From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from smtp120.iad3a.emailsrvr.com (smtp120.iad3a.emailsrvr.com [173.203.187.120]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 610BE3B2A4 for ; Fri, 14 Oct 2022 16:14:13 -0400 (EDT) Received: from app36.wa-webapps.iad3a (relay-webapps.rsapps.net [172.27.255.140]) by smtp16.relay.iad3a.emailsrvr.com (SMTP Server) with ESMTP id DC77C5B6B for ; Fri, 14 Oct 2022 16:14:12 -0400 (EDT) Received: from deepplum.com (localhost.localdomain [127.0.0.1]) by app36.wa-webapps.iad3a (Postfix) with ESMTP id C8D63603D7; Fri, 14 Oct 2022 16:14:12 -0400 (EDT) Received: by apps.rackspace.com (Authenticated sender: dpreed@deepplum.com, from: dpreed@deepplum.com) with HTTP; Fri, 14 Oct 2022 16:14:12 -0400 (EDT) X-Auth-ID: dpreed@deepplum.com Date: Fri, 14 Oct 2022 16:14:12 -0400 (EDT) From: "David P. Reed" To: starlink@lists.bufferbloat.net Cc: starlink@lists.bufferbloat.net MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_20221014161412000000_49943" Importance: Normal X-Priority: 3 (Normal) X-Type: html In-Reply-To: References: X-Client-IP: 209.6.168.128 Message-ID: <1665778452.8193981@apps.rackspace.com> X-Mailer: webmail/19.0.21-RC X-Classification-ID: 02d5964e-ab27-4679-b87b-c1e563d0445e-1-1 Subject: Re: [Starlink] The DoD "Transport Layer" X-BeenThere: starlink@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: "Starlink has bufferbloat. Bad." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 14 Oct 2022 20:14:13 -0000 ------=_20221014161412000000_49943 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable =0AHi Dave - =0A =0AWell, you may not be happy with my response, but I thin= k my views below are likely to play out in some form that is pretty predict= able. I think it will be a bad result in Space. (The idea of Space being "f= ree" is very unlikely to occur, just as unlikely as the current Internet wa= s to happen in 1975.)=0A =0A=0A> Date: Fri, 14 Oct 2022 12:31:35 -0700=0A> = From: Dave Taht =0A> To: "David P. Reed" =0A> Cc: starlink@lists.bufferbloat.net=0A> Subject: Re: [Starlink]= The DoD "Transport Layer"=0A> Message-ID:=0A> =0A> Content-Type: text/plain; char= set=3D"UTF-8"=0A> =0A> Dear David:=0A> =0A> Would it cheer you up any to le= arn, that 15+ years after the debate=0A> over UWB ended, that it's finally = seeing FAR more=0A> major uptake and reasonable standardization, and actual= working chips?=0A>=0A[DPR} not that much cheered up, actually. While that = disaster around UWB got me interested in how the politics worked, that isn'= t what depressed me. I won't be cheered up until the FCC stops treating the= spectrum like property and started basing its decisions on achieving fully= scalable wireless networking. UWB doesn't address that issue. It's unscala= ble for the same reasons - the misunderstanding of information theory and p= hysics of propagation that remains endemic in the whole framework of spectr= um "property rights". =0A=0A> It did me. I was pretty scarred by that mess = also, and what was it?=0A> the 272 notches the FCC demanded be cut out of i= t, which swamped=0A> circuit design capabilities at the time... but not as = bad as you.=0A> =0A> I didn't know until recently that it had hit iphones i= n 2019. and was=0A> part of the airtags, nor that the baseline latency on t= he things was=0A> 50us, with admittedly only a 1000 bit payload - Still cri= ppled as to=0A> distance, and total bandwidth to under 10mbits, but, power = usage is=0A> *amazing*.=0A> =0A> https://en.wikipedia.org/wiki/Ultra-wideba= nd=0A> =0A> The SPARK chips in particular have a nice looking devkit.=0A> = =0A> Anyway... just as the swamp of ipx and non-interoperable email systems= =0A> finally died...=0A> =0A> You can't in the end, keep a good idea down. = Maybe on average it takes=0A> 25 years to settle on saner things.=0A =0AIt'= s taken over 50, and counting, for the FCC to acknowledge that co-channel s= ignalling actually works, because in digital systems we have channel coding= since Shannon first wrote about information theory. And that is only the b= eginning of what is wrong with the "property" model, which assumes all wire= less signals require a perfectly clear channel.=0A =0AThe FCC still doesn't= acknowledge that Cooperative signalling protocols can create huge capacity= gains, or that it's possible that multiple co-channel signals can actually= create channel capacity that grows with the number of antennas (as long as= modest cooperation is ensured).=0A =0AThe FCC still doesn't acknowledge th= at the Internet is a unifying "service" that obviates almost all of the con= cepts of "allocating spectrum" to "services". (they still treat Broadcast s= ervices separately from telephony, and telephony separately from Land Mobil= e, etc.) So, for example, Emergency Communications is regulated as if the I= nternet cannot be utilized, as just one example. In other words, to a thoug= htful communications engineer, the FCC is a joke.=0A =0AMostly this is due = to two factors. 1. Property rights creates opportunity for scarcity based m= onopoly to be granted by the government to its friends. 2. The folks who ha= ve demonstrated these technologies (using information theory and propagatio= n physics and internetworking of wireless nets) are paid entirely by the wo= uld be monopolists (what used to be called "The Phone Company", the evil co= nspiracy of The President's Analyst, which you might have seen). The FCC is= a captured regulator. And its role, sanctified by Congress is to create si= loed monopolies. Not for the public good, but for the control of communicat= ions and enrichment of the controllers.=0A=0A> =0A> ... We have centuries t= o sort the solar system's internet out, and the more=0A> we can do to convi= nce the next generation as to the right principles=0A> to apply to it, the = better.=0A =0AI don't think the World Radio Conference (which manages all R= F services in the world, including the US), even has thought about Space, b= ut to be honest, what they want is to control all Space communications on b= ehalf of all governments, most of which derive substantial revenue by block= ing innovative new ideas.=0A =0AI am sad that is true, but it is almost cer= tainly gonna happen. The DoD will play the same role it did with radio in t= he beginning of the 20th century, buying up all the patents, blocking any n= ew entrants, and eventually creating RCA, a monopoly on all radio technolog= y. That will almost certainly happen to the Solar System's communications (= and property rights on messages from the earth to asteroids will be *owned*= by some company, backed by the coercive power of the governments colonizin= g space).'=0A=0A> =0A> =0A> ------------------------------=0A> =0A> Subject= : Digest Footer=0A> =0A> _______________________________________________=0A= > Starlink mailing list=0A> Starlink@lists.bufferbloat.net=0A> https://list= s.bufferbloat.net/listinfo/starlink=0A> =0A> =0A> -------------------------= -----=0A> =0A> End of Starlink Digest, Vol 19, Issue 7=0A> ****************= ***********************=0A> ------=_20221014161412000000_49943 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable

Hi Dave - 

=0A=

 

=0A

Well, you may not be= happy with my response, but I think my views below are likely to play out = in some form that is pretty predictable. I think it will be a bad result in= Space. (The idea of Space being "free" is very unlikely to occur, just as = unlikely as the current Internet was to happen in 1975.)

=0A

 

=0A
=0A

> Date: Fri, 14 Oct 2022 12:31:35 -0700
> From: Dave Taht &l= t;dave.taht@gmail.com>
> To: "David P. Reed" <dpreed@deepplum= .com>
> Cc: starlink@lists.bufferbloat.net
> Subject: Re= : [Starlink] The DoD "Transport Layer"
> Message-ID:
> <= CAA93jw4bZBDf3jJ-dboBbf9PS2TsYYJhW+myWHNUdOt7CJqWTw@mail.gmail.com>
> Content-Type: text/plain; charset=3D"UTF-8"
>
> Dear= David:
>
> Would it cheer you up any to learn, that 15+ y= ears after the debate
> over UWB ended, that it's finally seeing FA= R more
> major uptake and reasonable standardization, and actual wo= rking chips?
>

=0A

[DPR} not that much cheer= ed up, actually. While that disaster around UWB got me interested in how th= e politics worked, that isn't what depressed me. I won't be cheered up unti= l the FCC stops treating the spectrum like property and started basing its = decisions on achieving fully scalable wireless networking. UWB doesn't addr= ess that issue. It's unscalable for the same reasons - the misunderstanding= of information theory and physics of propagation that remains endemic in t= he whole framework of spectrum "property rights". 

=0A


> It did me. I was pretty scarred by that mess also, and w= hat was it?
> the 272 notches the FCC demanded be cut out of it, wh= ich swamped
> circuit design capabilities at the time... but not as= bad as you.
>
> I didn't know until recently that it had = hit iphones in 2019. and was
> part of the airtags, nor that the ba= seline latency on the things was
> 50us, with admittedly only a 100= 0 bit payload - Still crippled as to
> distance, and total bandwidt= h to under 10mbits, but, power usage is
> *amazing*.
>
> https://en.wikipedia.org/wiki/Ultra-wideband
>
> Th= e SPARK chips in particular have a nice looking devkit.
>
>= ; Anyway... just as the swamp of ipx and non-interoperable email systems> finally died...
>
> You can't in the end, keep a g= ood idea down. Maybe on average it takes
> 25 years to settle on sa= ner things.

=0A

 

=0A

It= 's taken over 50, and counting, for the FCC to acknowledge that co-cha= nnel signalling actually works, because in digital systems we have channel = coding since Shannon first wrote about information theory. And that is only= the beginning of what is wrong with the "property" model, which assumes al= l wireless signals require a perfectly clear channel.

=0A

 

=0A

The FCC still doesn't acknowledge t= hat Cooperative signalling protocols can create huge capacity gains, or tha= t it's possible that multiple co-channel signals can actually create channe= l capacity that grows with the number of antennas (as long as modest cooper= ation is ensured).

=0A

 

=0A

The FCC still doesn't acknowledge that the Internet is a unifying "ser= vice" that obviates almost all of the concepts of "allocating spectrum" to = "services". (they still treat Broadcast services separately from telephony,= and telephony separately from Land Mobile, etc.) So, for example, Emergenc= y Communications is regulated as if the Internet cannot be utilized, as jus= t one example. In other words, to a thoughtful communications engineer, the= FCC is a joke.

=0A

 

=0A

=0A


>
> ... We have centuries to sort the solar = system's internet out, and the more
> we can do to convince the nex= t generation as to the right principles
> to apply to it, the bette= r.

=0A

 

=0A

I don't thi= nk the World Radio Conference (which manages all RF services in the world, = including the US), even has thought about Space, but to be honest, what the= y want is to control all Space communications on behalf of all governments,= most of which derive substantial revenue by blocking innovative new ideas.=

=0A

 

=0A

I am sad that= is true, but it is almost certainly gonna happen. The DoD will play the sa= me role it did with radio in the beginning of the 20th century, buying up a= ll the patents, blocking any new entrants, and eventually creating RCA, a m= onopoly on all radio technology. That will almost certainly happen to the S= olar System's communications (and property rights on messages from the eart= h to asteroids will be *owned* by some company, backed by the coercive powe= r of the governments colonizing space).'

=0A


&= gt;
>
> ------------------------------
>
&g= t; Subject: Digest Footer
>
> ____________________________= ___________________
> Starlink mailing list
> Starlink@list= s.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink=
>
>
> ------------------------------
> <= br />> End of Starlink Digest, Vol 19, Issue 7
> ***************= ************************
>

=0A
 ------=_20221014161412000000_49943--