Does UL "go after" companies that produce unsafe devices. My guess would be no. As far as UL is concerned, companies voluntarily bring their products to them for certification. It is the consumers and legal authorities that give UL such a "big stick". And with this model, UL seems to be fairly free of legal hassles since (from their perspective) seeking certification is voluntary.
It seems to me a similar model could be followed by IETF or anyone with the business sense to start a company (e.g. the InterOperability Lab at UNH for Ethernet). Actually, the task should be very simple. The IETF is supposed to require two independent interoperable implementation before something can become a standard (and ideally even before becoming a draft standard). The "Internet Interoperability Lab" could just test voluntary submissions against the two versions used when approving the standard. Is the practice of requiring two independent implementations still being followed? Or is the problem that a generic RFC is being treated as an "approved standard" before it has actually become so (see RFC 2600). Tony
