From owner-freebsd-mobile Wed Nov 24 16:53:17 1999 Delivered-To: freebsd-mobile@freebsd.org Received: from dominator.eecs.harvard.edu (dominator.eecs.harvard.edu [140.247.60.28]) by hub.freebsd.org (Postfix) with ESMTP id 938E015553 for ; Wed, 24 Nov 1999 16:53:08 -0800 (PST) (envelope-from karp@eecs.harvard.edu) Received: (from karp@localhost) by dominator.eecs.harvard.edu (8.9.3/8.6.12) id TAA11655; Wed, 24 Nov 1999 19:53:07 -0500 (EST) Date: Wed, 24 Nov 1999 19:53:07 -0500 (EST) From: Brad Karp Message-Id: <199911250053.TAA11655@dominator.eecs.harvard.edu> To: matt@braithwaite.net Subject: Re: STRIP (was Re: richochet modems) Cc: freebsd-mobile@freebsd.org Sender: owner-freebsd-mobile@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Matt Braithwaite wrote: > That *is* pretty cool. On the other hand, STRIP is an adequate > substitute for PPP mode (i.e., mobile node to Internet gateway, rather > than mobile node to mobile node), and it runs on Linux as well, for > people who are into that sort of thing. HUMR, of course, works mobile-node-to-Internet-gateway *and* mobile-node-to- mobile-node. STRIP isn't *quite* a PPP substitute, even when used as a single hop from a mobile node to an Internet gateway. PPP can assign "pseudo" link level addresses to either side of a link, without either side knowing any sort of "MAC address" for the other. STRIP requires configuration of MAC addresses for the two sides of the link into an ARP table. You don't need STRIP to do PPP--you can "dial" the MAC address of the other modem, and get a reliable byte stream and *literally* run PPP over it. So if you know the MAC addresses, you don't need the STRIP code at all. StarMode's only added value is multi-access (and multi-hop without Metricom poletops in your area, though that's no longer a PPP substitute). As for Linux, HUMR is fully user-level (runs on tun), so it's more portable than STRIP, which is a Linux kernel driver. -Brad, karp@eecs.harvard.edu To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-mobile" in the body of the message