Samuel Smith

@pfeairheller I agree that we don't need to support all the verbs (at least at this time if ever) but there is something in me that wants to generalize the...

After some more thought, there is a third alternative. When one looks at how web frameworks work, the heavy lifting of URL composition encoding and decoding is done by the...

@seth I agree that the KERI core protocol does not need to be tied too closely to a Restful interface and may suffer for it. But I find the exercise...

@pfeairheller A suggestion to enable the more compactness is for the single element form to just have the "i" field . ``` { "v" : "KERI10JSON00011c_", "t" : "req", "r"...

# Authentication Support An issue not considered above but that may be timely for this proposal is access restrictions or access control for queries. Unlike the event messages which are...

@sethjback There are two logs. One is the log where events are listed in order by sequence number (sn). This log can be forked when there is a recovery rotation...

In python the first seen and sn tables are filled when an event is first seen. The Kever object does this with its .logEvent method. The first parameter indicates to...

@sethjback Well not exactly. Only interaction events may be recovered from (at least for non-delegated identifiers). So when a signing key is compromised the controller may perform a rotation to...

The problem is that SN are not unique on a log with forks. So the logic of replaying from a sequence number has to additionally specify is is th first...

Also with forks all the events with the same SN will not be contiguous. So how do you query a non-contiguous set of SN? You certainly can’t use them in...