[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: specification language?




>Rich Salz writes:
>> I thought we killed off this discussion...
>
>Indeed. I would say that the encodings discussion, although
>interesting, is a bit premature. Could we hold off on it?

What I thought was being proposed was using Java to specify the
trust relationships. Since this had been raised by two rather 
eminent persons that morning I was taking it seriously in that
light. Ie certificates would have the form something like:-

workout_trust (chain : list[certificate]) {
        for (cert = chain.head ...
        }

and so on...

This would then be plugged into the general, all purpose, all singing,
ever so secure Java engine inside the browser.

I don't think this is a good idea.


It is clearly premature to decide on bit formats at this point. 
We should work out the structure of the problem and what
information we want to express first. If this is very simple then
RCC822 will probably becon. If it is complex then ASN.1 or 
s-expr would seem in order.


        Phill



References: