Post 8: A Walnut Shell Isn’t a Walnut Shell

This may be the point where the entire project changed for me.

Being in the oil & gas business myself, I am familiar with how oil and gas operations have historically used materials such as ground up walnut shells in certain drilling applications. The industrial detail is less important here than the conceptual leap I am trying to make.

A walnut processor might see:

agricultural byproduct

An engineer might see:

a particulate material with a particular combination of hardness, density, particle size, compressibility and other physical properties

Those are completely different descriptions of the same thing.

One tells us where the material came from.

The other tells us what it can do.

Names can hide value

Suppose Industry A produces material X and considers it low-value.

Industry B buys material Y for hundreds of dollars per ton.

Nobody connects them because X and Y have different names, originate in unrelated industries, and appear in completely different databases.

But what if X and Y share the functional properties that actually matter?

Perhaps after modest processing, X can replace some portion of Y.

Now we have:

one industry’s excess can be equal to or greater than another industry’s expensive input

That is far more interesting than just ordinary recycling.

So we can stop cataloging objects and Catalog properties instead.

This very well may be one of the central ideas of the project.

So rather than tell the system:

“This is a walnut shell.”

We could tell it:

particle size hardness density porosity compressibility absorption thermal characteristics chemical stability biodegradability abrasiveness cost per ton annual availability geographic concentration

Then do the same thing on the demand side.

And we would not tell the system:

“This is an oilfield product.”

Instead, we say:

“We need a material possessing properties A, B, C, D and E within these cost and performance limits.”

Now, with the power of classification and machine learning the computer can ask a question humans rarely ask:

What things already exist in enormous excess that possess A+B+C+D+E?

The answer might be walnut shells.

But the important possibility is that tomorrow the answer could be something nobody has ever tried.

Cross-domain equivalence

I call this cross-domain equivalence.

Two things that markets consider unrelated possess sufficiently similar functional properties to substitute for one another.

And the great thing is that this can extend far beyond materials.

A parked electric vehicle and a stationary battery both contain electrical storage. In fact, there have been stories of people charging their Teslas at off peak times, then using that stored energy during peak times to charge other devices. Essentially using their vehicle as a large battery backup. A big huge portable power pack.

Another example that I recently encountered involved a ski resort during summer hiking activities. I had taken a ski lift to hike at the top of a mountain. Upon returning to take the lift back down to the base of the mountain, I was notified that there was lightning in the area and the lift was not going to be operating, however, they were sending a shuttle to pick myself and several other “passengers” up to take us back down the mountain. While I was initially expecting an ordinary “shuttle”, I was surprised to see a security guard show up with his security truck, filled with tools and equipment only for us to load up on our return journey. This was no ordinary shuttle but yet a security vehicle, excess capacity for security operations during the day, repurposed into a makeshift passenger shuttle. While this seems obvious now, the added expense of a separate shuttle operations for the rare times lightning strikes occur is impractical. These types of situations happen every day. It is in these unconventional areas that I am more interested in.

A delivery truck and another fleet traveling the same geography both contain transportation capability.

A warehouse and an underused commercial building both contain protected volume.

A phone and a dedicated sensor both contain sensing capabilities.

The labels are different.

The functions overlap.

The real invention may be the finder

If we discover one surprising substitution, that may be commercially valuable.

But something bigger is possible.

What if we can create a system that repeatedly discovers these relationships?

That would not be a marketplace for agricultural waste.

It would be a discovery engine for hidden economic equivalencies.

The walnut shell would merely be proof that the world contains relationships conventional categories prevent us from seeing.