PDCON:Conference/Self-replication: how to do more using less: Difference between revisions

PDCON:Conference/Self-replication: how to do more using less (view source)

116 bytes removed , 19 July 2011

no edit summary

8

edits

@@ Line 1: / Line 1: @@
 Author: Krzysztof Czaja
-One of the tiny annoying vices which still compromise the many virtues
+''Multi-instantiation'' is the process of taking a single
-of Pd, thus preventing its wider adoption by computer music community,
+declaration of an object, or a patch, and creating several instances
-is the lack of implicit multi-instantiation mechanism.  For example,
+of it, ''replicating'' the structure, while possibly varying the
-according to the common expectation, any implementation of a sound
+initial state.  An important application, among many others, is
-synthesis algorithm should qualify for seamless abstraction into a
+supporting the implementation of polyphonic instruments.
-polyphonic instrument.  For Pd, this is not possible, yet, and a
-proposal of how to fill the gap is described in this paper.
-An integral part of Pd is instantiation-time evaluation, although
+There are two distinct design options: the process of replication may
-currently, it is limited to a simple form of parametrization of object
+be initiated and controlled either from the outside, or from the
-arguments.  One can imagine other possibilities of user-guided
+inside of the replicated patch.  The latter possibility is explored in
-computation performed during instantiation phase. This proposal
+this paper in an attempt to advocate for ''self-replication'' as a
-explores one such possibility.
+conceptually simple, yet quite generic and powerful mechanism.