During a special session devoted to "Preservation and Reconstruction of Software-based Artistic Works" at the 14th IEEE Melecon 2008 conference in Ajaccio, Corsica, I presented the paper "Sustainability of 'live electronic' music in the Integra project". The paper is written by Jamie Bullock, myself and Lamberto Coccioli. The paper will soon be published online by IEEE (see ieee.org). Apart from our contribution there was one interesting paper from the Corsican university about computer aided design of traditional Corsican music, and no less than four from the CASPAR project.

Integra has much in common with the CASPAR project albeit much smaller (4 to 1 is probably a fair representation). One of the CASPAR presentations was focussed on a ontology-based temporal model for documenting artworks. In the project there has been a central decision to work with the CIDOC-CRM which provides "a common and extensible semantic framework that any cultural heritage information can be mapped to". Though similar to the ideas put forward within Integra it is also different in certain respects. From what I could gather it is a (very) hierarchic structure with an obvious entry point (the author) and a well defined exit (the work). I was reassured that the documentation could be ammended to and updated but my spontaneous thoughts are that this model may work very well for a certain kind of works but be equally problematic for a different kind of works. I should point out that these are comments made based on a conference presentation (I have not thoroughly read the paper) so I may be wrong.

I think the most interesting of the papers presented was the one by Rousseaux & Bonardi titled "ReCollection: a Disposal/Formal Requirement Based Tool to Support Collection Making" but also "Will Software Modules for Performing Arts be Sustainable?" by Bonardi et al. was interesting. The former discussed the relation between 'collection' and 'categorization' and the latter presented, among other things, a potenttially very useful tool for parsing Pd and Max/MSP patches.

A couple of reflections regarding the Integra system:

The current paper is great in that it gives an overview of, in particular the Integra module. We need to complement this with more specific information about the strength of the concept of module definition. For instance, though a module definition may not contain sufficient information in itself, the combination of definitions of many modules organized in a hierarchical manner with also their inter-relations defined and along with documentation about the module will allow for a recreation of the module implementation. And it allows for constructing new knowledge about existing modules. We need to be more explicit about the relations between definitions and implementations.

I also think we need to be more explicit about the relation between the overall goal and the different components in the system. The library is only *one* possible implementation to access and make use of modules. For instance, the life span of a module definition is not dependent on the library.

Henrik Frisk
May 10, 2008

We are pleased to announce that an Integra-related paper has been accepted for publication in the proceedings of the International Computer Music Conference this year.

The paper titled 'libIntegra: a system for software-independent multimedia module description and storage' is authored by Scientific working group members Jamie Bullock and Henrik Frisk, and outlines the new Integra XML file format and the software we have written for handling it.

The paper will be presented as a poster by Henrik in Copenhagen 27-31 August 2007. We hope to see you there!

Integra is now available on Sourceforge. The project URL is:

http://sourceforge.net/projects/integralive

This will eventually be the place to file support and feature requests, browse documentation and report bugs.

The SVN repository is also now based on sourceforge, and can be accessed by following the above link.

A small feasibility study has been conducted in order to test how well the Integra protocol works in different environments. A simple patch was devised, and implemented in PD, Csound, and Supercollider. The results have been written up in the attached report.

The files used with each piece of software can be found in the Integra SVN. To checkout a copy use:

svn co svn://integralive.org/repos/modules/branches/dev

From a command prompt.

In addition to this, a short basic comparison has been made between the environments. The results of this are also attached to this entry. This comparison was intended for my own purposes, but I thought I would make it available for possible interest.

Feasibility Study

Comparison of Existing Environments

An initial attempt at a namespace for the Yamaha DX7 can be found here:

/development/resources

I started by drawing up a UML class diagram, and derived the OSC address space from it. This is a proposed as a possible direction for Integra, i.e. we define an abstract module hierarchy, which describes module interfaces and derive other things (module implementations, OSC addresses) from it. The classes towards the top of the tree are temporary placeholders until they have been properly defined.

There are also some PD patches that attempt to implement some classes, and an inheritance mechanism in PD. IMO it is quite clunky, but some aspects of it may be of interest. A more in depth feasibility study will be made available shortly.