You are currently browsing the archive for the infrastructure category.

ripping up a contractLet’s reset our thinking to what a user’s expectations are, when operating a browser and interacting with pages and sites.

In my browser, when I visit a page, I am requesting that page. I am not requesting stuff other than that page itself. This is what the hypertext protocol (http) provides.

(Protocols are ritualized manners, like handshakes, bows and smiles. They also scaffold the social contract.)

Likewise, when I visit a site (such as a seller) with a service on the Web, I am not requesting stuff other than what that site presents to me in text and graphics.

So, for example, when I go to, I expect the browser to display that page and its links, and nothing more. And when I go to, I expect the browser to display the index page of the site — and, if I have some kind of relationship with that site, recognition that I’m a returning visitor or customer.

In neither of those cases do I expect tracking files, other than those required to remember state, which was the original purpose of Lou Montouli’s magic cookie, way back in ’94. Now known as just “the cookie,” it is in ubiquitous use today. In  Lou’s detailed history of that creation he writes, “The goal was to create a session identifier and general ‘memory’ mechanism for websites that didn’t allow for cross site tracking.”

Now let’s look at how we read a newspaper or a magazine here in the physical world. This time I’ll use my sister as an example of a typical reader. She’s a retired Commander in the U.S. Navy, and organized in the way she interacts with what we generally call “content.”

When a newspaper arrives, she “field strips” it. If it’s the Sunday paper, she pulls out all the advertising inserts and either throws them away or sets them aside, depending on whether or not they contain coupons that might interest her. Then she strips out sections that don’t interest her. The Travel section might go on one Sunday, the Sports section on another.

Then, when she reads the paper, she ignores most of the ads. One exception might be the magazine section, which tends to contain full-page brand ads by companies like Apple and Toyota. Those she might notice and like at some level. It all depends

My point is that she consciously blocks some ads and allows some others, some of which she pays attention to, but most of which she does not.

This kind of interaction is what the user expects the hypertext protocol (http) and good manners on the part of websites and services will provide. Websites that spy on users outside of their own domains (or use third parties to do the same) break the social contract when they do that. It’s that simple.

Yes, cases can be made for innocent forms of tracking, such as anonymized data gathering for analytics that improve what websites do. But they should be opt-in for users, not opt-out. Alas, that kind of tracking is a baby in the blocking bathwater. (The EFF’s Privacy Badger blocks many of these by default, and provides sliders for degrees of opting in or out of them.)

How did we get from the online world Lou Montouli sought to improve in ’94 and the one we have today? Check the metaphors for what we had and what we’ve lost.

Back in the mid-’90s we called the browser our car on the “information superhighway.” Cars, like clothing and shelter, are privacy technologies. They give us ways of operating in the world that conceal our most private spaces — ones where others are not welcome, except by invititation.

But, thanks to Zuboff’s Laws, our browsers became infected with spyware. Here is what those laws say:

  1. Everything that can be automated will be automated.
  2. Everything that can be informated will be informated.
  3. Every digital application that can be used for surveillance and control will be used for surveillance and control.

Sure, some of adtech’s surveillance is meant to give us a “better advertising experience” or whatever. Buy that’s beside the main point: it breaks the social contract in both the letter and the spirit of hypertext protocol. It gives us what none of us asked for and what most of us don’t want.

A few years ago, we tried to send a message to publishers and advertisers with Do Not Track, but it was fought, mocked and ignored by those to whom it spoke.

Fortunately, browsers support add-ons and extensions, so we took actions that can’t be ignored, by installing ad and tracking blockers. In doing so we acted as free and independent agents, just as we do in the everyday world with our clothing, our shelter and our cars.

What we need next are ways for us to engage constructively with publishers, in alignment with well-understood social contracts long established in the everyday world, and embodied in the hypertext protocol.

Engagement will also give us scale. As I explain in A Way to Peace in the Adblock War,

Some on the advertising side want to engage, and not to fight. In Dear Adblocking community, we need to talk, Chris Pedigo of Digital Content Next recognizes the legitimacy of ad blocking in response to bad acting by his industry, and outlines some good stuff they can do.

But they also need to see that it’s no longer up to just them. It’s up to us: the individual targets of advertising.

The only way engagement will work is through tools that are ours, and we control: tools that give us scale — like a handshake gives us scale. What engages us with the Washington Post should also engage us with Verge and Huffpo. What engages us with Mercedes should also engage us with a Ford dealer or a shoe store.

If we leave fixing things up to publishers and the adtech industry, all of us will be given different prosthetic hands, each of which will interact in different ways that are not of our choosing and give us no scale. In fact that is what we already get with the DAA’s Ad Choices and Ghostery’s massive opt-out list. We see how well that worked.

The road to personal independence and engagement scale is a long one.

In The Cluetrain Manifesto, we said,

we are not seats or eyeballs or end users or consumers. we are human beings and our reach exceeds your grasp. deal with it.

Except in 1999, when we wrote that, we didn’t yet have the reach. We just knew we would, sooner or later, as a native entitlement of the Net.

In The Data Bubble, I said,

The tide turned today. Mark it: 31 July 2010.

That’s when The Wall Street Journal published The Web’s Gold Mine: Your Secrets, subtitled A Journal investigation finds that one of the fastest-growing businesses on the Internet is the business of spying on consumers. First in a series. It has ten links to other sections of today’s report.

In fact it the tide didn’t turn, because we didn’t yet have the tools to turn it. The Journal’s series, titled “What They Know,” is still at The last entry is in 2013. They should fire it up again.

Because now, in late 2015, we have the first of those tools, with ad and tracking blockers.

But we have to do better. And by “we” I mean us human beings — and the developers working on our side for the good of everybody.

Note: This is the sixth post in a series covering online advertising, starting on 12 August. Here are the first five:

  1. Separating advertising’s wheat and chaff
  2. Apple’s content blocking is chemo for the cancer of adtech
  3. Will content blocking push Apple into advertising’s wheat business?
  4. If marketing listened to markets, they’d hear what ad blocking is telling them
  5. Debugging adtext assumptions

Here is the current perimeter of the Valley Fire, according to the USGS’ GEOMAC viewer:

ValleyFire 2015-09-13 at 3.10.24 PM_a

As you see, no places are identified there. One in particular, however, is of extremely special interest to me: Harbin Hot Springs. That’s where I met my wife and made more friends than I can count. It is, or was,  one of the most lovely places on Earth, inhabited and lovingly maintained by wonderful people.

I just matched up a section of the map above with Google Maps’ Earth view, and see that Harbin and its neighborhood are in the perimeter:

Screen Shot 2015-09-13 at 3.12.19 PM

After seeing this picture here, which looks northwest from downtown Middletown…

COyGRRHVAAEwC4w… I suspected the worse.

And now comes news that Harbin is “pretty much destroyed.” Damn.

Other places in the perimeter — or so it appears to me (please don’t take this as gospel):

  • Outer edges of Middletown and Hidden Valley Lake communities
  • Parts of Whispering Pines, Cobb, Holbergs and Glenbrook
  • Areas adjacent to McCreary Lake and Detert Reservoir

Watch here for official information about the fire.


ice-floes-off-greenland(Cross posted from this at Facebook)

In Snow on the Water I wrote about the ‘low threshold of death” for what media folks call “content” — which always seemed to me like another word for packing material. But its common parlance now.

For example, a couple days ago I heard a guy on WEEI, my fave sports station in Boston, yell “Coming up! Twenty-five straight minutes of content!”

Still, it’s all gone like snow on the water, melting at the speed of short term memory decay. Unless it’s in a podcast. And then, even if it’s saved, it’ll still get flushed or 404’d in the fullness of time.

So I think about content death a lot.

Back around the turn of the millennium, John Perry Barlow said “I didn’t start hearing the word ‘content’ until the container business felt threatened.” Same here. But the container business now looks more like plumbing than freight forwarding. Everything flows. But to where?

My Facebook timeline, standing in the vertical, looks like a core sample of glacier ice, drilled back to 1947, the year I showed up. Memory, while it lasts, is of old stuff which in the physical world would rot, dry, disintegrate, vanish or lithify from the bottom up.

But here we are on the Web, which was designed as a way to share documents, not to save them. It presumed a directory structure, inherited from Unix (e.g. domain.something/folder/folder/file.html). Amazingly, it’s still there. Whatever longevity “content” enjoys on the Web is largely owed to that structure, I believe.

But in practice most of what we pile onto the top of the Web is packed into silos such as Facebook. What happens to everything we put there if Facebook goes away? Bear in mind that Facebook isn’t even yet a decade old. It may be huge, but it’s no more permanent than a sand dune. Nothing on the Web is.

Everything on the Web, silo’d or not, flows outward from its sources like icebergs from glaciers, melting at rates of their own.

The one exception to that rule is the Internet Archive, which catches as much as it can of all that flow. Huge thanks to Brewster Kahle and friends for giving us that.

Anyway, just wanted to share some thoughts on digital mortality this morning.

As you were. Or weren’t. Or will be. Or not.

Bonus link: Locking the Web open.

Check this out:


I took that screen shot at the excellent Oakleaf restaurant in Pittsboro, NC a few days ago. Note the zero bars (or dots) of telephone service, and the very respectable (tested!) data service. To confirm what the hollow dots said, I tried to make a call. Didn’t work.

This seems to be a new thing for T-Mobile in North Carolina, where I spent much of this summer — or at least in the parts of it where I visited.

The company’s mobile phone coverage is pretty lousy to begin with, on the whole: great on highways and in the larger towns; but spotty when you head into the suburbs and countryside. What changed is the sudden near-disappearance of voice phone coverage in some places where it had worked before, and the improvement at the same time of data coverage.

At my sister’s house, near a major interstate highway, I could use my phone on the porch or in the yard, but not indoors, where I’d see the most dreaded two words in mobile telephony: “no service.” Or at least that was the case in July and early August.

Then something strange happened. I started getting data service indoors at her house, and in other places where before there was nothing. But all I got was data, identified by that little “LTE.” Telephony was five empty dots. At my sister’s place I also couldn’t make or get a call out in the yard, on the street, or anywhere in the neighborhood. But the data service was now terrific.

So I’m wondering if this is just me, or if T-Mobile is lately favoring data over telephony in some places. Anybody know? (I note that T-Mobile’s coverage maps only seem to deal with data, not telephony. But maybe I’m missing something.)

By the way, I should add that I wouldn’t trade T-Mobile for any other carrier right now, because I travel a lot outside the country. In addition to fine coverage in New York, Boston, and all the places I tend to go in California, T-Mobile gives me free data roaming and texting everywhere I go, and 20¢/minute on the phone. Yes, the data rates tend to be 2G rather than 3G or 4G/LTE. But it tends to be good enough most of the time. It also makes me tolerant of a less-than-ideal coverage footprint here in the U.S.


Right now every FM and TV station in Santa Barbara and San Diego can be heard in both places. Between them lays more than 200 miles of ocean across a curved earth. I’m not there right now, but I see what’s happening remotely over my TV set top box. (Thank you, SlingBox.) But, more importantly, John Harder‘s tropo map tells me so:


Tropo is tropospheric refraction of radio waves across a distance. Atmosphere has refractive properties that don’t matter most of the time. But we can see changes, for example, with mirages ahead of us above a hot road, which causes the air above to refract light at a low angle, essentially reflecting the sky, other cars and landscapes on the horizon. Something like this also happens over land and water.

I see by the map above that tropo is happening in other parts of California, Nevada, Utah and Arizona. I also see that it’s starting to happen here in north central North Carolina, where I can already pick up stations in South Carolina:


On 88.1fm, for example, I’m getting WRJA from Sumter, South Carolina, atop WKNC in Raleigh. WRJA is about 160 miles away while WKNC is only 40 miles away. But WRJA is 100,000 watts atop a thousand-foot tower, while WKNC is 25,000 watts on a 260 foot tower. (It’s actually as little as 35% of full power in most directions from the transmitter at NC State. They have a construction permit to change that a bit.) So they’re making a hash of each other here.

Back when I lived in the woods north of Chapel Hill, long before the Internet showed up and made all of this stuff irrelevant for listeners (who can get the same stations on the Net, anywhere), I had a directional Finco FM-5 antenna and a Channel Master Crossfire 3610 antenna (both salvaged from abandoned structures) on a pole next to my 1-story house. I rotated them by hand. If I had the same rig here I could point at either WRJA or WKNC and “null out” the other. I did this on hot summer mornings for fun back then, and eventually logged nearly every FM station from Pennsylvania to Georgia. (It’s a summer thing, and coincidental with heat waves over large areas.)

Here is a whole-country map that shows tropo happening pretty much everywhere:


I would love to have had this kind of resource back in those days. Now that I do, it’s hardly worth the trouble, since nearly all my radio listening is over the Net.

Anyway, if you’re wondering why your local station is being obliterated by a signal coming from a state or two away, or why you’re suddenly getting far-away stations where none were before, tropo is the most likely reason.

The second most likely one is Sporadic E skip, which brings in stations from distances of 800-1200 miles or so away. In that case the E layer of the ionosphere turns into something like a hot road surface, reflecting distant signals, but only at a certain angle. I’ll cover those in a later post when the phenomenon is actually happening. It’s not right now.

Meanwhile, if any of this intrigues you at all, check out William Hepburn’s amazing Worldwide Tropo Forecast Maps. Great eye and brain candy, because it exposes a real-world view of the world that isn’t what you see with your eyes.

Tags: , , , , , , , , ,

The Giant ZeroMany years ago, Craig Burton shared the best metaphor for the Internet that I have ever heard, or seen in my head. He called it hollow sphere: a giant three-dimensional zero. He called it that because a sphere’s geometry best illustrates a system in which every end, regardless of its physical location, is functionally zero distance away from every other end. Across the nothing in the Net’s hollow sphere, every point can “see” every other point, and connect to it, as if distance were not there. And at no cost.

It doesn’t matter that the Net’s base protocol, TCP/IP, is not perfect, that there are costs and latencies involved in the operation of connections and routers between end points — and that many people in the world still do not enjoy the Net’s graces. What matters is that our species’ experience of the Net, and of the world it creates, is of zero distance and cost. You and I can publish posts like this one, or send emails to each other, or even have live video conference calls, with little if any regard for distance and cost.

Our experience of this is as essential to our future as the discovery of language and fire was to our ancestors. The Net has already become as essential to human agency — the capacity to act with effect in the world — as the wheel and movable type. We are not going to un-discover it.

Yes, companies and governments can control or access to the Net, and sphincter what passes through it; but it’s too late for anybody or anything to keep our species from knowing what it’s like to be zero distance apart at zero cost. We now have that experience, and we will use it to change life on Earth. Hopefully for the better.

The Giant Zero of the Net has an analogue with the physical world, whose gravity pulls us all toward an invisible center we can’t see but know is there. As with the Net’s zero, we live on Earth’s surface. The difference is that, on the Earth’s zero, distance matters. So does the inverse square law. Sound, sight and radio waves fade across distances. We need to be close to hear and see each other. Not so on the Net.

The Giant Zero is also the title of my next book. Until then, if you dig the metaphor, you might also source World of Ends or NewClues, both of which are co-written by David Weinberger. For now I just want to post this so I can source something simple about The Giant Zero in one link.

HT to @dweinberger: every hyperlink travels across the zero. And thanks to Hugh McLeod for the image above. Way back in 2004, I asked him to draw me the Internet, and that’s what he did. I haven’t seen anything better since.

meerkatLook where Meerkat andperiscopeapp Periscope point. I mean, historically. They vector toward a future where anybody anywhere can send live video out to the glowing rectangles of the world.

If you’ve looked at the output of either, several things become clear about their inevitable evolutionary path:

  1. Mobile phone/data systems will get their gears stripped, in both directions. And it will get worse before it gets better.
  2. Stereo sound recording is coming. Binaural recording too. Next…
  3. 3D. Mobile devices in a generation or two will include two microphones and two cameras pointed toward the subject being broadcast. Next…
  4. VR, or virtual reality.

Since walking around like a dork holding a mobile in front of you shouldn’t be the only way to produce these videos, glasses like these are inevitable:


(That’s a placeholder design in the public domain, so it has no IP drag, other than whatever submarine patents already exist, and I am sure there are some.)

Now pause to dig Facebook’s 10-year plan to build The Matrix. How long before Facebook buys Meerkat and builds it into Occulus Rift? Or buys Twitter, just to get Periscope and do the same?

Whatever else happens, the rights clearing question gets very personal. Do you want to be recorded by others and broadcast to the world or not? What are the social and device protocols for that? (Some are designed into the glasses above. Hope they help.)

We should start zero-basing some answers today, while the inevitable is in sight but isn’t here yet.

It should help to remember that all copyright laws were created in times when digital life was unimaginable (e.g. Stature of Anne, ASCAP), barely known (Act of 1976), or highly feared (WIPO, CTEA, DMCA).

How would we write new laws for the new video age that has barely started? Or why start with laws at all? (Remember that nearly all regulation protects yesterday from last Thursday — and are often written by know-nothings.)

We’ve only been living the networked life since graphical browsers and ISPs arrived in the mid-90’s. Meanwhile we’ve had thousands of years to develop civilization in the physical world.

Relatively speaking, digital networked life is Eden, which also didn’t come with privacy. That’s why we made clothing and shelter, and eventually put both on hooves and wheels.

How will we create the digital equivalents of the privacy technologies we call clothing, shelter, buttons, zippers, doors, windows, shades, blinds and curtains? Are the first answers technical or policy ones? Or both? (I favor the technical, fwiw. Code is Law and all that.)

Protecting the need for artists to make money is part of the picture. But it’s not the only part. And laws are only one way to protect artists, or anybody.

Manners come first, and we don’t have those yet. Meaning we also lack civilization, which is built on, and with, manners of many kinds. Think about much manners are lacking in the digital world. So far.

None of the big companies that dominate our digital lives have fully thought out how to protect anybody’s privacy. Those that come closest are ones we pay directly, and are therefore accountable to us (to a degree). Apple and Microsoft, for example, are doing more and more to isolate personal data to spaces the individual controls and the company can’t see — and to keep personal data away from the advertising business that sustains Google and Facebook, which both seem to regard personal privacy as a bug in civilization, rather than a feature of it. Note that we also pay those two companies nothing for their services. (We are mere consumers, whose lives are sold to the company’s actual customers, which are advertisers.)

Bottom line: the legal slate is covered in chalk, but the technical one is close to clean. What do we want to write there?

Start here: privacy is personal. We need to be able to signal our intentions about privacy — both as people doing the shooting, and the people being shot. A red light on a phone indicating recording status (as we have on video cameras) is one good step for video producers. On the other side of the camera, we need to signal what’s okay and what’s not. Clothing does that to some degree. So do doors, and shades and shutters on windows. We need the equivalent in our shared networked space. The faster and better we do that, the better we’ll be able to make good TV.

doc036dLike the universe, the Internet is one thing. It is a World of Ends, comprised of everything it connects.

By nature it is as neutral as gravity. It favors nothing and is not partial to anything. Yes, there are exceptions to that rule, in the way Net access is provisioned, for example; but the basic nature of the Net — as a free, open and neutral shared space — is by now obvious to pretty much everybody who doesn’t have an interest in making it less. calls itself “a Facebook-led initiative bringing together technology leaders, non-profits and local communities to connect the two thirds of the world that doesn’t have Internet access.” But what it offers is not the Internet, but a sphinctered fraction of it: Facebook plus a few chosen others.

This is pure misdirection: a private fraction masked as a public whole. And it’s not fooling anybody. Especially India. See here, here, here, here, here, here — and every other place you’ll find piles of stories about it. (Start with the Critique section of the Wikipedia article on, and a search for

India is rejecting for one simple reason: They know sphincternet ≠ Internet, and that the sphinctered Net is not Neutral, meaning not the real thing.

Naturally, Mark Zuckerberg disagrees, and explains how in this post on the matter, which went up yesterday, and I’ll respond to, piece by piece:

Over the past week in India, there has been a lot written about and net neutrality. I’d like to share my position on these topics here for everyone to see.

First, I’ll share a quick story. Last year I visited Chandauli, a small village in northern India that had just been connected to the internet.

In a classroom in the village, I had the chance to talk to a group of students who were learning to use the internet. It was an incredible experience to think that right there in that room might be a student with a big idea that could change the world — and now they could actually make that happen through the internet.

Those students should know the whole Net. Not just a subset of it.

The internet is one of the most powerful tools for economic and social progress. It gives people access to jobs, knowledge and opportunities. It gives voice to the voiceless in our society, and it connects people with vital resources for health and education.

I believe everyone in the world deserves access to these opportunities.

Fine. Then either give them the whole thing, or call what you give them something else that’s clearly less: Facebook+, perhaps.

In many countries, however, there are big social and economic obstacles to connectivity. The internet isn’t affordable to everyone, and in many places awareness of its value remains low. Women and the poor are most likely to be excluded and further disempowered by lack of connectivity.

The Internet itself has no cost: on purpose. At its base is a protocol that nobody owns, everybody can use, and anybody can improve. (Not that anybody has yet — or ever will.) That’s one of the features of its inherent neutrality.

Yes, there are first-time and maintenance costs for the wires and waves that carry its bits. But, as Steve Kamman explains, “Bandwidth is dirt cheap. And bog-standard… This isn’t like electricity. There’s no power plant on the other end burning fuel to deliver those bits. Bits are nearly weightless and cost accordingly.”

Steve’s case is for where the Net ends up, everywhere: the effect implicit in its cause. Think about how to make that happen. Trust me: it’ll be good for Facebook too.

This is why we created, our effort to connect the whole world. By partnering with mobile operators and governments in different countries, offers free access in local languages to basic internet services in areas like jobs, health, education and messaging. lowers the cost of accessing the internet and raises the awareness of the internet’s value. It helps include everyone in the world’s opportunities.

But it’s not the whole Internet. It’s what you and your partners, in an exclusive and non-neutral way, have decided to provide.

We’ve made some great progress, and already more than 800 million people in 9 countries can now access free basic services through In India, we’ve already rolled out free basic services on the Reliance network to millions of people in Tamil Nadu, Maharashtra, Andhra Pradesh, Gujarat, Kerala and Telangana. And we just launched in Indonesia on the Indosat network today.

We’re proud of this progress. But some people have criticized the concept of zero-rating that allows to deliver free basic internet services, saying that offering some services for free goes against the spirit of net neutrality. I strongly disagree with this.

Zero-rating, or “toll-free data,” means not charging for some stuff on the Net, while charging the same fees for the rest. Simply put, it’s a form of price discrimination. Here’s what Wikipedia says about its reception and impact :

Zero-rating certain services, fast lanes and sponsored data clearly have their benefits for users of the subsidized services, but have also been criticised as anti-competitive and limiting open markets.[4] As many new internet and content services are launched targeting primarily mobile usage, and further adoption of internet connectivity globally (including broadband in rural areas of developed countries) relies heavily on mobile, zero-rating has also been regarded as a threat to the open internet, which is typically available via fixed line networks with unlimited usage tariffs or flat rates.[9] The Wikimedia Foundation and Facebook have been specifically criticized for their zero-rating programs, to further strengthen incumbent mobile network operators and limit consumer rights to an open internet.[10] (That’s as of today.)

Whatever else it is, it’s not neutral.

We fully support net neutrality. We want to keep the internet open. Net neutrality ensures network operators don’t discriminate by limiting access to services you want to use. It’s an essential part of the open internet, and we are fully committed to it.

But net neutrality is not in conflict with working to get more people connected. These two principles — universal connectivity and net neutrality — can and must coexist.

To give more people access to the internet, it is useful to offer some service for free. If someone can’t afford to pay for connectivity, it is always better to have some access than none at all.

Useful, yes. Neutral, no.

Non-neutrality is like Potter Stewart’s definition of porn: “I know it when I see it.” Which India does. doesn’t block or throttle any other services or create fast lanes — and it never will.

It just doesn’t carry them. It says “My way or no highway.”

We’re open for all mobile operators and we’re not stopping anyone from joining.

The Internet is not what just mobile operators carry.

We want as many internet providers to join so as many people as possible can be connected.

That’s fine. But what they provide won’t be the Internet if they don’t carry the whole thing. It will be a sampler box of rocks rather than all of geology.

Arguments about net neutrality shouldn’t be used to prevent the most disadvantaged people in society from gaining access or to deprive people of opportunity. Eliminating programs that bring more people online won’t increase social inclusion or close the digital divide. It will only deprive all of us of the ideas and contributions of the two thirds of the world who are not connected.

There wouldn’t be an argument if you didn’t call this thing “,” and if you didn’t represent a few Internet services as the whole thing. But you do, and that’s why you’re having trouble.

Every person in the world deserves access to the opportunities the internet provides. And we can all benefit from the perspectives, creativity and talent of the people not yet connected.

We have a historic opportunity to connect billions of more people worldwide for the first time. We should work together to make that happen now.

Fine. But make clear that what you’re offering isn’t the Internet, but a bunch of free services also found on the real thing.

Below that post are a zillion comments, some of which Mark answers. Here is the first Q&A:

Ritesh Pandya: We really appreciate your initiative on making the internet accessible to most remote part of the world, but the only question is why access only to selective websites and not all on

Mark Zuckerberg: It’s too expensive to make the whole internet free. Mobile operators spend tens of billions of dollars to support all of internet traffic. If it was all free they’d go out of business. But by offering some basic services, it’s still affordable for them and it’s valuable and free for everyone to use.

But it’s not the Net. It’s just a set of services that also happen to be on the Net.

The Internet is free. That’s its nature. So stop confusing access with the Net itself, and a few services with the whole thing. Nobody’s buying it.

Bonus links: New Clues,

[Later, May 4…] Wired says Zuckerberg has “expanded” to include more stuff. In other words, he’s dilated the sphincter.

This is about AM radio stations being worth less than the ground they stand on. Case in point: WMAL-AM in Washington, DC. You can see the problem with this Google Map:


The heart-shaped patch of green between the legs of I-495 and the I-270 spur is populated by four towers radiating the signal of WMAL, a landmark on Washington’s radio dial (at 630am) since 1925. The station’s 75-acre transmitter site is nearly as big as the nearby Bethesda Country Club golf course and the Westfield Montgomery Mall. It also sits deep in the suburbs, surrounded by trees and highways, most of which appeared long after WMAL erected the towers on cheap open land, far from the bustling Capitol, many decades ago. That land is worth a lot more now.

So it’s no surprise to read news (via The Sentinel) that Cumulus Media, which owns WMAL-AM & FM, has put the land up for sale. Says the report, “Local real estate experts estimate the property could be worth hundreds of millions.” I don’t know what WMAL-AM is worth, but I’m guessing it would be a few million, tops. So it makes financial sense to sell off the land. 

But what about the signal? Many AM stations have already “gone dark” (as they say in the business). Will WMAL do the same? In the first comment below, Jon Elbaz, who wrote the Sentinel piece, says Cumulus intends to keep WMAL-AM on the air somehow. But a question is raised: how long can any AM station on desirable land stay on the air? And by what means?

Back in radio’s golden age — when AM ruled the waves — the stations battling for the top of Washington, DC’s ratings heap were WTOP and WMAL. WTOP peaked when it went all-news in the 1960s, and has stayed at the top ever since. It did that by doing great work, and by wisely moving to FM a few years back, taking over the channel (103.5) long occupied by classical WGMS-FM, whose owners by then had unloaded its original 570am signal, which is still on the air as WSPZ. (More about that below.) WMAL also has an FM signal, on 105.9. That one is #9 in Nielsen’s latest figures, while WTOP is #1. WMAL-AM doesn’t show at all.

So I have to wonder about Cumulus commitment to keeping the signal on the air. Finding a new transmitter site is not a cheap undertaking. To explain, I’ll need to get technical.

To transmit, AM radio stations require a substantial sum of real estate. AM waves are hundreds of feet long, and require long radiating antennas. These take the form of towers. If a station has a directional signal, more than one tower is required to create the signal’s pattern. WMAL has two different asymmetrical patterns for use in the day and night. Here is how the four towers are arranged, and the patterns they produce:


Each tower is a quarter wavelength high, which at 630am makes them about 400 feet tall. Surrounding them is also a “ground system” of buried conductors running hundreds of feet in all directions from the towers. This is why WMAL needs those 75 acres. To stay on the air, WMAL will need to find replacement acreage, somewhere that allows the signals you see above (or slightly modified versions of them) to cross as much of the Metro area as possible, meaning it will have to be northwest of town. For that Cumulus will need to either buy land out that way, or co-site with some other station already operating there.

The only two stations with transmitters out there are WTEM (“ESPN 980″) and WSPZ, both sports stations (on 980 and 570 respectively) and owned by Red Zebra Broadcasting (in which the main stakeholders are also those of the Washington Redskins). (Here are aerial views, via Bing, of the WTEM and WSPZ sites.)

Of those, WSPZ’s site looks like it has more room. It’s in Germantown, about 22 miles from downtown Washington, more than twice the distance from downtown Washington as WMAL’s current site. I suspect the signal patterns could be “tightened” to concentrate energy toward Washington, though, and that might help. But ground conductivity — which matters hugely for AM signals — is poor in Maryland and Virginia, which is one reason AM stations there tend to suck in the ratings. (For evidence of how much ground conductivity matters, compare three AM stations, all 5,000 watts, and all on 570am: WSPZ in Washington, WNAX in Yankton, South Dakota and KLIF in Dallas. The latter two cover enormous territories, while WSPZ basically covers the District and bits of adjacent Maryland and Virginia. Ground conductivity in the middle of the country is about 15x better than the area served by WSPZ.)

In fact WSPZ is way out of town today because its owners a few years back did exactly what Cumulus is doing with WMAL today: they sold the land out from under the towers. This topo map shows where the WSPZ towers used to be, when it was still WGMS: a site in Potomac, across the Beltway about three miles from WMAL’s site. Believe me, the old signal from the old site covered the metro a helluva lot better than the new signal from the new site. Expect the same results if WMAL moves there.

So again, why keep WMAL-AM on the air at all?

One argument is that the WMAL-FM signal isn’t a great one. While it’s licensed for 28000 watts, it only hits that max to the northwest and southwest of its transmitter in Merrifield, outside the Beltway on the southwest side of town. Toward the district (northeast of the site) its signal has a huge dent, down to around 1/4 of what it puts out in the other directions:


So getting a bit of help on the AM side might still be worth the trouble.

Still, one wonders… How much time will pass before the land under WSPZ becomes far more valuable than the station — or even WSPZ and WMAL put together?

This kind of question sits in front of many AM station owners’ minds right now. I expect what we’ll have in the long run are AM stations standing on land with little or no market value. The rest will disappear along with their real estate.

[Later…] I also wonder about Cumulus’ commitment to saving the signal. In 2011 it acquired (by merger) KAAY/1090am in Little Rock, Arkansas — a 50,000 watt giant with rich history and a night signal that stretches from Cuba to Canada. Or used to. Wikipedia:

Unfortunately, owners of KAAY in later years allowed the stations famed transmission facilities in Wrightsville, AR to fall into disrepair. Copper thieves stole a large amount of transmission line, degrading the stations signal significantly. Roof damage allowed water to enter the 50,000 watt transmitter – knocking it off the air. Currently, KAAY has reestablished 50000 watt service during the day, but has yet to rebuild the 3 tower directional array, so nighttime service remains under an STA at 1250 watts non directional.

KAAY is the biggest AM station in Arkansas. If Cumulus cared, it would restore the station to full capacity. But the format is “brokered/Christian,” which is tends to be low-cost dial-filler. Only one AM station makes the published ratings for Little Rock, and it’s Cumulus’ KARN/920 “The Sports Animal.” Not KAAY. KARN is also at the bottom of the heap. Higher rated are four other Cumulus stations, all FMs.

So the Company isn’t suffering there. Its portfolio of stations does fine, and that’s what matters, right? If the market won’t miss WMAL-AM, why bother keeping it?

[Later again…] This story features an offer sheet on the property, and says offers need to be in by March 12. I also found this older story, about Cumulus’ plan to sell the land under KABC’s transmitter. I can find no evidence that the land has been sold, or is still on the market. KABC also has no construction permits to move to a new location.

[Later again…] Well, apparently they do think it’s worth the bother. Jonathan O’Connell reported this in a February 13 story in The Washington Post:

When the towers are torn down, it will not affect WMAL, said Cumulus spokesman Collin Jones. Jones said the company would lease transmission facilities elsewhere after the sale closed.

“Listeners will literally have no idea that it happened,” Jones said.

Well, some listeners. Others will. There is no way WMAL can move to another site without compromising the signal in some directions.

[May 5…] A buyer for the land has been found, stories in Bethesda Magazine and Radio World report. In a search on, however, I see no applications or construction permits at a new site, but perhaps Cumulus is still in negotiations for leasing space at other locations.





mutualmusiciansSo I just learned that a Kansas City Jazz station is headed toward existence. If you love any of these musicians, this should be very good news.

The story begins,

By this time next year, Kansas City-style jazz might be bebopping out of a new radio station near you.

The Mutual Musicians Foundation in the 18th and Vine jazz district announced this week it’s been granted a construction permit for a noncommercial, low-power FM radio station. The foundation is hoping the KC jazz station, at 104.7 FM, will be on the air by next January.

It will be called KOJH-LP. LP stands for low power, or what the FCC calls LPFM. Here’s the application for what’s now a granted CP, or Construction Permit.

In fact there is a jazz station called KOJH already — a streaming one in Oklahoma. Though it’s not a licensed radio station, it may have inherited those call letters from one. (I’ve looked, but haven’t been able to tell. Maybe the lazyweb knows.)

Here’s the station’s mission, filed with the FCC.

KOJH will broadcast from the Arts Asylum at Harrison and E. 9th Street. A new tower will go on the building. From there they will radiate a whopping 22 watts at 207 feet above the average terrain, at 104.7fm. It’s a tiny signal that will won’t reach far out of downtown.

Worse, most of Kansas City’s big FM stations have effective radiated powers (what’s concentrated toward the horizon, or populations) of 100,000 watts, and transmit from a collection of towers over 1000 feet tall, just a short distance east of downtown. One of those is KBEQ on 104.3, just two notches down the dial from KOJH. This means you will need a good radio to keep KBEQ from blasting KOJH sideways. Today’s car radios are good enough to keep that from happening. (And will likely get KOJH up to a dozen or more miles away.) Recent-vintage portable and home radios will have a hard time, unless they’re very close to the KOJH transmitter.

(Many manufacturers quit caring decades ago. And now Radio Shack has filed for bankruptcy. Even CEO Can’t Figure Out How RadioShack Still In Business, which ran in The Onion in 2007, has proven prophetic.)

So it is good to know KOJH plans to stream online, because that’s the future of radio.

But there are other stepping stones.

For example, something the Mutual Musicians Foundation might consider doing, while they get underway with KOJH, is buying an AM station that’s dropped out of the ratings. Some possibles, going up the dial:

    • KCCV/760. 6000 watts day, 200 watts night.
    • WHB/810. 50000 watts day, 5000 watts night.
    • KBMZ/980. 5000 watts day and night.
    • KCWJ/1030. 5000 watts day, 500 watts night.
    • KCTO/1160. 5000 watts day, 230 watts night.
    • KYYS/1250. 25000 watts day, 3700 watts night.
    • KDTD/1340. 1000 watts, day and night.
    • KCNW/1380. 2500 watts day, 300 watts night.
    • KKLO/1410. 5000 watts day, 500 watts night.
    • KCZZ/1480. 1000 watts day, 500 watts night.
    • KWOD/1660. 10000 watts day, 1000 watts night.

(Note that wattage is just one variable. Location of the transmitter, efficiency of the towers, directionality of the signal, ground conductivity and frequency all matter too. For example, the lower the station’s frequency, the longer the wavelength, and the better its signal travels along the ground.)

Only three AM stations show up in Kansas City’s latest ratings: KCSP, a sports station at 610am, KCMO, a right-wing talk station at 710am, and KPRT, a gospel music station at 1590am. (With 1000 watts by day and just 50 watts at night, I’m amazed KPRT makes the ratings at all.)

All the un-rated stations listed above put signals across all of KOJH’s coverage area, and then some. Some, such as WHB (a legendary station and signal), may never be for sale. But I’ll bet some others are on the market today, and will only get cheaper.

Music sounds awful on AM, unless the station radiates HD radio encoding. Most engineers I know in broadcasting dislike HD radio and consider it a gimmick. But it does sound quite good on both AM and FM. The difference it makes on AM is amazing.

Loyal listeners of a format will do the work required to get a signal. I’m sure that’s the case with KPRT’s gospel listeners, for example. Now, after stumbling for years, HD radio is picking up with manufacturers. There is a nice list on the HD Radio site. Meanwhile, the market value of AM radio stations, especially ones with no ratings, is crashing to the point where the cost of operating them exceeds their income. (An AM station sucks about twice the wattage off the grid as it radiates from its transmitter.) In coming years many of them will sell for a song.

So those changes — the rise of HD Rado and the decline of also-ran AM station prices — are factors the KOJH folks might want to keep in mind as they fire up their LP signal on FM. Think local, but think big too.

Bonus link.

« Older entries