The 2 Questions from Yorke/Godrich/Spotify

Thom Yorke and Nigel Godrich's decision to remove some of their music from Spotify hinges on 2 questions, both of which have been asked repeatedly since streaming services matured. One question is tactical and specific, the other more philosophical:

1. How much is a single stream play worth?
2. Are artists entitled to a living?

How Much Is A Single Stream Play Worth?

Busker image courtesy of Wikipedia

I have written and spoken about this before, to much controversy. I have yet to see anyone arguing against Spotify, Rhapsody, MOG, Music Unlimited, RDIO, etc. provide their answers to this question.

This is a critical point to resolve. If we're going to claim that (new) artists aren't getting paid enough for plays, we need to know what "enough" is.

A few years ago, I gave a talk where I looked at what similar industries were charging "per play":

• Jukebox in a bar: $0.25-$0.50 per play. Yeah, that's a lot...but how much of this actually goes to the artist?
• CDs: $0.10 per play. The average CD is played fewer than 10 times after purchase. Assume $10 price for 10 tracks, each played 10 times. That's $1 per track, divided by 10 plays. 10 cents. How much of that goes to the artist? Depends on the size of the band and the organization, but maybe 10% overall.
• Downloads: $0.10 per play. Assumes similar usage for purchased CDs, but the artist gets less money because Apple takes 30% of the retail price and pockets it, and because the labels typically pay artists less for downloads than CDs. 
• Lala.com: $0.10 per stream, infinite plays. Lala failed because they, in their words, could not make this model a viable business. Don't know how much went to the artist, but you can assume similar or worse economics compared to CDs and downloads.
• SiriusXM: $0.002 per stream play (on the internet). [Updated]. I am using SiriusXM's internet streaming rates, based on the NAB rates, not its satellite broadcast payment rate (yes, the two are different).
  
  SiriusXM's satellite rates went up in January 2013 as part of a new CRB ruling, and the company currently pays 12.5% of their retail subscription price to copyright holders (either $1.81 or $1.25). (I could write about how logically weird it is that the rate increases .5% each year, implying somehow that the value of music is rising).
  
  The only data I can find is a little bit old, but it suggests that satellite radio listeners play 44 hours a month. Assuming that 25% of that time is ads and drops (a generous estimate), that means 33 hours of music, or 495 4-minute songs. Divided by $1.81, that yields $0.0037 and divided by $1.25, that yields $0.0025 per play for satellite plays, which is close enough to the internet rates and doesn't change the argument much.
  
  SiriusXM is a relatively large company delivering a fairly lame product. Very limited choice for users. It's taken them a decade to get to scale, including hundreds of millions in advertising and merging the only two companies in the space together. They're barely profitable after all that, even with extremely low churn and now, no competition.
• Generic Internet Radio (USA): $0.0012-$0.0022 per stream play. Depends on a bunch of things, like whether or not you're rebroadcasting FM radio or other programming, what your revenues are, and how big you are.
• Pandora: $0.0012 per stream play. That's a tiny bit more than a tenth of a cent. Pandora is the undisputed king of internet radio, and they have not been able to make a profitable business of it, even at massive scale. 
• FM radio (in the USA): $0.00 per play. You might get something if you're the songwriter, but if you are just the performing artist, you get nothing. FM radio remains a relatively strong and viable business, even if the variety of music offers continues to shrink along with their target demographic.

This comparison deliberately ignores how much more difficult it is to get people to do things higher up on the list (buying anything) than items lower on the list (listening to something for free).

It also ignores the huge difference in scale between items. We're just focusing on the gross possible payment to the artist or cost to the user.

The intentionally inflammatory conclusion I drew? The optimal price per stream, in order to have a successful business, needs to be somewhere between $0.00 and $0.002, because anything higher all but guarantees your business will fail.

Even the high end of that admittedly meager scale means you, as a business, will barely be hanging on. It's possible all of the market leaders in their respective industries - Pandora, SiriusXM, and so on - are total idiots and are running their business badly. But it seems more likely these rates are just punishingly high . Even mighty Spotify is not really making any money.

But who cares about business! Fuck those guys! What about artists? What's a fair rate for artists to charge per play?

Well, you clearly can't charge more than what CDs offer - $0.10 per play - or users will just buy CDs and the question is moot. And we're talking about a single stream play, not ownership or "infinite" plays (which is what ownership gets you), and thus you can't really charge more than what ownership costs.

Assuming you stick the rate you need at $0.10 (using my $10 CD with 10 tracks gets played 10 times math), and assuming you can find companies that will pay you those rates to stream your content, you need to generate 11,600 plays per month (139,200 per year) to earn minimum wage of $1,160 per month, or $13,920 per year.

Is that a lot of streams? I guess it depends on how you figure it. If you assume someone listens to your song every week, you need nearly 3,000 fans reliably playing your song once a week all year to make minimum wage. Then again, if you're Daft Punk, you blow past those numbers no problem.

Based on conventional wisdom and misleading graphs, you'd probably decide that artists should sell CDs, and perhaps everything else should be outlawed.

But you'd be condemning most artists to starvation.

Here's why:
According to Soundscan, in 2009 there were 98,000 CDs released (that means physical CDs that sold at least one copy registered by SoundScan).

Of those 98,000 CDs, only 2.1% sold more than 5,000 copies for the entire year.
Of those 98,000 CDs, only 1% sold more than 10,000 copies for the entire year.

Assuming you are a solo performer on a label, you need to sell 1,161 CDs per month - 13,932 CDs per year - to earn minimum wage.

Which means that less than 1% of artists on labels selling CDs earn minimum wage.

(If you go totally indie and do everything yourself, it's much better. You can get by selling 1,716 CDs per year. But remember, you are now doing everything yourself - running the label, collecting payments, shipping, marketing, etc.)

Being an artist is a tough job. Which leads us to the next question...

Are Artists Entitled To A Living?

Daft Punk: Professional Artists

One problem with having this discussion: Some people will pillory you for even asking the question. "OF COURSE!", they shout. "Artists are special and magical people who speak what cannot be spoken and address the human condition and enrich life!" And so on.

To answer this question, you need a clear view of art and business. The way I see it, if you are truly an artist, you create because you want to or have to. Financial remuneration for said creation is a secondary consideration.

On the other hand, if you are doing it for the money - if "the money" is your primary concern - you are an entertainer or businessperson. And you're going to focus on giving your broadest possible audience whatever they will buy the most of. You will think about this as you craft your, uh, craft, and you will budget a ton of money for marketing (again, see Daft Punk)

If you're not targeting either of those edges, you're waffling in the middle, and risk being someone who delivers compromised work to a non-optimal audience.

People should get paid for their work. But is anyone entitled to a living just doing what they want to do? And what's "a living"? How much income is anyone guaranteed, regardless of what they do?

I think of this: If you started a restaurant and no matter how hard you worked, it failed, nobody would say "wow, we should revamp how restaurants work so you can get paid enough to live."

Instead, they'd say things like "maybe you should have served different or better food." Or been in a different location with better traffic and/or less competition. Or had better service. Or decor. Or done something else.

Businesses fail all the time. The reaction is generally a shrug and "that's the market". Why should professional artists be treated differently?

There have been plenty of starving painters, writers, actors, and other "creatives" throughout history as well. Why do (pop/rock) musicians merit special treatment?

Who gets to decide who qualifies as a professional supported artist? (In the United States, we don't even really have much government support for the arts, even for undisputed titans of creativity).

Let's say society did decide to support artists with some kind of stipend/minimum guarantee because as a society, we value whatever artistic contribution these people are making. How much do we pay them? Minimum wage? 2 x poverty level? $100,000?

The economist then asks "well, why should the category of artist be special? What about other jobs?You'll just get the entire world claiming they're artists, and asking for their payment."

Historically, what you see is effectively a kind of means testing or "quality gate" for public or private patronage.

Despite all of the doom and gloom coming from people like Thom Yorke and the record labels, the fact remains that more music is being created, recorded, and released now than any other time in human history. Some of this is because there are more people, but much of it has to do with the tools for creation (computers, cheap instruments) and distribution (the internet and various services).

This does mean we are drowning in music, most of it not very good. Personally, I would rather have music be democratized, in all its messy, not very good, mostly uninteresting glory than have it be limited to a precious few ordained as "real" musicians by a bureaucrat or Pitchfork or Reddit or whoever your proposed authority happens to be.

Music used to be something nearly everyone did, and most people didn't expect to be able to make a living at it. In that respect, it is no different than lots of other things people do...like writing blogs, for example.

I'm all for debate around the business of music. But it needs to be real debate, around facts and proposals and issues. I will keep pushing forward, trying to build businesses that are directly addressing these issues. I hope the rest of the industry, including thought leaders like Thom Yorke and Nigel Godrich, will answer the hard questions with some proposals of their own.

Afterword: Some Notes and Clarifications

• The fact that the average CD is played fewer than 10 times after purchase comes from NARM.
• The sales figures for 2009 come from NARM and Soundscan.
• 2009 was not an unusual year in terms of sales and albums - that distribution is typical and has been for quite some time. Even back in the late 90s, the average album by a band on a major label with a national push sold about 1,000 copies.
• All of the other numbers are publicly available data and close enough for argument's sake.
• For SiriusXM, it's worth noting their 12.5% of consumer price royalty is low, compared to the 60-80% that internet services pay. And let's mention they don't have to pay royalties on a whole bunch of content that internet services still do (read the above links). And SiriusXM has no idea how many songs users are listening to.
  
  The "percentage of revenue" model basically says it doesn't matter. If users listen to zero songs, somebody gets paid. Does any of it go to the artists?
  
  If users listen to 10,000 songs in a month, deriving maximum possible enjoyment from the service, artists aren't going to get very much.
  
  "Percentage of revenue" is a perverse system that effectively says "the more listeners like music, the less we pay the artists per play."

44

I raise my mask and stare at the light leaking around the blackout curtain. My phone says it's 6 AM. I roll out of bed and fumble for my workout clothes.

44 today.

I'm not even sure if it is my birthday yet at home. What is home, at 44? I think about how much of my past is gone, erased, lost. Names and faces changed. The houses of my childhood long sold. "Where are they now?"

California. Los Angeles, a lifetime ago.

"Where are we now?
The moment you know, you know
As long as there's sun..."

San Francisco. Home for 13 years, and yet sometimes I still feel like I just got there. I think about articles I've read claiming people like me are ruining the place.

Where am I now?

London. Got here yesterday. I was in Tokyo 2 or 3 days ago.

I am alone in the hotel gym. I step on the treadmill gingerly. Glutes still hurting from Saturday's training session. Hip sore from bursitis and yesterday's workout. I punch in numbers and run slowly, carefully, more focused on my gait than anything else. I make it 25 minutes before pain makes me switch to the elliptical. I can't get my heart rate up as high, but at least it won't aggravate things too much.

My MP3 player blasts a song by Last Amanda. I did a show with them several years ago at Altamont:

"I've got so much I want to tell you/I don't think so
Share my experience with others/I don't think so
...I can't stand myself"

Life has changed so much, so quickly. Just 4 years ago, I had a huge party. You were probably invited, and maybe even showed up. I sang and played guitar. I was working at Rhapsody. I felt like I was on top of the world.

My physical self has been on my mind a lot lately. My voice has been broken for just over 2 months.It seems to be recovering...slowly. I spend far too much time worrying about it, but literally every day I am reminded of the loss, and how much I loved singing, and feeling like I could sing anything.

Largely driven by doctors' orders and by vanity, I have been exercising and watching my diet for the last few months. I am 15-20 pounds lighter than I was in February. In better shape now, in terms of performance and appearance, than I have been since probably my mid-20s. But in recent weeks my right hip has started complaining about the 20 miles of running I've added, and I am not sure how careful I'll have to be or how long that will take to recuperate. I haven't missed too many workouts, and am watching what I eat, but I am still worried about backsliding.

At least my back and leg have substantially improved over the last few years. I still have problems, but they're much less common and more manageable. Getting old. It's all wear and tear. Damage control.

That right hip is stiff and sore as I walk through the unseasonably warm London morning (LONDON! How did I get here?) to work. I listen to music, in glorious high quality on the best headphones I've ever owned.

One of my own songs comes up, from the yet-to-be-released album "The Ghost Town", the last album I've finished to date. It sounds really good. I like my singing on this one.

"When I crossed the desert, there was nothing that I lacked. 
I left in the dead of night, and I never once looked back."

I smile to myself. I really did cross the Mojave desert at night, leaving childhood and my old life behind, and I've never regretted it once.

I have so much to be grateful for, so much to be happy about. Sometimes it is hard for me to see that clearly. Perhaps it's just my nature to focus on the things I wish were better - I think that is part of what makes me good at my job.

I think about my other birthdays. I have made the same wish for decades. I won't tell you what it is, because then it won't come true. But it has come true, time to time, more often than perhaps I deserve.

I already bought myself a birthday present. A fancy new synthesizer. I wish I had more time to explore it, and more time to create with it. I don't really need more stuff, but there are worse things to spend money on than instruments.

Things are good, even great. My biggest challenges all seem to be self-inflicted to some degree. If I could just ease up a bit. If only. I keep trying.

I'll spend most of today sitting behind mirrored glass in a dark room, watching people be questioned about their activities. "Like FBI?", the Eastern European hotel clerk asked. Yes, exactly like FBI, I told him. Maybe I'll have a nice dinner with a colleague tonight.

Business travel is lonely. I don't really mind being alone, even in unhealthy doses, as long as I know there's an end in sight. I flew back home between international trips to see Iran and my friends. A touch of fatigue and loneliness is worth it for the adventure. Totally.

I'll be home again soon enough, with time to noodle around on my instruments and sleep and think and write. I have so many projects in the works (including an unusual one I am very excited about), and so many people to see.

I think of my friends, my family, my wife, and how fortunate I am.

Happy birthday to me.

Thom Yorke Is Wrong, and 3 free solutions

3 Free Solutions

I'll start off with some answers...

For my colleagues in the digital music business, here are 3 totally free product solutions to the problems Thom Yorke and Nigel Godrich are complaining about. I may implement some of these myself. Each of these ideas does have its own challenges and problems, of course.

1. Editorial Programming
2. Tip Jar
3. Make it easy to add content

Editorial Programming

The simplest approach to pushing more new music is to promote it. Have the music service editorial staff take a proactive role in seeking out new and cool stuff, and feature it (on the home page, or playlists, or radio channels). Even just badge new stuff.

Today's reality, however, is the services are generally pushed to feature what the labels want them to feature, and are discouraged from picking favorites.

Additionally, despite what Pitchfork et. al. say, most of the "tastemakers" today write about the same few dozen albums every year. For better or worse, there's a lot of overlap between those few and what the industry is pushing. I suppose there's a blog post to be written about it.

But the easiest way to help out new artists is to promote new music aggressively. And it doesn't really cost a thing!

Tip Jar

This is not a new (or even particularly good) idea. But services could include a button to "donate" more money directly to artists they like.

One problem is "who is the artist?" In a few small cases, it's very clear who the artist is. But what if they're dead? What if they can't be found? What if the band has broken up, or is arguing with each other about money?

Where do you send the money? Even assuming you can resolve that issue, generally services only have contact with the label and publisher, not the "artist". Building that database would be difficult and contentious.

The bigger problem is just that most listeners won't even bother donating. They'll say "well, I am already paying for the service, so..." Most of these fans will go to the artist's site and buy stuff there.

(You could also include URLs to the artist's web site, but many of the above problems still apply).

Make it easy to add content

Each service could/should have prominent links "for artists" that give them instructions for how to get their music in the service fastest and cheapest. Consider adding support for direct deals (these days, most services require use of TuneCore or a similar aggregator).

Some stuff that doesn't help:

• Live shows. Not all bands play live, and those that do have finite capacity for it, and fairly large overhead. This will not solve the problem at all.
• Buying downloads instead. Same (or worse) contractual and accounting problems. Artists don't get that much cash from these, either. Plus you're supporting one of two giant companies that combined represent almost 90% of the music retail business. Apple makes more money on downloads than the artist does, by a huge, huge margin. How does that help? 
• Raising the price of services. Most people think music services are already too expensive, and cost is the #1 cited reason for users cancelling. 

Thom Yorke Is Wrong (and so is Nigel Godrich)

Once again, it is time to clear up some misconceptions about the streaming music business.

Super-cool Thom Yorke of Radiohead and Atoms For Peace took to Twitter, (as did his friend and music partner Nigel Godrich), broadcasting that he is pulling his solo album and his new band's albums from Spotify. 

It's not entirely clear why. Some of their tweets:

...It's bad for new music. 

The reason is that new artists get paid fuck all with this model.. It's an equation that just doesn't work 

The numbers don’t even add up for spotify yet.. But it’s not about that.. It’s about establishing the model which will be extremely valuable.  

Meanwhile small labels and new artists can't even keep their lights on. It's just not right.
Millions of streams gets them a few thousand dollars.. Not like radio at all.. 

Plus people are scared to speak up or not take part as they are told they will lose invaluable exposure if they don't play ball. 

If people had been listening to spotify instead of buying records in 1973... I doubt very much if [Pink Floyd's Dark Side of the Moon] would have been made.. It would just be too expensive.

Ok. Let's address these things.

First, it's hard to see exactly why music services (Spotify, Rhapsody, Google Play Music All Access, RDIO, MOG, Music Unlimited, etc.) are "bad for new music". Music services offer access to everyone, and most of them will automatically include "unknown" artists in recommendations and radio channel experiences. If you know what you're looking for, they make it easy to find.

As for payments, in general, all content is treated the same by these services. I've written extensively here and elsewhere about how those payments actually work. I will reiterate some key points:
Services pay the content owners: the labels and the publishers. For small artists, frequently both of these things are the artist themselves. For "new" artists on labels, how much the artist gets paid has little to do with the deal the service has with the content owner, and everything to do with the deal the artist has with the label and publisher.

This album, by a "new" artist, sold 2.2 million copies when
released in 2012. At least one new artist is doing fine. 

"New" artists having bad or unfavorable deals has been the status quo since the beginning of the record industry. Music services have actually provided an alternative - direct deals - which did not exist before.

It is entirely possible that labels (and publishers) are paying out money to the artists in ways that favor them rather than the artist. That's what they do. As an example, most music services have to pay a minimum to labels and publishers, even if users play nothing in a given month. Do the labels divide that up among all of the artists on their roster and give them a cut? Or do they pocket the whole thing. I'm pretty sure I know the answer, and I bet you can guess what it is yourself.

My takeaway? The point Godrich and Yorke are making here is simply invalid.

If they're complaining that new artists don't generate enough plays, well, that's possibly true for some of them, but not true for all of them. Same as any new artist dealing with sales: most new artists have no sales. In fact, most albums have no sales, period. This isn't news. And any decrease in sales has more to do with the sheer volume of music released.

And this is really what most of the artists involved in this discussion are really bothered about. The old way, including radio and sales of goods, was a "winner take all" model. If the average annual per capita expenditure on music was $35, a user who bought 2 CDs would be giving all of that money to 2 artists.

But in a subscription world, that vastly higher sum - ($10 x 12 months, or $120, less the 10-20% services actually get to keep) is divided amongst all the possible things the user can play. This results in many more artists making a much smaller sum. I'd argue it is more fair. It is just less remunerative for the artist.

This album had one good song. The rest was really bad.
Even the "good" song wasn't that good. 

Looked at from a different angle, prior to music services, users would frequently buy an album because they liked one song on it. And then they'd find they hated the rest of the album. The artist captured the full album amount "unfairly", and the user had no choice in the matter.

Is it hard to "keep the lights on" for small labels and artists? Yes, it is. Again, that's been true for the entire life of the business. And music services offer outlets that are far easier and more transparent than previous physical goods distribution systems.

It's not the same as radio. At all. And that's a good thing. New artists and new music have always been totally screwed by radio, except for the tiniest slice of new stuff designated as the next big thing by the machine. And how you get designated "next big thing" involves a lot of stuff that has nothing to do with how good your music is, and everything to do with how you play the game.

Streaming services put everyone on an equal footing. Yes, it's hard for new bands to rise above the noise and the competition. But at least they have a chance, and people can play the music when they want, and not wait for some DJ to make your music their own personal cause.

Yorke and Godrich seem to imply the services are strong-arming artists into participating. That is straight-up garbage. It is absolutely true that artists lose exposure if they don't participate: if people can't find your music because you've held it back, they can't very well play it, can they? It's the same as withholding your content from record stores because you don't like how little they pay you. You don't have to put your music in their store, but it's not fair or accurate to imply the store is pressuring the artist.

I am particularly irritated about these types of accusations because I have worked on several of these services, and without exception, the entire staff have been musicians and music-lovers dedicated to supporting artists and other music-lovers. I take it personally when people impugn their motivations and intents. We love music, and we love musicians.

If we wanted to make money, we'd have either not gone into music in the first place, or at the very least started some kind of dodgy safe-harbor-using-user-generated-content company that "complied with law" (wink, nudge) instead of burning up millions of dollars in cash advances and contracts.

Finally, there is the assumption that a great work like "Dark Side of the Moon" wouldn't have been made because it would have been too expensive.

*sigh*.

Factually, this is problematic. "Dark Side" wasn't a particularly expensive album to make. Yeah, they used Abbey Road, but it was affordable. And it didn't take that long. In many respects, while it was expected to be a "hit", it was not a "big bet" by anyone involved. It was just the next album by a successful band. Nobody knew or expected it to be a huge hit.

When you look at the albums that really were crazy expensive, or had outsized expectations, the resulting product has seldom lived up to those expectations, and most of the money was wasted or spent on marketing.

And nothing is stopping record companies from making similar bets today. Typically, when an album is "expensive", it's because the label is spending a ton on marketing, or paying talented "beatmakers".

But let's set that aside for a more important point: Are you an artist, or are you a businessperson?

I'm pretty sure Pink Floyd's motivation for creating that album wasn't "hey, this will make us a ton of cash". They wanted to make a record. They had been playing most of the songs on tour for months, so they already knew the material, and knew it would go over well for their target audience.

The album had no "hit single", and Pink Floyd didn't really care about such things. They were making art, meaning they were more concerned with "doing what they wanted" than "how it would sell".

If you are businessperson, the reverse is true. You have to care about "what will sell" more than "what you want". And you would do things like minimize your upfront (i.e. recording) costs and not take many risks.

I assume that Godrich and Yorke are basically saying "because there's so little cash coming in, nobody will invest in making expensive albums". Well, not true. People are still spending lots of money on albums.

But spending money has never correlated with success, and in today's world, you don't have to spend a lot of money to make a good record, much less a successful record.

Look at movies or painting or games or any other art form. If you are depending on someone else to give you massive amounts of cash up front on the hopes of massive returns, well, I think you're doing it wrong.

But more importantly, if all of the music fans that had been buying music had been subscribing to Spotify or other music services, the music business would have been drowning in money.

The main problem today, as I have stated several times, is that not enough people care about music to buy or subscribe. Until that changes, whether it is a music service or download store, the industry is in trouble. Where is the fault? With the artists? The labels? The shop owners? The fans? Maybe all of the above.

But the secondary problem is perhaps more intractable - how do you get fans to listen to more new music, and not just the artists and albums they already know?

That's a tough one, and if Godrich and Yorke can fit any ideas into 140 characters or less, I'm all ears. In the interim, I will continue to do what I have always done, and push for a better music business today.

Almost 44

It's hot in Tokyo. The jetway, the airport, the train platform are filled with warm air. I feel the sweat running down my chest behind my t-shirt as I drop into my seat on the Narita Express. Despite all the travel, I'm not crushingly tired yet.

I've been to Tokyo before in July. 23 years ago. I turned 21 here, sweating, alone, with my wisdom teeth tearing up the insides of my cheeks. I bought a beer and some junk food from the 7-11 with some of my precious funds to celebrate.

It seems like a lifetime ago. Then again, I suppose it was. More time elapsed since than the 21 years I was celebrating.

The announcer bot says we're leaving soon and the trains will separate in Tokyo. The empty train car adds people. The air conditioning whispers.

I snap a photo of myself reflected in the window.

The distant and slightly blurred view looks great, I think. A ghost or a memory. A reflection of another train or another time.

As we start to pull away, I run through the list of things on my mind these days. There's so little I can actually do anything about. So many of my concerns seem to have a solution of "take it easy, don't do anything." Oddly enough, that inaction somehow seems much more difficult.

I wipe my brow with my wristband and listen to some music, using the new offline feature in the Music Unlimited iOS app. It mostly works.

When I was 21, I had a Sony Walkman with me for music here. I remember listening to a lot of albums I didn't like at the time, trying to figure out why I didn't like them and if there was anything good about them. Danzig. Gene Loves Jezebel. Those stick out as the worst. I can't remember what else I had, other than a cassette of mixes for my first "real" solo album, Zero Summer Game. I'd lose that cassette in a club with some teenage models. Worth it.

The sun sinks quickly, and by the time I'm at Shinagawa eki, it's nearly dark. I don't even remember how I got from the airport to the city on my trip here 23 years ago. I don't even remember which airport. But like this time, there was nobody there to meet me. Just me, a guitar, a suitcase. Finding my way. Ready for adventure. Sweating.

The train station feels like the inside of someone's mouth, hot and moist. I roll through the rush hour crowds to the taxi stand. I use my limited Japanese language skills to ask the driver to take me to my hotel of choice. I have lost count of how many times I've been here in the last 18 months, but I've stayed at enough of the local hotels to know which one I like best. We glide off, and I try to relax, rolling down the window in the back seat as we roll through Shinagawa.

When I was here in the summer of 1990, I stayed briefly at the Keio Plaza hotel in Shinjuku, basically long enough to sleep off the trip before a brief stay with a friend's family and then taking over someone's apartment. I can't remember much about the hotel other than being exhausted.

At the hotel, I check in and dump my bags. It's nearly 7 pm Tokyo time. I've been awake for something like 22 hours so far. I change into my gym clothes and work out. My hip is complaining, so I stop running after a few minutes and switch to the stationary bike.

I get cleaned up, have some food, and drop into a deep and sudden sleep.

The next few days are a blur of meetings. It's a better trip than most, despite the infernal weather. The building is cooler inside than last year. It's bearable, pleasant. Sad to say farewell to a departing colleague. Some fun dinners, or at least fun until the hammer of jet lag comes down. At the hotel, I can barely get out of my clothes before collapsing from fatigue. There is absolutely no free time during the entire trip.

And almost before I know it, it's Friday afternoon. I'm back on the Narita Express, heading for the airport, the finish line, lifting up into the skies, memories of this trip, other recent travels, and my own ancient history swirling in my head.