gabriel rosenkoetter on Sat, 23 Nov 2002 16:20:05 -0500 |
On Sat, Nov 23, 2002 at 01:54:34PM -0500, Jesse Huestis wrote: > (DLink was the sharpest in MS OSs offering 256 bit encryption) Using what algorithm, peer-reviewed by whom? WEP's bad, but a vague, airy algorithm you can't look at is worse. > What are the standard card people are using with Linux? Anything based on a WaveLAN or PRISM2 chipset should work. Try it and see (or figure out what chipset your card uses; your vendor should be able to tell you). > Also, according to documentation in the Linksys manuals, WEP encryption > is not standard. Has anyone had problems with WEP not working mixing > and matching wireless companies cards and access points? No, LinkSys is saying that they're using a modified version of WEP so they don't generate weak keys, at the cost of reducing the key search space. It's a trade-off. Anyway, it'd really better interoperate with other things claiming WEP support, or they're advertising falsely. You can be sure that D-Link's 256-bit mode will only work with other D-Link hardware (and that quite possibly only if you're using D-Link's software to drive it). If you want secure wireless, hold on for 802.11i where it'll be done right. If you want a little better security on 802.11{a,b}, try Niels Ferguson's Michael algorithm over MIC. If you must have secure wireless today, you can do IPSec at the IP layer, using 128-bit WEP on the wifi just to obscure the source and destination of packets (the only thing IPSec can't encrypt, obviously). -- gabriel rosenkoetter gr@eclipsed.net Attachment:
pgpGtir7Fy1G3.pgp
|
|