Matrix calculations, Hill Ciphers and applications in machine learning

Getting your Trinity Audio player ready...

Get ready for a new direction in the series of Dr. Craig Wright lectures and discussions once known as “Theory of Bitcoin.” The new series is called “Bitcoin Class with Satoshi,” featuring Dr. Wright’s lessons on some of the deeper and more technical aspects of Bitcoin Script. In this first episode, you’ll receive a brief but detailed introduction to applications in machine learning, matrix calculations, Hill Ciphers, Queueing… with some discussion of how these principle could be used in the real world.

“Bitcoin Class” began a few weeks ago as a trio of Dr. Wright, Ryan X. Charles and sCrypt CEO Xiaohui Liu, and is now continuing as a spin-off series minus Charles, with Dr. Wright as the teacher and Liu as the main student.

Look at what Bitcoin can do natively

Amidst all the current talk about second-layer token protocols and contracts, Dr. Wright says, it’s time for people to start looking at something that’s natively there: Bitcoin‘s actual native data stack.

His first example is using queueing functions similar to those in Python to use Bitcoin to solve logistics problems. Giving the recent disruptive episode of the Suez Canal blocking, he says it’s possible to solve problems dealing with ship queues, delivery costs and times, employees required, what extra goods need to be purchased, requirements for online computing (i.e. AWS) or physical factory time.

Using Bitcoin Script not only allows these processes and calculations to be automated, but could help businesses solve problems when they don’t have all the answers. Past episodes of “Bitcoin Class” looked at the possibility of publicly offering bounties to solve complex problems, and automating payments for those who provide the best answers. Some challenges of course, are how you determine best answers if your own information is incomplete; or make only certain information available to certain people at certain times. Luckily, Bitcoin Script can help there as well.

You may require certain conditions to be met before a transaction “fires” (i.e. is broadcast to be written on the blockchain). It may only be valid if a pre-set time has passed (using nLockTime), or only parts of it may validate if only partial information is received.

More control over data and access to it

Wright and Liu discuss matrix calculations, or non-linear arrays of variables upon which a programmer could perform operations. What those variables represent, or whether the table of variables is complete, or semi-complete, can determine the eventual outcome.

The example Dr. Wright gives is that of a digital image that can be degraded. Perhaps a content creator or image hosting site would like to sell full or limited rights to use a particular image, but must first make that image available for potential users to view (without allowing them to copy). Usage could be granted for a limited period of time or number of uses, after which the image “decomposes.” Low-resolution samples are provided, but the recipient only receives the high-resolution version and usage rights after paying for them. Matrices could decompose or reconstruct the full image at any time.

Liu calls it “a kind of DRM,” and Dr. Wright agrees, but says images are only one example—these processes can apply to any kind of data, not just images. Perhaps researchers need to gain (or provide) access to specific sections of their total data, link it (or portions of it) to other applications, and more.

Such applications already exist in some form or another today, albeit on centralized servers. Liu asks the question many do, which is why one would need Bitcoin to do this. Dr. Wright explains that Bitcoin provides a way for the transactions to be provable, permanently prove ownership, make everything more programmable, and allow more control over access. It’s also a way to standardize these processes on a single ledger using a common programming script, rather than building entire new systems whenever these things need to be done.

Dr. Wright’s lessons are mainly theoretical and Liu’s questions mainly concern where they could be practically applied. But that’s the teacher and student performing their rightful roles; it’s up to the teacher to lay out the concepts, and the talented student to creatively apply them.

The “Bitcoin Class” series is a lot more technical than “Theory of Bitcoin,” and will likely appeal more to those interested in development (or who have existing degrees in data/computer science fields). But the segments are now presented in more easily-digestable lengths, allowing anyone to drop in and maybe discover something about Bitcoin they didn’t know before.

To watch previous episodes of the Theory of Bitcoin and Bitcoin Class, check out the Theory of Bitcoin YouTube channel here.