Alexis
Hi everyone. Welcome to another episode of Talk Tech with Data Dave. I’m Alexis, your host of this podcast, and I am here with my dear friend, Data Dave to talk about all things data, all things cloud, all things AI, all things technology, all things D3Clarity.
Dave before we get started, I’m going to remind our listeners, as I always do, that you can submit a question to data Dave by emailing us at talktech @d3clarity.com or by submitting a question right on the D3Clarity website. We are always interested in answering your questions on the podcast.
Dave, this morning we have a topic that I’m thinking is going to be pretty applicable to our listener population at large, mostly because it’s something that I deal with on a day to day basis.
So back at the very, very beginning, Talk Tech with Data Dave, episode one, we asked the question, what is data? And then over the last year, plus, we have been talking and I’ve been hearing, and one of the things I keep hearing is the word data and the word information used somewhat interchangeably. I know I’m guilty of it, and so I’m hoping today you can help me understand the difference between data and information.
What is the difference between the two? Hopefully we can walk away being able to use our vocabulary a little bit better. And when I say we, I mean me. All right, let’s learn. What’s the difference?
Data Dave
Well, first of all, Alexis, good morning. And this morning, hope everything is great.
Alexis
Oh yeah, I just went on that there.
Data Dave
Good morning.
So, the difference between data and information, that’s a really good question. The way I look at it, this will be a conversation. The way I look at it, and reasonably well understood approach, is that data is the raw material and information is the refined material. So, one way of thinking about it, say you’re mining iron ore, then data is the ore, the buckets of ore that come out of the ground, and then you refine it into the iron or something like that.
What we often think about is the information or the knowledge value chain that says data– and I think I used the definition last time we talked about data and said- to me, data is the evidence of a fact or the evidence of an event, evidence of history, if you like. Data is the particular evidence of history that says something happened at this point. And then that data that set of data leads to information. That information leads to knowledge, to the point where somebody can have knowledge, possess knowledge, and that knowledge can lead to action. “I know this, therefore I do.” And that combination of knowledge and action leads to experience.
Alexis
If I did something, I’d call it experience. Yeah.
Data Dave
If we follow it all the way, then that knowledge and that experience leads to wisdom, because I did this and this happened, therefore I am now wise, because I know that this action leads to this outcome and that then finishes the cycle up at wisdom, which, of course, if you take action, creates more events, which creates more data, which cycles it round.
Alexis
Stop there for a second. If I know something is going to happen, I have the wisdom that something is going to happen. So, I take action based on my wisdom. And taking action creates an event. Evidence of an event is data. It’s kind of cyclical.
Data Dave
It’s cyclical. Yes.
Alexis
Okay.
Data Dave
Cyclical. Cyclical. We can debate that one as well.
No, but there’s this chain where data is your raw material. You analyze the data to produce information, you use that information to gain knowledge, use that knowledge to make a decision, and that decision leads to action, which of course creates events.
Alexis
Okay, I’m following you there, but help me a little bit more. Can you give me like a business example of this? Something that I can relate to?
Data Dave
Let me try and do exactly that.
So the business example I would use would be- take a retailer. A retailer sold a bunch of things in the month of June. There’s a list of things that were sold, a list of events. That is just raw data. This person came into the store, bought one of these in the month of June. List of data. List of sales. If I compile that into information, I’m starting to say during the month of June, these were my sales for these different products. So now I’m starting to get information about my business performance using the data. That was the list of all the sales that occurred in the month of June.
Alexis
Yes.
Data Dave
Now that I’ve got that information, I can say, am I growing or am I shrinking? What is the path? What is the knowledge? I know that in summer I sell more t shirts than I do sweaters. I’ve now got knowledge. What action do I take based off that in July, I’m going to order more t-shirts than I did sweaters.
Alexis
The application of the knowledge.
Data Dave
So now I’ve gone from data, which is my list of sales in June and probably may, and then I’ve analyzed that to have the information about product movement through my store.
I now know what my product movement has been and my trends that are there. So now I can take action on that knowledge to set up my inventory for July. Not only does it create the event of deliveries to say, I now have the inventory that I can sell in July, but that also leads into all the July sales to build on the set from June to set up for August. So that gives you that thread of data, to information, to knowledge, to action, to wisdom. Next year, with your wisdom, with your experience of this year, it’s all going to work so much quicker and so much smoother.
Alexis
Because I’m going to know already how many t-shirts to order versus sweaters, which is going to create another action, which is going to create another event, which is going to create more data.
Data Dave
You could say that the data for this year forms information that can go into your decision making for next year and say the previous year behaved like this. Not only that is April, May, June, as data that I can analyze, I can also look at previous year’s data and say, is there likely to be a correlated pattern in my data that leads to better information, better knowledge, better action, better experience, better wisdom, better. Right.
Alexis
So it sounds like data and information are kind of the first two steps in a chain.
Data Dave
Yes, exactly. We often call this either the information value chain or the knowledge value chain.
Alexis
Because those are the two things in the middle of the chain. That makes sense.
Data Dave
Those are the two things that come at the end. If we play this into AI for a minute, that’s exactly what we’re doing in AI. We are inferring how the data supports the information, supports the knowledge, creates the decision. So, we’re trying to push automation up the chain. We have always collected data, use data to analyze business intelligence, business analytics, reporting, etcetera, to distill information from that data, to guide people into knowledge, therefore make a decision. As we look at the wave in AI, we’re simply trying to move automation up that chain to say, not only can we create information out of data, but we can create simulated knowledge out of the information and simulated decisions from that knowledge.
Now, you notice I say simulated in that, right? I say simulated because does a machine really have knowledge? Qith knowledge, there’s sort of a possessive nature, which you can only have knowledge if somebody possesses. A library doesn’t contain knowledge, it contains information.
Somebody might have written down their knowledge, but now it’s information about that person in the form of a book. So, you could argue that a library contains information, not knowledge. And there has to be some kind of possessive in order to have the knowledge..
Alexis
A human has to apply it.
Data Dave
And therefore there has to be an intelligence that applies that knowledge. So that’s why I use the word simulated with AI, because it’s kind of, does the AI really have knowledge? Does the AI really do that? No. In my mind, it simulates, it impersonates. It is a representation of what a person would do with this data, with this information. Therefore having knowledge and making a decision.
Alexis
I want to hop back to the very first thing you said during the example. You said that data is the record of sales, and then information is seeing that you had x amount of sales within the month of June.
That, to me, sounds a little bit like semantics. Those two things sound really, really similar, just kind of you saying them backwards. There seems to be quite a fine line between data and information.
Data Dave
There is a fine line between data and information. There is a very fine line between data and information. That’s why, in common vernacular, they’re almost synonyms. Almost.
Alexis
That’s why I’ve been using them interchangeably for so long.
Data Dave
They’re not quite synonyms, but they are almost synonyms. In common speak, you would almost use them synonymously. But when we get sort of semantic about it, the way I look at it is data is the fact it is kind of immutable. It can have a certain amount of quality, has structure to it often, and it describes a thing. Information is a little more nebulous. It’s kind of inferred from those facts. It is a refinement of the facts. I don’t dig iron out of the ground; I dig iron ore out of the ground.
I turn that iron ore into iron and refine it further into steel, and then I make a bridge or a house or whatever it might be. So there’s this thread of refinement and structure to it, and we’ve applied the same construct to using a global term of data or information. I’m struggling now with essentially written word, if you like, or the language or library content, really. I suppose to say that the raw material is the data. The raw thing that is produced is the data. Information is inferred from the data. Knowledge is gleaned from the information, and then that knowledge is used to make a decision.
Alexis
And that value chain that you’re describing can work the other way, that’s data science at its core, right?
Data Dave
So you can say, based off of my wisdom and my experience, I believe that this is going to happen. I’m going to form a hypothesis now. I’m going to read all the information that exists there to see if I can expose the facts that would support my hypothesis.
This is fundamental, really, if you think about it, we’ve just described the scientific method.
Alexis
Yes.
Data Dave
The scientific method is observe, form a hypothesis, experiment, generate data, does the data support my hypothesis? And the information is really that experimentation tier.
Alexis
I like that. I like thinking about it like that.
Data Dave
That’s using the same thing the other way. And it’s the cornerstone of the scientific method. If you do the scientific method without the experimentation bit, think of it more of an observation. “I’m going to observe a whole bunch of facts that occur in nature. So I’ve got this hypothesis. So I’m going to observe a whole bunch of facts that I think support my hypothesis. I’m going to analyze those facts for the discrete data that is relevant to my hypothesis and get rid of the sort of irrelevant pieces. And then I’m going to use that to form information and then use that to knowledge to support my hypothesis.” My hypothesis therefore becomes, “I know this, I don’t just believe this.”
Alexis
Right, right.
Data Dave
And that’s knowledge. I know is that transition from hypothesis to knowledge and you use information to support that, but within that is the raw material of data.
Alexis
There we go. I get it now. I’m hearing the difference. For the listeners in full transparency, Dave and I had a full half hour conversation prior to starting to record and I still wasn’t following him. But believe it or not, I’ve got it now.
This is making sense to me, finally. I like that. Thank you, Dave, for drawing that down. Listeners, if you haven’t listened to episode one, “What is data?” or “Data science versus data analytics,” we talked about a lot of these concepts in those two episodes as well. So, you might want to go back and listen to those two to help kind of build a core for what you just heard. Maybe I should have started the podcast with that instead of ending the podcast with that, but here it is.
I think I’m getting it now.
Data Dave
Yeah, this is one of the fundamental concepts that goes into those episodes. I’m kind of surprised we haven’t done this one before, to be honest with you. Yeah, but it’s a fairly well known sort of value chain to say, data, information, knowledge in the pure sort of abstract content structures, data, information, knowledge, wisdom. I like to insert actions, decisions into that to produce that wisdom and experience. So data, information, knowledge, action, experience, wisdom.
Alexis
And then we’ve hit it again. That kind of creates a cycle, and then also that value chain works both up the line and down the line kind of the same way, like supply and demand does up the line and down the line.
Data Dave
Think of it this way as well. You can think of this as the guild system.
So, you’ve got an apprentice at the bottom. An apprentice is just starting up. He doesn’t know anything. He’s just beating things with a hammer. I’m going to take a blacksmith approach. Right. So he’s just beating things with a hammer to bend metal. Doesn’t really know very much, doesn’t really have a great deal. He’s doing, just doing work, creating events, if you like.
The information becomes, what events do I need to do in order to create a horseshoe? So now he’s getting a little more experienced. He’s more of a journeyman, if you like, but he’s still working through that. He’s more of a journeyman.
And then you’ve got the tradesman who really, really knows, has the knowledge and is shoeing horses and making all kinds of things. And then at the top you’ve got the guild master who has the wisdom of the experience and the many years to be able to teach all the way down the thread.
Alexis
Yes.
Data Dave
I don’t know if that helps.
Alexis
It makes sense to me, and I know I’m putting a graphic into the pod that’s going to show it in a somewhat triangular shape, pyramid shape, if you will, to show the idea that you have a lot of data, and you take a lot of data and you make it a little bit smaller, and then you have an okay amount of information, and then you take it a little bit smaller and you have an okay amount of experience. And then you get to the top where all of that kind of makes sense and creates our wisdom, how we’re going to apply it, what we’re going to do, and how we’re going to make sure that all those little things continue to make sense. I like it. I am going to try to be better about using data and information correctly.
Listeners out there, if you hear me in another podcast using that incorrectly, I would love for you to call me out. That is the best way for me to learn. Failing is how I learn. So, call me out if I do it wrong in the future.
Dave, thank you for breaking this down in such simple terms for me. the fundamental aspect of this is actually making, like I said earlier, some of our other conversations even more clear at this point. So, thank you so much for helping me with this one.
Data Dave
You’re very welcome. And the thing I’m going to say is that in common language, they are often used as synonyms because we often live on the line between data and information and we don’t think of them as separate, but I think of them as separate because I think that gives a good thread for definitions and a good thread for the way we operate for sure.
Alexis
Well thank you Dave for chatting with me listeners. We appreciate you listening. Please like comment, subscribe, share and send us some questions at talktech@d3clarity.com.
Data Dave
Thank you.
