Jason Thorpe writes:
] Even if it is in "serious flux" (which I don't think it is), there has 
] traditionally not been a problem with checking in functional 
] works-in-progress into NetBSD-current.

Traditionally, on a branch, until such time as its deemed acceptable
to pull up.

Itojun never answered my question as to what branch tag he was
intending to commit it on.

 
] Honestly, I don't understand why people (well, one or two people) are 
] so rabidly against this going in.  Besides, the objections seem to be 
] more on the original source of the code rather than on the technical 
] merits.

It affections current functionality.  Its not like altq doesnt
exist in the kernel right now.  The origin of the code does matter.
Are you willing to be the developer liason in pushing architectural
changes back to them?  See previous discussion about the "overlap"
of developers.

I'd like to see an API.  I'd like some assurances we wont see
skunkworks[*] security fixes again.  I want to know that we wont
have problems that we've had in the past, and that we wont get
several features locked together that we have to fully accept or
reject.

-=erik.




[*] For people not familiar, SkunkWorks was a research group that worked
for the US Government developing top secret technology.  Very secretive,
even when the product was released.  They built the SR-71 Blackbird spy
plane.