blog.sinden.org

Last week I participated in a Human-Computer Interaction design class at the invitation of Indiana University informatics professor Marty Siegel. Our discussion was entitled "Human-Music Interaction"

Preparing for the session certainly got me thinking a lot about organ console design, and all of this came fresh on the heels of my discovery of a coupler on my organ at the cathedral. I've been working there for about two months and it took me that long to find this coupler.

Now, most organists are shaking their heads in disbelief at this point, so let me say a few things in my defense.

1. I had looked for the coupler, just not that hard.
2. I had only really hoped for it maybe once or twice.
3. Since finding it (stumbling across it, really), I haven't used it.

How is it that an organ console can be so poorly designed that standard feature could be hidden from it's primary user for hours of use?

Answer: organ consoles are generally poorly designed, and I think they could be a lot better.

For instance, on this organ the couplers are sort of scattered all over the place. Most of the pedal couplers are in the pedal division, but one of the pedal couplers is located in the division itself. Strange. And this is not the only instrument that is a little disorganized when it comes to the console.

To help us in our quest, I think it's important to look to organs of the past, and organs of questionable repute (theater organs) because the trajectory gives some hints as to where we might go. The theater organ, which is called on to play music of much greater complexity than the standard organ repertoire cannot afford to be poorly designed, and generally they are designed well.

Blockwerky beginnings

In the beginning, God created the heavens and the earth and Blockwerk, a formless void of organ sound. Organs were large and loud, and they were loud all the time. There was no control over what parts of the organ would play. The whole thing would play the whole time.

Then someone said, "Man, that high-pitched mixture sound is really annoying. I wish there were a way to stop it." And lo, stops were born.

GOOD DESIGN: Stops gave the organist control over the organ's resources. All stops operate independently of each other, and so the myriad resgistrational possibilities were born.

Pulling out all the stop styles

But organ builders were (and continue to be?) so enamored with this control, that they often fail to take into account visual and physical design aspects of the stop controls themselves.

Stops started as knobby things that could be grasped easily with the hand, and their arrangement, due to mechanical issues, generally had some connection to the divisions of the organ.

Examples in modern instruments: The Fisk organ at Old West Church, Boston, and the Jurgen Ahrend organ in the Musée des Augustins in Toulouse, France both have Positiv stops that are actually on the Positiv. While it is certainly difficult to make over the shoulder adjustments while playing, the location of the pipes controlled by these stop knobs is obvious.

Since the advent of electric stop action, however, things have changed. Stop knobs have become smaller mushroom shaped objects. They are no longer meant to be grasped with the hand, but flicked with the finger. On the other hand, multiple stops can be grasped at once. Meanwhile, stop layout has become more variable.

One might make the case that stop knob size, and the number of stops that can be engaged simultaneously with a single hand correspond to changes in historical voicing styles. North German Baroque instruments have large, widely-spaced stop knobs. These are meant to be drawn deliberately, one at a time. Large orchestrally-voiced instruments generally have small, clustered tabs. These are meant to be engaged in groups.

FUTURE DESIGN: Take a look at the Wannamaker organ in Philadelphia (pictured right). Here, the sheer number of stops (and divisions) created a serious design problem. Divisions are separated by color, rendering their contents clear to the organist. Here also, stop tabs replace knobs to save space. While this may be a step in the right direction, stops as such are not visually distinguishable within divisions.

Aristide Cavaillé-Coll, the famous French organ builder, also used color to his advantage. Spoons, mechanical controls that were operated by foot, were labeled in colors that corresponded to the divisions they affected.

For further discussion: Are stop knobs (pull to engage) preferable to stop tabs or rockers (press to engage), or is their another option that enables the organist to manually manipulate his sounds easily? Organists press the key to engage pitch. Should stops mirror this action (tabs) or should they distinguish stop selection by a pulling, gathering motion (knobs)?

Worthwhile wording

Whatever mechanism is chosen to engage sounds, that stop will need to be identified in some way. This is generally done with the stop name followed by the Arabic numeral indicating pitch (ex. Principal 8').

This is a poorly conceived system, especially between organs that all have essentially the same tonal options, but with different names.

It would be much more logical to rely on the number to convey sound information. Organists don't look for a specific stop name when they sit down at an organ, they look for the 8' principal (which is usually called 8' Principal -- 8' Montre on a French organ). The design of stop labels should reflect this. The number should be the most prominent, with the name included in smaller type as a means of largely esoteric information.

With a simple short-hand, all major stop families could be codified with symbols. Principals are underlined, reeds have a degree circle, strings are italicized (maybe not a good solution, but works well on the computer). Flutes, the most ubiquitous sound family on the organ, receive no symbol. Mixtures have long been easily identifiable by their Roman numbers.

This is merely a sketch of a system that could certainly be expanded to designate for short resonator reeds (·), celestes (±) or horizontal reeds (). Certainly one doesn't want to go overboard with this. The goal is not to show every detail of the tonal disposition but to easily identify stop family through a series of limited symbols (maybe five or six) rather than a slew of stop names (fifty or sixty). The only example I ever recall seeing of a symbol on a stop knob is Skinner's Flute Triangulaire which is sometimes labeled with a small open triangle.

To be continued: Topics for part 2 of this article: usable registration assists, combination action, artificially intelligent combination action, sequencers, James Higdon, thinking beyond single points of contact for input controls, thumb slides

Labels: organ