Archive for April, 2015

The Gap

April 29, 2015

In the AI of the ’80s, the goal was to solve the problem by various means, but mostly by making the problem small enough to solve. It turned out that most problems were too big. Consider that the point of HTML was to feed knowledge to AI machines without spending the money to encode the world’s knowledge on your dime. All this human reading, commerce and ad service was besides the point. Hell, a server log was an accident.

So we start out looking at the world. Actually, the world is large, so we start by focusing and tightening up our scope until we get to a comprehensible world.

00 A World

Yes, we’ll start where Euclid started. Well, he may have started with a point, instead of a line, but lines and points define each other. To get to a single point, we draw another line, not shown this time.

01 00 A Point In A World

We might think of a point, as being the result of an argument. And, while we are arguing we’ll stick with the real world, no concepts allowed. So the argument is all about taxonomy. “You’re an idiot.” “No, I’m not.” But then, idiot would necessitate that such a thing really existed, and no, not the concept of an idiot. Better to name it a rock, so we can keep our argument simple and non-conceptual.

01 01 A Point In A World

But,  somehow, we’ve admitted the concept of an idiot. So now are stuck with maintaining a taxonomy and an ontology. We end up with two worlds: the world of ideas, and the world of realizations. Realizations happen long after we get everyone on the same page as to the idea. There is some spatio-temporal notions of distance and time involved in getting everyone of the same page. And, that is pre-idea. Post-idea, post-implementation, that distance and time is tied up in the technology adoption lifecycle, even if we are talking product as opposed to the technology. Getting back to the taxonomy and ontology involved, they are different and separate worlds.

01 02 A Point In A World

Between those two worlds is a gap. We should be glad the gap is there since it’s where economic value comes from. Products reduce the impedance a constraint presents us with. Products might eliminate that impedance in its entirety. But, Goldratt’s Theory of Constraints tells us that there is always another constraints. We should be happy about that as well, because we won’t run out of work–ever. So why are we unemployed? Well, it’s not globalism, robots, computers, or laziness.

01 03 A Point In A World

But, back to the gap. Those lines are not straight. We might use matrix algebra to straighten them out, but really, they curve. We don’t even try to cross a gap until someone can see or imagine the other side.

01 04 A Point In A World

In the gap, we find value. We make the unknown a little more known. We generate few more bits in the crossing.

01 05 A Point In A World

When we can cross a gap without tossing aside yesterday’s world, when we innovate continuously, we capture cash. When the freight hauling train gets stuff to a port, billing captures some cash, eventually. But, wealth was created when the railroads were built, when railroads were a discontinuous innovation. Railroads might be a bad example, because they were vertically integrated and tended to capture all the cash involved. Today, we are no longer vertically integrated, so the cash is captured by each members of a value chain. Wealth doesn’t get captured in a single set of books. No entity gets all the cash.

One of the core jobs when getting an innovation, a discontinuous innovation, adopted is building that value chain and creating that wealth that feeds the coming cash capture. Too much of what we do today is about cashing out on yesterday’s wealth.

Back to those taxonomies and ontologies, they involve decisions. Those decisions define the terrain. The terrain isn’t even known. On that map of travel times out of New York, you got out west where the map went blank. There was terrain there, but nobody had surveyed it, mapped it, defined the features and the data that we encode in our maps. I’ve drawn the taxonomy and ontology used here as in the leaf nodes attaching to the terrain elevation lines. I’m left wondering if the taxons and ontons, the decisions, are a better place to run the terrain. Do we reach a place, or do we go up and down hills? That question seems to be the distinction between discontinuous and continuous innovation. Did we stop somewhere, or did we keep moving? Did we engage in trench warfare, or the war of fluid tank battles with no rear or forward areas? The point here is that you draw your on taxonomies and ontologies and put the terrain features where you want them. Just use a consistent set of rules for doing so.

01 07 A Point In A World

Once you have your map, you can put your value chain on the terrain as well. Here I’m using circles as a Fourier analysis of the value chain. I’ve followed the Styrofoam cup as microphone notion of saying the circles fit the largest area between the constraining elevations of both the taxonomy and the ontology. We end up with the largest possible circles, the highest frequencies you can get. Now, we might not sense that tightly. We might sense smaller. But, sensing larger is a fail in the game theory sense. We’ve gone too far. We won’t notice, except that our gut instinct will tell us something is wrong.

01 08 A Point In A World

In the figure, the purple points represent the points of contact between our sensors and our terrain. The small circle is our peak. Well, hopefully, it is our peak, because it represents the top of the value chain, where you want to be. The circles are eccentric. That means that depending on their direction of approach a competitor might surprise you.

Enjoy. Comments?

Spatio-temporal Product Management

April 23, 2015

Reading 101 Things I Learned in Fashion School, brought to mind Architectural Drawing, 2nd ed. The latter book talks about architectural drawing in the post-building information system era. CAD system changed drafting and illustration for architects. CAD systems went on to change how buildings got built. CAD systems went beyond all that. CAD systems disrupted drawing as thinking. But after so many generations of Gartner’s hype cycles, drawing made a come back. Drawing found it’s place in a mixed analog-digital process. Analog drawing found its value amid the digital.

One way to think about this is in the sense of value projected, as in the post on projected value. CAD started simply as a different way to draw. CAD went on to be different way to compute, nomography no more. CAD became CAD-CAM. With Eisenman, CAD became an animation and rules-based system that moved buildings from static to dynamic. Each of these projections through other systems.

But, what were these value projections projected through? Space. How long did it take? Time. Value was projected through a terrain having a spatio-temporal reality. CAD had to meet CAM. CAD-CAM had to meet other elements before building information systems were realized.

As a product manager, you can look into the future and make that future happen, or you can let the future surprise you. But, it will be the vendor that reaches out to that next population that gets economics of scale, that gets there first, in the marketing sense, that gets the monopoly position and the immunity that monopoly brings you from the promo spenders. That first company gets to create wealth. The followers get to capture cash. Oh, well.

Once you layout your spatio-temporal product roadmap, you’ll start to see the research agendas that already contribute to your product as spatio-temporal maps. How long will the next advancement of display resolutions take to arrive? How will that impact your offering, beyond richer colors? How will an n time sort algorithm change your product? When net neutrality goes away, how will you code around that?

Now, back to the fashion book. I’ve read other books in this series. Each book tells you about the point of view of a given domain. Design for art, design for fashion, design for the engineer, and design for the MBA are very different things. Each of these domains defines the world its way, aka each has its own culture. I’ve called these functional cultures long before I came across a journal on epistemic cultures.

I’ve gone back to a long ago interest in photography and ran into HDR, high dynamic range photography. I can’t help but wonder how HDR ties to data warehousing, and big data.

As a final note, I’ll answer a question tweeted to me by an old friend. We were tweeting about how globalism needs to be addressed by managers and investors today. We need discontinuous innovation. Continuous innovation won’t bring globalized jobs back to our shores, or make up for all the jobs lost to globalization. So where do you find the companies doing that?  Start by reading the research front of a field you want to invest in, make or get a  spatio-temporal map of that field’s research terrain, know who is who in the field, and harder, know who the students are. Then, watch the sales tax permit applications. Join the community. It will take a lot of work. It won’t be easy, but following Moore’s guerilla game strategy, get there.


More Spatio-Temporal Maps

April 15, 2015

Last night I came across a spatio-temporal map on the web showing travel time back 200 years ago or so. I can’t find the page, so we’ll have to let the when slide. I should have put the citation on the graphic itself. Note to processes.

As a kid, we’d be out of country for a long time. We’d maintain our connections to America via the Sears catalogue, the Armed Forces (radio) Network, and the movies that ran at the base theater. And, something called white bread. I was great stuff unlike white bread you buy at the grocery store today. We’d fly out of McGuire AFB in New Jersey. Once we flew out of somewhere else. But, we’d always fly back to McGuire. We’d end up in Newark, and Brooklyn Heights in NYC. More germane, we’d drive from New Jersey to the Alabama-Florida border where the stateside family lived.

The first time we came back it was on a boat. The next time we went it was on a prop plane. The next time it was on a jet. A week got compressed into a few hours…. But, here we were in New Jersey driving the interstates. They didn’t go everywhere. We never picked the places we were going so they were right off the interstates. No. We drove through lots of little towns. Lots of zigs and zags. Lots of late nights leaving the pitch black, entering the neon-lit burges, plunging back into the pitch black. There was a whole lot of spatio-temporal. Like that country song, “The world must be flat, people leave town and never come back….” Not really, but when they leave a product, they never come back.

So on to the map.

spatio-temporal map

Notice that we are talking land travel times, and earliest possible arrivals. Had they put travel by ship on here, the maps would be different. Travel and communications constraints made for dialect. You talked like all the people in your town. Widening that to say, your state wouldn’t make much sense, since different parts of your state where reached on different days. Your town might be isolated like far northwest South Carolina, which might be on the other side of the Appalachian mountains from the rest of the state. Those mountains were a considerable constraint. You went where the trails went, where the roads went, and before the railroads, you went where the canals and rivers went. Maybe you had a horse and went where the hell, you wanted to go.

Your travels where not a random walk. They were typically Levy flights. You were already in a fractal space. You were already in a world filled with Bayesian priors, unless of course, Joe of Joe’s Garage fame said, “Oh, no sir, you can’t get there from here.” Laugh, except that in the middle of the night, pre-interstate, east bound out of New Orleans, it was true. Dad always got lost there. But, don’t take my word for it. Try driving EB White’s coastal Maine.

Joe was just injecting some Gaussian noise, aka randomness into your Bayesian priors. So why do I not stop for directions?

But we are lucky, someone before us built roads, bridges, and tunnels. That “You can’t get there from here, not today.” was real. Today, we build stores and magazine stands on the side of an internet cable, on top of some API, running in one or more clouds somewhere on who knows how many servers at any given moment. We have a standard platform. We have whole product. When we improve it, those elevation lines on the map get closer together. When we improve it, we create wealth, create a monopoly, capture some cash, and become hero’s for some part of fifteen minutes. Being a hero twice doesn’t get you another fifteen minutes of fame. The terrain is out there. We just don’t have a mapping company willing to take on that project. No areal photos, no maps–apparently.

Notice the unknown, aka six weeks and beyond. There were keywords out there egging people on. The Santa Fe Trail, oh, yeah, a wide patch of grass in an otherwise rocky terrain. Follow the grass. Follow the wagon up ahead. Imagine having a teenager in your wagon. That fuzziness we reach at the six week mark is where that Gaussian normal and our Bayesian normal fight it out for statistical significance. Where our Poisson distributions and associated vectors leave us in the creative moment of creating a generative grammar for those following us. “Yeah, the lore says, cross this river to the south.” So another wagon shifts to the south. The rest of the wagons following that one shift as well. Tomorrow’s wagons, who knows.

Those elevation lines are still relevant today. I know that the drive through Chicago took three hellish hours, and that Erie, PA, I’d call it a day–a long spatio-temporal day of topological transforms. “It looked straight, until I changed maps.” Like, trying to drive west in Maryland or north in Nebraska. “You can’t get there from here.” Well, not until we put the interstate through. When we upgrade our roads, the steep, up and down two-lanes cut the hill tops off for fill, the bottoms get culverts and fill, the lanes multiply–the experience becomes flat. The same thing happened on the internet. Expectations rise. And, our look back machine considers the old stuff as lame, but back then it wasn’t lame–a historical elevation line runs through that.

The map above aggregated the data from tons of travelers. The data wasn’t just where New Yorkers went.

That added day running through north central Alabama was a mountain range between Huntsville and Birmingham. People in Huntsville don’t drive the hour or hour and a half to jobs in Birmingham. Yet, people make those kinds of drives in Houston all the time. Flat matters. In Houston, the roads a likely to be under water. Building culverts matter.

The map shows populations connecting to other populations. Your personal map is covered by a single normal with a lot of vectors an Poisson distributions under it. Look up into the sky. You don’t see the distributions you’re under., not a single one.  The technology adoption lifecycle connects a few highly related populations. The Bayesian terrain is there. The typical corporation have their economies of scale, reusable populations. Reusing populations means discontinuous innovations are out. It means you’re stuck in NYC even as every other non-local elevation line got erased.

Our product road maps tend to be straight lines. Having no topological map, we don’t built s-curves to gently get us over the next mountain. No, we drive straight ahead into the ditch. Well, at least you drove past the party at the rest area.

That one week travel time elevation line has moved all the way to the west coast with pockets still isolated as ever in between.

Innovation eliminates or reduces the impact of a constraint. What you see on the old map was time being lost to physical constraints, constraints on transportation and communications, constraints on getting there and getting done. Knowing what the situation today shows us how successful or innovations in roads, bridges, tunnels, and culverts have been. You don’t have to have taken the road trip or have gotten lost on Hwy-90 to know that those innovations have moved the constraints many times in your lifetime. It happens close to home, right in the street in front of your house, and the utility right-of-way behind your house. It happens in your house, in the walls, in the unused telephone connections, and wifi. It happens right behind your screen. All of it will continue to change. It’s not a matter of putting up another retail store, or serving the consuming masses. Mostly, it’s about getting a map. Find an elevation line and move it.

Lets hope we can get there from here.


Value Projected

April 11, 2015

When marketers talk about value, I’m like, “Oh, please.” None of these markets read Value Merchants or Value-based Fees. Besides the recession is over, so there is less downward pressure on price. Besides everything is free because we monetize everything in every way. Still, value persists independent of price, aka acquisition costs, all those other costs rolled into Gartner’s total cost of ownership (TCO), and all those costs that don’t get reflected in the TCO. And, the need to talk about value persists as well.

Out on twitter, somebody will ask if UX is where value lives. No. Value is out there somewhere.

I’m going to start far away from value and talk about data Then, I’ll talk about data and value at the same time–not that value comes from data. Oh, yeah, IT people think value comes from data.

So lets start with a point.

03 00 A Point

Yeah, that point. We all know that point. But, we might ignore the notion that a point is made. It took Euclid two lines to make a point. These days we use Excel or maybe R to make a point. Then, we use points to make a point, alas a different  species of point. We might keep a point under a distribution for later use.

Actually, a point is born in a sensor. We’ll call a survey a sensor just in case a thermocouple just doesn’t tell you what you need to know.

03 12 A Point - Sensor

But, a lot of sensors require the sensed entity to be illuminated. A camera doesn’t do much in the dark.

03 13 A Point- Illuminator - Sensor

So now we have an illuminator, the point being sensed and a sensor. A political pollster might ask a slanted question, provide slanted answers to that question, and expect you to fill in those dots all before sending those forms to the reader, aka the sensor. That was a lot of illumination to make some bad points.

We might even sense an alias that is related to the thing we really want to sense, but can’t. You were out of town, so we asked your significant other. We’ll put some words in your mouth. It was only ten questions.

03 14 Proxy Point-Relation-Point

I use something I call the triangle model to talk about made things. It’s a decision tree. The base of the triangle is the made stuff. It could be a UX. Or, it could be the illuminator, sensor, and the point we made via a relation. So I redrew the above diagram.

03 15 Made Data

We have three triangles in this figure. I drew another figure with a forth triangle, because the relation was made. But how do these triangles interact? So another figure.

03 16 Triangle Model

In this figure, the illuminator’s triangle is shown in light gray. The thick gray line is the illuminator’s emitter. The sides of the illuminator’s triangle extend outward to the base of the sensor’s triangle. The proxy being sensed must be between the illuminator and the sensor. The relation projects through the sensor data processing it as it goes. After some calculation the relation produces the point.

The figure is demonstrating how the value of the proxy is conveyed through each component of the system eventually providing us with a point that can be used later. The components projected their value into the future into the distance from the functionality, UX, or realization. The point can be seen as the summation of the value the system captured and packaged. The point will be used a long way from the illuminator and sensor. The point will be used a long way from the relation as well. The point won’t tell anyone how it was built, but some of those details are needed before other’s can rely on that point.

Here is another figure illustrating the projection of value.

03 17 Triangle Model

Here we have someone using out point in a report. The person writing the report projected the report through a word processing app. That app projected value through a machine. The point was projected through the report after the point was projected from the proxy thought the illuminator-sensor plane. Lots of value projected from various sources from various distances and points of view.

Yeah, not at the UX at all.


Noise and Knowlege

April 7, 2015

I started reading something on topology and statistical distributions. This is one of my research topics. My intuition tells me there is something to the notion that linear analyses fail when the space is hyperbolic, and succeed despite themselves when the space is spherical. In this reading the author said that the distribution sits on a manifold. Why a manifold? I’ll take that under advisement and wait for it to become more meaningful.

I’ve sketched tons of graphics. Some of those I’ve used in this series of posts on the normal distribution. We’ll look at one more, a normal distribution on a small sphere. The tails get longer, but like a squeezed water balloon, the noise that fills that additional length had to come from somewhere. We usually think of the distribution as sitting on a flat Euclidean plane. When the distribution is sitting on a sphere, flat goes out the window. The noise comes from the bottom of the distribution moving outward from the mean.

49 Normal on a Sphere

The figure shows how the shape of the distribution would change and how the tails reach around the sphere. In the figure, the increase in tail length is projected back a Euclidean plane. Lost in this is the height loss that happens as the number of standard deviations, or sigmas increase, as the firm grows larger.

Notice that in this particular figure, I’m taking the Frequentist point of view that of a distribution containing random noise, rather than knowledge.

Back in my radar mechanic days, the story about using a Styrofoam cup and a microwave transmitter as a bug was making it’s rounds. Microwaves are fun stuff. If you change the shape of the container, you change the frequencies emitted. That container is typically a metal waveguide. These waveguides are firm. They don’t get deformed in typical usage. But in this application as a bug, the lack of firmness is essential. A room vibrates, so the cup vibrates. Sound waves bounce around the room, so they eventually deform the Styrofoam cup. Those deformations clip frequencies from the square microwave pulse filling the cup.

Coffee Cup 03

The figure shows the opening of a Styrofoam cup in blue. The red sphere fits inside the cup. When you pick the cup up, the cup is deformed. The deformation makes the opening thinner, but longer. The frequencies that were the size of the red sphere are now clipped as only frequencies the size of the brown sphere can fit into the cup. You could use that Styrofoam cup to squeeze out some Morris code.

In the last two posts I wrote about how black swans clip the tail of the distribution. Black swans are typically big price/valuation losses on in the financial markets, aka missed quarters and such. But, these black swans are like setting an epsilon in calculus when you are trying to find the convergence of a function, so these tail movements happen all the time by tiny price fluctuations that happen every day once you’re a public company. Your normal distribution is like that Styrofoam cup. Your price constantly vibrates. This also means that your outliers might be under the distribution one day and not the next. Your real option tracking portfolio would act similarly, so the strategic decisions driven by those real options would oscillate as well.

Given that a statistical distribution has a surface defined by a function, and functions can be analyzed via a Fourier analysis, the shape of the distribution is doing the Styrofoam cup thing inside the distribution, not just relative to the outliers at the base of the distribution.

31 01 Randomness Structured

In this figure, I’ve drawn the largest sphere, the largest frequency, that could fit under the distribution. I also drew a much smaller sphere. Notice that with an unlimited budget, there is no physical limit to how small that smaller sphere could be. Alas, at some point you end up with a laser, aka another way to pick up the vibrations from a Styrofoam cup. But, budget and significant use keep us from getting smaller frequencies into our Fourier analysis. This is much like doing a factor analysis. There is always another factor, but the smaller those factors get, the more expensive it is to capture the underlying data. Beyond budget, you might want to ask just how much company cognition you can dedicate to ever finer factors.

I did not show the spheres between the largest and smallest that would fit or pack under or inside the distribution. Also not shown are the lifecycle of the distribution. The largest sphere gets smaller as the company gets larger. The spheres start out small in a Poisson distribution and get larger as those Poisson distributions tend to the normal.

32 02d Poisson Distributions Tending to the Normal

The above figure roughly lays out the Poisson distributions under the normal while those Poisson distributions tend to the normal. Frequentist probabilities use the law of large numbers to find macro-level behavior, and the law of small numbers to find micro-level behavior. Poisson distribution provide the basis for Markov chains. Markov chains begin chipping away at the notion of that a distribution only contains random noise. Markov chains begin to structure the contents of a normal  distribution, the normal distribution being large. Poisson distributions, aka small distributions constitute traversals of the area under the distribution, aka vectors, like a vector of differentiation.

32 02c Imposing a Structure Under the Normal

The Poisson distribution here starts as an outlier. It follows a chain of vectors until it gets to the mean. It could be a random walk under the normal. It need not pass through the mean. When we seek out our next discontinuous technology, our random walk would be a Levy flight. Imposing structure happens as we gain knowledge of the systems under the normal. As this structure is imposed, the probabilities become less Frequentist and more Bayesian.

The next few figures illustrate how the technology adoption lifecycle imposes structure on the contents of the normal distribution, on the once random variables.

32 02a Normal Distribution Footprint Outliers32 02b Imposing a Structure Under the Normal32 04 Imposing a Structure Under the Normal

Here I wanted to show how the B2B early adopter was an outlier. Yes, I called that early adopter a weird person. They don’t make a good reference case for other prospects in their vertical. The technology adoption lifecycle is organized by pragmatism. Marketing would set the width of each slice of pragmatism. Those widths can change. But, business cases and other reference data needs to be generated for each slice. Market to a few, since traversal across the market takes time, but try selling to just one at a time. Requirements collection needs to be bound by the width of the pragmatism slices as well.

32 03 Imposing a Structure Under the Normal

In the above figure I’ve put the B2B early adopter, who is an outlier in our category’s normal distribution, is also an outlier in the normal distribution of the vertical that that B2B early adopter does business in. In the bowling ally, the seats and dollars you will take from that vertical is the only workable stage gating you’d want to do before deciding to take on building that B2B early adopter’s product visualization.

Structure comes out of nowhere, well, out of the daily operations, experimentation, and other efforts to reduce the uncertainty of the innovating organization. Geometries change, the shapes of the noise change, distributions change, but they change in organized ways. Frequentist probabilities are not the only ones out there. Some people have found that they have topological problems. Know who to call.


Enjoy? Comments?


Learn and Forget

April 4, 2015

One of the ideas you hear a lot for a long time now is how you have to keep learning going into the future. Have fun with that.

The technology adoption lifecycle, aka the normal distribution says something else about the future. The transition from early to late moves the focus from carrier to carried, from IT to the departments doing other work, from the horizontals to the verticals, from geeks to consumers, from players to payers, from firewall disdainers to firewall payers. All of this happens at the mean of the normal. At the mean, we have to forget.

The transition from early to late also does other things. It switches your stock price outlook from growth to decline. IPOs were big, but now will be smaller, as the premium goes away. And, you miss a quarter, because you forecast continuing growth, but now having sold more than 50% of your available market, you have fewer prospects ahead, more pragmatic prospects, smaller companies ahead, and marketing pushing the wrong people down your pipeline. Sales might even feel like going for the CxO sale, which slows down the deal flow immensely.

So you missed your quarter, so stock price effects happen as well. You have to take your employees off of stock option compensation, since those stock options are worth less and will not grow, which means you’ll be renegotiating salaries upward. You look like an ordinary company. You are not a startup anymore. It happens. You’re VCs will be pushing for your M&A, if you took VC money. Yes, your company could have a 10x larger market after you’re acquired, but that accrues to the acquirer. Much happens crossing from early to late, aka crossing the mean.

But, lets get back to what your product team learns and forgets. Lets bring this home.

25 World as  Bits 02

In this figure we look at what the development team is doing right now. The x-axis is time. Below the x-axis is functionality provided by your value chain suppliers, say your cloud vendor. The world that development created includes their code, above the x-axis and the bits provided by the value chain, the whole product. Make is above the x-axis. Buy is below it. A lot of coordination is required at the x-axis, maintenance windows,

Now is a vertical line slicing through the distribution. Inside the distribution is what was just released, or what will be released the next time. It is what we are making now. Below the x-axis is what we are buying now. Above the normal is a population we will come to serve in the future. The buy portion of the line also has a segment looking to the future and what will be bought in the near future. Consider that your value chain providers likewise operate under their own normal distribution. Your future buy is their future sale.

To code for the future populations, aka make, and to buy for the future populations is learning via accretion or accumulation, as described in Stewart Brand’s, How Buildings Learn. Yes, your organization learned its way across the technology adoption lifecycle (TALC), as it sold its way across it as well. When I’m talking about the TALC as a normal, I’m talking Bayesian probabilities, where there is knowledge under the distribution, rather than the frequentist’s randomness.  Moore said that pragmatism organizes the TALC, so the lifecycle is not random. I did have a sales rep describe his lead-handling process as a random one. Marketing is Bayesian, aka not random, while Sales is Frequentist, aka random–A fundamental conflict. Likewise, content marketing lays out content under the normal, so consistent with marketing’s inescapable approach, it’s knowledge under the distribution where time and pragmatism is the coordinates of place, and learning.

So let’s cross the mean and look at the other side.

25 World as  Bits 03

Here now is right of mean, deeper into the future. I think of those bits I mentioned earlier as the size of the world. Here the world is large in the sense of total bits, but most of those bits are not relevant any longer. We have left customers behind. We have dropped the geeks as our market. Sure they are still there, but we don’t serve them. We’ve even been charging for our services, because our consumers pay for things. Free doesn’t fit anymore.

On the figure, I’ve labelled the forgetting, since that is what we are doing. As me move into laggard, phobic, and non-adopter spaces, we are disappearing into the stack. It’s hard to disappear if you are a media operation striving to capture attention in your content layer as your carrier layer is disappearing. Beware of your monetizations when they run contrary to forgetting. We have forgotten things in the past. We have ongoing forgetting operations. Making the future is just there to accommodate the forgetting. When we first crossed over the mean from early to late, we had to hide controls, while retaining the power of our application. Moore called this task sublimation. Look at all the things we don’t do in our mobile apps that we used to do. We don’t save files anymore. Everything is in a database that we don’t administer. We make a little for now. Our buying side undergoes the same forgetting pressures. Our supply chain has to survive on selling less, not more. Their bits are disappearing just like ours. So the right side of the normal is about forgetting.

Next, I’ve put both sides on the same figure.

25 World as  Bits 05

And, I sum up some points next.

25 World as  Bits 06

The world we made traverses the normal from left to right, from the past to the future. We consume bits converting them to dollars, as we consume customers. We have a scorched earth approach to consuming customers. We only have one initial sale. We retain them, but it is the initial sale that moves across the x-axis. The customer is the tick of our clock. The bigger the customer the sooner we get to the mean. Crossing the mean we begin to reduce the size of our world. We stop growing the number of bits. We undergo bit reduction. Oddly, our organization concocts a growth myth, so it can stay larger as it depreciates bits. The organization gets larger in terms of standard deviations, which lowers the ceiling under the distribution as our margins get thinner.

Notice that I labeled a particular point on the surface of the distribution with the word “Stop.” In calculus, we approach a limit defined by an arbitrary height epsilon. Epsilon is what I’m calling “min.” We need a minimum amount of cash to keep going. You can only layoff so much. “Stop” is where the normal converges with our burn rate.

The TALC is the lifecycle of a category. Blue oceans happen near that epsilon convergence. Crossing the mean should have us looking for the next discontinuous technology to ride. Today we are comfortable with capturing cash, instead of creating wealth, building entirely from other people’s whole product, starting out to the right of the mean. We are comfortable with stasis. We live in small worlds with rich pasts. We look back at that past and forget that it taught us. It fit the populations it served. If it didn’t we wouldn’t remember it at all. It didn’t lack design. The ethic was different.