epiphany webkit in fedora

[Jeff Spaleta]

2008/4/29 Martin Sourada <martin.sourada at gmail.com>:
>  Yes, that's right... But I thought early rawhide use mostly people who
>  are able to handle broken stuff quite well, so I'd expect it would not
>  be that bad...


There's a distinction that needs to be made between needs testing and
too new to be useful.  The people who make that call are the package
maintainer and the upstream developers.  If the upstream developers
have a roadmap and know they want to get to a certain point before
encouraging wider testing..then it would be irresponsible to include
into rawhide.

We have to remember the point of rawhide is NOT to get new things
built just because they are new. The point is to get them into rawhide
in a timely manner so to maximize the synergy between our release
cycle and upstream development efforts. If upstream development knows
they still have a ways to go on fundamental codebase work, and aren't
ready for wider testing yet... then we shouldn't overwhelm the
process.  if upstream knows they are not prepared to make use of the
feedback exposure in rawhide would provide..then its wasted effort to
expose it in rawhide...even if its a high profile application going
through major re-development.

-jef