I’m back from a 3 week visit to Europe. The time was spent in Genoa (Italy) for NIME and Albi (France) for a Jamoma development workshop, with some sightseeing in between.

Alexander has posted some photos and other info from the NIME conference here:

• http://www.arj.no/2008/06/08/nime-jamoma-workshop/
• http://www.arj.no/2008/06/16/nime-paper/

Yannick has posted some really nice photos from Albi, where we worked for 2 weeks on moving Jamoma over to Max 5 and started to address future plans, usability, and other topics.

• http://www.flickr.com/photos/yhancik/sets/72157605754235826/

It’s taking a bit of time to get caught up with everything. It’s time for work to resume on finishing Tap.Tools 3.

A couple days ago I checked some code into Jamoma for managing automated unit testing. It is written as a Ruby script that uses the ‘rosc’ library (which was mentioned to me by Trond Lossius) to set up an OpenSoundControl server, launch Max, and then start running a series of test patches.

I’m hoping that we can create library of test patches in the coming weeks, and then run this script regularly after we make changes to the Jamoma source. This will be a tremendous help to improve the quality-control of the Jamoma releases and the Jamoma development process.

The information and scripts can be found in the active branch of the Jamoma SourceForge repository.

Jamoma has been building up momentum fairly steadily ever since I began working on it in 2001 (at the time it was Jade). This year is definitively the most active year for work with Jamoma.

• Last December we held a intensive development workshop hosted by iMAL in Brussels, Belgium.
• We will be presenting a paper (poster) at the upcoming NIME this June in Genoa, Italy (Authored by myself together with Trond Lossius, Alexander Refsum Jensenius, Pascal Baltazar, and Nils Peters).
• There will be a Jamoma workshop as a part of NIME in Genoa, Italy.
• June will also feature an intensive development/testing workshop in Albi, France hosted by GMEA.
• In August, Jamoma will be represented in several places at the ICMC in Belfast, Ireland. This includes a paper focusing on Jamoma by Trond Lossius, Alexander Refsum Jensenius, Nils Peters, and myself. There will also be a panel discussion focusing on SpatDIF, which is an initiative that is affiliated with Jamoma.
• In November, there are plans for a Jamoma workshop in Oslo, Norway.

Wow. Exciting stuff!

Jamoma 0.4.6 has been released at http://jamoma.org/

Jamoma 0.4.6 is quite stable, and represents 5 months of bug fixes and minor tweaks since we released 0.4.5. It works with Max 4.6.3, and is definitely the best release we’ve assembled.

It does not, however, work with Max 5. We have also been working on a Max 5 version of Jamoma over the last number months. Now that Max 5 has been released, we will be putting significant effort into a Max 5-compatible version of Jamoma.

The upcoming NIME conference in Genoa, Italy will host a Jamoma Workshop.

This workshop will be a good place to find out about Jamoma, see it in practical applications, and give us feedback. It should be fun!

Jamoma version 0.4.5 is now available from http://jamoma.org.

A lot of really good things in it, including the new jcom.map object, rewritten ramp libraries (and the jcom.ramp object). Also new is interpolation between presets, a whole new preset interface accessible from the module menu, etc.

If you haven’t updated or tried Jamoma since 0.4.2, then this update is a must, as literally hundreds of bug fixes are included.

Dave Watson has posted pictures of the Jamoma workshop we had last spring in France. The link is here: http://photos.elephantride.org/v/jamoma_ws/ .

Looking forward to the next workshop!

http://www.bek.no/Members/lossius/trond_events/trond_events_2007/jamoma-spatialization-workshop-in-bergen

BEK will be hosting an international developer workshop on sound spatialisation.

The workshop will be focusing on development of modules for spatialisation in Jamoma with a particular emphasis on development of a Spatialisation Description Interchange Format.

Last night, Cycling ‘74 published the latest Radial article by Gregory Taylor. It focuses largely on the mapping interface in Radial.

In the various presentations at the Jamoma workshop in Albi, we saw a number of mapping interfaces. In some ways this interface reminded me the mapping that Pascal showed in his environments.

I wonder if there is anything that Jamoma can learn from Radial here…

A lot has been happening with Jamoma at a fairly quick pace during and since the Jamoma workshops in France. I thought I would highlight a couple of things that have been going on in more detail than is present in the change logs…

User Interfaces

One change that may be noticeable, but is otherwise trivial, is that we changed the default skin used for the look of the built-in user interface. I had been thinking about this for a few months, but it was sparked by some comments Trond made at the workshop. First, Trond was preferring for them to be as simple as possible. Ideally we could get a graphic designer to do them. The point is that they should not be so distracting or drawing the eye in ways that are not fruitful… The emphasis should be on the information that is important at the time. While that last point is quite difficult, I believe we’ve made a step in the right direction.

The difference colors between module types is also more subtle. And finally, we all realized a need at the workshop to place some emphasis on using modules without the built-in interface. For many this will mean building your own interface and just using the module like a normal Max object.

Snake Redux

Since some of our conversations toward the end of the workshop in Albi I have been trying to figure out what role, if any, the snake should continue to play in our development efforts. My current thoughts on it follow…

The current multi-cable is fast, but it cannot be used for dynamic routing.

The snake, on the other hand, should be used for dynamic routing, but will need to do internal buffering to make this possible. This means that the CPU cost of the snake will be equivalent to the cost for using send_/receive pairs on the signals that it is passing. In the limited case of a matrix router, this should not be a problem because a send~/receive pair must be used anyway to break the signal chain of logical feedback.

Thus it seems that there would still be use for both the multi-cable and the snake.

One idea which was raised when we were in Paris was to use Jitter objects and pass matrices for the snake. However, we don’t want to require Jitter for audio modules if at all possible. It isn’t clear though, because perhaps we can use simple matrices without Jitter somehow?

Window shade and screen layout

Trond made the analogy that modules being visible at a given time are like the staves visible in a cutaway score at a given time. They are visible or invisible are when they are communicating

Propaganda

One thing I was trying to get to in my presentation at the Albi workshop were some of the ideas in this: http://www.scribemedia.org/2006/07/09/dhh/

In particular, the 3rd segment of the video (03) from roughly 2:00 minutes to roughly 5:15. In this segment David Heinemeier Hansson is discussing the philosophical reasons for a structured framework approach to web development (specifically using Ruby on Rails). Particularly resonant is the idea that “Decisions are bad,” but many of the ideas from Ruby on Rails have been influential in how I think about Max development and Jamoma.