Re: [DC3] Community networks, IETF/IRTF work, the GAIA WG

My comments in the list, as I guess it may be still relevant to the DC3 audience and not only to Sivasubramanian and me.

In general, if there are proposals related to what IETF/IRTF can do in the area of community networks, let us know Jane and me. They can be discussed in the IETF GAIA WG meetings or we can plan for a workshop to discuss it elsewhere. On 3/9/18 17:35, Sivasubramanian M wrote: Dear Leandro, On Mon, Sep 3, 2018 at 7:59 PM Leandro Navarro <leandro(a)pangea.org> wrote: The existing infrastructure owned / closely shared by telecom (wireless) operators and submarine / terrestrial cable networks might be under-utilized or sub-optimally utilized in different parts of the world. This is also true of space missions, which at least increasingly share payloads, but here again, the equipments tends to be dedicated to imaging, satellite telephony, broadcasting, Internet etc. Leaving satellites aside, if we focus on cables, network computers and PtP towers and equipment, the resources could be better shared, between operators, not only between the major telecom operators, but also between content networks, IXPs, Community Networks, and other major captive users such as a large University campus or a Government / military organization. Agreed, let me know if anyone is interested in contributing to this topic. I presented some slides about this in the last GAIA WG (link below) based on work guifi did for a specific rule for sharing terrestrial cables, but that can be applied to undersea and other stakeholders. Deploying a network should not be restricted to the major telecom providers, but to anyone (according to some rules), as it has impact on fundamental human rights.

Spectrum sharing models could help the present licensees resell unused Spectrum, but some technologies and standards may have to be developed to enable licence holders use the assigned spectrum much the same way as submarine cable divides data streams between transmissions. ISOC is about to launch a document about spectrum, including spectrum sharing. However IETF is not focused on spectrum, maybe only in the implications of it, since they tend to focus around the IP protocol, and spectrum is not even digital :-) The IEEE 802 committee would be relevant regarding standardization of sharing mechanisms/protocols. Sharing seems to be hard for large orgs and human beings in general (it has the overhead of coordination). The natural strong trend in cable or spectrum owners is to keep idle resources for any potential future use, instead of sharing them, even as a return to the nearby population. Prescriptive regulation may break these barriers, since recommendations about sharing seems to have little effect.

The Convergence of technologies is not only about converging voice and data (phone and computer), but also broadcasting. Could the IETF pay attention to the possibilities of merging tower technologies for phone, IP Telephone, Data, Radio and Television Broadcast, Two way Radio, Satellite Telephony (whatever is technically feasible) and also recommend standards for equipment at the user's end, to make the phone double up as a two way radio (for use in emergency situations where a sim card may not work) and to make the cellular phone work as a satellite phone? Another idea would be to build in the capabilities for a low power emergency / always-on-for-all-for-a-lifetime Internet (say at 144 kbps) which will work free of cellular towers anywhere in the everyday mobile phone and correspondingly determine an architecture and infrastructure that would co-exist with the Gigabit Internet. That seems related to a WG in slow wireless networks, for instance https://datatracker.ietf.org/wg/roll/ and perhaps https://datatracker.ietf.org/wg/lpwan/ and the more general group on https://datatracker.ietf.org/wg/manet/

https://tools.ietf.org/html/rfc4542 and there was a WG now closed with several documents produced: https://datatracker.ietf.org/wg/ieprep/documents/

In fact some friends are working on this topic, in cases such as the "I am Alive (IAA) system". Emergency networks and community networks both share the focus on enabling all citizens under different circumstances. IETF / IRTF could work on synthesizing past and recent technologies to help develop Community Networks, with features that do not exist in a system of tense separation between the phone companies and Internet networks. Something like GAIAs RFC: https://datatracker.ietf.org/doc/rfc7962/

IETF could also break down the components of the 5G bundle, identify components (for example 1-20 GBPS bandwidth, the advances in P2P technologies etc) and determine what components are achievable independent of the 5G bundle, as for instance by building equipment based on 802.11 ac etc., and recommend open standards that match or surpass the 5G promises. Someone will have to propose, get participants and lead an effort like this.

Thanks for your contributions, Leandro. Sivasubramanian M > For instance, in the last GAIA meeting I had a talk about sharing of > fibre in submarine and terrestrial ( > https://datatracker.ietf.org/meeting/102/materials/slides-102-gaia-network-… > ) > > The WG page link lists all working groups - there are several - Which of > these groups are most pertinent to Community Networks? Does your question > pertain to existing working groups, or aspirationally suggest new topics > for new working groups? > > I was sharing the question, in general, about which IETF WG (a new WG is > starting too high) we should meet to ensure the Internet public > specifications they produce are "compatible" with community networks, or at > least they polish some specific obstacles. For instance, I saw in IETF WG > there was a discussion that one particular message in the protocol was > intended to facilitate flow management to ISPs at the risk of disclosing a > bit of info about usage, which means that specifications take into account > the needs of those that use them. > > I see, for instance, opportunities, but there must be many more: > > - in the routing and security area with trust in wireless mesh routing > protocols (http://people.ac.upc.edu/leandro/pubs/semtor.pdf) > > - or in the description (http://netjson.org/ or > http://en.wiki.guifi.net/wiki/CNML/en) and management of networks > (self-configuration, discovery if some adaptations needed for CN), > > - or in the use of decentralized algorithms for resource allocation (like > IP ranges or names) [there is a IRTF WG about that] > > - in ways to reduce the cost of Internet access, aggregating and sharing > N Internet gateways among M users (in mesh networks), with M >> N (instead > of the more expensive M == N for commercial interest) ( > http://people.ac.upc.edu/leandro/pubs/icnsc-acc17.pdf) > > - in considerations about human rights (HRPC), with an overlap with GAIA. > > I can help you to connect with to the corresponding contacts in IETF/IRTF. > > > Regards, Leandro. > > > Sivasubramanian M > > >> In the IRTF, the research branch of IETF, there are relevant WG (at >> least GAIA, HRPC, perhaps DINRG). >> As co-chair, together with Jane, of the GAIA WG ( >> https://datatracker.ietf.org/rg/gaia/about/), we would like to know >> from you and work to move forward. >> >> Leandro. >> On 28/8/18 23:02, Sivasubramanian Muthusamy wrote: >> >> Dear Leanardo, >> >> >> On Tue, Aug 28, 2018, 9:54 PM Leandro Navarro &lt;leandro(a)pangea.org&gt; >> wrote: >> >>> Nice to read about frugal 5G. I see you combine the Wi-Fi with mobile >>> operator models and centered into the SDN model. Nice, a bit confusing. I >>> don't see why centering all under SDN and a global controller. >>> >>> Just a general comment/clarification about wireless (mobile, Wi-Fi >>> ...). The G is about Generations, and there are several related but >>> separate things in the 5G bundle: >>> >> It is interesting that you have identified the components of the bundle >> of 5G promises. >> >>> - A lot of it (the "new radio") will come anyway to Wi-Fi (for instance >>> 802.11ax in the 2.4-5GHz, ad or ay in 60GHz) with speeds in the range of >>> 1-20 Gbps. One discussion from the IEEE Communications Society: >>> http://techblog.comsoc.org/2018/07/21/will-802-11ax-be-a-5g-contender/ >>> >> >> The aspiration here is the 1-20 GBPS speed. If ax or ay can deliver >> that, that is where our focus needs to be. >> >> Your link points to the IEEE paper that says >> >> "all such “5G” BWA deployments (e.g. Verizon, C-Spire, etc) are >> proprietary" >> >> ."once the cost curve comes down, 802.11ax Wi-Fi has the potential to >> deliver 5G-like user experiences at a fraction of the cost of similar >> cellular gear" >> >> >> Therefore, the target speeds of 1 Gbps of user experience and peak up to >>> 20 Gbps will come anyway and probably at the same time in mobile and Wi-Fi >>> networks (both types of radio chipsets more or less come from the same >>> sources). >>> >> >> If we have ax or ay standard access points at the last mile, what >> technologies we need in the middle that are not proprietary and not >> cartel-dependant? >> >>> - The other is about business models: 5G is about mobile operators, and >>> keeping as much as possible in their controlled networks, away from the >>> open Internet. >>> >> >> The idea is to break away from these business models, stay away from >> controlled networks, use open technologies, make an AP, build a tower of >> some other sort, connect to the wire on the sea shore and deliver 1-20 >> GBPS,, call it "CN nextG", forgive me for the pun :) >> >>> In that model, mobile operator networks offer very high quality (low >>> latency, performance guarantees) at a price inside their own network, while >>> keep both "eyeballs" and content providers as customers inside its network. >>> >> The Internet is outside (the operator's "walled garden"), >>> >> Not acceptable. This thread is about Internet as the lifeline, >> everywhere, in it's wholeness, for everyone, always. >> >>> slowerr and less predictable, in their model. In that centralized >>> (operator) model, software defined networks make a lot of sense, much less >>> in Internet networks like community networks. >>> >>> Wi-Fi and community networks clearly differ in the "business models" >>> from mobile operator networks. Your Wi-Fi access point at home may be >>> "yours" and can be meshed with others and become part of a community >>> network if you want to, but your 5G base station at home will be an >>> operator box, >>> >> which would be another perpetual trap >> >>> justt a service. >>> >>> - Regarding service cost and coverage, my bet is 5G, the big investment >>> required, may slow down the expansion of mobile access to rural and remote >>> areas, as the focus is in higher/premium quality service for a higher price >>> service, and higher infrastructure cost per customer (high margin), instead >>> of "best-effort"/commodity quality, at low price, low cost hw per customer >>> (thin or negative margin). >>> >>> If you want more, we have a research paper exploring some of these >>> issues http://people.ac.upc.edu/leandro/pubs/5G.pdf >>> >>> Comments are very welcome, >>> >> let's see if we can develop alternative models for "next generation" >>> models for everyone, >>> >> >> + + + 1 >> >> Sivasubramanian M >> >> not mainly focused on those with high-end mobiles in dense affluent >>> areas. >>> >>> Regards, Leandro. >>> >>> >>> On 28/8/18 05:41, Sarbani Banerjee Belur wrote: >>> >>> Dear Jane, >>> >>> Prof. Abhay Karandikar is working on 5G solutions for rural broadband. >>> His presentation can be viewed in the link below. >>> >>> >>> https://www.ee.iitb.ac.in/~karandi/talks/Rural%20Broadband%20and%20Frugal%2… >>> >>> With regards, >>> >>> Sarbani >>> >>> On 28-08-2018 03:54, Jane Coffin wrote: >>> >>> Hi Siva – >>> >>> >>> >>> Thank you for this. >>> >>> >>> >>> From what I have seen – 5G seems to be an urban solution with some >>> heavy equipment costs (still in projection phase). >>> >>> >>> >>> Have you seen any presentations on 5G and rural solutions? >>> >>> >>> Best, >>> >>> Jane >>> >>> >>> >>> >>> >>> Internet Society | www.internetsociety.org >>> >>> Skype: janercoffin >>> >>> Mobile/WhatsApp: +1.202.247.8429 >>> >>> >>> >>> *From: *&lt;dc3-bounces(a)listas.altermundi.net&gt; >>> &lt;dc3-bounces(a)listas.altermundi.net&gt; on behalf of Sivasubramanian M >>> &lt;6.Internet(a)gmail.com&gt; &lt;6.Internet(a)gmail.com&gt; >>> *Reply-To: *Dynamic Coalition on Community Connectivity >>> &lt;dc3(a)listas.altermundi.net&gt; &lt;dc3(a)listas.altermundi.net&gt; >>> *Date: *Monday, August 27, 2018 at 6:17 PM >>> *To: *&quot;steve(a)villagetelco.org&quot; &lt;steve(a)villagetelco.org&gt; >>> &lt;steve(a)villagetelco.org&gt; &lt;steve(a)villagetelco.org&gt;rg>, Dynamic Coalition >>> on Community Connectivity &lt;dc3(a)listas.altermundi.net&gt; >>> &lt;dc3(a)listas.altermundi.net&gt; >>> *Subject: *Re: [DC3] A Community-Run ISP Is the Highest Rated >>> Broadband Company in America >>> >>> >>> >>> I am not sure if the attached Qualcomm presentation by Yongbin Wei has >>> already been shared by any one in this mailing list. Found this, >>> interesting because it talks about horizontal and vertical spectrum >>> sharing, says MM bands are naturally more suitable for sharing, talks about >>> Spatial Division Multiplexing, others on this list might understand all >>> this better... >>> >>> >>> >>> Aspire and take CNs to 5G ???? >>> >>> >>> >>> On Mon, Aug 27, 2018 at 11:57 PM Sivasubramanian M < >>> 6.Internet(a)gmail.com&gt; wrote: >>> >>> Dear Steve, >>> >>> >>> >>> Thank you for your kind words. >>> >>> >>> >>> I took time to respond to your message as I was spending time on >>> http://villagetelco.org . Watched the video on the front page, and it >>> was fascinating to learn that mesh potato also doubles as some form of a >>> local telecom intranet. >>> >>> >>> >>> Long before messenger, skype and whatsapp, sometime during 2000, a >>> friend from Montreal told me that his telecom provider bundled wireless in >>> his cellular telephone, and provided the ability for the subscribers to use >>> the phone as a phone, and also as a wireless handset, with options (don't >>> recollect if it was for the extended cellular range or for the limited >>> wireless range, and not sure if the wireless range was different from the >>> cellular range) for each subscriber to choose five other subscribers as an >>> inner network, flip a button and talk to any of them, and the four others >>> in turn had the ability to choose their own circle of five friends. >>> >>> >>> >>> Mesh potato enhances the value of the Community Networks by enabling >>> local communications. When the Internet of Community Networks is bundled >>> with LAN features, and even more, with close circle network features, the >>> value of the Community Network surpasses that of commercial networks. Just >>> a thought. ( I am also reading your hardware specifications with interest.) >>> >>> >>> >>> Sivasubramanian M >>> >>> >>> >>> On Mon, Aug 27, 2018 at 10:50 PM Steve Song &lt;steve(a)villagetelco.org&gt; >>> wrote: >>> >>> Dear Siva, >>> >>> >>> >>> Thank you for taking the time to review and make such thoughtful >>> commentary on the draft paper. We will certainly take your feedback into >>> consideration in producing the final version. >>> >>> >>> >>> Regards... Steve (and Carlos) >>> >>> >>> >>> P.S. Love the parking lot analogy! >>> >>> >>> >>> On Mon, 27 Aug 2018 at 12:34, Sivasubramanian M &lt;6.Internet(a)gmail.com&gt; >>> wrote: >>> >>> Dear Jane, >>> >>> >>> >>> >>> >>> Please receive the attached comments on the ISOC Spectrum Paper >>> together with some unverified thoughts. >>> >>> >>> >>> Sivasubramanian M >>> >>> >>> >>> On Sat, Aug 25, 2018 at 2:13 AM Sivasubramanian M &lt;6.Internet(a)gmail.com&gt; >>> wrote: >>> >>> >>> >>> On Mon, Aug 20, 2018, 7:11 AM Jane Coffin &lt;coffin(a)isoc.org&gt; wrote: >>> >>> Hi Siva – >>> >>> >>> >>> Brian Hall from NYC Mesh is included in this thread. He can help >>> explain NYCMesh. >>> >>> >>> >>> I fully agree that different connectivity models need to be considered >>> and would add that different policy/regulatory models also need to be >>> considered. It is something we are keenly in favor of 😉 >>> >>> >>> >>> Links to some recent papers on Community Networks also are below and >>> attached is a Spectrum Paper we are looking for feed-back on by 24 August. >>> >>> midnight DC time? >>> >>> >>> >>> Thanks. >>> >>> >>> >>> >>> >>> Steve Song, Carlos Rey-Moreno, Mike Jensen are the primary authors with >>> direction/collaboration with our team at the Internet Society. Please send >>> me an email if you do have comments for us to consider. >>> >>> >>> >>> Other Resources: >>> >>> Case Study/Article re CN in Georgia: >>> https://www.nytimes.com/interactive/2018/01/05/technology/caucuses-mountain… >>> >>> African CN Paper – Partnership with Carlos Rey-Moreno: >>> https://www.internetsociety.org/wp-content/uploads/2017/08/CommunityNetwork… >>> >>> India – CN Paper – Partnership with DEF: >>> https://www.internetsociety.org/wp-content/uploads/2017/10/W4C-Policy-Paper… >>> >>> Licensing Brief: >>> https://www.internetsociety.org/wp-content/uploads/2018/05/Unleashing-Commu… >>> >>> Spectrum Approaches for CNs: >>> https://www.internetsociety.org/wp-content/uploads/2017/10/Spectrum-Approac… >>> >>> >>> >>> The inputs to the IGF from this Coalition are really great and Luca has >>> the links. >>> >>> >>> >>> Best, >>> >>> Jane >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> Internet Society | www.internetsociety.org >>> >>> Skype: janercoffin >>> >>> Mobile/WhatsApp: +1.202.247.8429 >>> >>> >>> >>> *From: *&lt;dc3-bounces(a)listas.altermundi.net&gt; on behalf of >>> Sivasubramanian M &lt;6.Internet(a)gmail.com&gt; >>> *Reply-To: *Dynamic Coalition on Community Connectivity < >>> dc3(a)listas.altermundi.net&gt; >>> *Date: *Thursday, August 16, 2018 at 1:47 PM >>> *To: *Dynamic Coalition on Community Connectivity < >>> dc3(a)listas.altermundi.net&gt;gt;, osama manzar &lt;osamam(a)gmail.com&gt; >>> *Subject: *Re: [DC3] A Community-Run ISP Is the Highest Rated >>> Broadband Company in America >>> >>> >>> >>> Dear Jane, >>> >>> >>> >>> Thank you for the informative messages and links. DEF has done >>> considerable work in the area of Community Networks, and several networks >>> such as the one that Sarbani has written about exist in India. However >>> there are issues related to the scale and scope of operation, which revolve >>> around the regulatory policies that make it a little difficult to create >>> and operate networks. >>> >>> >>> >>> If I understood well, the NY Mesh network has a supernode from the IXP >>> and its bandwidth comes from the IXP. I am copying this thread to Osama >>> Manzar of DEF to ask if there are examples of Community Networks in India >>> that do not depend on agreements with Telecom Operators / telecom related >>> ISPs, which may not wholeheartedly support Community Networks beyond the >>> notion of small rural networks. >>> >>> >>> >>> There are a few problems that need to be addressed, and a different >>> connectivity model may be required together with ample support from the >>> Telecom ministry for the spread of community networks in India. >>> >>> >>> >>> Thank you. >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> On Thu, Aug 16, 2018 at 11:01 PM Jane Coffin &lt;coffin(a)isoc.org&gt; wrote: >>> >>>