[manet] Re: Fwd: New Version Notification for draft-eastla ke-manet-babel-wi-fi-00.txt
Donald Eastlake <[email protected]> Tue, 14 Jul 2026 10:10:00 -0400
| Newsgroups | gmane.ietf.manet |
|---|---|
| Message-ID | <CAF4+nEFjRbTyncntr=creznCJbC6kr7HRuRnAnDpD46QjDvi5w@mail.gmail.com> |
--===============5476425434204201821== Content-Type: multipart/alternative; boundary="000000000000e21910065692c368" --000000000000e21910065692c368 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Hi Fred, See below for follow-up to your follow-up: On Mon, Jul 13, 2026 at 6:13=E2=80=AFPM Templin (US), Fred L < [email protected]> wrote: > > Hi Donald, see below for follow-up: > > > -----Original Message----- > > From: Donald Eastlake <[email protected]> > > Sent: Monday, July 13, 2026 2:50 PM > > To: Templin (US), Fred L <[email protected]> > > Cc: manet <[email protected]> > > Subject: Re: [manet] Fwd: New Version Notification for draft-eastlake-manet-babel-wi-fi-00.txt > > > > Hi Fred, > > > > On Mon, Jul 13, 2026 at 10:11=E2=80=AFAM Templin (US), Fred L > > <[email protected]> wrote: > > > > > > Thanks for the draft, Donald. A couple of quick comments: > > > > > > ... > > > > The draft is obviously still in early stages, but future versions may benefit from a discussion of multiple interfaces =E2=80=93 both multiple Wi= -Fi mesh > > interfaces and heterogeneous interface types (such as Wi-Fi mesh with 5G). > > > > As for multiple interfaces, I agree at a high level, though there are > > several complexities to consider: > > > > 1. While it is possible to have multiple IEEE 802 interfaces (such as > > Wi-Fi, Bluetooth, and Ethernet), Wi-Fi mesh and similar > > multi-interface extensions would likely continue to route based on MAC > > addresses. > > OK, so I see now what you mean by "mesh" is multi-hop routing based on MAC addresses and not "classical" MANET routing based on IP addresses. If yes, then I think we may need to reconsider use of the term "mesh" in the MANET Internetworking draft because the intention there was for IP address-based forwarding. I don't think the word "mesh" is limited to a particular layer in the OSI model. In the IETF it almost always means layer 3, IP mesh. But IEEE 802.11 (Wi-Fi) and 802.15.x (Bluetooth, etc.) and 802.1 (bridging) are all layer 2. (I'm not sure it makes sense to talk about a layer 1 "mesh" although I guess you could have a complicated acyclic layer 1 network with amplifiers, junctions, copper <-> optical fiber converters...) I think it's fine for the MANET internetworking draft to continue to use "mesh" for IP mesh and, if anything, the "Wi-Fi mesh" draft should make it clearer that it is NOT layer-3/IP. > > A Wi-Fi mesh appears to higher levels, like IP routers or > > hosts, as a single multi-access link. > > Yes, it looks like an Ethernet switch with all nodes as single-hop neighbors at the IP layer even though there may be multiple hops needed at the MAC layer - correct? Yes, although, of course, the cost/bandwidth/delay between any pair of interfaces attached to this synthetic multi-access link may be different from those measures for other pairs of interfaces. > > It is entirely possible to have > > an instance of Babel routing on IP addresses where some links are > > actually Wi-Fi mesh instances using disjoint Babel or other MAC > > address routing. I am not certain that 5G would be an appropriate > > interface type in this context. > > Sorry - I meant to specify 5G SideLink - as in, operating outside the context of 5G infrastructure. Then, with forwarding at the IP layer instead of the MAC layer the 5G SideLink and 802.11 in "ad-hoc mode" could be multiple interfaces belonging to the same MANET. > > > 2. A station with a single Wi-Fi radio can have multiple mesh > > peerings. However, if you are referring to multiple radios, my > > understanding is that 802.11 requires a logically separate station per > > radio=E2=80=94even if physically combined=E2=80=94each with a different= MAC address. > > This area requires further thought. > > I think multiple interfaces doing MAC-layer forwarding is a different analysis than multiple interfaces doing IP-layer forwarding. So, I think I will stop now to avoid introducing any more confusion. Ok. =F0=9F=99=82 Thanks, Donald =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D Donald E. Eastlake 3rd 2386 Panoramic Circle, Apopka, FL 32703 USA [email protected] > Thank you - Fred > > > > > Thanks, > > Donald > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D > > Donald E. Eastlake 3rd > > 2386 Panoramic Circle, Apopka, FL 32703 USA > > [email protected] > > > > > Thank you =E2=80=93 Fred Templin > > > > > > > > > > > > From: Donald Eastlake <[email protected]> > > > Sent: Friday, July 10, 2026 6:45 PM > > > To: manet <[email protected]> > > > Subject: [EXTERNAL] [manet] Fwd: New Version Notification for draft-eastlake-manet-babel-wi-fi-00.txt > > > > > > > > > > > > Hi, > > > > > > > > > > > > I have posted an early, incomplete draft on the use of Babel for Wi-Fi (IEEE 802.11) mesh. Sorry it isn't further along but it is a start. > > > > > > > > > > > > Thanks, > > > Donald > > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D > > > Donald E. Eastlake 3rd > > > 2386 Panoramic Circle, Apopka, FL 32703 USA > > > [email protected] > > > > > > > > > > > > ---------- Forwarded message --------- > > > From: <[email protected]> > > > Date: Mon, Jul 6, 2026 at 7:00=E2=80=AFPM > > > Subject: New Version Notification for draft-eastlake-manet-babel-wi-fi-00.txt > > > To: Donald E. Eastlake 3rd <[email protected]> > > > > > > > > > A new version of Internet-Draft draft-eastlake-manet-babel-wi-fi-00.txt has > > > been successfully submitted by Donald E. Eastlake 3rd and posted to the > > > IETF repository. > > > > > > Name: draft-eastlake-manet-babel-wi-fi > > > Revision: 00 > > > Title: Babel for Wi-Fi (IEEE Std 802.11) Mesh > > > Date: 2026-07-05 > > > Group: Individual Submission > > > Pages: 8 > > > URL: https://www.ietf.org/archive/id/draft-eastlake-manet-babel-wi-fi-00.txt > > > Status: https://datatracker.ietf.org/doc/draft-eastlake-manet-babel-wi-fi/ > > > HTML: https://www.ietf.org/archive/id/draft-eastlake-manet-babel-wi-fi-00.html > > > HTMLized: https://datatracker.ietf.org/doc/html/draft-eastlake-manet-babel-wi-fi > > > > > > > > > Abstract: > > > > > > The BABEL routing protocol (RFC 8966) is well applicable (RFC 8967= ) > > > to networks with unstable link metrics such as wireless networks. > > > Wi-Fi (IEEE Std 802.11-2024) is an example of such a network and the > > > Wi-Fi standard includes a mesh feature which was specified to be > > > configurable for different routing protocols and link metrics. This > > > document specifies how, in Wi-Fi mesh, to use BABEL and/or the delay > > > based link metric specified in RFC 9616. > > > > > > > > > The IETF Secretariat > > > > --000000000000e21910065692c368 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable <div dir=3D"ltr">Hi Fred,<br><br>See below for follow-up to your follow-up:= <br><br>On Mon, Jul 13, 2026 at 6:13=E2=80=AFPM Templin (US), Fred L <<a= href=3D"mailto:[email protected]">[email protected]</a>>= ; wrote:<br>><br>> Hi Donald, see below for follow-up:<br>><br>>= ; > -----Original Message-----<br>> > From: Donald Eastlake <<a= href=3D"mailto:[email protected]">[email protected]</a>><br>> > Sen= t: Monday, July 13, 2026 2:50 PM<br>> > To: Templin (US), Fred L <= <a href=3D"mailto:[email protected]">[email protected]</a>&= gt;<br>> > Cc: manet <<a href=3D"mailto:[email protected]">manet@ietf= .org</a>><br>> > Subject: Re: [manet] Fwd: New Version Notificatio= n for draft-eastlake-manet-babel-wi-fi-00.txt<br>> ><br>> > Hi = Fred,<br>> ><br>> > On Mon, Jul 13, 2026 at 10:11=E2=80=AFAM Te= mplin (US), Fred L<br>> > <<a href=3D"mailto:Fred.L.Templin@boeing= .com">[email protected]</a>> wrote:<br>> > ><br>> &g= t; > Thanks for the draft, Donald. A couple of quick comments:<br>> &= gt; ><br>> > > ...<br>><br>> > > The draft is obvio= usly still in early stages, but future versions may benefit from a discussi= on of multiple interfaces =E2=80=93 both multiple Wi-Fi mesh<br>> > i= nterfaces and heterogeneous interface types (such as Wi-Fi mesh with 5G).<b= r>> ><br>> > As for multiple interfaces, I agree at a high leve= l, though there are<br>> > several complexities to consider:<br>> = ><br>> > 1.=C2=A0 While it is possible to have multiple IEEE 802 i= nterfaces (such as<br>> > Wi-Fi, Bluetooth, and Ethernet), Wi-Fi mesh= and similar<br>> > multi-interface extensions would likely continue = to route based on MAC<br>> > addresses.<br>><br>> OK, so I see = now what you mean by "mesh" is multi-hop routing based on MAC add= resses and not "classical" MANET routing based on IP addresses. I= f yes, then I think we may need to reconsider use of the term "mesh&qu= ot; in the MANET Internetworking draft because the intention there was for = IP address-based forwarding.<br><br>I don't think the word "mesh&q= uot; is limited to a particular layer in the OSI model. In the IETF it almo= st always means layer 3, IP mesh. But IEEE 802.11 (Wi-Fi) and 802.15.x (Blu= etooth, etc.) and 802.1 (bridging) are all layer 2. (I'm not sure it ma= kes sense to talk about a layer 1 "mesh" although I guess you cou= ld have a complicated acyclic layer 1 network with amplifiers, junctions, c= opper <-> optical fiber converters...) I think it's fine for the = MANET internetworking draft to continue to use "mesh" for IP mesh= and, if anything, the "Wi-Fi mesh" draft should make it clearer = that it is NOT layer-3/IP.<br><br>> > A Wi-Fi mesh appears to higher = levels, like IP routers or<br>> > hosts, as a single multi-access lin= k.<br>><br>> Yes, it looks like an Ethernet switch with all nodes as = single-hop neighbors at the IP layer even though there may be multiple hops= needed at the MAC layer - correct?<br><br>Yes, although, of course, the co= st/bandwidth/delay between any pair of interfaces attached to this syntheti= c multi-access link may be different from those measures for other pairs of= interfaces.<div><br>> > It is entirely possible to have<br>> >= an instance of Babel routing on IP addresses where some links are<br>> = > actually Wi-Fi mesh instances using disjoint Babel or other MAC<br>>= ; > address routing. I am not certain that 5G would be an appropriate<br= >> > interface type in this context.<br>><br>> Sorry - I meant = to specify 5G SideLink - as in, operating outside the context of 5G infrast= ructure. Then, with forwarding at the IP layer instead of the MAC layer the= 5G SideLink and 802.11 in "ad-hoc mode" could be multiple interf= aces belonging to the same MANET.<br>><br>> > 2.=C2=A0 A station w= ith a single Wi-Fi radio can have multiple mesh<br>> > peerings. Howe= ver, if you are referring to multiple radios, my<br>> > understanding= is that 802.11 requires a logically separate station per<br>> > radi= o=E2=80=94even if physically combined=E2=80=94each with a different MAC add= ress.<br>> > This area requires further thought.<br>><br>> I th= ink multiple interfaces doing MAC-layer forwarding is a different analysis = than multiple interfaces doing IP-layer forwarding. So, I think I will stop= now to avoid introducing any more confusion.<br><br>Ok.=C2=A0=F0=9F=99=82<= /div><div><br></div><div>Thanks,</div><div>Donald<br>=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<br= >=C2=A0Donald E. Eastlake 3rd<br>=C2=A02386 Panoramic Circle, Apopka, FL 32= 703 USA<br>=C2=A0<a href=3D"mailto:[email protected]">[email protected]</a><b= r></div><div><br>> Thank you - Fred<br>><br>> ><br>> > Th= anks,<br>> > Donald<br>> > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<br>> > =C2= =A0Donald E. Eastlake 3rd<br>> > =C2=A02386 Panoramic Circle, Apopka,= FL 32703 USA<br>> > =C2=A0<a href=3D"mailto:[email protected]">d3e3e3= @gmail.com</a><br>> ><br>> > > Thank you =E2=80=93 Fred Temp= lin<br>> > ><br>> > ><br>> > ><br>> > >= From: Donald Eastlake <<a href=3D"mailto:[email protected]">d3e3e3@gmail= .com</a>><br>> > > Sent: Friday, July 10, 2026 6:45 PM<br>> = > > To: manet <<a href=3D"mailto:[email protected]">[email protected]</a= >><br>> > > Subject: [EXTERNAL] [manet] Fwd: New Version Notifi= cation for draft-eastlake-manet-babel-wi-fi-00.txt<br>> > ><br>>= ; > ><br>> > ><br>> > > Hi,<br>> > ><br>&g= t; > ><br>> > ><br>> > > I have posted an early, in= complete draft on the use of Babel for Wi-Fi (IEEE 802.11) mesh. Sorry it i= sn't further along but it is a start.<br>> > ><br>> > &g= t;<br>> > ><br>> > > Thanks,<br>> > > Donald<br>= > > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<br>> > > =C2=A0Donald E. Eastlak= e 3rd<br>> > > =C2=A02386 Panoramic Circle, Apopka, FL 32703 USA<b= r>> > > =C2=A0<a href=3D"mailto:[email protected]">[email protected]= </a><br>> > ><br>> > ><br>> > ><br>> > >= ; ---------- Forwarded message ---------<br>> > > From: <<a hre= f=3D"mailto:[email protected]">[email protected]</a>><br>&= gt; > > Date: Mon, Jul 6, 2026 at 7:00=E2=80=AFPM<br>> > > S= ubject: New Version Notification for draft-eastlake-manet-babel-wi-fi-00.tx= t<br>> > > To: Donald E. Eastlake 3rd <<a href=3D"mailto:d3e3e3= @gmail.com">[email protected]</a>><br>> > ><br>> > ><br= >> > > A new version of Internet-Draft draft-eastlake-manet-babel-= wi-fi-00.txt has<br>> > > been successfully submitted by Donald E.= Eastlake 3rd and posted to the<br>> > > IETF repository.<br>> = > ><br>> > > Name: =C2=A0 =C2=A0 draft-eastlake-manet-babel-= wi-fi<br>> > > Revision: 00<br>> > > Title: =C2=A0 =C2=A0= Babel for Wi-Fi (IEEE Std 802.11) Mesh<br>> > > Date: =C2=A0 =C2= =A0 2026-07-05<br>> > > Group: =C2=A0 =C2=A0Individual Submission<= br>> > > Pages: =C2=A0 =C2=A08<br>> > > URL: =C2=A0 =C2= =A0 =C2=A0<a href=3D"https://www.ietf.org/archive/id/draft-eastlake-manet-b= abel-wi-fi-00.txt">https://www.ietf.org/archive/id/draft-eastlake-manet-bab= el-wi-fi-00.txt</a><br>> > > Status: =C2=A0 <a href=3D"https://dat= atracker.ietf.org/doc/draft-eastlake-manet-babel-wi-fi/">https://datatracke= r.ietf.org/doc/draft-eastlake-manet-babel-wi-fi/</a><br>> > > HTML= : =C2=A0 =C2=A0 <a href=3D"https://www.ietf.org/archive/id/draft-eastlake-m= anet-babel-wi-fi-00.html">https://www.ietf.org/archive/id/draft-eastlake-ma= net-babel-wi-fi-00.html</a><br>> > > HTMLized: <a href=3D"https://= datatracker.ietf.org/doc/html/draft-eastlake-manet-babel-wi-fi">https://dat= atracker.ietf.org/doc/html/draft-eastlake-manet-babel-wi-fi</a><br>> >= ; ><br>> > ><br>> > > Abstract:<br>> > ><br>&= gt; > > =C2=A0 =C2=A0The BABEL routing protocol (RFC 8966) is well ap= plicable (RFC 8967)<br>> > > =C2=A0 =C2=A0to networks with unstabl= e link metrics such as wireless networks.<br>> > > =C2=A0 =C2=A0Wi= -Fi (IEEE Std 802.11-2024) is an example of such a network and the<br>> = > > =C2=A0 =C2=A0Wi-Fi standard includes a mesh feature which was spe= cified to be<br>> > > =C2=A0 =C2=A0configurable for different rout= ing protocols and link metrics.=C2=A0 This<br>> > > =C2=A0 =C2=A0d= ocument specifies how, in Wi-Fi mesh, to use BABEL and/or the delay<br>>= > > =C2=A0 =C2=A0based link metric specified in RFC 9616.<br>> &g= t; ><br>> > ><br>> > > The IETF Secretariat<br>> &g= t; ><br>><br></div></div> --000000000000e21910065692c368-- --===============5476425434204201821== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: base64 Content-Disposition: inline X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18KbWFuZXQgbWFp bGluZyBsaXN0IC0tIG1hbmV0QGlldGYub3JnClRvIHVuc3Vic2NyaWJlIHNlbmQgYW4gZW1haWwg dG8gbWFuZXQtbGVhdmVAaWV0Zi5vcmcK --===============5476425434204201821==--