>> law, technology, and the space between

All content by Kyle E. Mitchell, who is not your lawyer.

You can subscribe via RSS/Atom or e-mail and browse other blogs.

The Main EventAmazon versus Startups and the New York Times

And here we have it, ladies and gentlemen—the main event!

In one corner, The New York Times, Americans’ most or least favorite newspaper, depending on whether they read the New York Times, boasting people who actually pay to read its content on the Internet in 2019.

In the opposite corner, Amazon Web Services, weighing in at Way More Money Than Print Can Accurately Convey and boasting more corporate communications people than the Washington Post boasts journalists.

The bell rings. They saunter to the center. Glistening and shaking, they trade blows:

Preferencing its own products!

Plentiful partnerships!

Strip-mining open source!


Pull quotes and photo portraits!

Tweetstorms and unnamed analysts!

And slowly, inexorably, it dawns on the crowd at bout assembled. The contenders are heavyweights indeed. No doubt. But neither has brought the punch to knock the other down. Maybe to knock anyone down. This isn’t a title fight. It isn’t even much of an exhibition. It’s some kind of obligatory box-ticking exercise.

We can watch them huff and puff their way through twelve rounds and a few thousand words. We can tally up the points. We can even declare a winner, by diligent decision. But there’s no new narrative to note. No rhetorical route to recant. Not so much as a good, clean zinger to savor. Nothing that matters, that changes the game, that evolves the conversation.

The titans have met … and they got nothin’. It’s one for the stats, not one for the history books.

For the partisans of either camp reading here, a few point-by-points follow. Not that they’ll shake any minds, but a fool has to try.

Then some thoughts on what this all means, or, rather, what it doesn’t mean, and how little meaning there is here to worry about, or apparently to fight for.

And how that’s the point.

Strip Mining

“Strip-mining open source” is many-levels accurate. Miners strip for minerals near the surface, not at great depth. It takes enormous machines and vast economy of scale. It purges the ecology, scars the landscape, and disrupts local communities. It’s short-sighted and wrong, a bunch of mega machines burning diesel to dredge more coal out of the ground.

Delectable nuance. But nuance is for bookish saps and literary suckers. Saps and suckers like me.

For the rest of humanity, the gist of “strip-mining open source” is highly inaccurate. Strip mining is legal, albeit regulated. It is also productive. We look askance at coal now, when we remember to think about it. But it wasn’t so long ago that my progenitors had to clock in for their anthracite coughs down a deep, dark shaft in mountain country.

There aren’t any miners here. We’re talking about people, largely educated, affluent people, who twiddle smooth fingertips on plastic keys and stare at pretty screens all day. If they skew hippie and croon community Kumbaya instead of chasing that cheddar, there’s no great sympathy waiting when they fail to become spontaneously wealthy.

Taking your stuff isn’t wrong if you said they could. Especially since you knew they would, because this is America. You can start your commune, but you can’t secede from the market.


“Customer-centric” is demonstrably true. AWS apparently loves doing what AWS customers say they will pay AWS for. Of course, none of said customers purportedly asked for AWS. But now that they have it, it’s the cruel yoke of input-validated product-market fit unto death, come Hell or high regulatory exposure.

But “customer-centric” is manifold lame. Nobody outside the corporate envelope—the Amazon corporate envelope—cares about such pat internal business pablum. You kinda have to be there, repeatedly, at some form of internal meeting, likely mandatory, and well tied to compensation, for a slogan like that to resonate. The correct pronunciation is “AWS-customer-centric”. The “AWS” goes without saying. Which is the problem.

American corporate mission mumbo is like the fresh flesh of a ripe avocado: it goes brown and slimy on exposure to the outside world. The firm rhetorical bite and vibrant vocal color retains only with a thorough surface acid barrier of satire. Extemp:

customer-centrism, n.

The theory or belief that everyone is an Amazon customer and that Amazon is therefore situated at the center of the universe.

See Also: Bezomaic System

But Amazon hasn’t taken to its bland corporate blog for self-deprecation. It takes itself seriously.

But there’s nothing inspirational or countercultural about grubbing for purchase orders. That’s called work.

Open Source

The upshot here is that there’s nothing left to really fight for, so they’re not. The reason we’re here, the reason there’s a kerfuffle to flummox about, is a little meme-coup called “open source” from the dawn of the current Internet era. It is now old enough to drink. It’s no longer a blank slate of unblemished promise to write our boundless hopes and dreams upon. It has baggage. It commits bad behavior.

Notwithstanding the money that is and will be coursing through the computer economy’s veins for the foreseeable future, the volatile, intoxicating essence of “open source” as compu-cultural mythos has boiled off for anyone but the rawest neophytes and most committed old hands. The major players are positioned to fight for scraps, well shorn of boundless-plenty-awaits mentality. But the major players’ heavy weapons are nowhere to be seen when it comes time to fight for open source bona fides, because it is scraps, and their generals know better. They harbor no Internet-exceptionalist illusions, if they ever did.

Open source has been around too long, explicitly and implicitly promising too much to too many, not to engender real and inevitable disappointment. Slapping GPL on everything cool didn’t convert the world to software freedom fundamentalism or freeze computing in holy state circa 1998. That vision was Utopian only insofar as it required a functioning Utopia to make any sense. Neither did slapping an MIT license on the compulsive emissions of addicted hackers turn all the brave and true into Ubermenschen and famous names. That was mathematically and psychologically impossible, even for those relative few invested very early on. Stallman, Torvalds, Raymond, period.

For a time, those dreams were yet to be disproven by experience, and lo, the paper wealth did flow. If you didn’t get yours by now, those who did, and the rational market that blessed them, have come to agree that you just didn’t deserve it. We’re in the New Beetle era of open source, and you can’t afford one.


What didn’t appear in the Times or the AWS piece was any evidence of enlightenment. This is a fight neither side—AWS or startups—really stands to win. But better for AWS not to fight. Because fighting invites more of those whose open source flames keep burning to give in and burn out. The off-ramp to realism, again in the offing.

Harsh Reality: It’s good to be the king. The bigger the pile of great software free for literally any use laying around on the Web, and the more schmucks giving their life’s work away for nothing, the better it is to stand in the commanding heights of the market. When every decision you make, everything that you do, every dollar that you spend, matters more than anyone else’s, makes more than anyone else’s, what you choose to do with the free software largesse matters more than what anyone else does. It’s not winner-take-all. That’s too simple. It’s closer to winner-keeps-winning.

Usurping market top-dog position is damn difficult, and has little to commend it in the moral tones retconned onto open source history. It takes swell luck, and the preconditions for receiving such luck are largely business in nature. Not software brilliance. Not a sharing heart. Not ideological or rhetorical sparkle. Publishing, network effects, and some luck made O’Reilly. Support, network effects, and some luck made Red Hat. Hosting, network effects, and some luck made AWS. Whatever’s next, network effects, and more luck will make the next jackpot.

Harsh Reality: You don’t need open source to reach customers. And open source won’t make them love irrelevant code. If your code is quality and you take the time to reach the people who need it, open source is a bonus, and maybe more on the engineers writing it side than the engineers using it side. So hack up your feature grid. Write a new license. Mix it up in one repository. The cold, hard, practical value of good software is going up, and the fuzzy, acerbic, insular cachet of “open source” is going down. In part thanks to controversies like AWS v. Everybody that force the conflicted fallibility of open source under the popular eye.

Taking open source code and folding it into the AWS empire, repeatedly, is going to reduce the amount of open source code on offer, as AWS apparently defines “open source”. The market is going to iron that “irrationality” out. The more feedback provided, and the more the media reinforces it, the faster mentality will shift. But Amazon’s already walked away with the game. And they’re probably more concerned about AWS customer perception than their unaffiliated farm team of software trend setters.

In short, it’s all harsh realities. And reality is fast dissolving the airy, ephemeral fantasies of open source. For ambitious producer and opportunistic consumer alike.

Your thoughts and feedback are always welcome by e-mail.

back to topedit on GitHubrevision history