3 ms·

Very very good point. My research group and ISP federation tries to work together with all of these 100+ initiatives, but sadly, a large fraction suffer from th

by morphle 13y ago

Very very good point. My research group and ISP federation tries to work together with all of these 100+ initiatives, but sadly, a large fraction suffer from the "not invented here" syndrome or even worse, the "never read any scientific papers that solved this many years ago" syndrome. (See my profile).

femto 13y ago
I read your profile, and looked though the websites that it mentioned, but I'm not really any clearer on the connection to mesh networking. Do you mind giving a summary? I genuinely am interested. I'm interested in mesh networking and its relation to the physical layer. As a (time deficient) hobby, I've been noodling on a bit of free space optics for a number of years. In my day job, I design physical layer radio infrastructure for data/voice networks. (I just realised that "noodling" probably doesn't make sense outside Au/NZ. It comes from the practice of picking though mine tailings for opals, meaning: to toil in an ad hoc manner over a long period of time.)
- morphle 13y ago
  I will try to be succinct here, but that is hard to do on the spur of this moment. I must refer to my scientific papers (see my HN profile for email address) where I had a few months to describe it right. Indeed, the website is being worked on while I write this, we are just coming out of stealth mode next week. Our hardware switches are to build true mesh networks at hardware encrypted linespeeds up to 800 Gbps (Q1 2014). Currently, most ISP and end-user routers and switches are set up behind NAT and with default routes in an hierarchical star topology, all pointed (slaved) to the central ISP. Only that ISP has true internet that they buy or peer with other ISP's and the top tier 1 global internet backbones. None of these users, worldwide 99.9% or more, can be considered to be in a mesh network. This is why the [No Such Agency] and governments only need to copy or deep packet inspect at a few points to see most worldwide internet traffic. Most of that traffic is not encrypted at one end, like the ISP or website side you are using. Peer to peer traffic is not yet using strong encryption or is not anonymous. Except Tor tunnels are, but that is slow and is under attack from the NSA and many national firewalls. Open Libernet only proposes to incentivize such a tor-like tunnel system, but do not claim Tor strength anonymity or security. BTW, I exchanged email with the Open Libernet authors, they are computer scientist with master degrees from Beirut. Our system we designed over the last 20 years is based on a home switch with 8 ports at 1 to 3 Gbps, either gigabit copper, wifi or optical (or any connection including your radio and free space optics) would connect you to neighbors to the left, right, behind you, before you and wireless, at $50 mass produced ($200 for 8 ports at 10G). More importantly, it is free and open source hardware based on our own free and open source manycore microprocessors with free and open source software based on the new open openflow standards, so hackers, scientist and anyone can trust the encryption at linespeeds in these mesh switches and routers by constantly scrutinizing . The protocol we are making is similar to TOR, but always at linespeed. If you're house injects traffic, it is replacing less critical encrypted bittorrent or dummy traffic, in such a way that no change in bandwidth or content can be detected, providing you with even better anonymity than Tor does. It brings anonymous bittorrent and voice/email/skype-like traffic. The 60 manycore processor must be always on so your home router will pass on traffic from you neighbors in trade for them passing on your traffic. This allows the switch/router to also function as (web)server. One of the services will be an anonymous distributed cloud service, based on free cores and any terabyte harddisks you share by connecting to the switch. Most importantly, although you can run Unix/linux on the some cores, the switch and openflow protocols do not. This allows hardware protection of the e-capabilities that guard the free and open source encryption algorithms, so it is [No Such Agency] strength security. Linux, Intel and Arm processors has to many unknown security flaws and known backdoors and are not truly open hardware and software. We should be much more secure than Tor or Freedombox this way, although we do run these free software systems as well. So now you have a secure anonymous free and open source mesh network that is running at hardware speeds, preferably over optical fibers, within your neighborhoods and not in the hand of any company or government you can not trust. You can at any time check the software and hardware if it still can be trusted. Your neighbors are by default incentivized (forced by social peer pressure) to also get a mesh router and more importantly, to keep it always on. By adding and sharing harddisks, they gain a shared petabyte anonymous cloud storage. By having a core surplus in the encrypting switch, the are further paying for the shared cloud services running distributed on their home switches/server. By using our built-in manycore parallelism, they are incentivized further to not rely on Linux encryption, but our microcode/bytecode based encryption that has no backdoors and can be constantly be checked for it. So mesh every aspect, not just the tunnels. Incentivize everyone, so they must peer with each other to gain high speed at zero or lowest cost. Keep it out of the control of a single provider, so it will remain free and open, right up to the continental crossing points and beyond. It must be crowd funded, by the richest most powerful group on the planet, we the people who should hold these anonymity, privacy and mesh peering/sharing truths to be self evident. I hope I made it clear you should mesh at every level and provide all that without end-users installing a single thing. They should not want to turn of the devices, ever. So make it very low power. Next these neighborhoods need to have a mesh network with each other, city to city. So they should share the cost of private fibers between cities, also running the same strength encryption as the home routers but now at terabits speeds per fiber. They need to hook up through submarine cables with other continents, again with the same free and open hardware switches running at multiple terabits. That is why we design and built the entire system from scratch with the latest and best hardware, software and encryption systems science can come up with. At the internet exchanges, this local or global mesh network also peers with the current internet, where you will loose some or all encryption if you do not terminate with peers using the same mesh protocols or travel thru big companies or government networks. This is why we not just built mesh networks locally and have them financed and owned by the users themselves, but also provide community owned inter-city and inter-continental mesh fiber nets that do not fall under government laws or investor control. We welcome developers, scientist, Phd students and HN type co-founders to work with or for us to build out this system to regain anonymity and privacy while providing gigabit+ speeds. We welcome ISP's and communities to use our open hardware and backbone at cost. Edited for errors.
  - femto 13y ago
    Thanks for the reply. Is there a repository available with the design/source files for the mentioned open hardware, free software and papers?
    - morphle 13y ago
      Not yet, we are, right today, putting up price lists and descriptions. Expect all you want and need to appear through http://morphle.org http://morphle.org and http://morphle.com http://morphle.com before june 2014. Please come help us put up the papers/design/source code you ask for, we are swamped already. We are applying for YC, write if you want to join us.
    - morphle 13y ago
      We are interested to use your "noodling" experiences in free space optics and packet radio as an additional interface type in our switch hardware. We have a low cost design for 1 Gbps free space optics below 1 km distances, similar to http://en.wikipedia.org/wiki/RONJA http://en.wikipedia.org/wiki/RONJA. Can you contact us? The repositories and the design/source files have so far only been shared with local neighbornets that start with the minimal required 300 user group (for financing the required backbone) who buy our routers at today's non-mass produce prices.

by morphle 13y ago

femto 13y ago

I read your profile, and looked though the websites that it mentioned, but I'm not really any clearer on the connection to mesh networking. Do you mind giving a summary? I genuinely am interested. I'm interested in mesh networking and its relation to the physical layer. As a (time deficient) hobby, I've been noodling on a bit of free space optics for a number of years. In my day job, I design physical layer radio infrastructure for data/voice networks. (I just realised that "noodling" probably doesn't make sense outside Au/NZ. It comes from the practice of picking though mine tailings for opals, meaning: to toil in an ad hoc manner over a long period of time.)

morphle 13y ago
I will try to be succinct here, but that is hard to do on the spur of this moment. I must refer to my scientific papers (see my HN profile for email address) where I had a few months to describe it right. Indeed, the website is being worked on while I write this, we are just coming out of stealth mode next week. Our hardware switches are to build true mesh networks at hardware encrypted linespeeds up to 800 Gbps (Q1 2014). Currently, most ISP and end-user routers and switches are set up behind NAT and with default routes in an hierarchical star topology, all pointed (slaved) to the central ISP. Only that ISP has true internet that they buy or peer with other ISP's and the top tier 1 global internet backbones. None of these users, worldwide 99.9% or more, can be considered to be in a mesh network. This is why the [No Such Agency] and governments only need to copy or deep packet inspect at a few points to see most worldwide internet traffic. Most of that traffic is not encrypted at one end, like the ISP or website side you are using. Peer to peer traffic is not yet using strong encryption or is not anonymous. Except Tor tunnels are, but that is slow and is under attack from the NSA and many national firewalls. Open Libernet only proposes to incentivize such a tor-like tunnel system, but do not claim Tor strength anonymity or security. BTW, I exchanged email with the Open Libernet authors, they are computer scientist with master degrees from Beirut. Our system we designed over the last 20 years is based on a home switch with 8 ports at 1 to 3 Gbps, either gigabit copper, wifi or optical (or any connection including your radio and free space optics) would connect you to neighbors to the left, right, behind you, before you and wireless, at $50 mass produced ($200 for 8 ports at 10G). More importantly, it is free and open source hardware based on our own free and open source manycore microprocessors with free and open source software based on the new open openflow standards, so hackers, scientist and anyone can trust the encryption at linespeeds in these mesh switches and routers by constantly scrutinizing . The protocol we are making is similar to TOR, but always at linespeed. If you're house injects traffic, it is replacing less critical encrypted bittorrent or dummy traffic, in such a way that no change in bandwidth or content can be detected, providing you with even better anonymity than Tor does. It brings anonymous bittorrent and voice/email/skype-like traffic. The 60 manycore processor must be always on so your home router will pass on traffic from you neighbors in trade for them passing on your traffic. This allows the switch/router to also function as (web)server. One of the services will be an anonymous distributed cloud service, based on free cores and any terabyte harddisks you share by connecting to the switch. Most importantly, although you can run Unix/linux on the some cores, the switch and openflow protocols do not. This allows hardware protection of the e-capabilities that guard the free and open source encryption algorithms, so it is [No Such Agency] strength security. Linux, Intel and Arm processors has to many unknown security flaws and known backdoors and are not truly open hardware and software. We should be much more secure than Tor or Freedombox this way, although we do run these free software systems as well. So now you have a secure anonymous free and open source mesh network that is running at hardware speeds, preferably over optical fibers, within your neighborhoods and not in the hand of any company or government you can not trust. You can at any time check the software and hardware if it still can be trusted. Your neighbors are by default incentivized (forced by social peer pressure) to also get a mesh router and more importantly, to keep it always on. By adding and sharing harddisks, they gain a shared petabyte anonymous cloud storage. By having a core surplus in the encrypting switch, the are further paying for the shared cloud services running distributed on their home switches/server. By using our built-in manycore parallelism, they are incentivized further to not rely on Linux encryption, but our microcode/bytecode based encryption that has no backdoors and can be constantly be checked for it. So mesh every aspect, not just the tunnels. Incentivize everyone, so they must peer with each other to gain high speed at zero or lowest cost. Keep it out of the control of a single provider, so it will remain free and open, right up to the continental crossing points and beyond. It must be crowd funded, by the richest most powerful group on the planet, we the people who should hold these anonymity, privacy and mesh peering/sharing truths to be self evident. I hope I made it clear you should mesh at every level and provide all that without end-users installing a single thing. They should not want to turn of the devices, ever. So make it very low power. Next these neighborhoods need to have a mesh network with each other, city to city. So they should share the cost of private fibers between cities, also running the same strength encryption as the home routers but now at terabits speeds per fiber. They need to hook up through submarine cables with other continents, again with the same free and open hardware switches running at multiple terabits. That is why we design and built the entire system from scratch with the latest and best hardware, software and encryption systems science can come up with. At the internet exchanges, this local or global mesh network also peers with the current internet, where you will loose some or all encryption if you do not terminate with peers using the same mesh protocols or travel thru big companies or government networks. This is why we not just built mesh networks locally and have them financed and owned by the users themselves, but also provide community owned inter-city and inter-continental mesh fiber nets that do not fall under government laws or investor control. We welcome developers, scientist, Phd students and HN type co-founders to work with or for us to build out this system to regain anonymity and privacy while providing gigabit+ speeds. We welcome ISP's and communities to use our open hardware and backbone at cost. Edited for errors.
- femto 13y ago
  Thanks for the reply. Is there a repository available with the design/source files for the mentioned open hardware, free software and papers?
  - morphle 13y ago
    Not yet, we are, right today, putting up price lists and descriptions. Expect all you want and need to appear through http://morphle.org http://morphle.org and http://morphle.com http://morphle.com before june 2014. Please come help us put up the papers/design/source code you ask for, we are swamped already. We are applying for YC, write if you want to join us.
  - morphle 13y ago
    We are interested to use your "noodling" experiences in free space optics and packet radio as an additional interface type in our switch hardware. We have a low cost design for 1 Gbps free space optics below 1 km distances, similar to http://en.wikipedia.org/wiki/RONJA http://en.wikipedia.org/wiki/RONJA. Can you contact us? The repositories and the design/source files have so far only been shared with local neighbornets that start with the minimal required 300 user group (for financing the required backbone) who buy our routers at today's non-mass produce prices.