[lsb-discuss] In the "new" world
rjschwei at suse.com
Mon Mar 31 21:26:03 UTC 2014
I know this is reaching a bit as we have not published the meeting
minutes from the f2f last week. I did post a high level summary to the
Things we didn't get around to discuss in more detail but that need some
thought is the documentation/specification of existing stuff.
For example I don't think we want to create a problem statement and a
solution etc. for every interface that is in the LSB today. If we did
that we'd just recreate the specification in a different form. I do not
consider that worth our while. On the other hand we have all these tests
that could definitely be valuable to distributions and upstream. Thus,
in these cases I was thinking that we could potentially invent a
mechanized way to pull out the tet stuff and create stand alone tests.
We could then just add these stand alone tests to the repo. This I think
is a worthwhile exercise if we can script it.
A second example is distribution identification, i.e.
AFAIK is currently not in the standard but is
a.) relatively simple to sum up in a problem statement and a solution
b.) should not be very controversial
On the other hand it is already on freedesktop.org. Thus, the question
here is, should we poke at low hanging fruit such as this? Should we
even bother with stuff that is already "documented and standardized"
elsewhere? At the f2f we heard from Marcus that having the xdg-utils in
the LSB is really handy.
Low hanging fruit such as the OS identification may be good point to
start discussions on mailing lists and vet the new process. Additionally
this should be something where we get easy agreement and thus put
ourselves back into the cautiousness of distro and other developers.
P.S. I know LSB 5.0 is top priority and we are trying to get that out
the door and everyone is busy. Trying to ride the momentu we have from
the f2f for a bit ;)
Robert Schweikert MAY THE SOURCE BE WITH YOU
SUSE-IBM Software Integration Center LINUX
Public Cloud Architect
rjschwei at suse.com
rschweik at ca.ibm.com
More information about the lsb-discuss