Re: TLS API discussion

On Mon, 19 May 1997, Paul E. Hoffman wrote:
> I agree that it should be talked about, but the question is where it would
> get the most attention from the best people. My suggestion is to float it
> by the TLS Working Group, asking them if they want it talked about there as
> a work item. If not, you're welcome to bring it back here, although I hope
> you can drag a few more TLS experts with you on the round trip. If the
> applications developers on this list can help with this, all the better.

I think it would be a mistake to design it in the TLS group.  I suspect
the simplicity requirements for application programmers would get ignored
and junk like ASN.1 and OIDs might creep in too far and make it unusable.

Actually I think it's a mistake to design an API by committee.  We should
identify one visionary who's good at APIs and give him a few basic
requirements (one of which would be to hide ASN.1 and OIDs from the
caller as much as possible). I've never seen a good API from a committee
(and rarely seen a good API in general).

So how's the SSLeay API?  If it's good we could just document that and
standardize it.

---
Chris Newman <[email protected]>

A paragraph from ASN.1 standard: "The resulting type and value of an
instance of use of the new value notation is determined by the value
(and the type of the value) finally assigned to the distinguished
local value reference identified by the keyword VALUE, according to
the processing of the macrodefinition for the new type notation
followed by that for the new value notation."