Re: LSB and paper standards
On Wed, 12 Aug 1998, Ransom Love wrote:
> When I made the offer/invitation to keep LSB going, I was not aware of
> the posting of the announcement of LCS. I, nor anyone at Caldera, was
> approached on participating in LCS. I seriously doubt that SuSE or
> Pacific Hitech were either. To simply state that it was a couple of
> engineers who were acting on their own accord is hard to swallow. Red
Please remember the LSB situation at the time... Bruce had posted a note
to the LSB mailing list saying "the LSB is solely concerned with a sample
implementation and if others want a paper standard, go do it on your own".
Debian and Red Hat decided to announce a project to do that, and we didn't
approach other vendors solely because we felt like we wanted to announce
it as soon as possible. The *only* LCS mailing list is a public list, and
we fully intended on approaching otehr distributions after announcing.
The changes in the LSB which were announced yesterday completely changes
the complexion of this argument, and it's not fair to judge the LCS
announcement according to the Bruce-less LSB. The LSB engineering group
now seems interested in a paper standard (which Bruce saw as a very low
priority), which makes the LCS and LSB somewhat conflicting. Resolving
that conflict in a way which leaves us with a single standard is important.
> standard. Clearly the announcement of LCS shows alterior motives,
> motives which will divide effective attempts at providing a standard.
The LCS group was trying to provide a paper standard while Bruce's LSB
effort was focused on a software implementation, so yes, the two
efforts did diverge somewhat.
> and LSB need to come under one group. Announcing LCS as a seperate
> project is counter productive and shows an inability to cooperate which
> has created the need for an LSB in the first place. I would hope that
> Dale and Erik would be willing to make a new announcement along that
> I do not feel comfortable with Dale or Erik as the chairmen of the LSB
> paper specification unless it was a sub-committee and taking direction
> from the current LSB working group.
I'd like to see the LCS integrated as a working group of the LSB, with the
mission of writing a minimal standard. The LSB should serve as:
1) The steering group setting the priorities of the LCS
2) A source of a base platform, conforming to the LCS, if
members of the LSB want it.
> I agree completely. Caldera is not interested in supporting a paper
> specification only. I believe that both the specification and sample
> implemenation are needed for the ISVs to support this effort. Actually,
Red Hat and Debian both intend to provide implementations of the LCS.
> I disagree with Erik's comment about baggage with LSB. The commercial
> ISVs are very much interested in LSB. LCS will confuse and send the
> wrong message unless it is under LSB. LCS did not even have the
> blessing of LI. At least LSB was discussed and approved by a braoder
Read the LSB archives (if they exist) and then try to tell me there is
no ill-feeling there. Statements like "our direction is foo and we needn't
discuss it anymore" abounded, and made (at least) me feel disconnected
from the project.
To summarize: the changes in the LSB situation since Tuesday morning have
definitely changed the LCS/LSB relationship that I originally envisioned.
What that realationship should be now needs to be decided, and that may
involve abandoning the LCS. I'd like to see an LSB working group formed
with clear chairpeople to form a first-draft paper standard in a short
time frame. That paper standard may form the basis for an LSB sample
implementation; that standard will in all liklihood be implemented by Debian
and Red Hat as well.
However, judging the LCS announcement by today's criteria is not fair; the
announcement was made before any of us knew of the changes in the LSB
membership, let alone the possibility of the LSB changing directions to one
which dovetails with the LCS concept.
| "For the next two hours, VH1 will be filled with foul-mouthed, |
| crossdressing Australians. Viewer discretion is advised." |
| Linux Application Development -- http://www.redhat.com/~johnsonm/lad |