[manet] Re: [EXTERNAL] Re: Fwd: New Version Notifica tion for draft-eastlake-manet-babel-wi-fi-00.txt
Donald Eastlake <[email protected]> Thu, 16 Jul 2026 21:03:50 -0400
| Newsgroups | gmane.ietf.manet |
|---|---|
| Message-ID | <CAF4+nEGLg_+g175wHZ6GQuV8wuziuZTAeHvSc35W7KvnXy0fDw@mail.gmail.com> |
--===============0472624959578054044== Content-Type: multipart/alternative; boundary="000000000000d4e5040656c421cc" --000000000000d4e5040656c421cc Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable See below at <DE> <DE> Hi, On Tue, Jul 14, 2026 at 10:46=E2=80=AFAM Templin (US), Fred L < [email protected]> wrote: Hi Donald, asking AI: =E2=80=9CHow does IEEE802.11 IBSS mode differ from IE= EE802.11 mesh?=E2=80=9D produces illuminating results. Back when I was doing multi-h= op routing with TBRPF all we had to work with was IBSS mode, as IEEE 802.11s did not exist yet. <DE> I don't know what AIs say, but IBSS (Independent Basic Service Set) mode is a simple one-hop complete mesh. Every Station needs to be able to transmit to and receive from every other Station in the IBSS. The original target scenario was a set of people sitting around a conference table in the absence of any Access Points (APs) that wanted to communicate with each other to share files or whatever. But, in my experience, IBSS is mostly used by personal computers that want to tether to a cell phone to access their data over cellular. <DE> 802.11 standard mesh (MBSS) is quite different with a multi-hop forwarding model inside the mesh. It is my understanding that most 802.11 "mesh" products are not conformant to the 802.11 standard although there are implementations and an early version of the standard was incorporated into the One Laptop Per Child project. Being a mesh Station and being an AP are orthogonal and a Station can be both. I can picture how multiple (heterogeneous) MANET interfaces works with IBSS mode, but I am less certain about multiple interfaces and IEEE802.11s. <DE> I don't want to get too deep into the 802.11 logical architecture but there are a bunch of complexities including mesh gateways, portals, etc. But in any case, the idea is to use Babel inside the existing 802.11 mesh standard so the way a mesh Station interconnects with things that are not part of the mesh would not change. Thank you - Fred On Tue, Jul 14, 2026 at 12:54=E2=80=AFPM Juliusz Chroboczek <[email protected]> w= rote: >> Is a similar profile possible for other MANET routing protocols over >> Wi-Fi mesh =E2=80=93 for example, OLSRv2? Or is there something special = about >> Babel that makes it better suited for Wi-Fi mesh? In the original 802.11s, there were two routing protocols: HWMP, which was MTI, and RA-OLSR (a variant of OLSRv1), which was optional. RA-OLSR failed in the market, I'm not sure if it is still in the spec (I haven't read the recent revisions of 802.11s). <DE> There were a number of people in the 802.11s Task Group pushing RA-OLSR but there is always tremendous pressure to simplify standards in 802. For various reasons, a number of them dropped out. Technical votes in 802 require a 3/4 supermajority and there had been 25+% supporting RA-OLSR which kept it in the standards. But when several of its proponents stopped participating, very quickly there was a 3/4 in favor of stripping RA-OLSR out of the draft and it was gone. <DE> As usual in 802, the 802.11 WG periodically does roll-ups to include amendments into the main standard. So there is no more "802.11s" and the combined IEEE Std 802.11 still has only the HWMP routing protocol for mesh. (The most recent version is IEEE Std 802.11-2024 which is 5,954 pages long.= ) On Tue, Jul 14, 2026 at 1:12=E2=80=AFPM Christopher Dearlove < [email protected]> wrote: The biggest potential option that basing on OLSRv2 rather than OLSR(v1) would be link metrics. (You=E2=80=99d be building a new way to encode data I presume so the 5444 f= ormat wouldn=E2=80=99t apply. In an ideal world the authors would have liked to s= eparate the structure of OLSRv2 from its 5444 encoding, but having already spun off NHDP that was just a step too far.) <DE> Well, you want to envelope the routing protocol messages within appropriate 802.11 frame constructs. But you might be able to do that with 5444 format messages. <DE> 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] > --000000000000d4e5040656c421cc Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable <div dir=3D"ltr"><div dir=3D"ltr"><p class=3D"MsoNormal" style=3D"margin:0i= n;font-size:12pt;font-family:Aptos,sans-serif">See below at <DE></p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A0</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif"><DE> Hi,</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A0</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">On Tue, Jul 14, 2026 at 10:46<span style=3D"font-family:Arial,= sans-serif">=E2=80=AF</span>AM Templin (US), Fred L <<a href=3D"mailto:[email protected]">Fred.= [email protected]</a>> wrote:</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">Hi Donald, asking AI: =E2=80=9CHow does IEEE802.11 IBSS mode d= iffer from IEEE802.11 mesh?=E2=80=9D produces illuminating results. Back when I w= as doing multi-hop routing with TBRPF all we had to work with was IBSS mode, as IEEE 802.11s did not exist yet.</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A0</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif"><DE> I don't know what AIs say, but IBSS (Independen= t Basic Service Set) mode is a simple one-hop complete mesh. Every Station ne= eds to be able to transmit to and receive from every other Station in the IBSS. The original target scenario was a set of people sitting around a conferenc= e table in the absence of any Access Points (APs) that wanted to communicate = with each other to share files or whatever. But, in my experience, IBSS is mostl= y used by personal computers that want to tether to a cell phone to access th= eir data over cellular.</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A0</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif"><DE> 802.11 standard mesh (MBSS) is quite different with a multi-hop forwarding model inside the mesh. It is my understanding t= hat most 802.11 "mesh" products are not conformant to the 802.11 stan= dard although there are implementations and an early version of the standard was incorporated into the One Laptop Per Child project. Being a mesh Station an= d being an AP are orthogonal and a Station can be both.</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A0</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">I can picture how multiple (heterogeneous) MANET interfaces works with IBSS mode, but I am less certain about multiple interfaces and IEEE802.11s.</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A0</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif"><DE> I don't want to get too deep into the 802.11 logical architecture but there are a bunch of complexities including mesh gateways, portals, etc. But in any case, the idea is to use Babel inside th= e existing 802.11 mesh standard so the way a mesh Station interconnects with things that are not part of the mesh would not change.</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A0</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">Thank you - Fred</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A0</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">On Tue, Jul 14, 2026 at 12:54<span style=3D"font-family:Arial,= sans-serif">=E2=80=AF</span>PM Juliusz Chroboczek <<a href=3D"mailto:[email protected]">[email protected]</a>> w= rote:</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">>> Is a similar profile possible for other MANET routing protocols over</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">>> Wi-Fi mesh =E2=80=93 for example, OLSRv2? Or is there something special about</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">>> Babel that makes it better suited for Wi-Fi mesh?</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A0</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">In the original 802.11s, there were two routing protocols: HWMP, which was</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">MTI, and RA-OLSR (a variant of OLSRv1), which was optional.=C2=A0 RA-OLSR</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">failed in the market, I'm not sure if it is still in the spec (I haven't</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">read the recent revisions of 802.11s).</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A0</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif"><DE> There were a number of people in the 802.11s Task Group pushing RA-OLSR but there is always tremendous pressure to simplify standards in 802. For various reasons, a number of them dropped out. Techni= cal votes in 802 require=C2=A0a 3/4 supermajority and there had been 25+% suppo= rting RA-OLSR which kept it in the standards. But when several of its proponents stopped participating, very quickly there was a 3/4 in favor of stripping RA-OLSR out of the draft and it=C2=A0was gone.</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A0</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif"><DE> As usual in 802, the 802.11 WG periodically does roll-ups to include amendments into the main standard. So there is no more "802.11s" and the combined IEEE Std 802.11 still has only the HWM= P routing protocol for mesh. (The most recent version is IEEE Std 802.11-2024 which is 5,954 pages long.)</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A0</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">On Tue, Jul 14, 2026 at 1:12<span style=3D"font-family:Arial,s= ans-serif">=E2=80=AF</span>PM Christopher Dearlove <<a href=3D"mailto:[email protected]">= [email protected]</a>> wrote:</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">The biggest potential option that basing on OLSRv2 rather than OLSR(v1) would be link metrics.</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A0</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">(You=E2=80=99d be building a new way to encode data I presume = so the 5444 format wouldn=E2=80=99t apply. In an ideal world the authors would hav= e liked to separate the structure of OLSRv2 from its 5444 encoding, but having already spun off NHDP that was just a step too far.)</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A0</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif"><DE> Well, you want to envelope the routing protocol messages within appropriate 802.11 frame constructs. But you might be able = to do that with 5444 format messages.</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A0</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A0</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif"><DE> Thanks,</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">Donald</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=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</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A0Donald E. Eastlake 3rd</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A02386 Panoramic Circle, Apopka, FL 32703 USA</p> <p class=3D"MsoNormal" style=3D"margin:0in;font-size:12pt;font-family:Aptos= ,sans-serif">=C2=A0<a href=3D"mailto:[email protected]">[email protected]</a>= </p></div><div class=3D"gmail_quote gmail_quote_container"><blockquote clas= s=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid r= gb(204,204,204);padding-left:1ex"><div class=3D"msg8625558456962407680"><di= v lang=3D"EN-US" style=3D"overflow-wrap: break-word;"><div class=3D"m_86255= 58456962407680WordSection1"><div style=3D"border-width:medium medium medium= 1.5pt;border-style:none none none solid;border-color:currentcolor currentc= olor currentcolor blue;padding:0in 0in 0in 4pt"><div><div> </div> </div> </div> </div> </div> </div></blockquote></div></div> --000000000000d4e5040656c421cc-- --===============0472624959578054044== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: base64 Content-Disposition: inline X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18KbWFuZXQgbWFp bGluZyBsaXN0IC0tIG1hbmV0QGlldGYub3JnClRvIHVuc3Vic2NyaWJlIHNlbmQgYW4gZW1haWwg dG8gbWFuZXQtbGVhdmVAaWV0Zi5vcmcK --===============0472624959578054044==--