opinions on porting software in alpha state?

Peter Thoenen eol1 at yahoo.com
Mon Mar 6 13:28:05 PST 2006

>> Users "expect" the ports to work -- that part I'm

Just quick comment on this before I move on with my reply.  Everything
in the port tree *works* (or should), even if all the features aren't
there and / or its buggy / alpha's software.  I just don't see anybody
submitting a port that compiles and does absolutely nothing other than
compile.  If they do, they should be shot ;) 

Now as for submitting alpha software, go for it.  Software is always a
moving target and almost never a complete set.  Sure it might be
labeled 1.0 stable but lets face it, 7 years down the road with Version
20.4 stable 1.0 will look like an alpha release.  If it works, does
what you want it to do, then go ahead and port it.  Somebody somewhere
may find it useful (an example of this is mixmaster which has been
alpha for nearly 10 years but its in use).

Now if you looking at porting alpha software when a stable version is
currently available, I am against that.  Either port the stable version
or create port-devel version like so many other ports (tor / tor-devel
for example).  If the alpha version isn't that big of an improvement
over the stable version, then just wait.


More information about the freebsd-ports mailing list