[v3] system_error, round one

[Beman Dawes]

Benjamin Kosnik wrote:

 > ...
> I am wondering if it may make more sense to just implement a revised
> version of this spec, since it seems pretty weak at the moment, and
> then shoot back those implementation differences in paper form to the
> LWG pre-Toronto.

Or just send me issues for a unified issues paper I'll be writing for 
Toronto. I've got a half-dozen or so issues already that others have 
already sent me.

> For the traits stuff, Howard nicely added rationale bits in gray
> under the impacted parts of the proposal. I see no similar thing for
> the system_error proposal, even though the differences between the
> revisions is quite substantial. Is there any place where the
> rationale for the evolution of this code was tracked?

No, I didn't as good a job as Howard at tracking the rationale for 
changes. The general rationale for most of the changes was to deal with 
issues that arose when Chris Kohlhoff converted the Boost networking 
stuff to use the system error proposal. The detailed rationale could be 
recovered from the Boost developers mailing list or by asking me.

--Beman Dawes