Dan Geer's Penetrating Remarks
Hon. Marc R. Pacheco
State House, Room 312-B
Boston, Mass. 02133
re: OpenDocument Standards
Dear Sen. Pacheco,
My name is Dan Geer. I am one of the half dozen ranking world experts in matters of computer security. By virtue of a long career both in academia (MIT and Harvard) and the private sector (six times an entrepreneur), there is absolutely no one in the State House who is not using software that I had a hand in producing, including yourself. I am a trusted advisor to the Federal Trade Commission, the Departments of Justice and Treasury, the National Academy of Sciences, the National Science Foundation, the US Secret Service, and the Department of Homeland Security. I am a Board member for a number of promising startups and their funding sources, have forty-two refereed publications, books and book chapters, four patents, over two hundred fifty invited presentations twenty percent of which were keynotes, and have been five times before the US Congress -- twice as lead witness. I have taught ten thousand students in the aggregate.
As an Officer of the Commonwealth, you understand the monopoly power of Microsoft quite well as the Commonwealth was the last man standing in the most recent round of antitrust litigation. What perhaps you did not grasp is the degree to which a computing monoculture is a security risk of the highest sort. It is, and I and others in the security research community are on record in unassailable ways that a computing monoculture is a hazard, but that it is an avoidable hazard if you want it to be. Microsoft maintains its power through user-level lock-in, as the Commonwealth noted and which it so adequately opposed. So long as that lock-in persists, there will be no solution to the monoculture risk. That lock-in is centered on and wholly confabulated with the use of proprietary formats for all documents produced by the Office Suite. Therefore, as a matter of logic and logic alone, if you care about the security of the Commonwealth then you must care about the risk of a computing monoculture. If you care about the risk of a computing monoculture, then you must care about barriers to computing diversification. If you care about barriers to computing diversification, then you must care about user-level lock-in. If you care about user-level lock-in, then you must apply yourself to the task of breaking the proprietary format stranglehold on the Commonwealth.
Fortunately, that has already begun. The Enterprise Technical Reference Model and its call for Open Document standards is precisely what is needed and it is not a moment too soon. As a ranking security professional with a doctorate in statistics, I can provide any amount of technical, quantitative proof that Open Documents are the point of maximum leverage and that the risk of remaining as we are exceeds any non-specialist's understanding including, with respect, yours. Warning times before attacks take place have fallen to zero. There is a new Windows virus every four hours. Perhaps 15% of all desktop Windows computers are running malware of some sort and I'll bet you $100 that includes your office. There is a direct and demonstrable correlation between increasing complexity of the Windows system and the effectiveness of attacks. Jurisdictional boundaries are meaningless if not undetectable in an always-on, fully-networked world. And as you almost surely know, your opponents are no longer misanthropic isolates but are instead professionals. So long as the Commonwealth voluntarily allows itself to be locked-in by the proprietary document formats of a proven monopoly, the Commonwealth cannot diversify and therefore the Commonwealth cannot mitigate its risk in any but the most marginal and palliative ways.
I am ready to vigorously debate these points with any and all comers both privately and in any venue. This is, in other words, a matter on which I actually do stake my professional reputation, my fortune, and my sacred honor. How may I be of assistance?
Very truly yours,
Daniel E. Geer, Jr., Sc.D.
P.S. I have blind relatives and if genetics is any guide may have that in my future. My comments still stand.