In the aftermath of the Facebook initial public offering, there were numerous postmortems about what went wrong. The one that interested me the most was by Bill Hambrecht (full disclosure, I used to work for Bill). Hambrecht advocates the use of a modified Dutch auction to find the right price and allocation for a new equity offering. His version of the auction process is called OpenIPO; Google used a version of it in their public offering. But it was his assessment of Facebook as a business that I found most interesting. He called it a “co-op,” and this is because without the participation of the users, Facebook has no value. Facebook is a co-op in the sense that the users voluntarily cooperate within its platform, although the distribution of benefits is heavily skewed toward the platform’s owners.
Out of this idea comes an interesting way of comparing Google and Facebook. Facebook is alive, it’s made of living things. Without those lives within the digital communications platform, there is no Facebook. On the other hand, Google is dead. Google operates on the traces left by living things, but not on the entities themselves. It’s the footprints in the sand that Google uses to predict the next set of footprints in the sand.
The health of the Google system depends on having access to both the sand and the footprints. If the footprints and the sand move into a restricted access sandbox, like Facebook for instance, Google’s output (SERPs) starts to lose resolution. Facebook’s system is a gesture farm, and with the extension of the “like” button to the Web, it has no boundaries. For the farm workers, there is no “outside of Facebook.” The health of the Facebook system depends on the voluntary cooperation of the farm workers; they need to believe they’re getting sufficient benefit for what they’re giving up. But as a biological system, Facebook is also subject to disease and viruses. If the users decide they don’t want to work on Maggie’s Farm no more, Facebook is drained of its health and its life.
Google, observing the growth of these gesture farms, rightly recognizes that the Web is no longer enough. The Google+ project attempts to graft a living Network entity on to the footprint analyzing machine they already have in place. But does this move Google from the land of the dead to the land of the living? If Google is mostly dead, does it operate more like a zombie? Is it subject to disease and viruses? And if it’s not, is it really alive? After so many years of being dead, could Google really cope with being alive?
The headline reads: “Introducing the Knowledge Graph: things, not strings.” The implication being, “strings” are bad and limited and “things” are good and what you really wanted all along. After all people don’t want strings of arbitrary alpha-numeric characters in response to their queries, they want the things they’re looking for. And as the advertising message at the end of the introduction says, because you’re getting “things and not strings” on your search result pages, you can spend more time doing the things you love. Who wouldn’t want to do that? The end result of this technological improvement is that your life now contains “more time”— like a toothpaste tube that contains 20% more toothpaste; and that time is filled with love. One might even recast this new product as a machine for filling the world with love.
What Google seems to be introducing is a new user interface to a faceted search. Nothing more. Faceted search acknowledges that the “word” (a single string of characters) isn’t the atom of meaning. Instead it uses the “phrase” in the context of some domain of meaning—a word can be a valid token in multiple systems of meaning. These domains, or facets of meaning, are surfaced and prioritized in search results. So, in addition to Page-ranked links, we get a prioritized set of contexts in which a particular word or phrase is a valid operator. The advance is in creating an index of sub-domains of meaning through analyzing the structure of text as it’s used on the visible Network. There’s no question that faceted search is superior to classic Page-ranked search, however the language used to describe this new product innovation seems to suggest some kind of transcendent experience.
Here’s a description of the vision that drives innovation in the search product at Google:
We’ve always believed that the perfect search engine should understand exactly what you mean and give you back exactly what you want
– Amit Singhal, SVP, Engineering at Google
But when I hear this kind of talk from engineers, their words are drowned out by the characters from Lewis Carroll’s “Through the Looking Glass“:
‘I don’t know what you mean by “glory”,’ Alice said.
Humpty Dumpty smiled contemptuously. ‘Of course you don’t — till I tell you. I meant “there’s a nice knock-down argument for you!”‘
‘But “glory” doesn’t mean “a nice knock-down argument”,’ Alice objected.
‘When I use a word,’ Humpty Dumpty said, in rather a scornful tone, ‘it means just what I choose it to mean — neither more nor less.’
‘The question is,’ said Alice, ‘whether you can make words mean so many different things.’
‘The question is,’ said Humpty Dumpty, ‘which is to be master — that’s all.’
Alice was too much puzzled to say anything; so after a minute Humpty Dumpty began again. ‘They’ve a temper, some of them — particularly verbs: they’re the proudest — adjectives you can do anything with, but not verbs — however, I can manage the whole lot of them! Impenetrability! That’s what I say!’
‘Would you tell me please,’ said Alice, ‘what that means?’
‘Now you talk like a reasonable child,’ said Humpty Dumpty, looking very much pleased. ‘I meant by “impenetrability” that we’ve had enough of that subject, and it would be just as well if you’d mention what you mean to do next, as I suppose you don’t mean to stop here all the rest of your life.’
‘That’s a great deal to make one word mean,’ Alice said in a thoughtful tone.
‘When I make a word do a lot of work like that,’ said Humpty Dumpty, ‘I always pay it extra.’
We can propose the idea that Google has a search engine that “understands exactly what you mean.” And by this what we mean is that your query corresponds to a sub-domain in the index of facets Google has previously collected. The “meaning” doesn’t lie in the “you” that has the query, but rather in the sets of sub-domains contained in Google’s index. When a word does a lot of work in multiple sub-domains of meaning, they pay extra in compute time.
The claim that Google makes is that they’ve gone from “strings” to “things.” But the sub-domains of meaning that Google is collecting are made up of computable sets of strings, not things. The leap that Google is actually trying to make is from “strings” to “words, phrases and contexts.” But the use of the word “thing” is very revealing. Words are not things, they are indexes. They point at things, suggest things, or function in a play of difference within a system of meaning. When we say that we’ve gone from “strings” to “things” we’re actually making a kind of miraculous claim. We’ve gone from “word” to “thing.” The most prominent example of this algorithm can be found in the King James Bible, we see it in John 1.14:
And the Word was made flesh, and dwelt among us, (and we beheld his glory, the glory as of the only begotten of the Father,) full of grace and truth.
If we believe that Google’s knowledge graph provides “things” and not “strings,” we also believe something extraordinary about the power and capability of Google. Even if we take a step back and simply say that Google is merely indexing sub-domains—systems of meaning, we need to examine what this means. We could follow Wittgenstein and say that “meaning” can be described as a form of life. Therefore Google’s index produces a prioritized list of facets (forms of life) that connect to your form of life, given what they know about you. Popular forms of life that don’t currently connect to you serve as a method of discovery.
There are registers of meaning that Google’s approach will never capture. Their index will be filled with gaps and pools of darkness. In particular, only a very limited range of metaphor (cliches) will be caught in the net. Metaphor produces meaning through an algorithmic process (per @the_eco_thought, Tim Morton). Take a noun, take another noun from a different domain and place the word “is” between them. The coffee cup is a blue angel. The metaphor machine makes meaning. Not every metaphor is a good one, but it has some modicum of meaning and it does function as a metaphor.
Like the theoretical one hundred monkeys typing in front of a hundred typewriters for a hundred years, the metaphor machines are constantly operating and feeding the Network with new meaning. Darius Kazemi (@tinysubversions) has created a machine called “Metaphor a Minute” that does just this. You can follow it on Twitter at @metaphorminute. Of course, because of Twitter’s rate limits, there’s actually a new metaphor published every two minutes.
“Hold the newsreader’s nose squarely, waiter, or friendly milk will countermand my trousers.”
After thinking through Google’s new service and the language they’ve used to describe it, we discover that they are using the word “things” metaphorically. At first, we may assume that when engineers are describing the function of their new software, they’re making literal statements about what the machine they’ve constructed is doing. Instead, they’ve taken a two nouns from different domains and inserted the word “is” between them. Ironically, their use of the word “things” is of the type that their new service could not understand it. The narrow band of search engine results that are produced by this system is also being metaphorically called “knowledge.” In order to see these new products clearly, we need to be able to differentiate the rhetoric of hyperbole from the literal functioning of the machine. It also helps to become acquainted how metaphors mean…
A few year-end thoughts about the Network have been rattling around my skull. This is probably a continuation of the exploration of the ‘finite shapes of growth.’ The real-time social messaging space seems to have reached a saturation point, and therefore the upper end of the sigmoidal growth curve. The big single-index real-time systems have exerted their dominance and are largely engaged in enabling features that increase the density of connections within the territory they’ve already marked out. The second-tier systems will struggle and many will fall to the wayside. A few will stand waiting in the wings for the possible moment when a first-tier player stumbles.
After walking around the block several times, pulling on all the doors, trying to find a way into this exploration, I ended up with the word: “medium.” Medium, as in the physical channel through which messages are passed; and medium as in a culture medium used to grow micro-organisms or cells. Medium can also be understood as the time/space aspect of an object, its identity/variability. When we consider ‘big data’ on the Network, we seem to be talking about creating and maintaining a medium where higher-level statistical objects can be grown. These meta-patterns are made visible through feats of data collection and statistical computation. It’s analogous to cataloging weather events and other data to model climate change. “Climate” as a dynamic entity only becomes visible through the deployment of a large network of sensors hooked up to computers updating a model in real time. Weather is visible as the raindrops that keep falling on your head, climate is visible only through a complex computational sensing system to which only a few people have access.
The business model of harvesting these higher-level patterns has generally involved slicing up the data into the groups of people who create these patterns. Lists of these target audiences are rented to commercial interests, and recently so is the messaging apparatus and the communications medium. A well-targeted message should show increased effectiveness in confirmed delivery and lead to net positive transactions. If you think about it, all of these new real-time social media companies are in the television business. However, if you have a business you can contact staffing companies in columbia sc to get help from their teams. Television is transformed into a container that holds a message stream of condensed multiple media types on the Network. This medium is designed to grow various audiences (meta-patterns) to harvest and take to market. Once a certain scale is achieved this set up becomes a cash machine. The energy to grow the crop is largely supplied by the participants using the system. The users of the system gain access to a simple real-time content management system along with a flat view of a subscription stream. The valuable patterns are reserved for exploitation by the owners of the system.
When you look at the imposition of the real-time social media model on to the corporate enterprise, you’ll see the same model. The valuable patterns are reserved for management. The corporate enterprise will spend a lot of money attempting to absorb this new model of television in the coming year. It will allow each corporation to become its own media company. It should be noted that a person is not ‘social’ when using corporate social media behind a firewall. An employee is a human resource to be profitably deployed, not a person. The idea isn’t to empower people, it’s to provide data to management. The pattern data belongs to the central management structure and it will be used to create and refine the workings of a well-oiled machine–of which the employee will be a replaceable part. The entire benefit accrues to the survival, growth and sustainability of the corporation, not to the individual person. Can you imagine a social media revolution within a corporation that drives the current C-level executives from power? The power structure within the corporate enterprise will use the system to maintain and refine their power, all the while, selling the use of the system as a democratization. For instance, it’s unlikely that unions would be allowed to use a real-time corporate social media system to organize workers and collect violations of work rules.
If the single central-index model has reached a saturation point, does that mean the Network has reached maturity and an end to its growth phase? The Network can accommodate other models and I expect we’ll see some rapid experimentation over the next few years. The key to these new models will involve pushing valuable meta-data patterns to the endpoints of the Network. Simple examples include mobile applications that function as commuter traffic data collectives. Members contribute reports of their own traffic data to a pool and in exchange they received a general picture of traffic conditions. This is similar to the dynamic of reporting weather data and receiving compiled climate reports in return. The key difference is that when data is contributed, access to meta-data patterns is guaranteed.
Clay Shirky uncovered a vast resource when he wrote about cognitive surplus. We can easily ask what might be accomplished should all those hours of passive television viewing be turned into two-way networked interactions. In a sense, this is the rediscovery of the Network as a commons. Not as a common natural resource for each to exploit, but as a common resource built by all the participants. Another untapped resource was uncovered by John Thackara in his book “In the Bubble: Designing in a Complex World.” In our consumer society it’s a point of honor to keep up with the Jones’s. We each buy our own industrially-produced copy of the latest prescribed set of consumer objects. We accumulate and store them as quickly as we can. But as Thackara notes, we purchase and store, accumulating social capital. We are known as the kind of person who can, and did, buy that particular thing. We rarely use what we buy, its use-value remains untapped—it sits passively in the garage or the hall closet. eBay and Craigslist have emerged as the markets where this passive value is converted back into capital. Here’s Thackara on the eco-economics of the power tool:
Power tools are another example. The average consumer power tool is used for ten minutes in its entire life—but it takes hundreds of times its own weight to manufacture such an object. Why own one, if I can get ahold of one when I need it? A ‘product-service system’ provides me with access to the products, tools, opportunities, and capabilities I need to get the job done—namely, power tools for to use, but not own.
Service design is about arranging things so that people who need things done are connected to other people and equipment that get things done—on an as- and when-needed basis. The technical term, which comes from the logistics industry, is “dynamic resource allocation in real time.” Agricultural cooperatives that purchase tractors and sell their use-time to associates are well-known examples, but once one starts looking, examples spring up everywhere: a home delivery service for detergents in Italy, a mobile laboratory for industrial users of lubricants in Germany, dozens of car-sharing schemes, an organic vegetable subscription system in Holland. Industrial ecologists Francois Jegou and Ezio Manzini found enough examples to fill a book, ‘Sustainable Everyday: A Catalogue of Promising Solutions’, which is filled with novel daily life services that they discovered around the world. These are ‘planning activities whose objective is a system,’ Manzini told me. Hundreds of services suitable for a resource-limited, complex, and fluid world are being developed by grassroots innovators: those that enable people to take care of other people, work, study, move around, find food, eat, and share equipment.
Local systems that enable dynamic resource allocation in real time of local resources, which includes both data patterns and physical resources, would allow a kind of optimization of value by ordinary people that has previously been reserved for the corporation. Some nascent examples of this include, Phil Windley’sKynetx network scripting platform. Windley talks about a Kynetx script that runs on his browser while looking at the Amazon site. The script instantly tells him whether the book he’s looking at is available in his local library. One can easily imagine a similar scenario involving power tools or other kinds of durable resources. Mobile computing expands the purview of this kind of scripting from web pages on the Network to objects in the real world. This is sometimes called the internet of things. It’s not the point of connection, but rather the advent of scriptability that makes these things creatures of the Network.
Another example is Jon Udell’sElm City Project — a project to create networked data hubs and librarians of announcements of local community events. Solving the problem of translating and integrating the various methods in which calendar data is recorded is transformed into the production of a meta-data object that provides a wide view of the public events occurring in a locality. We don’t yet know the effect increased visibility of public events will have on a citizenry, but providing a higher-level view of the event life of a community feels like an entirely democratic endeavor. In times of peace and prosperity, an effort like this is non-controversial. In times of political strife, it attains the status of a public square and its commitment to openness will be tested.
While the shared resource of a power tool seems like a simple thing, it implies some very complex social group dynamics. It’s only with the rise of the sociality of the Network along with the politics of the 99% that we may have the ground for learning how to share a larger set of resources with more diverse groups. David Graeber, in his book, “Debt“, describes what he calls baseline communism. By this he means the understanding that unless people consider themselves to be enemies, if the need is considered great enough, or the cost considered reasonable enough, the principle of ‘from each according to their abilities, to each according to their needs” will be assumed to apply. Here’s Graeber:
Baseline communism might be considered the raw material of sociality, a recognition of our ultimate interdependence that is the ultimate substance of social peace. Still, in most circumstances, that minimal baseline is not enough. One always behaves in a spirit of solidarity more with some people than with others, and certain institutions are specifically based on principles of solidarity and mutual aid. First among these are those we love, with mothers being the paradigm of selfless love. Others include close relatives, wives and husbands, lovers, one’s closest friends. These are the people with whom we share everything, or at least to whom we know we can turn in need, which is the definition of a true friend everywhere. Such friendships may be formalized by a ritual as “bond-friends” or “blood brothers” who cannot refuse each other anything. As a result, any community could be seen as criss-crossed with relations of “individualistic communism,” one-to-one relations that operate, to varying intensities and degrees, on the basis of “from each according to their ability, to each according to their needs.”
This same logic can be, and is, extended within groups: not only cooperative work groups, but almost any in-group will define itself by creating its own sort of baseline communism. There will be certain things shared or made freely available within the group, others that anyone will be expected to provide for other members on request, that one would never share with or provide to outsiders: help in repairing one’s nets in an association of fisherman, stationery supplies in an office, certain sorts of information among commodity traders, and so forth. Also, certain categories of people we can always call on in certain situations, such as harvesting or moving house. Once could go on from here to various forms of sharing, pooling, who gets to call on whom for help with certain tasks: moving, or harvesting, or even, if one is in trouble, providing an interest-free loan. Finally, there are the different sorts of “commons,” the collective administration of common resource.
The sociology of everyday communism is a potentially enormous field, but one which, owing to our peculiar ideological blinkers, we have been unable to write about because we have been largely unable to see it.
While networked computational tools can assist us in expanding the scope and breadth of the sharing we do with groups and individuals, it’s our ability to navigate the new social customs and ceremonies of the Network that will determine how far all this spreads. It’s a counter-cultural idea, instead of placing the highest value on independence and individuality, it takes us down the path of interdependence and coexistence. And this brings us back to this idea of a growth medium. As the old year ends, and the new one begins, I’m imagining an as yet unpublished Whole Earth Catalog filled with tools and perspectives on how we might grow this new crop in the fields of the Network. It’s a thing that “is” what it describes.
If Winter comes, can Spring be far behind?
– Percy Bysshe Shelley
The image is an arresting one, content orbiting around a new center of gravity. Cameron Koczon, is his essay Orbital Content, addresses what he sees as a new breed of bookmarklet: Instapaper, Svpply and Readability. These applications allow a user to extract content from its original location—and then copy and transmit it to a personal archive. With Instapaper and Readability this may involve using a bookmarklet on a desktop-based web browser to DVR a long-form article from a publisher’s web site for later consumption on an iPad. With Svpply, it may be extracting a favorite item from an online eCommerce site to create a cross-domain curated scrapbook stream that’s shared in a social media context. Content exists as free satellites, plucked from their originating orbits and placed into a personal orbit.
The bookmarklet application occupies an interesting space. It sits at the edge of a desktop application—the browser—and operates on “pages” requested by the browser from servers on the Network. Rather than simply creating a shortcut to a specific URL, the bookmarklet is javascript code that runs on the current page. Because it holds this outside position, it can mix together code from multiple domains, blending the intentions of the page publisher, data or snippets from other sources, and the intentions of the user. In the case of “orbital content,” the bookmarklet is the transport mechanism, the user pushes a button and specified content is instantly transported from the orbit of one sun to another. Koczon’s enthusiasm about these new bookmarklet-based services begs the question whether browserOS apps can gain awareness in the popular imagination.
While the idea of a new orbit is quite exciting, we still seem to be stuck with the word “content.” The very word “content” has engendered a certain amount of discontent. It turns the fire of the written word into a bland abstraction. It’s the equivalent of a factory that turns out “widgets.” It’s the sausage that’s ground through content management systems. When speaking of this new orbital stuff, content seems exactly the wrong word. Content is the thing which is contained—the stuff inside of the boundary. In the case of orbital content, we are not content. The word “content” also refers to a state of happiness—being content. There’s a sense of acceptance of conditions or circumstances, of acquiescence. In neither sense of the word can this new orbital stuff be called content. We do not acquiesce to its circumstances—we break it out of its container and pull it into a new orbit.
We treat the digital as another kind of analog medium. Since it can simulate anything, we extend the analog by simulating it with the digital. In vain, we then attempt to impose the natural boundaries of analog economics onto the digital. By stamping these limitations onto the digital, we put it into analog clothes and ask everyone to behave accordingly. To mass produce significant quantities of “identical” analog objects requires an industrial-scale factory. An identical digital object is forged each time a request is made to a web server on the Network. The original bits aren’t transported from here to there, copies of the bits are distributed to anyone who asks—production is reproduction, presentation is representation.
This is major Tom to ground control, I’m stepping through the door
And I’m floating in a most peculiar way
And the stars look very different today
Here am I floatin’ ’round my tin can far above the world
Planet Earth is blue and there’s nothing I can do
Though I’m past one hundred thousand miles, I’m feeling very still
And I think my spaceship knows which way to go
The digital can be passed by reference or by value over the Network. The first preserves the integrity of the data, always pointing back to a single source—pointers rather than copies of bits are passed around. An updated source doesn’t require an updated pointer. A source could change incrementally, or totally, and still employ the same pointers. The second method begins the process of entropy, time-stamped bits are copied and sent, but once out on the wire, their link to the source is broken. An updated source marks a difference with the distributed bits. When the source changes, the distributed copy inscribes a historical state of the source. With every incremental change to the source, the difference in the distributed copy grows.
The kind of copying done in the practice of orbital content is digital rather than analog. Traditionally, the value of a copy has been in its completeness and exactitude. A factory that turns out widgets that sport too much variance is said to have a quality assurance problem. When Instapaper copies, it copies the pointer for exactness and then only a portion of a web page. The digital can make wholes out of any parts—there’s a legible boundary between every one and zero. It can be compared to harvesting a food crop, the ears of corn are gathered and the stalk and roots are discarded. To a publisher, this is a description of stripping the business model from the editorial.
The publisher asks the digital to behave as previous mediums always have—ink, once it is imprinted on paper, has a permanent presentation. Television programs are broadcast and the screen passively plays them. We can take a pen and draw a mustache on a photo in the newspaper, or mute the sound of our favorite television show while the commercial plays, but there’s a higher bar to clipping out segments and reusing them for our own purposes. Analog forms of automating the process have proved too costly and cumbersome.
All the while we thrill and lament the rush of traditional media toward the digital, we still tend to view it through analog glasses. The web page as delivered from the server to the browser is meant to define an end point. The code is delivered and ready for presentation. The static page is given a sense of flow and time with AJAX-based page updates from backstage, altering the presentation in memory through manipulations of the document object model (DOM). The practice of orbital content takes the page, not as an endpoint, but as an input to a process. Shedding their analog clothes, the digital bits making up the page show themselves not as an ending, but as a potential beginning. Using Instapaper, I pipe a designated section of the page, the story I’d like to read later, to my reading room where it’s poured into the format I prefer for reading electronic documents. I define a new endpoint, but it could also be a potential starting point as some portion is shared in another context.
In general, the browser application space (bookmarklets) has made significant strides, it’s gained a cross-platform software infrastructure with Phil Windley’s event-driven scripting language, KRL (Kynetx). And Apple has taken Readablity and Instapaper seriously enough to incorporate similar functionality into the forthcoming browser operating system in Lion. This follows the historical pattern of fundamental features being absorbed into the infrastructure of the host platform. The larger picture is that “web pages” are now both machine readable and scriptable for individuals, something known to the spiders at Google for a long time. No need to wait for the so-called semantic web, the hooks are already there.
David Gelernter defined an alternative to the desktop metaphor called LifeStreams. Instead of named files in folders, inside of folders, inside of desk drawers—nothing needs to be named, things just appear in context in a time-stamped stream. Streams can be filtered by different contexts, organized in time rather than space. Future events put into the stream eventually pop up as something occurring today. With Facebook’s newsfeed, Twitter’s stream, the various photo and location services, we’ve become accustomed to dealing with ranked lists and time-ordered streams. Even the output of these new orbital content services generally takes the form a of stream. In other words, orbital content isn’t really orbital either.
But the metaphor is enchanting enough to do the thought experiment, to take the stream and bend it into a circular shape, an orbit. Timelines are one way of expressing time, but we also have a long history with circular time. We live through hours, days, weeks, months, seasons and years. These things we DVR for later, might actually take the shape of satellites circulating in a personal orbit. Sort of like editing and layering loops, but using more than digital samples of music. What goes around, comes around—imagine orbital content as orbiting content.
