A glimpse of some of the thinking I’ve been doing about the impact of “data” on our culture. I am close (so damn close) to sealing myself off and into only thinking about this, for my book (OpenCoSF is my last big project till I do). But thanks to the Vibrant Data project for taking an interview I did at TED earlier this year, and making it into something that almost makes me look like I have my shit together. I attest, I do not. I hope soon, I will.
(image shutterstock) I’ve been watching the news for tidbits which illuminate a thesis I’ve been working up for my book. Today the New York Times provided a doozy: Swiss Cows Send Texts to Announce They’re in Heat. As James Gleick, author of The Information, noted in a Twitter response to me: That’s one heckuva headline.
So what’s my thesis? It starts with one of the key takeaways from Gleick’s book, which is that we are, as individuals and a society, becoming information. That might seem a rather puzzling statement, because one could argue that we’ve always been information, it’s only recently that we’re realizing that fact. So perhaps a better way of putting it is that we’re exploring the previously unmapped world of information. In the 1400s, the physical world was out there, much as it is today (perhaps it had a few more glaciers…). But we hadn’t discovered it, at least, not in any unified fashion. Now that we’ve discovered, named, and declared the outlines of most of the physical world, we are rapidly moving into a new era, one where we are coloring in the most interesting bits of information in our world with what we now call “data.”
As we survey, chart, and claim this new territory, a truth is emerging: when we discover some set of information might be valuable, we turn that information into data. Information is a slippery concept – one that gives Gleick “the willies.” But data? That’s information we can manipulate.
So here’s my thesis: We create new data wherever we can find value. Put another way: If it’s valuable to know, new data will flow. Not to everyone, of course – as with oil, control of data is power. But the world is hell-bent on finding new data resources that unleash value. We’ve got wildcatters, we’ve got Exxon/Mobiles (think Facebook, Google, Amazon, the NSA, etc.), we’ve got pipes. And we’ve got incredible stories of the things folks will do to unlock the value of data.
Which takes us back to the cows of Switzerland. As the Times’ piece explains, a Swiss research team has created a system, comprised of implanted sensors and radio beacons, that measures a cow’s movement and internal body temperature. It converts these measurements into data, runs the data through an algorithm, and when the resulting computation indicates the cows are in heat, it sends a text message to the rancher. The net result: The rancher has a better chance of getting that cow pregnant (er, that didn’t quite come out right – but you know what I mean).
Net net: a pregnant cow is a more valuable cow. And to get a cow pregnant more reliably, one needs the data. Previously, that data was buried in a bovine’s unexplored nether regions (literally – the sensor is placed in the cow’s genitals). But given the value that data carries, these Swiss data wildcatters have tapped a new gusher. This data exploration is now happening over and over, in nearly every imaginable corner of our world. We’ve just tapped the tip of this data iceberg, of course; we’re just stepping onto the shores of the New World. We’d be wise to remember that as we move forward.
(image GigaOm) Like many of you, I’ve been fascinated by the ongoing drama around Twitter over the past few months (and I’ve commented on part of it here, if you missed it). But to me, one of the most interesting aspects of Twitter’s evolution has gone mostly unnoticed: its ongoing legal battle with a Manhattan court over the legal status of tweets posted by an Occupy Wall St. protestor.
In this case, the State of New York is arguing that a tweet, once uttered, becomes essentially a public statement, stripped of any protections. The judge in the case concurs: In this Wired coverage, for example, he is quoted as writing “If you post a tweet, just like if you scream it out the window, there is no reasonable expectation of privacy.”
Twitter disagrees, based on its own Terms of Service, which state “what’s yours is yours – you own your Content.”
As the NYT puts it:
Twitter informed the (Occupy protestor) that the judge had ruled his words no longer belonged to him: (he) had turned them over to Twitter, in other words, to be spread across the world.
(Twitter’s) legal team appealed on Monday of last week. Tweets belong to the user, the company argued.
I find this line of argument compelling. Twitter is arguing that its users do not “turn over” their words to Twitter, instead, they license their utterances to the service, but retain rights of ownership, those rights remain with the person who tweets. It’s a classic digital argument – sure, my words are out there on Twitter, but those are a licensed copy of my words. The words – the ineffable words – are still *mine.* I still have rights to them! One of those rights may well be privacy (interesting given Twitter’s public nature, but arguable), but I can imagine this builds a case for other ownership rights as well, such as the right to repurpose those words in other contexts.
If that is indeed the case, I can imagine a time in the not too distant future when people may want to extract some or all their tweets, and perhaps license them to others as well. Or, they may want to use a meta-service (there’s that idea again) which allows them to mix and mash their tweets in various ways, and into any number of different containers. Imagine for a minute that one of those meta services gets Very Big, and challenges Twitter on its own turf. Should that occur, well, the arguments made in this Manhattan case may well come into very sharp focus. And it’s just those kind of services that are nervous about where Twitter is going.
Just noodling it out. I may be missing some key legal concept here, but this strikes me as a potentially important precedent. I plan to speak with folks at Twitter about all this soon, and hopefully, I’ll have some clarity. Stay tuned.
I’m at least three books behind in my reviews, so I figured I’d bang out a fun one today: The Victorian Internet: The Remarkable Story of the Telegraph and the Nineteenth Century’s On-line Pioneers by Tom Standage. This 1998 book is now a classic – written as the Web was exploding on the scene, it reminded us that this movie has run before, 150 years in the past, with the rise of the telegraph. He writes:
The rise and fall of the telegraph is a tale of scientific discovery, technological cunning, personal rivalry, and cutthroat competition. It is also a parable about how we react to new technologies: For some people, they tap a deep vein of optimism, while others find in them new ways to commit crime, initiate romance, or make a fast buck age- old human tendencies that are all too often blamed on the technologies themselves.
Standage chronicles the history of the telegraph’s many inventors (Morse was just the most famous “father” of the device), and the passions it stirred across the world. Nowhere, however, did the invention stir more excitement (or bad poetry) than in the United States, where it can be convincingly argued that the telegraph’s ability to conquer distance and time almost perfectly matched the young country’s need to marshall its vast geography and resources. Were it not for the telegraph, the United States may never have become a world power.
Expansion was fastest in the United States, where the only working line at the beginning of 1846 was Morse’s experimental line, which ran 40 miles between Washington and Baltimore. Two years later there were approximately 2,000 miles of wire, and by 1850 there were over 12,000 miles operated by twenty different companies. The telegraph industry even merited twelve pages to itself in the 1852 U.S. Census. “The telegraph system [in the United States] is carried to a greater extent than in any other part of the world,” wrote the superintendent of the Census, “and numerous lines are now in full operation for a net-work over the length and breadth of the land.” Eleven separate lines radiated out from New York, where it was not uncommon for some bankers to send and receive six or ten messages each day. Some companies were spending as much as $1,000 a year on telegraphy. By this stage there were over 23,000 miles of line in the United States, with another 10,000 under construction; in the six years between 1846 and 1852 the network had grown 600-fold.
Standage writes with the amused eye of a British citizen – he currently works for the Economist as digital editor. One can sense a bit of English envy as he tells the telegraph’s tale – just as with television, the telegraph had early roots in his native country, but found its full expression in the United States. Thomas Edison started his career as a “telegraph man,” Alexander Graham Bell was inspired by the invention, the Associated Press grew out of the telegraph’s impact on newspapers, “e-commerce” was invented across the device’s wires, and huge corporations were born from its industries – Cable & Wireless, for example, began as a company that sourced insulation for telegraph lines.
The Victorian Internet is a must read for anyone interested in the history of technology, and in the cycles of hype, boom, and bust that seem to only quicken with each new wave of innovation. Highly recommended.
Other works I’ve reviewed:
Lightning Man: The Accursed Life of Samuel F. B. Morse by Kenneth Silverman (review)
Republic, Lost: How Money Corrupts Congress–and a Plan to Stop It by Larry Lessig (review)
Where Good Ideas Come From: A Natural History of Innovation by Steven Johnson (my review)
The Singularity Is Near: When Humans Transcend Biology by Ray Kurzweil (my review)
In The Plex: How Google Thinks, Works, and Shapes Our Lives by Steven Levy (my review)
The Next 100 Years: A Forecast for the 21st Century by George Friedman (my review)
Today I answered a question in email for a reporter who works for Wired UK. He asked smart questions, as I would expect from a Wired writer. (Some day I’ll tell you all my personal story of Wired UK – I lived over there for the better part of a year back in 1997, trying to make that magazine work. I mostly failed – but it’s up and running strong now.)
In any case, one question in particular struck me. The writer is preparing a piece on the future of search. (I’ll link to it when it comes out). What big problems, he asked, still plague search?
That got me thinking. Here’s my answer:
The largest issue with search is that we learned about it when the web was young, when the universe was “complete” – the entire web was searchable! Now our digital lives are utterly fractured – in apps, in walled gardens like Facebook, across clunky interfaces like those in automobiles or Comcast cable boxes. Re-uniting our digital lives into one platform that is “searchable” is to me the largest problem we face today.
It may be worth expanding on that sentiment. When it broke out in the mid 1990s, the web was society’s first at-scale digital artifact. It spread in orders of ten, first thousands, then millions, then hundreds of millions of pages – and on it went, to the billions it now encompasses. Everybody wanted to “be” on the web – a creator class started making pages and companies and services, a consumer class started “surfing” this vast new digital object, and our collective conscience marvelled at what we had created together: millions of small pieces loosely joined. And the key and unappreciated point is this: those pieces were indeed joined.
It was that joining – through links, of course – that made search possible, that created what is unquestionably the most powerful and lasting new company of the past 20 years – Google.* But as I wrote in Why Hath Google Forsaken Us? A Meditation, Google’s core model – built on the open, linked world of the web – is under threat from the advance of the iPhone and the app, the Facebook and the Path, the automobile console, the Xbox, the cable box, and countless other “unlinked” digital artifacts.
Google knows this. Why else invest so much in Android, in Google+, in Motorola (it’s not just phones, it’s also cable boxes), in self-driving cars, for goodness sake? Google wants a foothold wherever digital information is created and shared, and man, are we creating a sh*t ton of it. Problem is, we’re not making it easy – or even possible – to link all this stuff together, should we care to.
Which takes me back to that core question the Wired reporter asked me: What’s the biggest problem plaguing search? In short, it’s that our digital world is no longer small pieces loosely joined. It’s also big chunks separate and apart. And that makes search – in its most broadest interpretation – damn near impossible.
Which leads to another question: What then, is search? Of course, the Wired reporter asked me that as well. My answer:
Search is now more than a web destination and a few words plugged into a box. Search is a mode, a method of interaction with the physical and virtual worlds. What is Siri but search? What are apps like Yelp or Foursquare, but structured search machines? Search has become embedded into everything, and has reached well beyond its web-based roots.
So we all search now, all the time, across all manner of artifacts, large and small. But our searches are not federated – we can’t search across these repositories, as we could across that wonderful, vast, loosely joined early world of the web. We’ve lost the connection.
Call me a fool, but I think the need for that connection will be so strong, that in time, we’ll sew all our digital artifacts back together again. At least, I certainly hope we will. Right now, it ain’t looking so likely – what with patent wars, wagon circling by big platforms, and the like. But I’m an optimist – and I hope you are as well.
* Sorry but Facebook isn’t there – yet. And Microsoft and Apple, well, they may make a play for that crown either 20 years ago or 20 years hence, but if you ask me for the most important company ever that launched as a native web business, the answer is indisputably Google.
I’ve spent a lot of time thinking about data recently. It’s not just reading books like The Information or Mirror Worlds (or Super Sad True Love Story, a science fiction novel that is both compelling and scary), it’s my day to day work, both at FM (where we deal with literally 25 billion ad calls and associated data a month), and in reporting the book (I’ve been to MIT, Yale, Amazon, Microsoft, Facebook, Google, and many other places, and the one big theme everyone is talking about is data…).
We are, as a society and as individuals, in the process of becoming data, of describing and detailing and burnishing our dataselves. And yet, we haven’t really joined the conversation about what this all means, in the main because it’s so damn abstract. We talk about privacy and fear of big brother, or big corporations. We talk about Facebook and whether we’re sharing too much. But we aren’t really talking – in any scaled, framed way – about what this means to being humans connected in a shared society, to be in relationships, to be citizens and consumers and lovers and haters….
There are so many wonderful micro conversations going on about this topic, spread out all over the place. I’m hoping that when my book appears, it might be a small step in joining some of these conversations into a larger framework. That’s the dream anyway.
….most people have no idea what the cloud is, have pretended to know what it means on first dates, and yet effectively all respondents are active cloud computing users.
And that’s the way this stuff should work.
I get the point, but in a sense, I utterly disagree. If we as as society do not understand “the cloud,” in all its aspects – what data it holds, how it works, what the bargains are we make as we engage with it, we’ll all be the poorer for it, I believe. (For one aspect of this see my post on the Cloud Commit Conundrum). More on this as the Fall approaches, and I settle into a regular habit of writing out loud for the book.
So, casually reading through this Fast Company story about sexy female Twitter bots, I come across this astounding, unsubstantiated claim:
My goal was to draw a straight line from a Twitter bot to the real, live person whose face the bot had stolen. In the daily bot wars–the one Twitter fights every day, causing constant fluctuations in follower counts even as brands’ followers remain up to 48% bot–these women are the most visible and yet least acknowledged victims…
There it was, tossed in casually, almost as if it was a simple cost of doing business – nearly half of the followers of major brands could well be “bots.”
The article focuses on finding a pretty woman whose image had been hijacked, sure, but what I found most interesting (but sadly unsurprising) was how it pointed to a site that promises to a thousand followers to anyone who pays…wait for it…about $17. Yes, the site is real. And no, you shouldn’t be surprised, in the least, that such services exist.
It has always been so.
Back when I was reporting for The Search, I explored the gray market that had sprung up around Google (and still flourishes, despite Google’s disputed attempts to beat it back). Fact is, wherever there is money to be made, and ignorance or desperation exists in some measure, shysters will flourish. And a further fact is this: Marketers, faced with CMO-level directives to “increase my follower/friend counts,” will turn to the gray market. Just as they did back in the early 2000s, when the directive was “make me rank higher in search.”
Earlier this week I got an email from a fellow who has been using Facebook to market his products. He was utterly convinced that nearly all the clicks he’s received on his ad were fake – bots, he thought, that were programmed to make his campaigns look as if they were performing well. He was further convinced that Facebook was running a scam – running bot networks to drive performance metrics. I reminded him that Facebook was a public company run by people I believed were well intentioned, intelligent people who knew that such behavior, if discovered, would ruin both their reputation as well as that of the company.
Instead, I suggested, he might look to third parties he might be working with – or, hell, he might just be the victim of a drive-by shooting – poorly coded bots that just click on ad campaigns, regardless of whose they might be.
In short, I very much doubt Facebook (or Twitter) are actively driving fraudulent behavior on their networks. In fact, they have legions of folks devoted to foiling such efforts.Yet there is absolutely no doubt that an entire, vibrant ecosystem is very much engaged in gaming these services. And just like Google had at the dawn of search marketing, Twitter and Facebook have a very – er – complicated relationship with these fraudsters. On the one hand, the gray hats are undermining the true value of these social networks. But on the other, well, they seem to help important customers hit their Key Performance Indicators, driving very real money into company coffers, either directly or indirectly.
I distinctly recall a conversation with a top Google official in 2005, who – off the record – defended AdSense-splattered domain-squatters as “providing a service to folks who typed the wrong thing into the address bar.” Uh huh.
As long as marketers are obsessed with hollow metrics like follower counts, Likes, and unengaged “plays,” this ecosystem will thrive.
What truly matters, of course, is engagement that can be measured beyond the actions of bots. It is coming. But not before millions of dollars are siphoned off by the opportunists who have always lived on the Internet’s gray edge.
Early last month I wrote a piece entitled Do Not Track Is An Opportunity, Not a Threat. In it I covered Microsoft’s controversial decision to incorporate a presumptive “opt out of tracking” flag in the next release of its browser, which many in the ad industry see as a major blow to the future of our business.
In the piece, I argued that Microsoft’s move may well force independent publishers (you know, like Searchblog, as well as larger sites like CNN or the New York Times) to engage in a years-overdue dialog with their readers about the value exchange between publisher, reader, and marketer. I laid out a scenario and proposed some language to kick that dialog off, but I gave short shrift to a problematic and critical framing concept. In this post, I hope to lay that concept out and offer, by way of example, a way forward. (Caveat: I am not an expert in policy or tech. I’ll probably get some things wrong, and hope readers will correct me if and when I do.)
The “concept” has to do with the idea of a first-party relationship – a difficult to define phrase that, for purposes of this post, means the direct relationship a publisher or a service has with its consumer. This matters, a lot, because in the FTC’s recently released privacy framework, “first-party marketing” has been excluded from proposed future regulation around digital privacy and the use of data. However, “third-party” marketing, the framework suggests, will be subject to regulation that could require “consumer choice.”
OK, so in that last sentence alone are three terms, which I’ve put in quotes, that need definition if we are going to understand some pretty important issues. The most important is “first-party marketing,” and it’s damn hard to find a definition of that in the FTC document. But if you go back to the FTC’s *preliminary* report, issued in December of 2010, you can find this:
First-party marketing: Online retailers recommend products and services based upon consumers’ prior purchases on the website.
Later in the report, the term is further defined:
Staff proposes that first-party marketing include only the collection of data from a consumer with whom the company interacts directly for purposes of marketing to that consumer.
And in a footnote:
Staff also believes that online contextual advertising should fall within the “commonly accepted practices” category (Ed. note: Treated as OK, like first party marketing). Contextual advertising involves the delivery of advertisements based upon a consumer’s current visit to a web page or a single search query, without the collection and retention of data about the consumer’s online activities over time. As staff concluded in its 2009 online behavioral advertising report, contextual advertising is more transparent to consumers and presents minimal privacy intrusion as compared to other forms of online advertising. See OBA Report, supra note 37, at 26-27 (where a consumer has a direct interface with a particular company, the consumer is likely to understand, and to be in a position to control, the company’s practice of collecting and using the consumer’s data).
The key issue here for publishers, as far as I can tell, is this: “the delivery of advertisements based upon a consumer’s current visit to a web page or a single search query, without the collection and retention of data about the consumer’s online activities over time…where a consumer has a direct interface with a particular company, the consumer is likely to understand, and to be in a position to control, the company’s practice of collecting and using the consumer’s data.”
Whew. OK. We’re getting somewhere. Now, when that 2010 report came out, many in our industry freaked out, because of the next sentence, one which refers to – wait for it – third party marketing:
If a company shares data with a third party other than a service provider acting on the company’s behalf – including a business affiliate unless the affiliate relationship is clear to consumers through common branding or similar means – the company’s practices would not be considered first-party marketing and thus they would fall outside of “commonly accepted practices” … Similarly, if a website publisher allows a third party, other than a service provider, to collect data about consumers visiting the site, the practice would not be “commonly accepted.”
Now, this was a preliminary report, and the final report, which as I said earlier came out this past Spring, incorporates a lot of input from companies engaged in what the FTC described as “third party” marketing – companies like Google that were very concerned that the FTC was about to wipe out entire swathes of their business. And the fact is, it’s still not clear what’s going to be OK, and what isn’t. For now, my best summary is this: it’s OK for websites that have a “first party” relationship to use data collected on the site to market to consumers. If, however, those sites was to let “third parties” market to consumers, then, at some point soon, the sites need to figure out a way to give “consumers a choice” to opt out. If they don’t, they may be subject to regulation down the road.
Which brings us back to “Do Not Track,” or DNT. Now DNT has been held up as the easiest way to give consumer a choice about this issue – if a consumer has DNT enabled on their browser, then that consumer has very clearly made a choice – they don’t want third-party advertisements or data collection, thank you very much. See how easy that was?
Wrong, wrong, wrong!!! As implemented by Microsoft in IE 10, DNT is an extremely blunt instrument, one that, in fact, does *not* constitute a choice. It’s defaulted to “on,” which means that a consumer is not ever given a choice one way or the other. And once it’s on, it’s the same for every single site – which means you can’t say that you’re fine with third-party ads on a site you love (say, Searchblog, naturally), but not fine with a site you don’t like so much (say, I dunno, You Got Rick Rolled).
That’s pretty lame. Shouldn’t we, as consumers, be able to chose which sites we trust, and which we don’t? That’s pretty much the point of my post on DNT last month.
Fact is, we don’t really have a way to demonstrate that trust. Many in the industry – including the IAB, where I am a board member – are working to clarify all this with the FTC. The working assumption is that it’s far too much to ask of most publishing sites to give consumers a choice, much less give them access to the data used to “target” them.
Well, I’m not so sure about that.
Check out this screen shot from independent site GigaOm (yes, FM works with GigaOm):
A few other sites are starting to do similar notices – and I applaud them (this is already becoming standard practice in the UK, due to strict regulations around cookies). GigOm is saying, in essence, that by simply continuing to read the site, you agree to their privacy policies. Now, take a look at what GigaOm’s policy has to say about “third party advertising:”
GigaOM may allow third party advertising serving companies, including ad networks (“Advertisers”), to display advertisements or provide other advertising services on GigaOM. These third party Advertisers may use techniques other than HTTP cookies to recognize your computer or device and/or to collect and record demographic and other Information about you, including your activities on or off GigaOM. These techniques may be used directly on GigaOM….Advertisers may use the information collected to display advertisements that are tailored to your interests or background and/or associate such information with your subsequent visits, purchases or other activities on other websites. Advertisers may also share this information with their clients or other third parties.
To summarize: By reading GigaOm, you’ve made a choice, and that choice is to let GigaOm use third-party advertising. It’s a nifty move, and one I applaud: GigaOm has just established you as a first party to its content and services just like….
….Facebook, which just announced revenue of more than a billion dollars last quarter. Facebook, of course, has a first-party relationship with 955 million or so of us – we’ve already “opted in” to its service, through the Terms of Service we’ve all agreed to (and probably not read.) We’ve made a choice as consumers, and we’ve chosen to be marketed to on Facebook’s terms.
The same is true of Apple, Amazon, eBay, Yahoo, and any number of other large services which require registration and acceptance of Terms of Service in order for us to gain any value from their platforms. Google and Microsoft have been frantically catching up, getting as many of us as they can to register our identity and agree to a unified TOS in some way.
But what about independent publishers? You know, the rest of the web? Well, save folks like GigaOm (and AllThingsD, which warns its audience about cookies), we’ve never really paid attention to this issue. In the past, publishers have avoided doing anything that might get in the way of an audience consuming their content – it’s a death sentence if you’re engaged in the high holy art of Increasing Page Views. And bigger publishers like Time or Conde Nast don’t want to rock the boat, they’ll wait till a consensus forms, and then follow it.
But I like what GigaOm has done. It’s a very clear notice, it goes away after the first visit, and it reappears only if you’ve cleared your cookies (which happens a lot if you run an anti-virus program).
I think it’s time the “rest of the web” follows their lead. We rely on third-party advertising services (like FM) to power our sites. We live in uncertain times as it relates to regulation. And certainly we have direct relationships of trust with our audiences – or you wouldn’t be reading this far down the page. It’s time the independent web declares the value of our first-party relationships with audiences, and show the government – and our readers – that we have nothing to hide.
I plan to look into ways we might make easily available the code and language necessary to enact these policies. I’ll be back with more as I have it….
*Now, the other two terms bear some definition as well. I think it’s fair to say “consumer choice” means “give the consumer the ability to decide if they want their data used, and for what purposes,” and “third party marketing” means the use of data and display of commercial messages on a first party site by third-party companies – companies that are not the owner of the site or service you are using.
Look, I’m not exactly a huge fan of grimy greenbacks, but I do feel a need to point out something that most coverage of current Valley darling Square seems to miss: The “Death of Cash” also means the “death of anonymous transactions” – and no matter your view of the role of government and corporations in our life, the very idea that we might lose the ability to transact without the creation of a record merits serious discussion. Unfortunately, this otherwise worthy cover story in Fortune about Square utterly ignores the issue.
And that’s too bad. A recent book called “The End of Money” does get into some of these issues – it’s on my list to read – but in general, I’ve noticed a lack of attention to the anonymity issue in coverage of hot payment startups. In fact, in interviews I’ve read, the author of “The End of Money” makes the point that cash is pretty much a blight on our society – in that it’s the currency of criminals and a millstone around the necks of the poor.
Call it a hunch, but I sense that many of us are not entirely comfortable with a world in which every single thing we buy creates a cloud of data. I’d like to have an option to not have a record of how much I tipped, or what I bought at 1:08 am at a corner market in New York City. Despite protections of law, technology, and custom, that data will remain forever, and sometimes, we simply don’t want it to.
What do you think? (And yes, I am aware of bitcoin…)
BTW, this mini-rant is very related to my last post: First, Software Eats the World, Then, The Mirror World Emerges.
(image Edge.org) A month or so ago I had the pleasure of sitting down with Valley legend Marc Andreessen, in the main for the purpose of an interview for my slowly-developing-but-still-moving-forward book. At that point, I had not begun re-reading David Gelernter’s 1991 classic Mirror Worlds: or the Day Software Puts the Universe in a Shoebox…How It Will Happen and What It Will Mean.
Man, I wish I had, because I could have asked Marc if it was his life-goal to turn David’s predictions into reality. Marc is well known for many things, but his recent mantra that “Software Is Eating the World” (Wall St. Journal paid link, more recent overview here) has become nearly everyone’s favorite Go-To Big Valley Trend. And for good reason – the idea seductively resonates on many different levels, and forms the backbone of not just Andreessen’s investment thesis, but of much of the current foment in our startup-driven industry.
A bit of background: Andreessen’s core argument is that nearly every industry in the world is being driven by or turned into software in one way or another. In some places, this process is deeply underway: The entertainment business is almost all software now, for example, and the attendant disruption has created extraordinary value for savvy investors in companies like Amazon, Netflix, and Apple. Further, Marc points out that the largest company in direct marketing these days is a software company: Google. His thesis extends to transportation (think Uber but also FedEx, which runs on software), retail (besides Amazon, Walmart is a data machine), healthcare (huge data opportunity, as yet unrealized), energy (same), and even defense. From his Journal article:
The modern combat soldier is embedded in a web of software that provides intelligence, communications, logistics and weapons guidance. Software-powered drones launch airstrikes without putting human pilots at risk. Intelligence agencies do large-scale data mining with software to uncover and track potential terrorist plots.
That quote reminds me of Wired’s first cover story, in 1993, about the future of war. But in 1991, two years before even that watershed moment (well, for me anyway), Yale scholar Gelernter published Mirror Worlds, and in it he predicted that we’d be putting the entire “universe in a shoebox” via software. Early in the book, Gelernter posits the concept of the Mirror World, which might best be described as a more benign version of The Matrix, specific to any given task, place, or institution. He lays out how such worlds will come to be, and declares that the technology already exists for such worlds to be created. “The software revolution hasn’t begun yet; but it will soon,” he promises.
As we become infinite shadows of data, I sense Gelernter is right, and VCs like Andreessen and the entrepreneurs they are backing are leading the charge. I’ll be reviewing Mirror Worlds later in the summer – I’m spending time with Gelernter at this home in New Haven next month – but for now, I wanted to just note how far we’ve come, and invite all of you, if you are fans of his work, to help me ask Gelernter intelligent questions about how his original thesis has morphed in two decades.
It seems to me that if true “mirror worlds” are going to emerge, the first step will have to be “software eating the world” – IE, we’ll have to infect our entire physical realities with software, such that those realities emanate with real time and useful data. That seems to be happening apace. And the implications of how we go about architecting such systems are massive.
One of my favorite passages from Mirror Worlds, for what it’s worth:
The intellectual content, the social implications of these software gizmos make them far too important to be left in the hands of the computer sciencearchy…..Public policy will be forced to come to grips with the implications. So will every thinking person: A software revolution will change the way society’s business is conducted, and it will change the intellectual landscape.