Blockchain: What's Not to Like?
“Blockchain: What’s Not to Like?”
Speaker: Rosenthal, D.
Date: 30 July 2019
Speaker Session Preview
SMA hosted a speaker session presented by Dr. David Rosenthal (Retired) as a part of its SMA General Speaker Series. During his presentation, Dr. Rosenthal adopted a skeptical view of cryptocurrencies and blockchain technology. He stated that the goal of his presentation was to explain how the way in which the pieces of a system (e.g., one that conducts money transactions or stores files) fit together makes the problems that the technology encounters in practice difficult to fix. In other words, the problems are “inherent in the underlying requirements.” Dr. Rosenthal stated that one should not trust any single centralized entity, but rather a decentralized system. A decentralized system has checks and balances; the consensus of a large number of entities is required in order to allow a state transition to occur. He also stressed the importance of having a “permissioned” system, where a central authority authorizes entities’ involvement. Dr. Rosenthal highlighted the pitfalls of having a permissionless system (e.g., a permissionless system is vulnerable to Sybil attacks, in which one actor creates many seemingly independent voters who are actually all under his control). However, he stated that in the end, both permissionless and permissioned blockchains are fatally flawed. He spoke about proof-of-work in a permissionless system and presented the Brunnermeir and Abadi’s Blockchain Trilemma, which argues that blockchain must choose between two of the following three attributes: correctness, decentralization, and cost-efficiency. Dr. Rosenthal explained that Bitcoin in particular sacrifices cost-efficiency in favor of the other two attributes. Next, he stated that the security of a blockchain depends “not merely on the security of the protocol itself, but on the security of both the core software and the wallets and exchanges used to store and trade its cryptocurrency.” He spoke further about how to guarantee the security of cryptocurrency-based blockchains, Cryptokitties (a game that claimed that it could handle unlimited decentralized applications but ultimately collapsed), “smart contracts,” and “pump-and-dump” cryptocurrency schemes. To conclude, Dr. Rosenthal explained why, despite both permissioned and permissionless systems’ flaws, these systems are perceived as huge successes and highlighted the software supply chain security implications associated with the misuse of Certificate Authorities (CAs).
To access Dr. Rosenthal’s slides and some additional notes on this presentation, please visit https://blog.dshr.org/2019/07/blockchain-briefing-for-dod.html
David S. H. Rosenthal Bio
In 1998 at the Stanford Libraries David Rosenthal and Vicky Reich started the LOCKSS Program with funding from NSF. It is aimed at long-term preservation of web published materials (e-journals, books, blogs, websites, archival materials, etc). From 1999-2002 he worked on it at Sun Labs. From 2002 until he retired in 2017 he worked on it at the Stanford Libraries.
David built and tested the initial prototype, developed the OpenBSD-based network appliance technology that LOCKSS peers used for the first 5 years of production, and was part of the research team that developed the award-winning fault- and attack-resistant peer-to-peer network technology that underlies the LOCKSS network. This was a decentralized consensus system using proof-of-work published more than five years before Satoshi Nakamoto published the Bitcoin protocol, for a different application.
David started blogging in 2007 at https://blog.dshr.org/. As well as the LOCKSS technology, he has written on economic models for long-term storage, emulation as a preservation strategy, DNA as a storage medium, the decentralized Web, the economics of peer-to-peer systems, and many other topics. The blog features the complete text of his many presentations, with links to the sources, and expanded versions of some of his recent technical publications.
David joined Sun Microsystems in 1985 from the Andrew project at Carnegie-Mellon University, where he had worked on window systems with James Gosling. He worked on window systems with James at Sun, and was part of the teams which developed both NeWS and the X Window System, now the open-source standard. He also worked on graphics hardware, the operating system kernel, and on system and network administration.
David left Sun in 1993 to be Chief Scientist and employee #4 at Nvidia, now the leading supplier of high-performance graphics chips for the PC industry, where he worked on I/O architecture. In 1996 he joined Vitria Technology, now a leading supplier of e-business infrastructure technology. There, he worked on reliable multicast protocols and on testing industrial-strength software.
David was educated at the Haberdashers' Aske's School, Elstree at a time when it was a "direct grant school". He received an MA degree from Trinity College, Cambridge and a Ph.D. from Imperial College, London. From 1976 to 1983 he was a post-doc at EdCAAD, Edinburgh University's computer-aided architectural design group run by Aart Bijl, including spending 1982 on sabbatical teaching at the Universiteit van Amsterdam and researching at what was then the Mathematisch Centrum, and is now the Centrum Wiskunde & Informatica. He is the author of many technical publications and holds 23 patents. His interests include backpacking and the theater.
Site-wide Search
Search all site content, including all NSI and SMA publications, SMA Speaker series, NSI Team member bios, services, portfolio projects, company info, and more.
