Re: Service Identity (Re: Machine Identity)

[Jeroen Massar <[e-mail]>]

Juergen Schoenwaelder wrote:
[..]
> Some 10+ years ago, SNMP decoupled the identity of an SNMP service
> from the transport used to access the service by introducing the
> notion of an engineID. It was architecturally the right thing to
> do. At that time, people hoped that management applications would at
> start using engineIDs since they are architecturally the right thing
> to deal with proxies, NATs and all sort of issues. Ten years later, we
> must simply observe that this did not happen. Perhaps HIP as a more
> general solution to this problem has more luck.

I don't think that programs are going to adopt this, simply because the 
programmers don't know about it and because there is no easy way to have 
it available on the platforms they want to program for. If there was a 
simple API for this which worked on all platforms and had a BSD license 
+ commercial licenses if needed then it would help quite a bit, still 
though changing people, hard thing.

I would love to, but unfortunately I don't see it happening.

Greets,
  Jeroen