Storing (E-)optimal designs
r.a.bailey at qmul.ac.uk
r.a.bailey at qmul.ac.uk
Fri Aug 27 01:15:41 BST 2004
Looks good to me,
Ro in Brisbane
>
> Dear All,
>
> This email contains a proposal for how to organise the E-optimal designs
> which JP is producing into our file-based "database" of block designs.
> The proposal follows discussions with JP and Peter D.
>
> I propose a new content indicator E, with E1, E2, E3, ... having the
> meanings defined in the external representation. For example, the file
> v6-b8-k3-E1.a.xml would contain all (up to isomorphism) E-optimal binary
> connected block designs with v=6, b=8, k=3. Note that, for optimality
> criteria, *optimal* always means best in the class of binary designs
> with the same (v,b,k).
>
> We could, in due course, have new content indicators phi and mv
> for the other optimality criteria.
>
> I should emphasize that our present file-based "database" is an
> intermediate (and imperfect) mechanism to structure our large and growing
> collection of lists of designs, to make them accessible to the outside
> world in categories of interest (e.g. t-designs), and to be able to
> make assertions about the content of a file (e.g. this file contains
> all binary connected block designs up to isomorphism with this (v,b,r,k)).
>
> So for now, we should have a new directory v-b-k-E in which JP can
> place his E1-optimal designs. In due course we may decide to store lists
> of optimal designs (with respect to one or more optimality criteria)
> together in the same file for a given (v,b,k) and will need further
> conventions -- we shall see.
>
> Suggestions for improvements are welcome.
>
> Regards, Leonard
>
>
>
> _______________________________________________
> Developers mailing list
> Developers at designtheory.org
> http://designtheory.org/cgi-bin/mailman/listinfo/developers
>
--
R. A. Bailey
Snail: School of Mathematical Sciences Tel: (+44) 20 7882 5517
Queen Mary, University of London
Mile End Road Email: r.a.bailey at qmul.ac.uk
London E1 4NS
U.K.
More information about the Developers
mailing list