well as being a RFID-like access card.
I last read this about 20 years ago. Re-reading it now - this is probably the best depiction of a "smart phone" in pre-web science fiction that I can think of.
I last read this about 20 years ago. Re-reading it now - this is probably the best depiction of a "smart phone" in pre-web science fiction that I can think of.
Subscribe to:
Posts (Atom)