[Foresight-devel] Re: Foresight 1.x Plans

[Ken VanDine]

So where do you come up with 5% for functionalitly?  I would estimate on
a base install it is on par... missing a couple things and has a few
extra things.  The only real gap is the "extra" stuff, which should be
able to be ported very quickly.  The current stuff in 2-devel is stable
enough for use, the only problem we have seen has been update related as
we worked out some organization/arch stuff... which is about done.  That
stuff is expected in an alpha.  We wouldn't do that on the release
label.  Once we call it a beta, it will be on fl:2-qa and that will not
get as many changes... so the churn will be on fl:2-devel.

So once we put a fork in it an call it "ready" we can very rapidly port
the missing apps over, and if we put a few people on it we could
probably get ~80% of them over in just a few days... there will surely
be a few stragglers that give us grief building with the new toolkit or
or x86_64. 

I also wouldn't want to redirect users over to fl:2 as soon as we call
it 2.0, we can give it a few weeks... but I wouldn't want to drag that
out for long, and certainly wouldn't want to put effort into maintaining
it.  We should instead focus on the update path from fl:1 to fl:2.

Moving to fl:2 will be a good thing for everyone, it is already more
stable (from a usability standpoint).   Many things that have needed
attention for a long time are now resolved, since we had a new place to
start.  Plus, getting into the rhythm of using the new release process
will drive our quality way up.  

And the process of getting things ported from fl:1* to 2-devel is pretty
well hammered out now and documented.  Most things only take a few
commands to get them fully ported, especially if they have been built
properly with rmake before.  

--Ken
 
On Thu, 2008-01-03 at 18:38 +0100, Thilo Pfennig wrote:
> Am Thu, 03 Jan 2008 09:52:56 -0500
> schrieb Ken VanDine <ken at vandine.org>:
> 
> > My hope would be to redirect the groups in fl:1 to fl:2 shortly after
> > 2.0 is released... which i hope is before GNOME 2.22 :)
> 
> Uhm, right now my estimate is that Foresight 2 has about 5 % of the
> functionality of FL1. The goal was 100% - right now FL2 cant really
> compete with FL1 if you look at the number of packages and also it
> rather looks to be in a flow rather than in a state where one could
> begin doing actual packaging besides the core stuff. The whole building
> process is yet untransparent so I would fear that instead of having
> 100% functionality we would get a massive degression if we would really
> release on AND force users to switch. I always feared that we might
> ending up supporting two branches for a longer time  but now I dont
> really see another way than doing just that.
> 
> So I am saying a release id Ok for me but automatically switching I
> would veto as long as there hasnt been happening a massive improvement
> in FL2. I also like to suggest that we could freeze 1.4.x in the 2.22
> state and then give it another 6 months and then let it die - I
> also would volunteer in maintaining some packages.
> 
> Maybe I am pessimistic or my vision is not that strong but i cant
> imagine a stable FL2 on March 2008. But I depend on a stable system.
> 
> 
> Thilo
> 
> 
>