Forums

Hi Hiro,

You ask some very good questions:

kuwahara> Hi, I am just curious as to what is the
kuwahara> significance of this survey in terms of the next
kuwahara> SBML spec. If the voting results are meant to
kuwahara> reflect the voices of the SBML community and to
kuwahara> be taken into account seriously, then I think a
kuwahara> better means should be used. With the current
kuwahara> survey program, anyone can easily cast his/her
kuwahara> vote multiple times. So those who have very
kuwahara> strong feelings towards certain topics can
kuwahara> influence the results by casting their votes
kuwahara> multiple times.

Only the last vote is taken. (I check for duplicate names
and take only the last vote from each given person, ignoring
any previous votes.) I mentioned this point in a message
that was probably buried under other discussions:
http://www.sbml.org/forums/index.php?t=msg&th=722&start=0&rid=2


kuwahara> I am also wondering if there exists a
kuwahara> standardization process for the SBML
kuwahara> specification. For example, what is the
kuwahara> procedure for a porposal to become a new feature
kuwahara> in the next spec and who are involved in the
kuwahara> process? And who decides which features to be
kuwahara> added? Can those who have proposed new features
kuwahara> also be on the specification committee (if such
kuwahara> things exist)? If so, isn't that a conflict of
kuwahara> interest?

All of this has been in the hands of the "SBML editors",
which are currently Andrew Finney and myself. Up till now,
there has not been a more formal process (though we tried to
propose and introduce one circa 2003, to lukewarm
reception).

The voting approach you're seeing is a new development, a
move on the way towards a better process for SBML. I
presented an initial proposal at the October meeting (see my
slides at http://www.sbml.org/workshops/tenth/index.html on
"Restructuring the SBML process") and there will be formal
proposals circulated to this list soon. Ideally we would
have had the process in place in time for development of
SBML L2v2, but <insert excuses about lack of time here>, so
the current newly-introduced voting approach are a
half-measure, but one that also provides everyone with some
experience in the voting process and will hopefully let us
design something better for the more formal SBML process.

I promise there will be proposals and discussion on this
topic. The current scheme of having a handful of editors
(me, Andrew, Herbert and Hamid) do everything and make all
evaluations more-or-less worked when there weren't many
other people and projects involved, but it's unsustainable
today. We need something more open, formalized and
standardized -- closer to a standards-body organization like
the W3C (but simpler, because there aren't *that* many
people in the SBML world).

MH