comments on StatusNotifier spec

[Matthias Clasen]

I must say I am a bit taken aback by your responses here. 

Dan's mail made me look at the spec for the first time, and I have to
say he is entirely right in his criticism. The spec is full of awkward
naming (StatusNotifierHost, ServiceRegistered, etc), internally
inconsistent (e.g there's a ProtocolVersion property, but the document
has no version information that I can find), ignores decades old lessons
from X (when you pass on events, _always_ include the timestamp), and it
is entirely void of expected behaviors for either side of this
protocol. 

But instead of addressing these criticisms, you seem to be only
interested in selling your abstraction layer philosophy.

Not a good sign for reaching a consensus.
 

Matthias