jwilzewske

Saturday, January 19, 2008

The Rest is Not ‘Noise’: Can Concerts with Performers’ Explications between Works Be ‘Mashups’?

Georges Braque, Violin & Candle Mashup, 1910

W o Menschen singen lass dich ruhig nieder.
Denn böse Menschen signen keine Lieder.
[Where people are singing, rest easy.
Because bad people sing no songs.]”
— German proverb/rhyme.

The ‘unity’ of a concert experience—its musical components and its narrative components—occasioned a remark by Allan Kozinn in the New York Times last week, in his review of a Lark Chamber Artists concert. The concert was comprised of a diverse array of [excerpts of] works, ranging from Brahms’s Quintet Op. 111, Jennifer Higdon’s ‘Soliloquy’, John Adams’s ‘Book of Alleged Dances’, and Giovanni Sollima’s ‘Waves’. Allan wrote, “That’s right, just the Allegro [movement of the Brahms Quintet]. Several works were represented by excerpted movements. Generally, this is a bad idea, unless you can argue that individual movements from diverse works add up to a kind of classical mashup. But that’s an assertion you can’t make if, as in this case, each work is given a spoken introduction.”

[In popular multimedia/web-software vernacular, a ‘mashup’ is a ‘composite’ mix of elements created and edited from more than one source, so as to appear as one thing.]

Why would Allan say those things? What authority ever defined once and forever what mashups are or could be? (If anything, the term ‘mashup’ deeply resists any crisp or immutable definition because the inherently ‘populist’ nature of the mashup idiom defies elites and top-down authority, top-down ‘definitions’.)

And, anyhow, presenters’ ‘spoken introduction’ elements are regularly written or extemporized with great poetical and editorial care. Presenters’ remarks are ‘composed’ as programming elements in their own right, to complement and inform the music. Presenters should not be backhandedly demeaned as mere shills/impresarios/educators making ‘noise’, as Kozinn’s comment appears to do!

What’s more, a presenter’s remarks need not be didactic, in the way that Kozinn’s comment implicitly presumes. The presenter may say things that are instead ironical or provocative or enigmatic—for the express purpose of adding drama or stimulation or poignancy to the concert experience.

Or the presenter (or the author of Program Notes, or others) may construct the narrative so as to manifest certain ‘illocutionary’ expressive acts, ones that complement the meanings and modulate the meanings of the musical elements in the conglomerated ‘mashup’. ‘Illocutionary’ is a fancy word that means indirect effects that go beyond the literal text. But it also means that the indirect effects or implications may be multi-layered or different from the superficial or literal sense of the expressions. In English, for example, the interrogative mood is supposed to indicate that the utterance is intended as a question; the directive mood indicates that the utterance is intended as a ‘directive’ illocutionary act (an order, a request, etc.); the words ‘I promise’ are supposed to indicate that the utterance is intended as a commitment.

R ecall that, after Schubert’s death, his brother cut some of Schubert’s scores into small pieces and gave each piece, consisting of a few bars, to Schubert’s favorite pupils. And this act, as a sign of deep piety, is just as understandable as the different act of keeping the scores untouched, accessible to no-one. And if Schubert’s brother had burned the scores, that, too, would have been understandable as a different act of piety.”
— Ludwig Wittgenstein.

Possible illocutionary gestures in English include word-order, stress, intonation contour, punctuation, the mood of the verb, and ‘performative’ verbs. And if I say ‘I promise’ in with an upward pitch inflection at the end (as though asking a question about promising) or with a sarcastic tone, then my expression may not be a commitment at all; it may be a repudiation of any commitment. ‘Illocutionary act’ is a technical term introduced in 1962 by John L. Austin in philosophy-of-mind and linguistics investigations concerning what he called ‘performative’ and ‘constative utterances’. [Of course, ‘utterances’ can be music or text or imagery, not just speech.]

According to Austin’s ‘How to Do Things With Words’, an illocutionary act is an act (1) for the performance of which I make it clear to some other person that the act is or has been or soon will be or should be performed, and (2) the performance of which involves ‘consequences’ like rights, commitments, interpretations, or obligations. For example, in order to successfully ‘perform’ a promise I must make clear to my audience that the promise occurs, and then undertake the obligation to do the promised thing (or induce somebody else to do it). It’s a compound expression. It’s a type of mashup.

Austin introduced the concept of ‘illocutionary [speech] act’ by means of a contrast with other kinds of acts: the illocutionary act, he said, is an act performed in the course of saying something else, as contrasted with a ‘locutionary act’ (the act of saying some literal thing), and also contrasted with a ‘perlocutionary act’ (an act performed by saying something).

Kozinn’s glib ‘mashup’ comment, though, inadvertently relegates presenters’ narratives and the Program Notes authors’ texts to the dust bin—as something to be ignored, not part of the concert experience-proper—if we follow the logic of his remark to its conclusion.

Backhandedly, Allan’s attitude and comment also exalt a kind of traditional and deliberate ‘monumentality’ in classical music programming—a sort of ‘unitary executive’ politics, somewhat like Bruckner’s politics. “There is a pecking-order, and I shall tell you what that pecking-order is, and you shall passively receive it and like it and abide by it.” It is a composer pecking-order and a tradition-bound, conventionality-obsessed pecking-order.

I’m sure that Allan means well, but his comment also seems broadly defamatory and hostile to the whole idea of “individual movements from diverse works, adding up to a kind of classical mashup”—as if the only legitimate way, the only aesthetically and morally acceptable way, to perform a composer’s work is in its entirety. You must perform all of its movements, with all of the movements in the original sequence, and with no apocryphal or extraneous material appended or inserted or layered on top of it, palimpsest-style. Well, damn. I guess all of us electroacoustic folks should go shrivel up and die then. I guess all of us who like to hear chamber music excerpts played in bars should roll over and die.

No. I won’t curl up and die. I like my eclectic tastes, and I bet you do, too. I like to create peculiar playlists on my iPod—sequences that no sane presenter or chamber music ensemble would ever put on a program together. I like the quirkiness of my Last.fm dynamic impromptu mashups. I like to explore different combinations and emotional paths that were never envisioned by the composers whose pieces I put back-to-back. I like to attend concerts by expansive-minded artists and presenters who push the envelope, chart new waters.

Don’t get me wrong: the composer’s intentions do have moral standing and should receive due respect; misusing/abusing a work is never a good idea. But I do favor a ‘multiple drafts’ model of musical interpretation and meaning production, like Steven Jan advocates in his book, ‘Memetics of Music’. The individual (performer; audience member; composer; presenter) is regarded as the sum of her/his genotype and its phenotypic effects, together with her/his ‘memotypes’ and their ‘phemotypic’ effects. Each individual is endowed with inalienable musicality, albeit of different types and in different amounts. Each of us brings to our musical experiences her/his genetic and memetic heritage, conditioning, and intentions. Instead of the composer’s musicality and genetic and musical memetic ‘fitness’ always trumping all the rest of us, the ‘multiple drafts’ model contends that musical performances inevitably shape themselves—through interactions with the intentionality and musicality that each of the rest of us possesses. Mashups empower the rest of us, not just the composers, presenters, and code-heads.

P opfly is a mashup tool for end-users, not for code-heads.”
— Steve Ballmer, Microsoft.

The idea of the composers and artists and audience members as vessels or interconnected conduits for the flow and production of individualized meaning is a centuries-old theme in music history. Indeed, Mozart’s rants against the Archbishop of Salzburg’s provincialism include implications that composers are memetic ‘vessels’, whose expressive authority and merit depend not only on talent but on what memes fill the vessels.

A fellow of mediocre talent will remain a mediocrity, whether he travels or not; but one of superior talent, which without impiety I cannot deny that I possess, will go to seed if he always remains in the same province.”
— E. Anderson, tr. Letters of Mozart and his Family. 3e. Macmillan, 1985. P. 612, letter 331.

We have web mashups and music mashups. The former combine data from more than one source into a single integrated app. An example is cartographic data from Google Maps plus concert data from IONARTS blog, thereby creating a new web service that was not originally provided by either source.

Mashups generally combine content-elements from multiple sources and hide the disparate sourcing behind a unified user interface that obscures the bricolage. ‘Mashups’ and ‘portals’ are both content-aggregation technologies. Portals are older technology designed as an extension to traditional dynamic web applications, in which the process of converting data content into marked-up web pages is split into two phases: generation of markup ‘fragments’. and software-based automatic composition of the fragments into ‘pages’ to be served up to the client app. Each of these markup fragments is generated by a ‘portlet’, and the portal app synthesizes them into a single web page. Portlets may be hosted locally on the portal server or remotely on another server.

Portal technology is server-side, presentation-tier aggregation. It can’t be used to drive higher-level app integration or handle illocutionary interactive user gestures, two-phase transaction commits and roll-backs, or messaging-based redirection of the app-based conversation or search. Portal technology is heavy-handed composer/presenter-dominated politics.

Mashups are Web 2.0 composer/presenter/performer/audience politics. Maybe understanding and accepting and programming mashups is a ‘generational’ thing, an ‘age’ thing, a ‘class’ thing. I hope not. Mashups differ from portals in the following respects:

	Portal	Mashup
Classification	Older technology, extension to web services model; data models coerce semantics	Newer Web 2.0 distributed agent/broker/swarm technology; RDF/OWL data models do not impose top-down ontology and semantics
Compositional Method and Form	Approaches aggregation by splitting role of web services into two phases (markup generation and aggregation of markup fragments), privileging the role of programmer/composer	Adopts an egalitarian approach to content aggregation without regard to markup or authorship
Content Dependencies	Aggregates presentation-oriented markup fragments (HTML, WML, XML, etc.)	Can operate on pure XML content and also on presentation-oriented content (e.g., MusicXML)
Location Dependencies	Content aggregation takes place on the server	Content aggregation can take place either on the server or on the client
Aggregation Style	‘Salad Bar’ style (aggregated content is presented ‘side-by-side’ without overlaps)	‘Melting Pot’ style (individual content may be combined in arbitrary ways, resulting in flexible hybrid content)
Event Model	‘Read’ and ‘update’ event models are defined through a specific portlet API	CRUD operations are based on REST architectural principles, but no formal API exists
Relevant Standards	Portlet behavior is governed by standards JSR 168, JSR 286 and WSRP, although portal page layout and portal functionality are undefined and vendor-specific	Base standard is XML Data Interchange. RSS and Atom are commonly used. More specific mashup standards are expected to emerge.

Incidentally, MusicXML does not yet have an explicit ‘mashup’ ontology and syntax and semantics spec. (Maybe it needs one, preferably as an RDF/OWL ontology, not just low-level XML!) MusicXML 2.0 does have an implicit way to compose a mashup. Currently, MusicXML™ has standard specs to represent—Partwise DTD, where the partwise.dtd file represents a score by part/instrument/voice, and Timewise DTD where timewise.dtd represents a score by time/measure. The partwise.dtd contains measures within each part, while timewise.dtd contains parts within each measure. XSLT stylesheets are provided to couple the two DTDs.

The partwise and timewise score DTDs represent a single movement of music or track. Multiple movements or other musical collections are presented using an Opus DTD opus.dtd file. An opus.dtd MusicXML file contains XLinks to the individual scores, and, by default, they are serial. But there is nothing currently in the MusicXML spec that forbids Timewise DTD from coupling two or more scores, instead of two or more parts. Or, alternatively, you could just instantiate two or more scores in a merged partwise.dtd to create a synthetic mashup with the excerpts and sequences and synchronization you want.

This post has been a very long-winded way of saying, Sing! Play! Devise exotic musical mashups, in whatever ways suit you! Rest easy! People are doing music! (Even us people who use MusicXML for more than just writing and publishing music with Finale™ and Sibelius™ and LilyPond™ software.)

Remember, as they say in Germany, bad people sing no songs. Bad people just don’t make chamber music mashups.

Ken Ueno v. Aaron Jay Kernis composer Flickr Fight Web 2.0 Mashup

L ast night we were finishing off a jam cake and a cheesecake that were in the fridge so my wife and I both had slim slices. I ate my jam cake first, then my cheesecake. My wife ate them alternating bites of one with bites of the other. I guess her approach stopped short of stacking them and mashing them irrevocably together with her fork. She commented on our different methods. And I said, ‘But I like cheesecake, and I like jam cake. And eating them the way you do, I don’t think I get either.’ But there is a flaw in this analogy. In jam cake or cheesecake, a single bite gives us the entire essence of the cake (you’ve got to get some icing on the fork too, though, with the former, and crust for the latter), whereas I’d have to spend time with Bach or Led Zeppelin—to the extent of a movement, in the case of a suite or concerto (arguably) or the whole of the Goldbergs; and, on the other hand, through the entirety of ‘The Rain Song,’ unless one is of the feeling that an entire album constitutes the artwork (which, in Led Zeppelin’s case is not so much the case). So it would depend, in the case of the cakes, on being able mentally to separate the two, and on how quickly one can make that happen. Perhaps my wife can make the shift completely much more quickly than I, and enjoy, savor, each essence completely with a each single bite; perhaps my capacity is such that I need the saturation of a greater amount of cake, given the imminent distraction of the next cake on my plate, or the previous one. Maybe I could be satisfied with the first four bars, or just the theme, of Goldberg, with memory providing the whole of the essence, before moving on to just the guitar solo of the live version of ‘Stairway to Heaven.’ Or maybe you could, but I couldn’t... I’d be sorry to hear just one movement of Brahms in concert, but what if one doesn't know any better? What if one hasn’t trained oneself ... to expect the other movements and not feel the presentation is complete without them? Nineteenth-century listeners would’ve been much less concerned about it, and twenty-first-century listeners who aren’t primarily immersed in this so-called concert music tradition aren’t concerned about it, given the fact that their iPod shuffles through their Minor Threat, Death Cab for Cutie, and Haydn playlists without value/stylistic judgment. [Even though I don’t feel like changing my way of listening], I can acknowledge that mine isn’t the only way to listen.”
— Robert Kirzinger, email, 19-JAN-2008.

Iron Man v Snake Eyes (Part 1)

“At Graham County jail up in Safford,” Iron Man said, “they threw me in with the hardcore federal prisoners. It’s a pod-type setting. There’s one big TV at the top of a long hallway. Everyone’s always arguing over the TV. The whites had it every third day, then the blacks, then the Mexicans and Indians.
This one guy showed up. A big dude. Six-one. About two-hundred-and-twenty-five pounds. Half Indian, half Mexican. They called him Snake Eyes. He was always running his mouth about how tough he was and how he could kick so many peoples asses.
I had my own issues. My marriage was crumbling. My business was crumbling. All my equipment and tools were being stolen by fucking tweakers! So I was stressed. The first plea bargain they offered me was twenty years – and I couldn’t believe it ’cause all I did was knock some asshole’s door in that owed me five-hundred dollars, and he’d been dodging me for a month and a half!
Snake Eyes started saying he was gonna watch TV whenever he wanted, and he didn’t care who wanted to watch it, that there were more Mexicans and Indians in the pod than white boys, so they could control the TV any time they wanted. Well, the Superbowl was coming up, so I told him,‘We’re watching the Superbowl on Superbowl Sunday and that’s that.’”
“Was Sunday the white’s TV day?” I asked.
“It was the black’s day and we had an agreement with them.
Snake Eyes said,‘You’ll watch the Superbowl if we let you watch the Superbowl.’ I said,‘No. We will be watching the Superbowl, and that’s that.’
I went and worked out with two buddies I’d made. A white guy comes out and says,‘Hey, Iron Man, I think the Mexicans are gonna jump us when we go back into the pod.’ And I said,‘Let’s go find out right now.’
We go in. Snake Eyes is standing there with ten Mexicans. He says,‘We’re gonna settle this shit right now. We’re gonna fuck you guys up right now.’ He’s trying to pump up all the Mexicans saying,‘Let’s get these white boys.’
Right then I realized it would not turn out well if the four of us had to fight these eleven guys. I felt rage well up in me. Every single bit of rage that had come from everything happening in my life welled up in an instant – the plea bargain, the marriage, my business, all my fucking equipment that I’d invested every single penny I’d saved and could borrow was being stolen by tweakers and now I’ve got this loudmouth blowhard wannabe tough guy trying to get ten Mexicans to beat me and my friends up, so I….”

What would you do next if you were Iron Man?

Click here for Iron Man v Snake Eyes Part 2

Email comments to writeinside@hotmail.com or post them below

Copyright © 2007-2008 Shaun P. Attwood

Thursday, January 17, 2008

Humidifying Captive Singers Breathing at High Tidal Flow rates >10 L/min

The reaction surprised me. My December CMT post about humidity effects on performance hall acoustics—an exotic topic and one that I addressed with some physics and a spreadsheet—has elicited quite a few emails. The post concerned the timbre and attenuation of sound as a function of distance from the stage. That is, my remarks were from the perspective of the audience and presenters and architects. But so far the comments and emails about the post have been almost entirely from singers, most of them objecting to my “Hypothetically [from an acoustics perspective], it would be wonderful to perform in very low-humidity air” remark. All of the commenters opine that relative humidity (RH) equal to 45% (the nominal target value that HVAC systems in most public buildings aim to maintain) is too low for their vocal health—their throats become dry, their vocal cords become inflamed. The commenters further say that even the higher RH range 55%-65% that I was arguing for (for acoustics reasons, not performer health reasons) in the previous CMT post is lower than they would like. [Absolute humidity (AH) is the amount of water vapor present in a gas. Relative humidity is the ratio of the absolute humidity to the maximum absolute humidity. Relative humidity is perhaps most important, as any deficit then becomes an evaporative ‘challenge’ to the tracheobronchial tree, which has to deal with the humidity gradient—100% RH deep in the alveoli in the lungs, and ambient room RH at the mouth and nostrils.]

I agree with your comments, about preferring more humid air to sing in! I sing tenor myself! So I think, in this post let’s look at the health rationale for performance hall humidity that’s high. How high is compatible with the comfort and health of all the people in the hall, performers and audience members alike? Can architects do things to make performance halls safer and healthier for singers? Below I suggest some answers to those questions.

Human skull, view head-on through nose, showing turbinate bones

First, think about what’s different when you’re singing vs. when you’re not singing. When you’re singing, far less of the liters per minute of air that you breathe in and out goes through your nose. The nose has several functions: it warms, cleanses, and humidifies inhaled air, detects odors, and serves as a resonating chamber to modify the voice. Air enters through the external nares (nostrils) to the nasal cavity. The nasal cavity is divided by the nasal septum into right and left chambers called nasal fossae. The chamber inside the fossa is called the vestibule. The vestibule is lined with stratified squamous epithelium. The air flow over 3 pairs of structures: inferior, superior and middle nasal conchae. The conchae consist of mucous membranes supported by thin, scroll-like turbinate bones. The passage through the nasal cavity warms, moisten and filters the air.

But when you’re singing much of the air you inhale goes through your mouth, and less of the air goes through your nose than when you’re not singing. And the moist surfaces in your mouth and oropharynx have less surface area from which moisture can evaporate, compared to the moist surfaces on the conchae in your nasal passages. What’s more, the velocity of the air going past the moist surfaces in your mouth is greater and the time that the air has to pick up moisture from those surfaces is less than it would be in nose breathing. The net result is that the air reaching your vocal cords is far drier—less humid—when you are singing than it would be if you were not singing and predominantly nose-breathing. The rate of evaporation of moisture from your larynx and your lower respiratory tract is greater when you are singing than it would be at the same air flow rate in L/min by nose-breathing. This situation not only dehydrates the airway lumen, but also increases the effort required to produce sound, as Sivasankar and others have shown.

You can’t bring your shower or a rainforest to the concert hall. Or can you? Could an architect design passive humidifiers (a back-stage waterfall?; an on-stage ‘water feature’?) that would create a microenvironment capable of transferring enough moisture? No. Even if you wanted to do it, it would not be adequate. Passive transfer can’t get enough kg of water into the air, even in a small area, to cope with the losses as the air is mixed and exchanged by the HVAC system. ‘Active’ humidifiers (atomizers/nebulizers, ultrasonic foggers, steam-based systems, etc.) are needed.

On an architectural scale, the situation has a lot in common with the heat and moisture exchanger (HME) elements [for example, Tyco DAR Hygrobac, Pall BB2215, EdithFlex] that are used in intensive care unit (ICU) mechancial ventilators in hospitals. Active heat and moisture exchangers (HMEs) represent a simple and effective way to replace one of the most important upper airway functions: they retain the heat and moisture of expired air and return it to the inspired gases. As the nasal cavities normally play a very active role in this conditioning, bioengineers jokingly refer to HMEs as ‘artificial noses’. And the ability of any nose, natural or artificial, to prevent drying of secretions in the respiratory tract and drying of the respiratory mucosal surfaces depends on the delivered gas temperature and relative humidity. We look at this aspect in more detail below.

HMEs are made of cellulose or synthetic materials and have an active surface area that is about 2,000 cm². This provides absolute humidity (AH) levels of 30 to 35 mg/L (H₂O content) in the air going from the endotracheal tube into the patient’s trachea—an AH that is somewhat above the level required by the ISO-8185 and ISO-9360 standards, which specify performance guidelines for active humidifiers in medical mechanical ventilator circuits. Temperature of the air at the HME output into the ventilator breathing circuit is between 27°C and 34°C, depending on the device type and the flow rate. And, of course, the amount of moisture the air can hold is temperature-dependent: air at 20°C can only hold a maximum of 17.3 mg/L, whereas at 37 °C (normal core body temperature, 98.6°F) it can hold 43.9 mg/L. So the air at the output of a well-functioning HME is pretty close to saturated with water—a relative humidity of 80% or higher (in the upper right-hand corner [white cells] of the spreadsheet below). But as the air gets warmer in the respiratory tract of the patient, the relative humidity in that inspired air drops—to 70% or less in the trachea and mainstem bronchi. Click on this spreadsheet screenshot to download the spreadsheet and see the 4th-order statistical regression equations for the quantitative relationships between RH, temperature, and AH. What we’d be aiming for in an architectural / HVAC design is to take room air (about 20 °C) along the olive-shaded humidity “isobar” or “isopleth”(lower right-hand ellipse, arrow toward upper left-hand ellipse in the spreadsheet screenshot). The best we could possibly do is get the air to be about 35% RH when it hits your vocal cords.

Spreadsheet to Calculate Absolute Humidity as a Function of Relative Humidity and Temperature

In fact the flow-rate/moisture-transfer-effectiveness point made above about velocity and residence-time in the mouth-vs-nose breathing discussion above is also a very real issue in mechanical ventilator management and HME design. Moisture content (absolute humidity, AH) in mid-airway at 2 hours (ISO-9360) is a fuction of ventilator tidal volume, and, with typical ICU ventilator and typical HME-type humidification of ventilator circuits, we get values like these:

V_t = 250 mL: AH = 34.4 mg H₂O/L @ 32°C
V_t = 500 mL: AH = 33.6 mg H₂O/L @ 32°C
V_t = 750 mL: AH = 33.1 mg H₂O/L @ 32°C
V_t = 1000 mL: AH = 32.9 mg H₂O/L @ 32°C

Other things being equal, the greater the flow-rate through the HME per minute, the less moisture gets transferred to the air going through it—and the lower the AH in the air that gets into the respiratory tract, and the more moisture the respiratory mucosae will lose to the air.

The aim is to lose as little moisture by evaporation from the respiratory mucosal surfaces as possible—to maintain the mucociliary function of the cells in the respiratory tract that get rid of debris and micro-organisms; and to prevent injury to vocal cords, trachea, and oropharyngeal mucosa.

The biophysics and bioengineering principles that have been used for designing mechanical ventilators and humidification devices for very sick ICU patients might seem to be pretty remote from respiratory design considerations pertinent to healthy singers. But they are not. When you’re singing, the higher air velocity, larger tidal volumes, and less convoluted path (smaller average pathlength) and smaller surface area in the larger-caliber airway reduce moisture transfer into the inspired air, compared to breathing through your nose. Let’s say your concert performance is nearly 3 hours long, and your part has you singing nearly continuously. Physiologically speaking, that situation is actually not too different from what the air path and flow rates would be if you were intubated and on a ventilator without an HME, running on only dry room air for 3 hours. If you were on a ventilator for a 3-hour surgery, there’d surely be an HME or similar device maintaining the humidity in the air you breathe and keeping your respiratory tract moist. It takes very little time inspiring dry air to cause damage!

And look at the microscopic picture below! This shows the acute inflammation that comes with just 3 hours of breathing dry air. This is from Hirsch’s 1975 journal article, on the physiology of trachea mucus velocity and humidification in dog trachea. Obviously, you don’t get this sort of histologic data in humans, in autopsy pathology data or otherwise. The only way to get this kind of data is in experimental animal models, as Hirsch did. The amazing thing about this is how fast the onset of inflammation happens—the activation of the mast cells underneath the mucosal surface, the release of cytokines, the infiltration by polymorphonuclear leukocytes and other cells in the inflammatory response—over the course of only several hours exposure to dry air. Your subjective impressions as a singer (about your throat feeling injured or inflamed after performing in dry air) are dramatically confirmed by this microscopic photo!

Hirch 1975, Fig. 3, Photomicrograph of trachea after 3 hours of breathing dry air

This evaporative drying and inflammation process is, in fact, why proper humidification of ICU ventilators has been such a prominent topic over the past 30 years in critical care medicine. Keeping the respiratory tract from drying out is essential for preventing ventilator-acquired pneumonia (VAP) and avoiding injury to the respiratory tract. Heating and humidifying the inspired gas has for many years been an established standard of care for patients on mechanical ventilators.

Keeping the vocal cords and surrounding tissues from drying out is also important with regard to the biomechanics of phonation and musical sound production, as Verdolini and colleagues have shown.

A study by Tanner and coworkers examined whether any of several inhaled nebulized ‘mist’ products could help alleviate the effects of trans-oral breathing of dry air. The answer was, basically, ‘No.’ Whatever benefit they provide lasts only a few minutes—far too short a time to be of much help for a 2-hour performance.

So, if personal-use products won’t help, could systematically improving building designs help? Yes. Architects could design or retrofit performance halls to provide proportional zone humidity control. The humidity is sensed in each ‘zone’ and compared to a setpoint, similarly to what’s done with zone thermostats and temperature control for different temperatures in different parts of the building. If the humidity in the singers’ zone is below the preferred setpoint level, a control action is taken to add moisture in that zone. The humidity is sensed periodically or continuously. One or more hygrometer sensors in each zone and the hygrostat control circuitry enable precise and accurate control.

The hygrostat is typically a microprocessor-based device that implements a software algorithm-controlled feedback loop, and is capable of communications over a local area control network (LACN). The control sequences can expand on basic proportional humidity control and include ‘integral-derivative’ control. In this case, the integral (area under the measured humidity-time curve) is used to calculate the amount (kg) of moisture that the humidity has deviated from the setpoint. The control action is limited to avoid overshooting the setpoint and oscillations and delays in control response, as often occur with proportional control. ‘Derivative’ or ‘humidity rate-of-change’ control can be used for dynamic applications where the ambient conditions change relatively frequently and dramatically, due to local weather or other factors—very much in the same way as derivative temperature control is done. Derivative control measures the slope of the humidity trend, and it adjusts the parameters of the control algorithm to respond to the slope changes.

As one possible approach, architects could create forced-air active-convection ‘air curtains’ between zones—adding ducting to create thin, downward planar ‘jets’ of air traveling tens of centimeters per second, directed from the ceiling to the floor, to create and maintain separate zones or microenvironments—one for the stage and one for the audience? Too radical, you think? Well, air-curtaining is routinely done in industry—such active ‘air curtains’ are used in clean rooms for medical device and pharmaceuticals manufacture, for example. And they could be practical in studios, to create separate ‘zones’ (a high-humidity one for the area occupied by the singers, separated from other zones for audience or other performers). But such air curtains would not be very practical for performance hall stages with very high ceilings. And there would be tremendous problems with acoustic baffling and duct noise. The high air velocities required for effective ‘air curtaining’ are normally noisy—they are designed for industrial purposes where turbulent flow patterns and noise are not important design considerations. To design air curtains for large halls that would achieve quiet, near-laminar flow patterns and acceptable acoustics for chamber music would require huge effort and expense. It could be done—but it would be very difficult. I’m unaware of performance facilities where multi-zone humidity control has been done.

But passive-convective multi-zone humidity control is done all the time in zoos. The reptiles and other animals have the atomizers, nebulizers, ultrasonic foggers, etc., controlled by multiple hygrometer sensors to create suitable microenvironments for them. Do singers merit less attention from architects and HVAC engineers than snakes?

How high should the humidity target level be, to be of significant help to singers? Well, normal breathing through the nose adds about 75% of the total water content before the inspired gas reaches the larynx, whereas with mouth-breathing, inspired gas is only about 25% saturated above the pharynx. Returning to our ICU ventilator analogy for guidance in suggesting an architectural design value, we know that, for respiratory tract health, the inspired gases should be delivered to the endotracheal tube or tracheostomy close to saturation at body temperature (i.e., giving a minimum RH = 75% when warmed [by passage into the airways] to 37 °C, AH = 32 mg/L). The new-generation HMEs are capable of providing 30 to 35 mg/L AH at 27 to 34 °C.

The rate of moisture loss from respiratory tract mucosa is proportional to the difference (P_sat - P_a) in water vapor partial pressure at the respiratory tract mucosal surface temperature and the average ambient temperature in the airway (mixed bulk air, near the airway centerline), respectively. If the evaporation rate exceeds the moisture secretion rate by the glands lining the respiratory tract, then drying of the mucosal lining will occur. For example, in a cool room, P_a will be low enough that moisture can easily evaporate from the respiratory tract mucosa even if RH = 100%. In fact, you lose about 7 g of water per m² of body surface area each hour via normal breathing at normal body temperature. This evaporative loss of moisture from the respiratory tract is nearly independent of ambient temperature below 30 °C (because the temperature differential is large, P_a is small, and the (P_sat - P_a) differential is therefore large). There’s no way around it! Unless you’re singing in the shower or under a waterfall, you’re always losing moisture.

Even the most efficient HME also allows a net loss of heat and moisture from the respiratory tract. But an efficient HME reduces the rate of loss to a modest, sustainable figure that the body can “keep up” with, without the mucosal surfaces getting dry or injured. HMEs constructed out of hygroscopic materials outperform hydrophobic HMEs in terms of keeping the respiratory tract well-hydrated. A singer-friendly architect’s design and HVAC engineering design would be ones that deliver air to zones that singer’s perform in such that the singers’ respiratory tract mucosal glands can “keep up” with the evaporative losses without getting dry or injured.

Assessing the adequacy of humidification is difficult. Excessive humidification may cause an increase in respiratory secretions volume, and insufficient humidification may result in a decrease in secretion volume because mucus becomes encrusted in the airways and retards further secretion. In my experience, there’s never “too much” humidity when it comes to singers’ vocal comfort and health. The biggest impediments to optimizing room humidity for singers are that (a) the comfort of others and the environmental requirements for some musical instruments (e.g., strings) entail lower RH, in the 45% to 65% range, and (b) the existing HVAC systems don’t have the necessary humidifier capacity to go above this range. Ideally, though, I believe we singers would want RH in the 65% to 70% range, if one of the HVAC approaches mentioned above could fit within building budgets. Secondly, we’d prefer that the performance hall temperature be set higher than normal, maybe 23 °C or so. That way, the air at RH 65%-75% will have a higher AH, will carry more moisture, resulting in higher RH values in our vocal tracts. But first, you need to find architect/engineering firms that’re ready to think a bit “outside the box,” to begin to serve the needs of singers.

Thanks again to all of you who emailed me, for your comments and questions!

DSM. Humidity and Performance Venue Timbre. CMT blog, 28-DEC-2007.
AARC. Clinical Practice Guideline: Humidification during mechanical ventilation. Resp Care 1992; 37:887–90.
Adolph E, Visscher M, Bronk D, Landis E, Ivy A, eds. Physiology of Man in the Desert. Wiley Interscience, 1947.
Branson R. Secretion management in the mechanically ventilated patient. Resp Care 2007; 52:1328-42.
Branson R. The effects of inadequate humidity. Respir Care Clin N Am 1998; 4:199–214.
Branson R, Davis K, Campbell R, Porembka D. Humidification in the intensive care unit: prospective study of a new protocol utilizing heated humidification and a hygroscopic condenser humidifier. Chest 1993; 104:1800–5.
Branson R, Davis K. Evaluation of 21 passive humidifiers according to the ISO-9360 standard: moisture output, deadspace, and flow resistance. Respir Care 1996; 41:736–43.
Gilmour IJ, Boyle MJ, Rozenberg A, Palahniuk RJ. The effect of heated wire circuits on humidification of inspired gases. Anesth Analg 1994; 79:160–4.
Hirsch J, Tokayer J, Robinson M, Sackner M. Effects of dry air and subsequent humidification on tracheal mucous velocity in dogs. J Appl Physiol 1975; 39:242-6.
Nakagawa H, Fukuda H, Kawaida M, Shiotani A, Kanzaki J. Lubrication Mechanism of the Larynx during Phonation: An Experiment in Excised Canine Larynges. Folia Phoniatr Logop 1998; 50:183-94.
Nishida T, Nishimura M, Fujino Y, Mashimo T. Performance of heated humidifiers with a heated wire according to ventilatory settings. J Aerosol Med 2001; 14:43–51.
Rankin N. What is optimum humidity? Respir Care Clin N Am 1998; 4:321–8.
Sivasankar M, Fisher K. Oral breathing challenge in participants with vocal attrition. J Speech Lang Hear Res. 2003; 46:1416-27.
Sivasankar M, Fisher K. Oral breathing increases Pth and vocal effort by superficial drying of vocal fold mucosa. J Voice. 2002; 16:172-81.
Sottiaux T. Consequences of under- and over-humidification. Respir Care Clin N Am 2006; 12:233–52.
Tanner K. Roy M, Merrill R, Elstad M. Effects of three nebulized osmotic agents in the dry larynx. J Speech Lang Hear Res 2007; 50: 635-46
Verdolini K, Min Y, Titze I, Lemke J, Brown K, van Mersbergen M, Jiang J, Fisher K. Biological mechanisms underlying voice changes due to dehydration. J Speech Lang Hear Res. 2002;45:268-81.
Verdolini K, Titze I, Fennell A. Dependence of phonatory effort on hydration level. J Speech Hear Res 1994; 37: 1001-7.
Vintturi J, Alku P, Lauri E. Objective analysis of vocal warm-up with special reference to ergonomic factors. J Voice 2001; 15:36-53.
Williams R, Rankin N, Smith T, Galler D, Seakins P. Relationship between the humidity and temperature of inspired gas and the function of the airway mucosa. Crit Care Med 1996; 24:1920–9.
Crane Ultrasonic Humidifier
Sunpentown SU-2081 Ultrasonic Humidifier
ISO-9360 Anaesthetic and respiratory equipement heat and moisture exchangers. ISO, 2002.
EN-13544-1 nebulizer standards (CEN Technical Committe TC215WG2)
ISO-8185. Humidifiers for medical use. (BS-5724-2.24) ISO, 1998.
ISO-8185. Humidifiers for medical use. ISO, 2007
Respironics Inc.
Rotronic hygrostats and hygrometers
DAR Filters (Nellcor)

Ceiling-suspended atomizers, Copenhagen grocery

06 Sep 07

Xena Influences a Youngster

“I’ve been working out,” Xena said, showing off long legs tattooed with trails of ants.
“Still doing dick-lifts?” I asked.
“I quit those ’cause my prostate hurts. I just need to have my nuts cut off then all my problems will be over with – other than what wheelbarrow to carry my breasts in.”
A member of the American Indian people of northeastern Wisconsin joined Xena, causing Xena to sing,“Menominee doop do de do do, Menominee doop do de do. He’s a cutie pie, like a Menominee doll. That’s just like a China doll except much darker.”
“My sister’s boyfriend,” I said, “almost got shot by the Taliban. I wonder if he questioned his life.”
“I got shot and stabbed and questioned my life,” Xena said.
“What did you ask yourself?”
“Why the fuck am I here? Can anyone see the wet spot between my legs? My God, why does blood smell a lot like shit?”
“Xena,” a passer-by yelled,“don’t forget your hood.”
“What hood?” I asked.
“The Gay Triad Hood,” Xena said. “It’s all over the world. We don’t need a street number. It’s like South Side Posse Bloods – we let everyone join.”
A youngster was brought to Xena.
“Whoowhee,” Xena said. “He new booty!”
“He’s twenty-one. First time in prison. Only got two in.”
Filled with awe, the youngster looked up at Xena.. His face turned red. He tried to speak but no words came out.
With mischief radiating from his face, Xena said,“We’ll turn him out, tear him up, and send him back to the hood – then, maybe, it’ll be his last time inside!”
The youngster’s eyes widened with horror. He looked over his shoulders, and hastened away.
The onlookers laughed.
“Xena, I have to go back to Yard 1,” I said.
“When you dream about me tonight,” Xena said, “just say good things and feed me Hershey’s Chocolate Kisses while I’m layin’ on my bed – a queensize bed of course – and tell the flower boys, ‘Rose petals. Red and purple,’ ’cause the cosmos make me sneeze.”

Email comments to writeinside@hotmail.com or post them below

Copyright © 2007-2008 Shaun P. Attwood

Tuesday, January 15, 2008

15 Jan 08

On Shanks (Part 3)

“What kinds of circumstances lead to prisoners getting shanked and what’s your advice to a new prisoner?”
“The circumstances include drug debts, plain stupidity, disrespect, or trying to get in the mix with the gangs. My advice is: do your best to stay out of the gangs and things that bring problems, especially drugs.”
“If things have developed to the point where a prisoner suspects he’s about to get shanked, what defensive measures can he take?”
“You take magazines (preferably National Geographic, that’s the best one) and you strap them around your waist using either Saran Wrap or garbage bags, and you tie it on with clothes or whatever works. Saran Wrap out of the kitchen is best. You have to tie it up to cover your middle-torso area. The magazines act like body armour. Then, if you haven’t got big fists, get a big stick, a rock, batteries or a padlock in a sock, or anything else like that you can lay your hands on.”
“Say someone catches you offguard and shanks you in the heart, how long have you got left to live?”
“It’s just a matter of minutes.”
“You touched on how slow it is for the prison to get a prisoner to a hospital. Exactly how slow is it?”
“If they life flight you out, it’s usually at least half an hour. It takes time for them to figure out what’s going on. Then they gotta call for the helicopter. The helicopter has to fly in, land, pick you up, fly out. If they wait for an ambulance it takes a lot longer. Plenty leave on the helicopter and don’t come back. I’ve seen cops leave on the helicopter and not come back, and they get the helicopter here a whole lot faster for the cops.”
“How common are shankings in the prison system?”

Email comments to writeinside@hotmail.com or post them below

Copyright © 2006-2007 Shaun P. Attwood

Monday, January 14, 2008

Freund: Yet Another Partita Compiler (YAPC), Refactorings Noch Nach Bach

[50-sec clip, Don Freund, Sonapartita, Adagio, Daniel Rubenstein & Muhiddin Dürrüoglu-Demiriz, 1.3MB MP3]

[50-sec clip, Don Freund, Sonapartita, Passepied, Daniel Rubenstein & Muhiddin Dürrüoglu-Demiriz, 1.3MB MP3]

[46-sec clip, Don Freund, Sonapartita, Presto, Daniel Rubenstein & Muhiddin Dürrüoglu-Demiriz, 1.1MB MP3]

[50-sec clip, Don Freund, Sonapartita, Siciliano, Daniel Rubenstein & Muhiddin Dürrüoglu-Demiriz, 1.3MB MP3]

This exciting and beautiful piece contains extensive musical references to Bach, and other modern composers like Lukas Foss and Stravinsky. The title alludes to ‘Nach Bach’, a harpsichord piece by Georges Rochberg. The ‘Noch Nach’ is ‘Yet Another, After Bach’. For example, the third movement (Presto) replicates the violin part of the final movement of J.S. Bach’s G minor Sonata but has Freund’s imaginative piano part interacting in curious ‘noch nach’ ways with the violin.

Besides Rubenstein’s and Demiriz’s superb and sympathetic performance of it, the Freund work is interesting on multiple levels as relates to compositional methods and musical analysis.

The structures of the Bach sonatas and partitas that this ‘sonapartita’ mimics can be cast in computational form—using humdrum, for example. And, although I do not know whether Freund used a computer software to help generate this composition algorithmically, it’s possible to see from his finished result how a series of Java constructors and ‘refactoring’ transformations could be written to do so. Freund’s german title for this piece ‘Noch Nach Bach’ reminds me, in fact, of the old Stephen Johnson AT&T Bell Labs compiler generator utility, YACC (Yet Another Compiler Compiler), part of the original UNIX operating system.

An early decision in writing a compiler is whether the language will be treated in terms of patterns of characters (individual notes) or as patterns of tokens (horizontal motifs and vertical clusters). How do you create a compiler to generate the result you want/expect? We use refactorings such as safe rename and move refactorings, plus advanced refactorings like ‘ExtractMethod’ or ‘ExtractSuperclass’—or in a music composition application, we would use more complicated refactorings to be performed across large spans of each movement such as ‘ApplyMotifSupertype’ or ‘InferTypeArguments’ or ‘GeneratePartitaCounterpoint’.

The refactoring performed on the method A#partita() above first searches for all method invocations to partita. The only occurrence can be found in the constructor of class B. The Freund refactoring inserts the new indirection method indirection(A), which delegates to the original bach partita method. The partita-generating method invocation to partita in constructor B() is redirected to call the new indirection method. This object-oriented refactoring program amounts to a transformational network compiler—in this case, one that automatically composes partitas mimicking the patterns represented in Bach.

David Lewin’s wonderful book covers ‘transformational network’ theory from the point of view of post-composition analysis, not from the point of view of algorithmic composition. Of course, Klumpenhouwer, Lambert, Stoecker, Headlam, Rahn, and Mazzola and many other musicologists have productively applied transformational nets in analyzing music and for music information retrieval (MIR).

To date, though, software composition tools and ‘compiler compilers’ to create composition tools that can generate music that faithfully emulates design and orchestration patterns in established idioms (like Bach’s) have lagged behind. But Eclipse and other platforms for creating compilers and ‘compiler compilers’ have gotten significantly better and easier to use over the past 5 years or so. It is now far more feasible to create decent software-based composition tools than it was just a few years ago.

That side-steps the question of whether using such a tool—even as an ‘assistant’—violates aesthetic principles you hold dear. For my part, I would use such tools, provided their musicality was acceptable. After all, I use ProEngineer™ CAD/CAE software for the engineering that I do in my day-job. I use AdobeIllustrator™ and other computer-assisted art software for graphic art that I create. Those are not ‘crutches’ that in any way replace or undermine the integrity of my role as a designer. Why wouldn’t I happily use a good software composition tool? But the ‘compose’ widget in Finale and the GenJam applet for jazz improvisation aren’t that helpful. And the computational challenges are so daunting that I won’t hold my breath. Twenty years ago, William Schottstaedt worked on software-based counterpoint voice-leading and, since then, most people have used Markov-chain or other stochastic algorithms. These all focus on ‘small-scale’ decisions—Markov is, after all, a nearest-neighbor algorithm for generating a set of choices from which to choose. Schottstaedt used a best-first search algorithm to select a preferred candidate at each step. More recently, genetic-algorithm and simulated-annealing techniques are being used. But none of them is yet, so far as I know, very satisfactory. And none of them has an effective ‘scope’ that extends beyond about a 10-note or 12-note phrase. If there are 10 notes and, say, just 16 ways of moving from one note to the next in a given part or voice, then a phrase of a mere 10 notes begets 16¹⁰ possibilities—a combinatoric explosion, even for just one part and just one 10-note phrase. So, if a partita compiler compiler is going to succeed within the memory constraints on currently-available systems (even with 64-bit machines and terabytes of main memory and pagefile store), it will clearly have to go back to Fux’s ‘Gradus ad Parnassum’ or other rule-based processes to generate choices and to supervise polyphonic orchestration. Algorithms to predict the weather or earthquakes have better odds!

I wonder whether Freund was thinking about YACC when he chose his title. I wonder whether he used any computational tools in creating this nice composition.