On the top left is a photo taken with my trusty old (also much used and abused) Canon 5D Mark III. On the top right is one taken by a borrowed new Sony a7Riii. Below both are cropped close-ups of detail. The scene is in a room illuminated by incandescent track lighting. It is not an art shot, though it does contain photo art by our good friend Marian Crostic, whose Sony a7R she is kindly remanding to my custody tomorrow. (Her main camera is now an a7Riii like the borrowed one I used here.)

Both photos were shot with Canon and Sony’s best 24-105 f4 zoom lenses, at the 105mm end. Both were also set to automatic, meaning the camera chooses all the settings. In both cases the camera chose ISO 3200 at f4. The only difference was shutter speed: 1/125 sec on the Canon and 1/160 sec on the Sony. While 3200 is not the prettiest ISO, I wanted to compare both cameras indoors under less than ideal lighting, because that’s typical of situations where I shoot a lot of people.

One difference between these cameras is the pixel density of the sensor: the Canon’s shot is 5760 x 3840 pixels, while the Sony’s is 7952 x 5304. While that difference accounts for some of the higher detail in the Sony’s shot, it’s clear to me that the Sony lens is simply sharper, as Ken Rockwell kinda promised in this glowing review. (Also, to be fair, the Canon lens has had a lot of use.)

All the images above are screen shots of RAW versions of the photos (.CR2 for the Canon and .ARW for the Sony). Though I don’t have the time or patience to show differences in the .JPG versions of these photos, it’s clear to me that the Canon’s JPGs look less artifacted by compression. The obvious artifacts in the Sony shots have me thinking I may only shoot RAW with the a7R, though I’ll need to test it out first.

The main difference overall, at least in this setting, is in the warmth of the color. There the Canon has a huge advantage. I could say it’s also because the Sony is slightly less exposed (by the higher shutter speed); but I noticed the same difference in test shots I took outdoors as well, under both overcast and sunlit skies, and at ISO 100. The Canon seems warmer, though the Sony has far more detail one can pull out of shadows.

I should add that neither camera got the color of the wall (a creamy white) right in these photos, with the Canon leaning hot and the Sony leaning cool.

Anyway, I just thought I’d share that much before I pick up the a7R, and start using it to shoot stuff in New York, where I’m headed Wednesday night after more than a year away.



My post yesterday saw action on Techmeme (as I write this, it’s at #2) and on Twitter (from Don Marti, Augustine Fou, et. al.), and in thoughtful blog posts by John Gruber in Daring Fireball and Nick Heer in Pixel Envy. All pushed back on at least some of what I said. Here are some excerpts, with my responses. First, John:

Doc Searls:

Here’s what’s misleading about this message: Felix would have had none of those trackers following him if he had gone into Settings → Privacy → Tracking, and pushed the switch to off […].

Key fact: it is defaulted to on. Meaning Apple is not fully serious about privacy. If Apple was fully serious, your iPhone would be set to not allow tracking in the first place. All those trackers would come pre-vaporized.

For all the criticism Apple has faced from the ad tech industry over this feature, it’s fun to see criticism that Apple isn’t going far enough. But I don’t think Searls’s critique here is fair. Permission to allow tracking is not on by default — what is on by default is permission for the app to ask. Searls makes that clear, I know, but it feels like he’s arguing as though apps can track you by default, and they can’t.

But I don’t think Searls’s critique here is fair. Permission to allow tracking is not on by default — what is on by default is permission for the app to ask. Searls makes that clear, I know, but it feels like he’s arguing as though apps can track you by default, and they can’t.

I’m not arguing that. But let’s dig down a bit on all this.

What Apple has here is a system for asking in both directions (apps asking to track, and users asking apps not to track). I think this is weird and unclear, while simply disallowing tracking globally would be clear. So would a setting that simply turns off all apps’ ability to track. But that’s not what we have.

Or maybe we do.

To review… in Settings—>Privacy—>Tracking, is a single OFF/ON switch for “Allow Ads to Request to Track.” It is by default set to ON. (I called AppleCare to be sure about this. The guy I spoke to said yes, it is.) Below that setting is a bit of explanatory text with a “Learn more” link that goes to this long column of text one swipes down four times (at least on my phone) to read:

Okay, now look in the fifth paragraph (three up from where you’re reading now). There it says that by turning the setting to OFF, “all apps…will be blocked from accessing the device’s Advertising Identifier.” Maybe I’m reading this wrong, but it seems plain to me that this will at least pre-vaporize trackers vectored on the device identifier (technically called IDFA: ID For Advertisers).

After explaining why he thinks the default setting to ON is the better choice, and why he likes it that way (e.g. he can see what apps want to track, surprisingly few do, and he knows which they are), John says this about the IDFA:

IDFA was well-intentioned, but I think in hindsight Apple realizes it was naive to think the surveillance ad industry could be trusted with anything.

And why “ask” an app not to track? Why not “tell”? Or, better yet, “Prevent Tracking By This App”? Does asking an app not to track mean it won’t?

This is Apple being honest. Apple can block apps from accessing the IDFA identifier, but there’s nothing Apple can do to guarantee that apps won’t come up with their own device fingerprinting schemes to track users behind their backs. Using “Don’t Allow Tracking” or some such label instead of “Ask App Not to Track” would create the false impression that Apple can block any and all forms of tracking. It’s like a restaurant with a no smoking policy. That doesn’t mean you won’t go into the restroom and find a patron sneaking a smoke. I think if Apple catches applications circumventing “Ask App Not to Track” with custom schemes, they’ll take punitive action, just like a restaurant might ask a patron to leave if they catch them smoking in the restroom — but they can’t guarantee it won’t happen. (Joanna Stern asked Craig Federighi about this in their interview a few weeks ago, and Federighi answered honestly.)

If Apple could give you a button that guaranteed an app couldn’t track you, they would, and they’d label it appropriately. But they can’t so they don’t, and they won’t exaggerate what they can do.

On Twitter Don Marti writes,

Unfortunately it probably has to be “ask app not to track” because some apps will figure out ways around the policy (like all mobile app store policies). Probably better not to give people a false sense of security if they are suspicious of an app

—and then points to P&G Worked With China Trade Group on Tech to Sidestep Apple Privacy Rules, subtitled “One of world’s largest ad buyers spent years building marketing machine reliant on digital user data, putting it at odds with iPhone maker’s privacy moves” in The Wall Street Journal. In it is this:

P&G marketing chief Marc Pritchard has advocated for a universal way to track users across platforms, including those run by Facebook and Alphabet Inc.’s Google, that protects privacy while also giving marketers information to better hone their messages.

Frustrated with what it saw as tech companies’ lack of transparency, P&G began building its own consumer database several years ago, seeking to generate detailed intelligence on consumer behavior without relying on data gathered by Facebook, Google and other platforms. The information is a combination of anonymous consumer IDs culled from devices and personal information that customers share willingly. The company said in 2019 that it had amassed 1.5 billion consumer identifications world-wide.

China, where Facebook and Google have a limited presence, is P&G’s most sophisticated market for using that database. The company funnels 80% of its digital-ad buying there through “programmatic ads” that let it target people with the highest propensity to buy without presenting them with irrelevant or excessive ads, P&G Chief Executive Officer David Taylor said at a conference last year.

“We are reinventing brand building, from wasteful mass marketing to mass one-to-one brand building fueled by data and technology,” he said. “This is driving growth while delivering savings and efficiencies.”

In response to that, I tweeted,

Won’t app makers find ways to work around the no tracking ask, regardless of whether it’s a global or a one-at-a-time setting? That seems to be what the
@WSJ is saying about  @ProcterGamble ‘s work with #CAID device fingerprinting.

Don replied,

Yes. Some app developers will figure out a way to track you that doesn’t get caught by the App Store review. Apple can’t promise a complete “stop this app from tracking me” feature because sometimes it will be one of those apps that’s breaking the rules

Then Augustine Fou replied,

of course, MANY ad tech companies have been working on fingerprinting for years, as a work around to browsers (like Firefox) allowing users to delete cookies many years ago. Fingerprinting is even more pernicious because it is on server-side and out of control of user entirely

That last point is why I’ve long argued that we have a very basic problem with the client server model itself: that it all but guarantees a feudal system in which clients are serfs and site operators (and Big Tech in general) are their lords and masters. Though my original metaphor for client-server (which I have been told was originally a euphemism for slave-master) was calf-cow:

Here’s more on that one, plus some other metaphors as well:

I’ll pick up that thread after visiting what Nick says about fingerprinting:

There are countless ways that devices can be fingerprinted, and the mandated use of IDFA instead of those surreptitious methods makes it harder for ad tech companies to be sneaky. It has long been possible to turn off IDFA or reset the identifier. If it did not exist, ad tech companies would find other ways of individual tracking without users’ knowledge, consent, or control.

And why “ask” an app not to track? Why not “tell”? Or, better yet, “Prevent Tracking By This App”? Does asking an app not to track mean it won’t?

History has an answer for those questions.

Remember Do Not Track? Invented in the dawn of tracking, back in the late ’00s, it’s still a setting in every one of our browsers. But it too is just an ask — and ignored by nearly every website on Earth.

Much like Do Not Track, App Tracking Transparency is a request — verified as much as Apple can by App Review — to avoid false certainty. Tracking is a pernicious reality of every internet-connected technology. It is ludicrous to think that any company could singlehandedly find and disable all forms of fingerprinting in all apps, or to guarantee that users will not be tracked.

I agree. This too is a problem with the feudal system that the Web + app world has become, and Nick is right to point it out. He continues,

The thing that bugs me is that Searls knows all of this. He’s Doc Searls; he has an extraordinary thirteen year history of writing about this stuff. So I am not entirely sure why he is making arguments like the ones above that, with knowledge of his understanding of this space, begin to feel disingenuous. I have been thinking about this since I read this article last night and I have not come to a satisfactory realistic conclusion.

Here’s a realistic conclusion (or at least the one that’s in my head right now): I was mistaken to assume that Apple has more control here than it really does, and it’s right for all these guys (Nick, John, Augustine, Don and others) to point that out. Hey, I gave in to wishful thinking and unconscious ad hominem argumentation. Mea bozo. I sit corrected.

He continues,

Apple is a big, giant, powerful company — but it is only one company that operates within the realities of legal and technical domains. We cannot engineer our way out of the anti-privacy ad tech mess. The only solution is regulatory. That will not guarantee that bad actors do not exist, but it could create penalties for, say, Google when it ignores users’ choices or Dr. B when it warehouses medical data for unspecified future purposes.

We’ve had the GDPR and the CCPA in enforceable forms for awhile now, and the main result, for us mere “data subjects” (GDPR) and “consumers” (CCPA) is a far worse collection of experiences in using the Web.

At this point my faith in regulation (which I celebrated, at least in the GDPR case, when it went into force) is less than zero. So is my faith in tech, within the existing system.

So I’m moving on, and working on a new approach, outside the whole feudal system, which I describe in A New Way. It’s truly new and small, but I think it can be huge: much bigger than the existing system, simply because we on the demand side will have better ways of informing supply (are you listening, Mark Pritchard?) than even the best surveillance systems can guess at.

This piece has had a lot of very smart push-back (and forward, but mostly back). I respond to it in Part II, here.

If you haven’t seen it yet, watch Apple’s Privacy on iPhone | tracked ad. In it a guy named Felix (that’s him, above) goes from a coffee shop to a waiting room somewhere, accumulating a vast herd of hangers-on along the way. The herd represents trackers in his phone, all crowding his personal space while gathering private information about him. The sound track is “Mind Your Own Business,” by Delta 5. Lyrics:

Can I have a taste of your ice cream?
Can I lick the crumbs from your table?
Can I interfere in your crisis?

No, mind your own business
No, mind your own business

Can you hear those people behind me?
Looking at your feelings inside me
Listen to the distance between us

Why don’t you mind your own business?
Why don’t you mind your own business?

Can you hear those people behind me?
Looking at your feelings inside me
Listen to the distance between us

Why don’t you mind your own business?
Why don’t you mind your own business?

The ad says this when Felix checks his phone from the crowded room filled with people spying on his life:

Then this:

Finally, when he presses “Ask App Not to Track,” all the hangers-on go pop and turn to dust—

Followed by

Except that she gets popped too:

Meaning he doesn’t want any one of those trackers in his life.

The final image is the one at the top.

Here’s what’s misleading about this message: Felix would have had none of those trackers following him if he had gone into Settings—>Privacy—>Tracking, and pushed the switch to off, like I’ve done here:

Key fact: it is defaulted to on. Meaning Apple is not fully serious about privacy. If Apple was fully serious, your iPhone would be set to not allow tracking in the first place. All those trackers would come pre-vaporized. And Apple never would have given every iPhone an IDFA—ID For Advertisers—in the first place. (And never mind that they created IDFA back in 2013 partly to wean advertisers from tracking and targeting phones’ UDIDs (unique device IDs).

Defaulting the master Tracking setting to ON means Felix has to tap “Ask App Not To Track” for every single one of those hangers-on. Meaning that one click won’t vaporize all those apps at once. Just one at a time. This too is misleading as well as unserious.

And why “ask” an app not to track? Why not “tell”? Or, better yet, “Prevent Tracking By This App”? Does asking an app not to track mean it won’t?

History has an answer for those questions.

Remember Do Not Track? Invented in the dawn of tracking, back in the late ’00s, it’s still a setting in every one of our browsers. But it too is just an ask—and ignored by nearly every website on Earth.

Here is how the setting looks, buried deep on Google’s Chrome:

It’s hardly worth bothering to turn that on (it’s defaulted to off), because it became clear long ago that Do Not Track was utterly defeated by the adtech biz and its dependents in online publishing. The standard itself was morphed to meaninglessness at the W3C, where by the end (in 2019) it got re-branded “Tracking Preference Expression.” (As if any of us has a preference for tracking other than to make it not happen or go away.)

By the way, thanks to adtech’s defeat of Do Not Track in 2014, people took matters into their own hands, by installing ad and tracking blockers en masse, turning ad blocking, an option that had been laying around since 2004, into the biggest boycott in world history by 2015.

And now we have one large company, Apple, making big and (somewhat, as we see above) bold moves toward respecting personal privacy. That’s good as far as it goes. But how far is that, exactly? To see how far, here are some questions:

  • Will “asking” apps not to track on an iPhone actually make an app not track?
  • How will one be able to tell?
  • What auditing and accounting mechanisms are in place—on your phone, on the apps’ side, or at Apple?

As for people’s responses to Apple’s new setting, here are some numbers for a three-week time frame: April 26 to May 16. They come from FLURRY, a subsidiary of Verizon Media, which is an adtech company. I’ll summarize:

  • For “Worldwide daily op-in rate after iOS 14.5 launch across all apps,” expressed as “% of mobile active app users who allow app tracking among uses who have chosen to either allow or deny tracking” started at 11% and rose to 15%.
  • The “U.S. Daily opt-in rate after iOS launch across all apps,” expressed as “% of mobile active app users who allow app tracking among users who have chosen to either allow or deny tracking” started at 2% and rose to 6%.
  • The “Worldwide daily opt-in rate across apps that have displayed the prompt,” expressed as “% of mobile active app users who allow app tracking among users who have chosen to either allow or deny tracking” started at 31% and went down to 24%.
  • The “Worldwide daily share of mobile app users with ‘restricted’ app tracking” (that’s where somebody goes into Settings—>Privacy—>Tracking and switches off “Allow Apps to Request to Track”), expressed as “% of mobile active app users who cannot be tracked by default and don’t have a choice to select a tracking option” started and stayed within a point of 5% .
  • And the “U.S. daily share of mobile app users with ‘restricted’ app tracking,” expressed as “% of mobile active app users who cannot be tracked by default and don’t have a choice to select a tracking option” started at 4% and ended at 3%, with some dips to 2%.

Clearly tracking isn’t popular, but those first two numbers should cause concern for those who want tracking to stay unpopular. The adtech business is relentless in advocacy of tracking, constantly pitching stories about how essential tracking-based “relevant,” “personalized” and “interest-based” advertising is—for you, and for the “free” Web and Internet.

It is also essential to note that Apple does advertising as well. Here’s Benedict Evans on a slope for Apple that is slippery in several ways:

Apple has built up its own ad system on the iPhone, which records, tracks and targets users and serves them ads, but does this on the device itself rather than on the cloud, and only its own apps and services. Apple tracks lots of different aspects of your behaviour and uses that data to put you into anonymised interest-based cohorts and serve you ads that are targeted to your interests, in the App Store, Stocks and News apps. You can read Apple’s description of that here – Apple is tracking a lot of user data, but nothing leaves your phone. Your phone is tracking you, but it doesn’t tell anyone anything.

This is conceptually pretty similar to Google’s proposed FLoC, in which your Chrome web browser uses the web pages you visit to put you into anonymised interest-based cohorts without your browsing history itself leaving your device. Publishers (and hence advertisers) can ask Chrome for a cohort and serve you an appropriate ad rather than tracking and targeting you yourself. Your browser is tracking you, but it doesn’t tell anyone anything -except for that anonymous cohort.

Google, obviously, wants FLoC to be a generalised system used by third-party publishers and advertisers. At the moment, Apple runs its own cohort tracking, publishing and advertising as a sealed system. It has begun selling targeted ads inside the App Store (at precisely the moment that it crippled third party app install ads with IDFA), but it isn’t offering this tracking and targeting to anyone else. Unlike FLoC, an advertiser, web page or app can’t ask what cohort your iPhone has put you in – only Apple’s apps can do that, including the app store.

So, the obvious, cynical theory is that Apple decided to cripple third-party app install ads just at the point that it was poised to launch its own, and to weaken the broader smartphone ad model so that companies would be driven towards in-app purchase instead. (The even more cynical theory would be that Apple expects to lose a big chunk of App Store commission as a result of lawsuits and so plans to replace this with app install ads. I don’t actually believe this – amongst other things I think Apple believes it will win its Epic and Spotify cases.)

Much more interesting, though, is what happens if Apple opens up its cohort tracking and targeting, and says that apps, or Safari, can now serve anonymous, targeted, private ads without the publisher or developer knowing the targeting data. It could create an API to serve those ads in Safari and in apps, without the publisher knowing what the cohort was or even without knowing what the ad was. What if Apple offered that, and described it as a truly ‘private, personalised’ ad model, on a platform with at least 60% of US mobile traffic, and over a billion global users?…

Apple has a tendency to build up strategic assets in discrete blocks and small parts of products, and then combine them into one. It’s been planning to shift the Mac to its own silicon for close to a decade, and added biometrics to its products before adding Apple Pay and then a credit card. Now it has Apple Pay and ‘Sign in with Apple’ as new building blocks on the web, that might be combined into other things. It seems pretty obvious that Privacy is another of those building blocks, deployed step by step in lots of different places. Privacy has been good business for Apple, and advertising is a bigger business than all of those.

All of which is why I’ve lately been thinking that privacy is a losing battle on the Web. And that we need to start building a byway around the whole mess: one where demand can signal supply about exactly what it wants, rather than having demand constantly being spied on and guessed at by adtech’s creepy machinery.

What does it mean when perhaps hundreds of thousands of one’s photos appear in articles, essays and posts all over the Web?

It means they’re useful. That’s why I posted the originals in the first place, and licensed them to require only attribution. Because of that, I can at least guess at how many have been put to use.

For one example subject, take Lithium, a metal in the periodic table. Lithium is making news these days, because it’s both scarce and required for the batteries of electric and hybrid vehicles. At issue especially is how and where lithium is extracted from the Earth. As Ivan Penn and  put it in The Lithium Gold Rush: Inside the Race to Power Electric Vehicles (6 May in The New York Times), extraction “might not be very green.”

But it is blue. Or turquoise. Or aqua. Or whatever colors you see in the photo above.

I took that shot on a 2010 flight over Nevada. Looking out the window, it’s hard to miss lakes of bright colors on the desert floor, looking like stained glass windows into the Earth. I didn’t know at the time that the puddles were lithium, but I did know they’d be useful when I published them, along with whatever information a little bit of research would reveal about them. After I did the research, I put 17 photos in an album on Flickr titled Lithium Mines in Nevada and added the same set to another album called Mines and Mining, which is now 329 photos long.

Also on that flight, which produced 130 photos now in an album called 2010_08_06 rno-phx-bos, other topics of interest are the Shoshone Mountains, Yucca Lake and Yucca Flat (with “subsidence craters” over underground nuclear bomb explosions), the Nevada Test Site, (where hundreds of atomic bomb tests took place, among other interesting things, “Doom Town” on Frenchman Lake, Broom Lake in Area 51, Creech Air Force Base (from which military drones are remotely controlled), Grand Canyon, and Buffalo at night. None of the photos of mine at those links (all in Wikipedia) are especially artistic. In fact most of them make me cringe today, because I hadn’t yet mastered Photoshop when I posted them in the first place. Back then I shot only .jpgs, rather than RAW photos, which means I can’t go back do much to improve them. But all are useful, especially to writers and publications covering the topic of lithium mining. For example, my photos of those lithium lakes appear in—

And those are just the first six among 23,200 results in a search for my name + lithium. And those results are just from pubs that have bothered to obey my Creative Commons license, which only requires attribution. Countless others don’t.

Google also finds 57,400 results for my name + mining. On top of those, there are also thousands of other results for potash, river, geology, mining, mountains, dunes, desert, beach, ocean, hebrides, glacier, and other landforms sometimes best viewed from above. And that’s on top of more than 1500 photos of mine parked in Wikimedia Commons, of which many (perhaps most) are already in Wikipedia (sometimes in multiple places) or on their way there.

And those are just a few of the many subjects I’ve shot, posted and annotated to make them useful to the world. Which is why I’m guessing the number of photos actually being used is in the hundreds of thousands by now.

I have placed none of those photos in any of those places. I just put them up where they can easily be found and put to use. For example, when I shot Thedford, Nebraska, I knew somebody would find the photo and put it in Wikipedia.

Shots like these are a small percentage of all the photos I’ve taken over many decades. In fact, most of my photography is of people and scenes, not stuff like you find in the links above.

But apparently my main calling as a photographer is to push useful photos to the edge of the public domain, and to describe and tag them in ways that make them easy for researchers and journalists to find and use. And so far that has been a very successful strategy.


So I have a camera question for the fellow photographers out there.

My main camera is a 2012-vintage Canon 5D Mark III , which replaced a 2005-vintage Canon 5D (source of the lithium lake shots), which replaced a Canon 30D of the same generation, and a Nikon Coolpix before that. All of these are retired or beat up now. Being um, resource constrained, every camera and lens I’ve used in this millennium I’ve either rented or bought used.

Now, out of great kindness, an old friend is giving me a Sony a7R that has been idle since she replaced it with a Sony a7Riii. I’ve played with her newer Sony, and really like how much lighter mirrorless full-frames can be. (And the a7R is lighter than the a7Riii.) The question now is what kind of lens I want to start with here, given that my budget is $0 (though I will spend more than that). The Sony equivalent of the lens I use most, a Canon 24-105 f4 L, runs >$1000, even used.

I suppose I could get non-Sony lenses for less, but … I’m not sure that’s the way to go. I’m kinda tempted to get a telephoto zoom or prime for the Sony and keep using the Canon for everything else. But then I’m carrying two cameras everywhere.

But I just looked at Ken Rockwell’s take on the Sony 24-105mm f/4
FE G OSS Full-Frame E-Mount
, which appears to outperform the Canon equivalent (two links back) so maybe I’ll find a bullet to bite, and spend the grand.

[25 May…] And I did that. The lens just arrived. Now I just need to match it up with a7R, which will probably happen next Tuesday. I trust you’ll see some results soon after that.

Tags: , , ,

Have you ever wondered why you have to consent to terms required by the websites of the world, rather than the other way around? Or why you have no record of what you have accepted or agreed to?

Blame the cookie.

Have you wondered why you have no more privacy on the Web than what other parties grant you (which is none at all), and that you can only opt in or out of choices that others provide—while the only controls you have over your privacy are to skulk around like a criminal (thank you, Edward Snowden and Russell Brand, for that analogy) or to stay offline completely?

Blame the cookie.

And have you paused to wonder why Europe’s GDPR regards you as a mere “data subject” while assuming that the only parties qualified to be “data controllers” and “data processors” are the sites and services of the world, leaving you with little more agency than those sites and services allow, or provide you?

Blame the cookie.

Or why California’s CCPA regards you as a mere “consumer” (not a producer, much less a complete human being), and only gives you the right to ask the sites and services of the world to give back data they have gathered about you, or not to “sell” that personal data, whatever the hell that means?

Blame the cookie.

There are more examples, but you get the point: this situation has become so established that it’s hard to imagine any other way for the Web to operate.

Now here’s another point: it didn’t have to be that way.

The World Wide Web that Tim Berners-Lee invented didn’t have cookies. It also didn’t have websites. It had pages one could publish or read, at any distance across the Internet.

This original Web was simple and peer-to-peer. It was meant to be personal as well, meaning an individual could publish with a server or read with a browser. One could also write pages easily with an HTML editor, which was also easy to invent and deploy.

It should help to recall that the Apache Web server, which has published most of the world’s Web pages across most the time the Web has been around, was meant originally to work as a personal server. That’s because the original design assumption was that anyone, from individuals to large enterprises, could have a server of their own, and publish whatever they wanted on it. The same went for people reading pages on the Web.

Back in the 90s my own website, searls.com, ran on a box under my desk. It could do that because, even though my connection was just dial-up speed, it was on full time over its own static IP address, which I easily rented from my ISP. In fact, that I had sixteen of those addresses, so I could operate another server in my office for storing and transferring articles and columns I wrote to Linux Journal. Every night a cron utility would push what I wrote to the magazine itself. Both servers ran Apache. And none of this was especially geeky. (I’m not a programmer and the only code I know is Morse.)

My point here is that the Web back then was still peer-to-peer and welcoming to individuals who wished to operate at full agency. It even stayed that way through the Age of Blogs in the early ’00s.

But gradually a poison disabled personal agency. That poison was the cookie.

Technically a cookie is a token—a string of text—left by one computer program with another, to help the two remember each other. These are used for many purposes in computing.

But computing for the Web got a special kind of cookie called the HTTP cookie. This, Wikipedia says (at that link)

…is a small piece of data stored on the user‘s computer by the web browser while browsing a website. Cookies were designed to be a reliable mechanism for websites to remember stateful information (such as items added in the shopping cart in an online store) or to record the user’s browsing activity (including clicking particular buttons, logging in, or recording which pages were visited in the past). They can also be used to remember pieces of information that the user previously entered into form fields, such as names, addresses, passwords, and payment card numbers.

It also says,

Cookies perform essential functions in the modern web. Perhaps most importantly, authentication cookies are the most common method used by web servers to know whether the user is logged in or not, and which account they are logged in with.

This, however, was not the original idea, which Lou Montulli came up with in 1994. Lou’s idea was just for a server to remember the last state of a browser’s interaction with it. But that one move—a server putting a cookie inside every visiting browser—crossed a privacy threshold: a personal boundary that should have been clear from the start but was not.

Once that boundary was crossed, and the number and variety of cookies increased, a snowball started rolling, and whatever chance we had to protect our privacy behind that boundary, was lost.

Today that snowball is so large that nearly all personal agency on the Web happens within the separate silos of every website, and compromised by whatever countless cookies and other tracking methods are used to keep track of, and to follow, the individual.

This is why most of the great stuff you can do on the Web is by grace of Google, Apple, Facebook, Amazon, Twitter, WordPress and countless others, including those third parties.

Bruce Schneier calls this a feudal system:

Some of us have pledged our allegiance to Google: We have Gmail accounts, we use Google Calendar and Google Docs, and we have Android phones. Others have pledged allegiance to Apple: We have Macintosh laptops, iPhones, and iPads; and we let iCloud automatically synchronize and back up everything. Still others of us let Microsoft do it all. Or we buy our music and e-books from Amazon, which keeps records of what we own and allows downloading to a Kindle, computer, or phone. Some of us have pretty much abandoned e-mail altogether … for Facebook.

These vendors are becoming our feudal lords, and we are becoming their vassals.

Bruce wrote that in 2012, about the time we invested hope in Do Not Track, which was designed as a polite request one could turn on in a browser, and servers could obey.

Alas, the tracking-based online advertising business and its dependents in publishing dismissed Do Not Track with contempt.

Starting in 2013, we serfs fought back, by the hundreds of millions, blocking ads and tracking: the biggest boycott in world history. This, however, did nothing to stop what Shoshana Zuboff calls Surveillance Capitalism and Brett Frischmann and Evan Selinger call Re-engineering Humanity.

Today our poisoned minds can hardly imagine having native capacities of our own that can operate at scale across all the world’s websites and services. To have that ability would also be at odds with the methods and imperatives of personally targeted advertising, which requires cookies and other tracking methods. One of those imperatives is making money: $Trillions of it.

The business itself (aka adtech) is extremely complex and deeply corrupt: filled with fraud, botnets and malwareMost of the money spent on adtech also goes to intermediaries and not to the media you (as they like to say) consume. It’s a freaking fecosystem, and every participant’s dependence on it is extreme.

Take, for example, Vizio TVs. As Samuel Axon puts it in Ars TechnicaVizio TV buyers are becoming the product Vizio sells, not just its customers Vizio’s ads, streaming, and data business grew 133 percent year over year.

Without cookies and the cookie-like trackers by which Vizio and its third parties can target customers directly, that business wouldn’t be there.

As a measure of how far this poisoning has gone, dig this: FouAnalyticsPageXray says the Ars Technica story above comes to your browser with all this spyware you don’t ask for or expect when you click on that link:

Adserver Requests: 786
Tracking Requests: 532
Other Requests: 112

I’m also betting that nobody reporting for a Condé Nast publication will touch that third rail, which I have been challenging journalists to do in 139 posts, essays, columns and articles, starting in 2008.

(Please prove me wrong, @SamuelAxon—or any reporter other than Farhad Manjoo, who so far is the only journalist from a major publication I know to have bitten the robotic hand that feeds them. I also note that the hand in his case is The New York Times‘, and that it has backed off a great deal in the amount of tracking it does. Hats off for that.)

At this stage of the Web’s moral devolution, it is nearly impossible to think outside the cookie-based fecosystem. If it was, we would get back the agency we lost, and the regulations we’re writing would respect and encourage that agency as well.

But that’s not happening, in spite of all the positive privacy moves Apple, Brave, Mozilla, Consumer Reports, the EFF and others are making.

My hat’s off to all of them, but let’s face it: the poisoning is too far advanced. After fighting it for more than 22 years (dating from publishing The Cluetrain Manifesto in 1999), I’m moving on.

To here.

I wrote this fake story on January 24, 2005, in an email to Peter Hirshberg after we jokingly came up with it during a phone call. Far as I know, it was the first mention of the word “iPhone.”

Apple introduces one-button iPhone Shuffle

To nobody’s surprise, Apple’s long-awaited entry into the telephony market is no less radical and minimalistic than the one-button mouse and the gum-stick-sized music player. In fact, the company’s new cell phone — developed in deeply secret partnership with Motorola — extends the concept behind the company’s latest iPod, as well as its brand identity.

Like the iPod Shuffle, the new iPhone Shuffle has no display. It’s an all-white rectangle with a little green light to show that a call is in progress. While the iPhone Shuffle resembles the iPod Shuffle, its user interface is even more spare. In place of the round directional “wheel” of the iPods, the iPhone Shuffle sports a single square button. When pressed, the iPod Shuffle dials a random number from its phone book.

“Our research showed that people don’t care who they call as much as they care about being on the phone,” said Apple CEO Steve Jobs. “We also found that most cell phone users hate routine, and prefer to be surprised. That’s just as true for people answering calls as it is for people making them. It’s much more liberating, and far more social, to call people at random than it is to call them deliberately.”

Said (pick an analyst), “We expect the iPhone Shuffle will do as much to change the culture of telephony as the iPod has done to change the culture of music listening.”

Safety was also a concern behind the one-button design. “We all know that thousands of people die on highways every year when they take their eyes off the road to dial or answer a cell phone,” Jobs said. “With the iPhone Shuffle, all they have to do is press one button, simple as that.”

For people who would rather dial contacts in order than at random, the iPhone Shuffle (like the iPod Shuffle) has a switch that allows users to call their phone book in the same order as listings are loaded loaded from the Address Book application.
To accommodate the new product, Apple also released Version 4.0.1 of  Address Book, which now features “phonelists” modeled after the familiar “playlists” in iTunes. These allow the iPhone Shuffle’s phone book to be populated by the same ‘iFill’ system that loads playlists from iTunes into iPod Shuffles.

A number of online sites reported that Apple negotiating with one of the major cell carriers to allow free calls between members who maintain .Mac accounts and keep their data in Apple’s Address Book. A few of those sites also suggested that future products in the Shuffle line will combine random phone calling and music playing, allowing users to play random music for random phone contacts.

The iPhone Shuffle will be sold at Apple retail stores.

As a follow-up to what I wrote earlier today, here are my own favorite podcasts, in the order they currently appear in my phone’s podcast apps:

Note that I can’t help listening to the last two, because I host one and co-host the other.

There are others I’ll listen to on occasion as well, usually after hearing bits of them on live radio. These include Radiolab, This American Life, 99% Invisible, Snap Judgement, Freakonomics Radio, Hidden BrainInvisibilia, The Moth, Studio 360. Plus limited run podcasts, such as Serial, S-Town, Rabbit Hole and Floodlines.

Finally, there are others I intend to listen to at some point, such as Footnoting History, Philosophize This, The Infinite Monkey Cage, Stuff You Should Know, The Memory Palace, and Blind Spot.

And those are just off the top of my head. I’m sure there are others I’m forgetting.

Anyway, most of the time I’d rather listen to those than live radio—even though I am a devoted listener to a raft of public stations (especially KCLU, KPCC, KCRW, KQED, WNYC, WBUR and WGBH) and too many channels to mention on SiriusXM, starting with Howard Stern and the NBA channel.


KPCCNPR, which turned 50 yesterday, used to mean National Public Radio. It still does, at least legally; but they quit calling it that in 2010. The reason given was “…most of our audience — more than 27 million listeners to NPR member stations and millions more who experience our content on NPR.org and through mobile or tablet devices — identify us as NPR.” Translation: We’re not just radio any more.

And they aren’t. Television, newspapers and magazines also aren’t what they were. All of those are now experienced mostly on glowing rectangles connected to the Internet.

Put another way, the Internet is assimilating all of them. On the Internet, radio is also fracturing into new and largely (though not entirely) different spawn. The main two are streaming (for music, live news and events) and podcasting (for talk and news).

This sidelines the radio sources called stations. Think about it: how much these days do you ask yourself “What’s on?” And how much do you listen to an actual radio, or watch TV through an antenna? Do you even have a radio that’s not in a car or stored away in the garage?

If you value and manage your time, chances are you are using apps to store and forward your listening and viewing to later times, when you can easily speed up the program or skip over ads and other “content” you don’t want to “consume.” (I put those in quotes because only the supply side talks that way about what they produce and what you do with it.)

This does not match the legacy structure of radio stations. Especially technically.

See, the purpose of stations is to stand in one place radiating sound (called “programs”) on signals, in real time (called ‘live”), around the clock, for a limited geography: a city or a region. Key thing: they have to fill that time.

For this stations can get along without studios (like companies in our current plague have found ways to get along without offices). But they still need to maintain transmitters with antennas.

For AM, which was born in the 1920s, the waves are so long that whole towers, or collections of them, radiate the signals. In almost all cases these facilities take up acres of real estate—sometimes dozens of acres. For FM and TV, media born in the 1940s, the waves are short, but need to radiate from high places: atop towers, tall buildings or mountains.

Maintaining these facilities isn’t cheap. In the case of AM stations, it is now common for the land under towers to be worth far more than the stations themselves, which is why so many AM stations are now going off the air or moving off to share other stations’ facilities, usually at the cost of lost coverage.

This is why I am sure that most or all of these facilities will be as gone as horse-drawn carriages and steam engines, sometime in the next few years or decades. Also why I am documenting transmitters that still stand, photographically. You can see a collection of my transmitter and antenna photos here and here. (The image above is what radiates KPCC/89.3 from Mt. Wilson, which overlooks Los Angeles.)

It’s a safe bet, for a few more years at least, that stations will still be around, transmitting to people mostly on the Net. But at some point (probably many points) the transmitters will be gone, simply because they cost too much, don’t do enough—and in one significant way, do too much. Namely, fill the clock, 24/7, with “content.”

To help get our heads around this, consider this: the word station derives from the Latin station- and statio from stare, which means to stand. In a place.

In the terrestrial world, we needed stationary places for carriages, trains and busses to stop. On radio, we used to need what we called a “dial,” where radio stations could be found on stationary positions called channels or frequencies. Now those are numbers that appear in a read-out.

But even those were being obsolesced decades ago in Europe. There a car radio might say the name of a station, which might be received on any number of frequencies, transmitted by many facilities, spread across a region or a country. What makes this possible is a standard called RDS, which uses a function called alternative frequency (AF) to make a radio play a given station on whatever channel sounds best to the radio. This would be ideal for the CBC in Canada and for regional public stations such as WAMC, KPCC, KUER and KCRW, which all have many transmitters scattered around.

Alas, when this standard was being worked out in the ’80s and early ’90s, the North American folks couldn’t imagine one station on many frequencies and in many locations, so they deployed a lesser derivative standard called RDBS, which lacked the AF function.

But this is now, and on its 50th anniversary public radio—and NPR stations especially—are doing well.

In radio ratings for New York, Los Angeles, San Francisco, Washington, San Diego, and dozens of other markets, the top news station is an NPR one. Here in Santa Barbara, about a quarter of all listening goes to non-commercial stations, led by KCLU, the most local of the NPR affiliates with transmitters here. (Best I can tell, Santa Barbara, which I wrote about here in 2019, is still the top market for public radio in the country. Number two is still Vermont.)

But I gotta wonder how long the station-based status quo will remain stationary in our listening habits. To the degree that I’m a one-person bellwether, the prospects aren’t good. Nearly all my listening these days is to podcasts or to streams on the Net. Some of those are from stations, but most are straight from producers, only one of which is NPR. And I listen to few of them live.

Still, it’s a good bet that NPR will do well for decades to come. Its main challenge will be to survive the end of station-based live broadcasting. Because that eventuality is starting to become visible.

On Quora, William Moser askedWould the KPIG radio format of Americana—Folk, Blugrass, Delta to modern Blues, Blues-rock, trad. & modern C&W, country & Southern Rock, jam-bands, singer/songwriters, some jazz, big-band & jazz-singers sell across markets in America?

I answered,

I’ve liked KPIG since its prior incarnation as KFAT.I’ve liked KPIG since its prior incarnation as KFAT.

It’s a great fit in the Santa Cruz-Salinas-Monterey market, anchored in Santa Cruz, which is a college/beach/hippie/artist kind of town.

Ratings have always been good, putting it in the top few. See here.

It has also done okay in San Luis Obispo, for similar reasons.

For what it’s worth, those are markets #91 and #171. Similar in a coastal California kind of way.

The station is also a throwback, with its commitment to being the institution it is, with real personalities who actually live there, aren’t leaving, and having a sense of humor about all of it. Also love. And listener participation. None of that is a formula.

Watch this and you’ll get what I mean. It’s what all of radio ought to be, in its own local ways, and way too little of it is.

William replied,

I have played tha brilliant ‘Ripple’ since it was released; the editing is spot-on.I have played tha brilliant ‘Ripple’ since it was released; the editing is spot-on.

With respect to DJ personalities, there are at least two that, in my fantasy of owning that station, would be gone before the ink was dry. (You might even have an idea of whom). Paradoxically, that’s probably part of what makes KPIG work.

It’s really the cross-country market appeal of the ‘Americana’ music format that is my question.

My response:

I think it’s a local thing. KPIG (and KFAT and KHIP before it) is a deeply rooted local institution. It’s not a formula, and without standing one up in some other region like it, and funding it long enough to see if it catches on, it’s hard to say.

All of radio is in decline now, as talk listening moves to podcasts and music listening moves to streaming. The idea that a city or a region needs things called “stations,” all with limited geographical coverage, and with live talent performing, and an obligation to stay on the “air” 24/7, designed to work through things called “radios,” which are no longer sold in stores and persist as secondary functions on car dashboards, is an anachronism at a time when damn near everything (including chat, telephony, video, photography, gaming, fitness tracking and you-name-it) is moving onto phones—which are the most persistently personal things people carry everywhere.

There are truly great alternative stations, however, that thrive in their markets: KESP in Seattle and WWOZ in New Orleans are two great examples. That KPIG manages to persist as a commercial station is especially remarkable in a time when people would rather hit a 30-second skip-forward button on their phone app than listen to an ad.

So I guess my answer is no. But if you want Americana, there are lots of stations that play or approximate it on the Internet. And all of them can be received on your phone or your computer.

That was a bit tough to write, because yesterday I was poised to enjoy Ron Phillips‘ long-standing Saturday morning show on WWOZ when I learned he had died suddenly of a heart attack. Ron has been a great friend since the 1970s, when he was a mainstay at WQDR in Raleigh, and I was a partner in its ad agency (while still being a funny guy at cross-market non-rival WDBS), and I had planned to give him a call after his show, to see how he was doing. (He’d had carpal tunnel surgery recently.)

Though WWOZ is alive and thriving, and there persist many radio stations that are vital institutions in their towns and regions, radio on the whole has been in decline. See here:

The slopes there are long, and a case can be made, on that low angle alone, that radio will persist forever, along with magazines and TV. But it’s clear that our media usage is moving, overall, to the Internet, where mobile devices are especially good at doing what radio, TV and magazines also do—and in some ways doing it better.

But back to William Moser’s question.

There are already many ways to stream Americana (aka American roots music) on the Internet, whether from stations, streaming services like Spotify, Pandora, or a channel or two on SiriusXM. But those are largely personality-free.

What KPIG has (once described to me as “mutant cowboy rock & roll”) isn’t a format. It’s an institution, like a favorite old tavern, music club, outdoor festival, or coffee shop—or all of those rolled into one. Can one replicate that with an Internet station, or a channel on some global service?

I think not, because those services are all global. You need to start with local roots. WWOZ has that, because it started as a radio station in New Orleans, a place that is itself deeply rooted. After years of living all over the place, Ron moved to New Orleans to be with those roots, and near its greatest radio voice.

Radio is geographical. All the stations I mentioned above, living and dead, are not the biggest ones in their towns. KPIG’s signal is almost notoriously minimal. So is KEXP’s in Seattle. They are local in the most basic sense.

I suppose that condition will persist for another decade or two. But it’s hard to say. Mobile devices are also evolving quickly, getting old within just a few years.

I’m not sure we’ll miss it much, the succession of generations being what it is. But we are losing something. And you can still hear it on KPIG.

KSKO radio

On Quora, somebody asked, Which is your choice, radio, television, or the Internet?. I replied with the following.

If you say to your smart speaker “Play KSKO,” it will play that small-town Alaska station, which has the wattage of a light bulb, anywhere in the world. In this sense the Internet has eaten the station. But many people in rural Alaska served by KSKO and its tiny repeaters don’t have Internet access, so the station is either their only choice, or one of a few. So we use the gear we have to get the content we can.

TV viewing is also drifting from cable to à la carte subscription services (Netflix, et. al.) delivered over the Internet, in much the same way that it drifted earlier from over-the-air to cable. And yet over-the-air is still with us. It’s also significant that most of us get our Internet over connections originally meant only for cable TV, or over cellular connections originally meant only for telephony.

Marshall and Eric McLuhan, in Laws of Media, say every new medium or technology does four things: enhanceretrieveobsolesce and reverse. (These are also caled the Tetrad of Media Effects.) And there are many answers in each category. For example, the Internet—

  • enhances content delivery;
  • retrieves radio, TV and telephone technologies;
  • obsolesces over-the-air listening and viewing;
  • reverses into tribalism;

—among many other effects within each of those.

The McLuhans also note that few things get completely obsolesced. For example, there are still steam engines in the world. Some people still make stone tools.

It should also help to note that the Internet is not a technology. At its base it’s a protocol—TCP/IP—that can be used by a boundless variety of technologies. A protocol is a set of manners among things that compute and communicate. What made the Internet ubiquitous and all-consuming was the adoption of TCP/IP by things that compute and communicate everywhere in the world.

This development—the worldwide adoption of TCP/IP—is beyond profound. It’s a change as radical as we might have if all the world suddenly spoke one common language. Even more radically, it creates a second digital world that coexists with our physical one.

In this digital world, we are at a functional distance apart of zero. We also have no gravity. We are simply present with each other. This means the only preposition that accurately applies to our experience of the Internet is with. Because we are not really on or through or over anything. Those prepositions refer to the physical world. The digital world is some(non)thing else.

This is why referring to the Internet as a medium isn’t quite right. It is a one-of-one, an example only of itself. Like the Universe. That you can broadcast through the Internet is just one of the countless activities it supports. (Even though the it is not an it in the material sense.)

I think we are only at the beginning of coming to grips with what it all means, besides a lot.

« Older entries