Firewall Wizards mailing list archives
Re: The Future of Security
From: David LeBlanc <dleblanc () mindspring com>
Date: Mon, 06 Dec 1999 10:59:40 -0800
At 01:33 PM 12/6/99 EST, Randy Witlicki wrote:
David wrote:At 04:54 PM 12/2/99 -0600, Don Helms wrote:
To many times, it's the simple stuff that folks forget to lock down.
[left in message for emphasis] [I said:]
Then what bothers me even more is that these Java-based gizmos want to work across my house wiring. So once I get all my appliances running that, now I need a firewall for my exterior electrical sockets or some kid will plug a gizmo into the outside of my house and have the kitchen looking like Mickey and the brooms in the Sorcerer's Apprentice...
Well, in *theory* at least, things like the Jini initiative are trying to do what Marcus has ranted about over the past few years - Throwing out all the current bad practices which emphasize perimeter security and rebuilding our protocols and so on at the individual host level. Of course, we have to make sure the design is correct and that we implement it correctly, but what else is new ?
[included by Randy]
AR.2.1.4 Security
The design of the security model for Jini technology is built on the twin notions of a principal and an access control list.
Although I agree that you can't make perimeter security the ONLY thing that keeps you secure, stuff like this is why I don't want to abandon it entirely. Further, we've got some very real conflicts going on when we're talking appliances. I have a friend who has a way to make a common kitchen appliance work much better, and he's trying to get the manufacturers to implement it. These people seriously worry about cost differentials of $1 or less. So, I expect the processing power of these Java-enabled gizmos to be as small as possible, which also tells me that the IP stack isn't going to be sophisticated, and experience tells me that means they are probably prone to DoS attacks. Next, we're talking about basing the security of these devices on some sort of ACL, yet they are supposed to 'discover' one another. Discovery implies 2 things - one is that they will be chatty, and the other is that they will respond to requests for at least a minimal amount of information about themselves. Both of these aspects tend to cause security issues. Basing security on an ACL means that they will have to ship with some default (* in the .rhosts file, anyone?, or how about Domain Users with the right to log on from the network?) that makes for a nice out-of-box experience. Then we'll have most people never changing the default, which leads me back to a situation where I could very easily have Mickey and the dancing brooms. Prediction: Securing the household appliances from the Internet will prove much too difficult for the average person who can't program their VCR, and some sort of very cheap and limited firewall appliance will be part of whatever is used to connect your average household to the net. I also predict that at some point in the future, teenagers will be able to cause mayhem by plugging a device into an exterior household power socket, and that this will be deemed far superior to smashing mailboxes or lighting bags of dog feces on fire and ringing the doorbell. David LeBlanc dleblanc () mindspring com
Current thread:
- RE: The Future of Security, (continued)
- RE: The Future of Security Eric Budke (Dec 03)
- RE: The Future of Security David LeBlanc (Dec 06)
- Re: The Future of Security Rick Smith (Dec 03)
- Re: The Future of Security David LeBlanc (Dec 06)
- RE: The Future of Security Scott, Richard (Dec 03)
- RE: The Future of Security Scott, Richard (Dec 05)
- RE: The Future of Security R. DuFresne (Dec 06)
- Re: The Future of Security ark (Dec 06)
- RE: The Future of Security Rick Smith (Dec 06)
- Re: The Future of Security Randy Witlicki (Dec 06)
- Re: The Future of Security David LeBlanc (Dec 06)
- Re: The Future of Security Damir Rajnovic (Dec 07)
- Re: The Future of Security David LeBlanc (Dec 06)
- RE: The Future of Security LeGrow, Matt (Dec 08)