Re: END_SEGMENT and END_STREAM redundant

This is similar to my earlier question about duplicate segment codings. The spec needs to specify which is significant (observable by the application): the flag bits, or the segmentation.

On 2014–04–20, at 6:23 AM, Roberto Peon <grmocg@gmail.com> wrote:

> Arguably these are not redundant, since the message may not have been completed when the stream was terminated.
> -=R
> 
> On Apr 19, 2014 8:08 AM, "Adrian Cole" <adrian.f.cole@gmail.com> wrote:
> Hi, all.
> 
> The combination of END_SEGMENT and END_STREAM seem redundant to me.
> It seems like you wouldn't need to support both flags at the same
> time, as ending a stream surely ends the segment.  Is it worthwhile
> calling this out?
> 
> -A
>