- From: Ted Hardie <hardie@merlot.arc.nasa.gov>
- Date: Wed, 21 Feb 1996 11:59:09 -0800 (PST)
- To: Ned Freed <NED@innosoft.com>
- Cc: pjc@trusted.com, paulle@microsoft.com, fielding@avron.ICS.UCI.EDU, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
> > > Here are some requirements that I am asked to provide today... > > That is I need auth that can be demonstrated in a court of law as > > being "spoof proof" > > > Can we do it? > > Of course not. I am frankly appalled that you would even ask this question. > Actually, it may not be as bad as you think. Long, long ago, when I was a tiny little programmer, I worked on a billing system for a law firm. Security, oddly enough, was a major part of the system. Because the law sometimes allows one party to recover its legal costs from the party it's suing, the courts have laid down a number of principles about billing records, to keep the charges from being inflated when such recovery takes place (after all, the party hiring the lawyer probably wants to stick it to the other party, and might well collude with the lawyer to do so). The original methods the courts set down for all of this had to do with paper records and appropriate data archiving of those records; we had to adapt those to a billing system that would be run electronically. The principles were actually fairly easy to enforce, and would be even easier today. They asked for auditablity, proof of oversight, and off-site archiving of all billing records. (I understand that now off-site archiving is waived if the records are stored on write-once media.) It required a little translation to fit the principles into the systems we had available at the time, but it was very doable. In other words, the legal requirements for this type of transaction were established, were not onerous, and could be translated into the electronic realm. To return the case being posed, the question as originally asked can't really be answered, because you don't say what legal standard is going to be applied. If you know what legal standard is being applied (like financial non-repudiability, or auditability, or whatever), figure out what the requirements would be on any system, electronic, paper, or in-person, then re-ask the question. We may be able to do a better job of pointing out the points of risk. If the legal standard to be applied is not established, you have a much bigger problem; you then need to figure out what the strictest standard would be (or the union of all standards, which ever is worse), and then meet that--or you risk helping set the standard. Regards, Ted Hardie Disclaimer: I am not a lawyer, but I was raised by wolves.
Received on Wednesday, 21 February 1996 11:55:52 UTC