InfraOps Insights from Gleick’s 'The Information'

Hello there,

James Gleick’s: The Information - A History, A Theory, A Flood is a solid introductory book on “Information”. Anyone working in our space would get a lot out of it. “IT” does in fact include the word/term after all. 

I’m trying a new format this time. I’m working to find my footing on what is the best format with my reading workflow. 

I’m sympathetic to authors when book reviewers cover most of what’s in the work. Unless told otherwise by the author I will not cover most of the book. What I’ve tried to do is to find a few notes that would allow me to provide some modicum of “flow” while not oversharing the content. 

I’m thinking about setting up a system that would provide ALL of the notes from the books that I’m covering with Justindustrious. But a requirement would be that I either have gifted you a book through our monthly give-away, or that you provide proof of purchase. 

I’m thinking that I can set up a Shopify account with $0 purchase options for the notes. I would then just have your receipt / picture whatever verified to send the digital notes over.

Here is an example of one of my book canvases.

We did the drawing to give away last month's book - Cal Newport’s: A World Without Email. We will be reaching out soon if we’ve not already so that we may get you the book.  

Anyhow thank you for being an early reader here. I look forward to getting better and better at this over time. 

Cheers,

Justin. 

Mechanization

To know that Leibnitz's vision to mechanize computation in general was proclaimed in the 1680s is potent stuff. In most, if not all, knowledge work, all that we “do” is computation and InfraOps is no exception. This resonates with my goal of "mechanizing" most of the “atoms” side of infrastructure operations work, so that we may release more of our people’s potential. By using automation and synchronization, InfraOps pros can escape most manual, repetitive administrative work. The shift to industrialize (mechanize) that which can be computed on our behalf empowers professionals, as their work on average will be more impactful. They can spend more of their time using their expertise to make progress and deliver valuable outcomes. This helps the entire CIO organization to drive greater business value and operational excellence.

As a simple example, let's imagine that you ask five different stakeholders the following question. We have a Bill of Materials (BOM) that needs to find its way to data center "C" in India. How do we make that happen? You will most likely get five different answers, all of which are the outcomes of computation. No matter what solution is proposed, it includes the following base computation (a logistical triangulation). You have an origin, a destination, and a BOM. All of the computing downstream of this can and should be "mechanized". That's not really a debate at this point. Although almost no MNCs do this yet within the InfraOps realm. 

The above said, I'm very interested in going UPSTREAM from the triangulation with industrialization, but for now we all need to put in the work downstream. 

Shaping Reality

John Archibald Wheeler

Imagine you are in a room with a light switch.

1. Posing a Yes-No Question: You ask, "Is the light on?" This is a yes-no question because there are only two possible answers: yes or no.

2. Reality Arising from the Question: Depending on your answer (yes, the light is on, or no, the light is off), your perception of the room changes. If the light is on, you see a brightly lit room. If the light is off, you see a darker room.

3. Information-Theoretic Origin: The state of the light (on or off) represents information. This information (whether the light is on or off) forms the basis of your perception of reality in that moment.

4. Participatory Universe: By asking the question and observing the light, you are participating in creating that specific reality. Your interaction with the light switch and the room shapes your experience of the room.

5. Cosmic Information-Processing Machine: Expand this idea to the entire universe. Just like the light switch example, every interaction and observation in the universe can be seen as processing information. The universe is like a giant computer where every event and observation contributes to the overall state of reality.

This is a fun framing. InfraOps professionals, as a group, manage vast amounts of data/information to make informed decisions that shape their ITosphere. I can see how good InfraOps systems and protocols do this by treating the IT environment as an information machine. At the end of the day, we need hardware made of silicon, metal, and plastic. It needs to be in a certain place at a certain time. It must be powered on with network access. No matter whether you rent the hardware from an intermediary, i.e., the cloud. Whether you use VARs, MSPs, consultants, distributors on and on... It's all yes/no questions. in the end.

Decreasing the amount of yes/no questions that are required is where the juice is.  

Standardization and Automation

The principles of efficient encoding and transmission in telegraphy have direct parallels in our atom-based InfraOps world. Telegraphy uses shorter sequences for common letters and phrases. This optimizes communication. As an example, InfraOps professionals can optimize their deployment model by automating all data center base builds. For this exercise, let's say that you have settled on a storage partner to be utilized for most if not all of your net new rack builds. You will often deploy this sub-set of storage devices. So, you are incentivized to codify your storage deployment "message". Then, you can trigger it with a dot or a dash.

The more I work on the above problem in my day job, the more I see my work as the enabling of more efficient “messaging” in a customer’s noisy IT business environment. 

Information Theory

Claude Shannon

1. “Information is closely associated with uncertainty.” Uncertainty, in turn, can be measured by counting the number of possible messages. If only one message is possible, there is no uncertainty and thus no information.

2. Some messages may be likelier than others, and information implies surprise. Surprise is a way of talking about probabilities. If the letter following t (in English) is h, not so much information is conveyed, because the probability of h was relatively high. 

   1. Cn yu rd ths?

3. “What is significant is the difficulty in transmitting the message from one point to another.” Perhaps this seemed backward, or tautological¹ , like defining mass in terms of force needed to move an object. But then, mass can be defined that way.

4. Information is entropy. This was the strangest and most powerful notion of all. Entropy - already a difficult and poorly understood concept - is a measure of disorder in thermodynamics, the science of heat and energy.

This is a tricky one. The more information you need to make a decision or action the more entropy and disorder you have. This is a non-intuitive way to think about the systems that we design to serve a particular purpose. If a question or decision includes a bunch of uncertainty, the number of yes/no questions that you and or your system must run through is much higher. And thus, entropy remains high or even increases. 

Okay, so we want to decrease the amount of information that must be encoded in a “message” for execution. Shouldn’t we then be able to measure the efficiency, thus “usefulness” of the message itself?

Logical Depth

Bennett's concept of "logical depth" is relevant to our InfraOps work, where the usefulness and value of information are paramount. In IT infrastructure management, it is of course not the quantity of information that matters but its relevance and utility. 

Earlier we talked about the moving of information from one place to another as a message. We in turn can then think of that message as a: trigger, automation, synchronization, instruction etc… The usefulness of the information encoded in the message can be measured. As per Mr. Charles, this measurement is called “Logical Depth”. 

The Greater the logical depth the more “complicated” executing the message is. Ideally we would like to build our systems in such a way that the recipient of messages in all contexts will have to execute through a set up steps that has the lowest logic depth that is practical. 

1  tautological - saying or expressing the same thing twice over in different words.