You can read the story on the LA Times here. The gist is: they’re moving it online only, replacing it with TED talks.

TED talks started out being available on-line only, now are being repurposed for radio.

So — is on-line “hipper” than radio? Or is it a parking place for the old and tired radio shows we don’t want to deal with anymore?

Some people, it’s true, don’t have access to the Interwebs, or don’t know how to download a podcast. Or they’re in their car, stuck on the 101. Here’s what Harry said on the air (this from the Times article:)

“Drivers, the poor and the elderly — who wants that audience? Well, gee, I do!” Shearer said while opening his show Sunday, briefly explaining the dispute between him and KCRW.

“I love my online audience,” he said. “Yes, this is a podcast and an online stream and all of that. But it starts off proudly as a radio program.”

Proudly. I like that.

If you’re a public radio network with a successful radio show, one heard on over 400 stations nationally, why on earth would you cancel it?

NPR “executives” say it’s because

… public radio has a glut of vibrant call-in shows involving national issues — and that they sought a news magazine with a mix of interviews and prepared stories to bridge the hours between Morning Edition and All Things Considered.

Well, they had one. It was called Day to Day. And they cancelled it. In 2009. At that time it was about the money, but NPR swears its current seven million dollar deficit is not the reason Talk of the Nation is getting the axe.

I say there’s something else going on. We may never know what it is. But at least I get a chance to reminisce.

Talk of the Nation began as an outgrowth of a call-in special NPR aired around the time of the 1991 Gulf War. The success of that program caught the news suits by surprise. Apparently NPR was unaware of talk radio’s phenomenal growth, already well underway since the late 80s. Seems like public radio listeners, given the chance, would love to pick up the phone and call a toll-free number for a chance to bloviate about the issues of the day.

So NPR decided it needed its own brand of talk radio. Talk of the Nation began airing in November of 1991 with the person who’d booked the guests for the Gulf call-in as producer. Staff included Sean Collins (plucked from Morning Edition); Danielle Mattoon, a young intern from San Francisco (now Culture editor at the New York Times) … and the mercurial genius/bad boy of radio, John Hockenberry, as host.

But no one on the staff had ever actually produced a news show before! Let’s think about that for a second. NPR’s Special Events department had been in charge of the Gulf call-in. Talk of the Nation‘s nominal producer was, in fact, a booker — someone with a very large Rolodex who knew whom to call and how to get through to that person. Invaluable asset for talk radio. But not a producer.

Therefore it was decided to bring me on as a consultant, fresh from my six-month stint producing Morning Edition earlier that year. (Whether or not anyone can actually “produce” ME is a matter of some debate. But anyway.) At the time, Hockenberry and I lived a few blocks from each other on Manhattan’s Upper East Side so we became commuting buddies from New York to DC each week.

There was no little chaos in those early days. We threw a lot of spaghetti at the wall to see what would stick. My claim to a lasting contribution is that I insisted the show had to have some kind of a “clock” — a predictable format, and not be an hour-long free-for-all. I also wanted it to be about the news, and not become the John Hockenberry Show. Luckily, John agreed.

I could go on (and on) but my point is that Talk of the Nation got off to a shaky and uncertain start, like many NPR programs. Unlike some (still upset over Bryant Park Project), TOTN was allowed its growing pains and progressed, through four hosts plus Ira Flatow, to become the mature, trusted, call-in powerhouse it is.

Or was.

Anyone who worked on this program, please add your comments while I go away and sulk.

Some students got it — one read a naval adventure to Richard Rodgers’s Victory At Sea soundtrack — and others had no idea. Many just chose a random pop song of the day, which not only sounded incongruous but created the problem of music vocals fighting with the narrator’s voice.

This production technique is used from time to time in radio, especially by those of us who do arts reporting. Voice over music can work well or it can be annoying. It can seem natural or it can feel gratuitous. There’s not very often a middle ground.

Journalists producing video documentaries grapple with a similar set of issues when putting music to pictures. Is the music genuinely appropriate or is it being used to make a subtle editorial point? When is or isn’t that ethical?

It’s fun to experiment. Fortunately, computers make that easy for us to do. Take any random set of pictures — cats, for example — put them in some sort of order to create a slideshow and pick some music to play along.

Sometimes it works:

Cats 1 from Lars Hoel on Vimeo.

Some rights reserved by Kevin H.

Yesterday, New Yorkers woke up to news of the death of New York City Mayor Ed Koch. If, like me, they were listening to 1010 WINS, they heard newscaster Lee Harris repeatedly telling us that flags throughout the city were “flying at half-mast.”

No they weren’t, Lee … not unless those flags were displayed aboard ships in the harbor.

Boats have masts. Here on land we have flagstaffs. Flags in the city are flying at half-staff.

At the end of the film The Year of Living Dangerously, Mel Gibson’s journalist character Guy Hamilton is forced to flee Indonesia during the 1965 revolution. He’s almost stopped and arrested at the airport but manages to distract the officials with his fancy Nagra III recorder. As baffled soldiers clumsily unspool tape from the reels, Hamilton eases out the door and makes his escape.

And watching this, all I could think was, “What are you doing? Are you crazy? You’re leaving a Nagra behind!”

All this comes to mind because of the recent death of Stefan Kudelski, who in 1951 invented what many consider to be the best piece of audio recording technology ever.

His passing has been largely ignored by the mainstream press in this country, but believe me, we audio grayhairs are dragging out our old photos and sharing memories of what it was like to use, if not to own, this fabulous device.

About 30 years ago my colleague Robert Malesky and I took a borrowed Nagra to the Maryland seashore to record ocean waves for a New Age audio relaxation cassette.

[The box to the right (facing me) is one of those fancy digital converter things … I was trying out the new technology along with the old.]

For over half a century the Nagra was the ne plus ultra of field recorders for film, radio and television, and it’s still in use in many places around the world. Stefan Kudelski won plenty of awards, including four Oscars and two Emmys, so I don’t need to add my pitiful encomiums to the list. But I do want to step back a moment and consider what these fastidiously-engineered, aesthetically beautiful and technologically precise recorders can teach us about audio today.

Although Nagra did (and does) produce digital recorders its principal fame rests on its analog machines. And its product was all about refinement: Nagra strove to enable engineers to make the best possible portable analog recordings.

That’s analog in a nutshell: perfection. How to make the analogy between waves in the air and the storage medium as nearly, perfectly, identical as possible.

Digital, on the other hand, is all about approximation. We have an ingenious method of converting waves in the air and storing them as numbers, so we count ourselves successful as long as our approximation sounds good enough.

The pursuit of perfection is expensive: a new Nagra IV-S would set you back $1,700 in 1971 dollars. That’s a little over $9,600 today!

Digital is much cheaper because the cost of “good enough” goes down constantly as long as Moore’s Law lets us make solid-state technology that’s simultaneously more powerful and less expensive. So you have digital audio recorders that are “good enough” available for under a hundred bucks!

But I predict that, forty or fifty years down the road, there’ll be little of the nostalgia for today’s digital devices that we’re seeing for the Swiss masterpieces of Stefan Kudelski. I was lucky enough to have used one, to have touched the exquisitely chamfered metal, marveled at the action of the modulometer.

To quote another film, 1981′s Diva, ”C’est un Nagra. C’est suisse, et tres, tres precis.”

When you’re producing a work for hire, the vetting process is unavoidable. This job — a podcast for the New York Public Library’s Lunch Hour NYC exhibition — was no different.

With only a vague idea of how the final product would sound, I’d recorded the principal voices along with a walk-through of the exhibition itself. Then I’d edited and mixed the elements together in a way that made sense to me.

Rather than e-mail a large MP3 file to my contact at the Library for her feedback, I created a “Lunch Hour NYC” folder in my Dropbox account and parked the file there.

Her first reaction was that the podcast didn’t really have an introduction; it just started. Fair enough (I’d noted that myself). I whipped together a short intro identifying the podcast as a production of the New York Public Library, and Danielle Linzer as the writer and narrator. The Sound Devices 722 and a Neumann KMR-81 set up on my desk, I read my copy in a couple of takes. Popped the digital file into Pro Tools, matched its level with the rest of the mix, bounced the new podcast, uploaded to Dropbox.

Next it seemed the ending was a bit abrupt. Was there something I could put there — perhaps more of the ambience I’d recorded? No problemo. Grabbed some ambi, mixed it in, faded it out slowly, bounced the new podcast, uploaded to Dropbox.

Then things began to get sticky.

Digression: I once produced a thirteen part series of roundtable discussions on a variety of literary topics for a well-known publishing house. The half-hour radio format seemed to me to need a break in the middle. No matter how articulate the guests, no one, I reasoned, really wants to hear talking heads natter on for thirty minutes.

So I would grab a few minutes of sound that related to the show’s subject and insert it midway in the production. For a program on spy novels, for example, I used a clip from the 1965 Martin Ritt film, “The Spy Who Came In From The Cold,” with Richard Burton. Just something appropriate to add a little texture.

At first the publishers loved the idea and the execution. But then a change of staff came along and my new overlords went ballistic. “How could you do this without permission? You’re opening us up to copyright infringement lawsuits!”

Years of working in public radio spoiled me with regard to rights and permissions. On NPR you can use just about any old audio from anywhere with no worries. But my stint in audio publishing also made me aware that you had to be careful if something was going to be sold. So I’d asked the publisher if this show, destined for public radio, would ever be made available commercially. “Oh, no” I was told.

Now it seemed their attitude had changed. I was to take out all the offending interstitial material from all thirteen shows. I pleaded the Fair Use provision of the Copyright Act, even got a lawyer to draft an opinion letter. No good. So hours and hours of work had to be undone by hours and hours more.

End digression, but I’m guessing you see where this is going.

On my exhibition walk-through I’d heard Irving Berlin’s “Let’s Have Another Cup Of Coffee” playing from an unseen speaker. I added it to the ambience I was recording, along with some video clips of scenes from old movies and TV shows on location at the Automat.

The song was a perfect production element, as was a clip from “That Girl” with Marlo Thomas reading the Automat prices aloud as she looked for a cheap meal.

But the Library was worried. Once again I cited Fair Use. I also appealed to logic: these were very short excerpts being used for educational purposes by a not-for-profit organization. And they were recorded at the very exhibit where they were being played for the public at large!

Still worried, the Library asked me to cut the song and the Automat clip to five seconds … down from their original durations of about ten seconds. And also, would I create a second version with no music or audio from video, just ambience?

No problemo. Remix, reedit, rebounce, reupload.

With the song only in the clear for five seconds, there was no way to fade it under gracefully at the end of a musical phrase, and — sure enough — that was the next criticism. Was there a way to have it go down more gradually?

No problemo. Remix, reedit, rebounce, reupload.

Finally, approval. Time to send the invoice. Now another dilemma: do I bill for the time involved in creating (bouncing) and uploading each iteration of the podcast? It wasn’t time spent actively working and it would send the project significantly over budget, so I decided against. After all, I might want to work for these people again!

And then … just when I thought I was out … they pull me back in. Turns out I’d had a brain freeze from day one and had been unconsciously referring to the exhibition as “Lunchtime NYC” instead of “Lunch Hour NYC.” Not a big deal when it’s a filename on a computer, but more so when you misidentify the exhibition in the podcast intro!

Whoops! My bad. Rerecord. Remix, reedit, rebounce, reupload. For reals.

Feel free to vet the finished result yourself: Lunch Hour NYC

The job is a podcast for the New York Public Library’s Lunch Hour NYC exhibition. It’s designed for visually-impaired visitors to download for a description of the show, although anyone can listen and benefit. At this stage in the production I had a rough mix in Pro Tools, which looked something like this:

Now it was time to make it sound good. The procedure’s exactly the same as it is for mixing a radio piece. You start by level-matching all the elements so that no one’s voice sounds too loud or soft compared to the rest of the production.

I had only two voices to worry about, narrator and curator. The narration recording was nice and clean (if I do say so myself), although I did hear a very slight high-frequency buzzing in the background – probably one of the lights. (You can’t make a location recording sound like a studio; there will always be some stray sound.) I tried getting rid of the buzz with a notch filter, a frequency-specific EQ with little effect on adjacent frequencies. However once I notched the main buzz I heard its harmonic frequency, and that was was down in the upper range of the narrator’s voice. So I gave up on the buzz, which I suspected no one else would hear anyway.

But I did work hard to get rid of sibilance. One of the few drawbacks of the Neumann KMR-81 microphone is its relentlessly detailed high end; not kind to women’s voices, especially the “esses.” And the narration had quite a few of those.

Which is why the audio gods invented the “de-esser.” You could call it a dynamic equalizer – it does nothing until the offending frequency comes along, then it brings its level down in proportion to its loudness. Basically it squashes all the “esses.” You dial in the location and amount of squashing and the de-esser does the rest. Neat.

Some gentle compression over all of the narration helped even out its level – not too loud, not too soft. A second compressor was used on the curator’s track. Once these settings were in place, then (finally!) it was time to match levels.

Levels matched; check. Next, we mix in ambience and other sounds. I’d recorded some music at the exhibition (Irving Berlin’s “Let’s Have Another Cup Of Coffee”) and well as some movie and video clips, and I used these to buffer the curator’s segments from the body of the podcast. Fade in, fade out. Fade in, fade out. And so on.

Another hour in the studio and we bounce the whole thing to MP3 (sigh) and upload to Dropbox for vetting.

Next: Them changes

Load the audio into the computer, fire up Pro Tools and it’s — magic time.

The New York Public Library had hired me to produce a podcast for their Lunch Hour NYC exhibition. The idea was to describe the show in audio to anyone who might need visual assistance.

I came away from my day at the Library with three 44.1 KHz, 24-bit sound files:

• the podcast narration read by its author, Danielle Linzer
• an interview with the co-curator, Rebecca Federman
• and a walk-through of the exhibition itself.

Job one involved not magic so much; more like sit-on-your-butt, slog-through-it editing. Parts of Danielle’s script would have challenged a professional and, though she did a fine job, there were numerous retakes and pickups. To piece it all together into one unbroken fifteen-minute narrative took about two hours.

Why so long? When you’re editing the temptation is always to pick the last take as the keeper and discard the earlier ones. But that’s not always the best play – you have to listen. Sometimes earlier takes have more energy and, except for one little flub, are better overall. To make the narration sound its best I did a lot of comping and even some editing within words.

After sending the narration (in MP3 format) to the library for approval I turned my attention to the interview. Discarding the questions left about ten “answers” ranging in length from thirty seconds to two minutes. It made sense to pick the ones that related directly to topics in Danielle’s script and put them after (or before) those topics.

Before each interview clip I used one of those intros I’d had Danielle record (“Rebecca Federman, co-curator of Lunch Hour NYC”). But there was a problem.

The IDs didn’t match the script reading in energy or tone, so they sounded funny edited at the end of a script sentence, even separated by room tone or a long breath. Also Rebecca and Danielle sounded very similar. If you weren’t listening carefully you might not know who’s talking, ID or no ID.  The answer? Mix in a little ambience from the exhibition to “buffer” the IDs and the clips.

Another hour or so of editing and moving chunks of audio around and I was ready to create a rough mix.

I’d just finished recording a narration and an interview at the New York Public Library for a podcast I was producing in conjunction with their “Lunch Hour NYC” exhibition. The podcast would be used to describe the show for anyone who might need visual assistance. Before leaving the Stephen A. Schwartzman building (that’s the one with the lions, at 5th and 42nd), I wanted to walk through the exhibition itself while recording some ambience. Since there was no master plan I wasn’t sure how it would be used but, as I always tell students, you can’t use ambience if you don’t have it.

You can read the Times review of the exhibition here. It’s a highly interactive, multimedia affair with areas devoted to lunch carts, home cooking, the automat, school lunches, power lunches, lunch rooms and lunch counters, the history of lunch and much more.

To record the sounds of Lunch Hour NYC I used my faithful Sound Devices 722 digital recorder and Neumann KMR-81 microphone — the same kit that recorded the spoken parts. Wearing my headphones and holding the mic on a boom pole I must have looked odd but, this being New York, no one gave me a second glance.

The script had mentioned the many clocks scattered throughout the exhibition. Unfortunately, none was of the ticking variety … the only clock sound I could get was a neon buzz. I wandered into an alcove where an old newsreel showed Depression-era religious leader Father Divine beginning one of his worship services with a free lunch. I recorded the soundtrack through the low-fi headphones attached to the installation.

In another room I came across what’s arguably the crown jewel of Lunch Hour NYC — a restored section of an actual Horn and Hardart Automat. I captured the sounds of its doors being opened and closed, and a woman reading the signage aloud.

Photo by Donna K.

A little further on, a father and son were also reading out loud from a list of diner ordering lingo: “Adam and Eve on a raft,” “cup o’ Joe,” “whiskey down,” etc. The KMR-81 picked up their lively conversation with no trouble.

A woman pointed out her childhood lunch box to a younger man as they admired a colorful display of the iconic metal containers. I engaged her in conversation; again, not knowing if any of it would be useful.

In hindsight I should have asked all these people for their names and gotten their permission to use their voices. But this was not a hard news story and I was going through the exhibition for my own benefit as much as anything.

A snippet of Irving Berlin’s “Let’s Have Another Cup Of Coffee” played from a hidden speaker:

Just around the corner,
there’s a rainbow in the sky,
So let’s have another cup of coffee,
and let’s have another piece of pie.

I added it to the reel, now about fifteen minutes long. On my way out I passed a video installation showing a continuous loop of movie and television scenes shot in the Automat. Again I pressed the microphone to the earpiece and was reward by an incredibly low fidelity recording. A scene with Doris Day and Audrey Meadows from 1962′s That Touch Of Mink was wonderful but meaningless without the imagery.

Walking out of Lunch Hour NYC, I wondered, as always, if I’d recorded enough ambience, gotten enough sound. If the three or four people I was working with hadn’t been waiting patiently for me to finish I might have gone back in for another loop around. As it was I said my thanks for all their help … and exited through the gift shop.

Next: post-production

Not long ago the New York Public Library asked if I’d help them create a podcast for their “Lunch Hour NYC” exhibition. The idea was to use the podcast to describe the installation for those who might need visual assistance. The Library hadn’t done anything like this before so it would be up to me to provide the technical and production know-how.

The project began at the Library’s main branch on Fifth Avenue. Since I believe in capturing the highest-quality sound possible no matter how the audio will be used, I packed my full reporter’s kit — Sound Devices 722 digital recorder, Neumann KMR-81 short shotgun microphone with Neumann shock mount, Sony MDR-7506 headphones, LTM carbon fiber boom pole. I’d roll at a 44.1 KHz sample rate but at 24 bits for maximum dynamic range.

At my request for a quiet place to record the podcast narration, we set up in a green room attached to the South Court Auditorium. The room’s many chairs and wall shelves nicely broke up any echoes and standing waves. NYPL maintenance also turned off the HVAC for the duration of the recording — this on advice of the building’s A/V guru, Brian Hurley. I didn’t even have to ask! That only left me to douse the buzzy overhead fluorescent lights and use the room’s floor lamps instead.

(Never, ever record in an environment with background noise. You can’t – repeat, can not – get rid of it later. I wasn’t able to make the green room into a real studio (more on that in the post-production section) but with a little effort it sounded quite good.)

To write and narrate the podcast, the library had engaged Danielle Linzer from the Whitney Museum. She’s their Outreach Services Specialist, with experience describing events for audio. We’d also be joined by the exhibit’s co-curator, Rebecca Federman, and by Fotis Flevotomos, the library’s Fulbright Visiting Artist.

Danielle read the script directly from her laptop. Her narration was enthusiastic and more than adequate. As anyone who’s taken a class from me knows, I’m not a fan of descriptive clauses and long sentences in copy that’s meant to be read aloud. Danielle’s script had quite a few of these … as she discovered along the way. (We did several re-takes.) The script also lacked an opening or title. It did have a serviceable close.

Sitting a short distance from Danielle, I held the Neumann about six to eight inches away and at a perpendicular angle to the room’s only door. That would keep any stray hallway sounds completely off-mic.

Rebecca joined us later. The idea was to have Danielle interview her and use the responses – somehow – in the podcast. I’d brought a second Neumann but Danielle didn’t want to be mic’ed. After asking permission I threw in a few questions of my own, as did Fotis.

Last task in the green room was to ask Danielle to record a few wild Rebecca IDs. It occurred to me that their voices sounded remarkably similar, so it would make good production sense to have Danielle say something like “Co-curator, Rebecca Federman” before each interview clip.

Next stop, the exhibition itself.