Join RedMonk Co-founder James Governor and Alois Reitbauer, Chief Technology Strategist at Dynatrace, atop the London Eye where they discuss the experience of observing a city from different vantage points while unpacking one of observability’s most pressing challenges: how do you make sense of a complex, ever-changing system without drowning in data? The conversation opens with a deceptively simple question, how do you orient yourself in a city you don’t know?, then builds into a rich discussion about context, reference points, and the difference between raw data and actionable insight. Just as navigating London requires knowing which landmarks to look for and updating that mental map when something changes, effective observability is less about collecting everything and more about knowing what actually drives decisions, because true situational awareness requires not just more data, but also better-calibrated, agreed-upon points of reference, and the wisdom to ignore everything else.
This RedMonk video is sponsored by Dynatrace.
Links
Transcript
James Governor (00:00)
The more points of reference you have, the more effective the context is.
This is James Governor from Redmonk and I’m with Alois Reitbauer from Dynatrace and we are here to talk about observability. It’s the London Eye. It’s huge. Here we are at ground level. This thing looks big. Obviously the city, we can’t really see much of what’s going on. Well, what we’re to do is talk about observability, context, the situation so that you can better understand what it is that you’re actually seeing.
Now, Alois, you work at an observability vendor, Dynatrace right? You don’t build things like-
Alois Reitbauer (00:42)
But we don’t build things like this. That
would be cool though.
James Governor (00:46)
But you do build infrastructure so that you can see things more effectively. So yeah, we’re here to see something. One of the first things as we think about what does it mean to get that kind of context is here we are. We’re in a city that I know very well.
Alois Reitbauer (01:01)
And I don’t know that well, I know that there is Big Ben over there, that’s what I do know.
James Governor (01:06)
Big Ben’s just over there, but in general the question is what’s the orientation of the city? How do we get around? How do we understand what’s going on? And that’s what we’re here to discuss and see.
Okay, so we’re talking about context. Yes. We’re in the city. For you, you’re looking around, you don’t know where you are.
Alois Reitbauer (01:26)
I where Big Ben is, where the London Eye is and where Tate and St. Paul’s is, but that’s pretty much
James Governor (01:31)
That’s about the limits of your context. think what’s interesting for me from an observability and observing the city perspective is I think I’ve the context extremely well. But you have to make sure that you’re actually looking around because it’s very easy sort of looking this way. think, well, look, I’m on the Thames. It’s east-west. House of Parliament is there. So therefore, east is that way. But in fact, it turns out that until you’ve got more points of reference,
Alois Reitbauer (01:34)
Pretty much yes.
James Governor (02:00)
better understand how everything fits together, then you begin to understand that East is actually further that way. I think that’s super interesting and important that it’s that you get the context through having multiple, that the context, the more points of reference you have, the more effective the context is.
Alois Reitbauer (02:17)
I think at some point it’s also too many different types of context, so how much do you need? Because otherwise it gets just a distraction, it’s just overwhelming. think that’s the problem with those are…
James Governor (02:25)
Getting higher in this London Eye, more
more overwhelming. On that note though, I’ve got to say, think it’s super interesting. At ground level, you don’t really have any trees there, quite how green it is. You want a vantage point to actually see the parks and life of the city. The higher you get, the more you see. But yeah, okay, well how much context is too much context, Alois?
Alois Reitbauer (02:48)
I think at a point where it doesn’t necessarily affect any decisions that you’re taking at this conversation a lot, like how much data do you actually need to collect? And I think for the first time ever entering the age of we want less data because data obviously also costs money because we’re no longer talking about megabytes or gigabytes, we’re talking about pet petabytes, at some point exabytes. Which data are you actually looking at? What actually drives decisions? If you don’t care about the data, why do you actually collecting it? Why do we even have it? When does it become in this direction? think.
We are right now to some extent almost on the transition from okay data towards events. I don’t care about the data. I only care about if there’s something interesting in the data that eventually makes me look at it, but not looking at the raw data.
James Governor (03:30)
And so how are you managing that from a hierarchical perspective? Or how do you decide when we actually have enough data? What’s the approach at Dynatrace?
Alois Reitbauer (03:38)
I think it’s an approach that we figured out for us like a very long time ago when you talk to most people in operations the number one question they ask is what changed? Like what is normal and what is now different than to what normal is? So reducing it more or less to the change points or what changed in your data. Like even here if you have to drive right now we would most likely know how to get from here to your place. But suddenly if there is like a
graphic jam, blockade, or it’s just a street closing? You would have to rethink like what changed? How did my perception of how I get from A to B change? And what are the actions that I now need to take? So it’s about information that is actionable in the sense it requires you to take action. And how much do you need? So that’s why we moved into the anomaly detection, trying to combine things and leaving the rest out of it and showing you, okay, this is what it was before and this is how it is right now.
and then trying to make more sense out of it. But for like just the normal stuff that didn’t change at all, even taking this out of the equation because people might just spend a lot of time looking at things that actually do not matter.
James Governor (04:45)
Well historically in London I mean I think that anomaly, you’re in a city where there are new buildings so if we look over there, see where is St Paul’s.
Yes, we’ve got St Paul’s over there behind the Oxo building. Now historically, St Paul’s was line of sight everywhere in the city. St Paul’s was just the biggest building. Whereas now, of course, you’ve got all these towers. Now the first place you look for is the Shard over there. Or some of the taller, the walkie talkie. We’re looking for different reference points. Where historically in the city, yeah, wherever you were, you’d be like, oh yeah, there’s the dome of St Paul’s.
So all of these buildings, think, are anomalies. In some ways, gets harder to know what’s going on. But then you end up with even more impressive waypoints.
Alois Reitbauer (05:34)
When we were standing down, we thought that this thing is terribly slow. Now we’re in it, we’re going round, we perceive it as being way faster. So which of our reference points is now correct? As we’re going back to ground level, I for me it’s like the whole perception of how fast the thing is going. Just by having the building right next to you. So more precision actually, where the sky and like the smaller buildings in the background. It is more accurate, it even feels faster right now than it did.
James Governor (05:57)
I think it’s more accurate because we’re close.
Alois Reitbauer (06:02)
Once we started to see the building think the closer your reference point is, your perception changes. But the question is now what is actually true? Because all of these things were true at a certain point in time for us. Because it was the best perception we could get out of it. So reference points do matter and agreement on reference points do matter. And that’s the interesting thing, we wanted to look at London as a city, but we realized just our means of observability, which is the London Eye, we were even not entirely sure how that thing is behaving.
James Governor (06:33)
We’d have to go on it a few more times to understand the behaviour and we’d be in better shape. There might be some people that didn’t mind going on it again.
Alois Reitbauer (06:41)
Now we see the plumping of observability. It is just as pretty as most people would think, right?
James Governor (06:46)
Well, that’s your world. You love living in that. Somebody does have to do the plumbing, the nitty gritty work. Yeah, but we can do We’ll leave that to you, Alois. You can do the plumbing. We could do You can do the infrastructure. That’s what you do at Dynatrace, isn’t it? So that was excellent. I’ve lived in London for many years. Certainly even before the London Eye was built, I’ve never been on it before. So in fact, you did bring a new context for me on that journey. So, yeah.
Alois Reitbauer (06:57)
That’s what we can do.
James Governor (07:15)
We’ve talked about context, we’ve talked about data volumes, we’ve talked about situational awareness, and basically that’s about observability.
