Comment on gTLD auction proceeds
Dear all, I agree with those that have commented before me that it may be wise to treat these proceeds as a one time opportunity to strengthen the operational stability, reliability, security, and global interoperability of the Internet. There is technical debt in many of these areas, and it is in the interest of the entire global internet community that there is a budget to identify and efficiently address issues. ICANN's Articles of Incorporation mention that it in case of dissolution it would seek to distribute its assets (if possible), to a charitable organization "organized and operated exclusively to lessen the burdens of government and promote the global public interest in the operational stability of the Internet". For a surplus ICANN does not need, a similar destination seems logical. Please allow me to share the experiences of NLnet foundation, a charitable trust fund entirely dedicated to the open internet since the mid nineties derived from a similar unique opportunity at the time. [ I will not go into a lot of detail on NLnet history, for those who are interested in how Dutch internet pioneers selflessly made our endowment possible see below . ] Our original endowment was about half of the sum that would be available now, and I believe it has served the internet well and made a difference disproportionate to the amount of money. The design choice of the NLnet board was to keep the organisation very small and not create big projects ourselves which would make oversight very complex, but to focus on (micro)grants through a global open call designed to be lightweight and with very small administrative burden to the submitters. NLnet operationalized its mission by financially supporting those with great ideas to improve the internet that do not have large organisational backup, and that are willing to share them with the world in a way everyone can contribute to their work). Not large abstract projects with huge consortia but knowledgable SME's, not-for-profits, open source projects and individuals with concrete ideas convincingly showing the urgency for their work. As with the original ARPA work, NLnet chose to put a high trust in its projects, and it pays off. Technical work typically fits into the following categories (between brackets I give some examples of projects NLnet funded in these categories, but I would think that the organisation that ICANN would support/create would see similar input to an open call) : * research and development into important internet-related standards and their maturation (NLnet examples: DNSSEC, RPKI, TLS, DKIM, DMARC, SIP, XMPP, HIP, WebRTC, Real Time Text, PPSPP, SCTP) * funding new tools and/or improvements in the core of important open source platforms where lack of adoption of new standards is harming the collective interest (NLnet examples: the IPv6 implementation in FreeBSD, Linux Kernel netfilter, OpenDKIM, OpenDMARC, NAT64, NSD, OpenMSRP, BIND, DNSCCM, etc) * Proof of concept/reference implementations created during the research into a standard (NLnet examples: the Comprehensive Queue Management Made Easy (CAKE) in CeroWRT, OpenFlow NBI, GNUnet, * Advanced security * Quality assurance activities such as interoperability testing events, EFF's SSL observatory. * The occasional travel support for senior standards contributors that are temporarily without means to attend e.g. IETF, W3C TAG meetings or interop events. * In addition, there are many small but very useful projects that result from an open call that one could never predict or fit into a meaningful category beforehand - yet clearly benefit the whole internet ecosystem. Examples from our experience include rebootless kernel updates for Linux (which allow live kernel patching of systems running on the internet, which reduces vulnerability to botnet harvesting), tools for systematic firmware analysis (allowing discovery of unknown shared bugs in firmware of networked devices) or the internationalisation of the GPLv3 open source license (removing a lot of legal uncertainty for users outside of its country of origin), anti-DDoS tooling, intrusion prevention and security testing. These were all small projects but with very interesting output. Put in another way: there are many things that the internet needs but does not know it needs, something which the funds now available to ICANN could facilitate. These are some of the lessons we learned from our work: * there are very few organisations that pay for cleaning up 'garbage', as in proving that technologies are no longer secure and need to be phased out. This is vital in the technology lifecycle. * there is an asymmetry in the cost and benefits of individual investments into the security and stability of the internet. If other organisations benefit just as much, many organisations - even those with enough means - will just wait for others to solve something * if you want to progress technology at an internet scale, this at some point involves open source software that allows permissive-less deployment and incremental innovation. * there is a real world gap between the way academics are funded (based on citations, publications and student numbers) and the work they do on internet standardisation - and that needs to be solved. * deployment of new technologies on the internet is really hard, especially when born in a small organisation. Small actors need some help with it. There is a lot of value to society we get for free. The internet and the world wide web are part of what makes our modern world run, yet none of us can't buy or own them. They were handed down to us, generously, by the people who created them. These pioneers had no business model in mind when they created the technologies we now use, and that made just about everything else that happened possible. With the money ICANN has in its hands, it can do great things. I wish the board of ICANN a lot of wisdom in its decision what to do, and hope it will benefit the internet in the way NLnet's board decided for its endowment to benefit the global internet society. Best, Michiel Leenaars Dorector or Strategy NLnet foundation Science Park 400 (Matrix 2) 1098 XH Amsterdam The Netherlands https://nlnet.nl sip/xmpp: michiel [@t] nlnet.nl  The history of NLnet goes back to in the early eighties, to Dutch internet pioneers like Teus Hagen, Ted Lindgreen, Jaap Akkerhuis, Frances Brazier and Wytze van Raay who played an active role in setting up the European pre-internet computer networks with the national research institute for mathematics and computer science CWI as its center point under the auspices of the national Unix user association. Together with many volunteers they bootstrapped a low cost grass roots network infrastructure with national coverage by all kinds of unusual and creative means - for instance leveraging the signaling infrastructure of the national rail network to accommodate modem banks - that consumed 25% of the traffic of the EUnet. Steep growth put an enormous pressure on the small organisation and in 1997 NLnet decided to sell off all operational activities. The proceeds of the sale were dedicated to the maintenance and health of the internet.