Sean Sebring:
Howdy everyone and welcome to another insightful episode of SolarWinds TechPod. I’m your host, Sean Sebring, joined by my brilliant co-host, Chrystal Taylor.
Chrystal Taylor:
Hello.
Sean Sebring:
In this episode, we’ll be discussing the realm of possibility. We’ll look at trend predictions published by analysts such as Gartner, Forrester, and VMblog. We’ll be joined by none other than our very own Sascha Giese, Senior Technical Product Marketing Manager at SolarWinds.
Sascha Giese:
Hi, guys. That is a very long title. It’s actually my inofficial title. These days I’m called Global Tech Evangelist for Observability or something crazy. As you probably hear from my voice, I’m from Germany, but don’t you worry, I bring good news and we talk about the future, about next year I think is our topic, isn’t it?
Sean Sebring:
Yeah, so today could be 2024 and beyond as Gartner likes to focus on beyond as well. But yeah, we’ll be looking at trend predictions from a handful of different sources and just discussing our own thoughts on these predictions.
Chrystal Taylor:
Well, let’s kick things off with the AI predictions because there are many. Even amongst the ones that I pulled, there are several for each, analyst organization had several around AI and I think it’s a pretty obvious place to start. I mean, even just looking at how much AI has grown this year, generative AI being used more frequently, everyone getting so on board with ChatGPT and all of that stuff. I mean, there are complications around that, which I think that is very interesting as well. There’s not a lot of regulation there yet.
Chrystal Taylor:
There are some predictions which I’ll start with.
Chrystal Taylor:
The first one around regulation is AI trust and risk management from Gartner. So they’re saying, this is a beyond. By 2026, enterprises that apply the chosen controls to AI applications will increase accuracy of their decision-making by eliminating 80% of faulty and illegitimate information. And all I have to say to that is, “Duh.” AI is only as good as the data its based upon. Everyone should know that, and if you don’t know that, please look into it a bit more because it is very important for the data that it’s backed by to be trustworthy and legitimate.
Chrystal Taylor:
It’s really interesting because AI is having to be dealt with with base consumers. It’s not just tech that is dealing with it. I mean, I’ve seen stories of them having to come up with controls for schools because kids are using generative AI to write their school papers for them and things like that, and so it’s being used everywhere and it is obviously helpful. I mean, Sean, you and I have talked about how sometimes we use them to help us write an introduction or something boring where you have to go and make alterations.
Chrystal Taylor:
Obviously, it’s not perfect the first time around and it doesn’t have any personality for you, but it can help you get started and I know quite a few people that use it to help them write code because it’s much faster than you having to go through and write. They’d be like, “Oh, write me a script to do this,” and then you just make alterations to it where it’s needed. And I find that really interesting. In my mind, that’s what we should be doing is using technology to make our tedious work easier and not write our papers for us as children.
Sean Sebring:
Well, I a hundred percent agree and the school thing is actually exactly where I was hoping to go with this because I do use it for work and it’s the same argument, but now in the future of you’re not going to have a calculator in your pocket, and guess what we do? So if that mindset was 20 years ago, then today being frustrated about the use of AI as cheating is equally as silly to me almost because why would we not want to leverage it? I mean it’s not to take away from the fact that assessments can still be done without AI. We can have proctored exams and oral exams do live testing. There’s nothing against that. I think in fact, the ethical education of utilizing AI would make more sense than trying to say that it’s potentially damaging education because also just as someone who does use it, I use it to help me format check, spell check, grammar, all that stuff. It’s got a wealth of knowledge to help check against that stuff. And so it can be insightful.
Sean Sebring:
I’ll say, here’s what I want to say and I’ll structure it as best I can just to see if it comes back with little to no recommendations. But when it does correct, it’s just an instant correction for me. And so I think you can learn from it just the same as if a teacher had proofed your paper and gave you notes and feedback on it.
Sascha Giese:
I think that quite often people as well as big companies don’t even realize how much AI they already use because it’s in so many tools and solutions that we all use. Hey, if I just look at my browser, there’s a browser tab by Grammarly the spelling correction thingy, and that is to a small percentage AI driven. Microsoft Office has new tool, what’s it called Co-pilot, autopilot or something like that. I don’t remember. It’s using AI. So sometimes I guess we don’t even realize, and that could be potentially scary because one day someone wakes up like, “Oh, I was using this machine all the time.” But on the other hand, it’s a natural progression. But you said something interesting there about the ethical use, that is a big topic in itself actually because before an AI can be used, it needs to be trained and the process of training an AI is basically how you explain your kid, how the world works when it’s a small kid. And unfortunately most AIs are just trained by a very small group of people of like-minded people and humanity is diverse, so the teams who create AI should be diverse too. So the AI has a better understanding of how the world works. We’re not there yet, but I remotely attended a conference in the UK a couple of weeks ago who are strong supporters of getting more female input in those teams just for more diversity.
Chrystal Taylor:
Yeah, it’s definitely a topic we’ve been seeing a lot over the past year or two, whereas it’s become more popular that there is a problem with bias control in AI, and it’s like anything else. When you’re creating it, you start with a small team. I mean even if you have a larger team, everyone’s in the same field. There’s a lot of people with similar interests, so you lose some of that diversity even if you think that you’re trying to, you still lose it because you don’t have the lived experiences that other people have that are different from you. So it’s interesting to see where it’s going. I think that we are, because it is a topic of conversation for humanity, as consumers are more using it that we are going to see some additional controls around that. We’re going to see conversations continue around improving because there’s no way to continue with them being completely biased.
Chrystal Taylor:
I mean, as I said earlier, it’s only as good as the data it’s backed by. So if the data is biased, then its results are always going to be biased. So it’s something to keep in mind as people use AI. And like you said, it’s used pretty much everywhere now. I mean our applications use it. It’s all over the place, so you’re going to have AI, you can’t ignore it.
Chrystal Taylor:
I do want to talk about real quick the fear that people have. AI is going to take my job kind of a thing. I think that that’s an interesting topic because it happens every time a new technology comes out. New technology comes out and people all of a sudden there’s a fear of redundancy and as we advance, it gets more and more. And I think that what’s interesting to me if you look through history is this does happen all the time. It’s just become a lot more frequent because we are advancing at a much more rapid rate than we used to. And so in my opinion, you should always be aware of the fact that you could be made redundant at any time. So you should always be sharpening your skills or learning something new or being prepared to go take on a new job or role or whatever.
Chrystal Taylor:
A friend of mine used to say, “If you are only a DNS admin you’re going to be made redundant if you never tried to learn anything out.” Yeah, just don’t fear, don’t fear the AI, but be cautious with it would be what I would say.
Sascha Giese:
Well, I don’t think a DNS admin will ever be redundant because it’s always DNS, right? But anyway, I had an interesting discussion with a guy from Microsoft maybe half a year ago. We met at a random conference. It’s my job to go to conferences and he told me some interesting stuff. They had two or three AIs and trained them to create their own language and this was so successful after a few days they couldn’t understand anymore how the AIs communicate and that was the point.
Sascha Giese:
Yeah, that was the point where they shut down this experiment because it was absolutely crazy. But he also said that his prediction was like in 20 years the majority of us in IT will become AI prompters. Could be possible.
Chrystal Taylor:
Interesting.
Sean Sebring:
I can totally see that. And that’s kind of where I was talking about with education, being educated on how to leverage it because just to go back to ChatGPT, because I’m much more familiar with it, I use it in my work to help me with formatting again and a little bit of content creation from time to time structuring things and the input, your question that you ask AI is going to influence what it tells you. So there could be two people that go to the AI with the same goal in mind, but based off of what they ask it to do and how they ask it to do it will definitely give them different results. And I think that there’s a skill to leveraging AI. And so, one of the predictions from Forrester says AI will, which they also contradict this later, but AI will shake up the cloud market.
Sean Sebring:
I don’t necessarily personally agree with that. I think AI is just a tool that all things will be able to leverage. It is a cog, it is a piece of all things moving forward from a technology perspective. It itself isn’t a standalone thing that you really have to worry about. It’s just going to help enhance, improve, make more efficient and potentially change. The only thing like Chrystal was saying is change how we look at things. The jobs will not necessarily disappear, but they will change from what were you doing now you’re going to leverage AI to do it.
Sean Sebring:
I think about the self-checkout instead of there being tellers to scan your items for you and ask you for the money, you do it, but now that same person stands there and monitors the situation. So their job didn’t go away, it just changed a little bit. Instead of swiping the groceries across the scanner, they’re watching for you or now even loading your stuff into bins and wheeling it to your car for you because you did a pickup. So things are just changing. They’re not going away necessarily.
Sascha Giese:
Well, I was in Lithuania two weeks ago for business and across the road from my hotel was a very small shop, like a convenience store 24/7 open, AI-driven. So you enter the shop and the first thing you have to do is to swipe your master card or whatever and there’s like a million cameras on the ceiling so they actually follow and whatever you pick from the store and you go to the machine to pay, it’s already listed there. It was a fascinating experience. I never moved into such a place. So as I went to paying it already sat at two bottles of water and a small croissant. That was totally amazing experience. No humans. I guess there were humans somewhere hiding and occasionally show up to refill the store. This was really an interesting experience, but to the point of AI will shake up the cloud market if replace cloud market and just call it data centers.
Sascha Giese:
I do see a little bit more of AI there in regards to converting speech to CLI commands. So the system will be a tool, as you said, Sean, and the tool talks to Cisco and Juniper devices the same time as if there’s no difference. And also obviously two orchestrators. So far, we would need to create our commands, whatever, the AI could just do this for you, “Hey, spin off 20 more machines for customer, blah blah, blah.” I would see that happening.
Chrystal Taylor:
Yeah, you brought up data centers and I think that sticking with the AI theme that we’ve got going right now, there was another prediction which I found interesting that it was an alternate prediction from the same analyst agency that AI will hit a wall due to hardware limitations. And this I think could happen. We’re already starting to see limitations because of GPUs on the market and things like that with crypto and the supply chain problems we’ve experienced over the past couple of years. And I don’t think that that’s gone away. It has lessened. You can get your PS5 now and things like that, but it’s certainly lessened over the years. But I think that the prediction there that it’s going to be hitting hardware limitations because they can’t get the hardware. I think this is true in a lot of cases, in a lot of different areas.
Chrystal Taylor:
You can’t get the silicon you need, for instance, for a computer chips sometimes. They had the same problem with getting the chips for vehicles two years ago. They had the same problem with… My dad works in controls for climate controls and things like that for air conditioning and that kind of stuff, and they would have a hard time sometimes getting their controls because they have computer chips and it’s the same kind of problems. There is a limit to what we have available and where we can get it from and how long it takes to get there. So now it’s taking even longer to get those things, which may put a limiter on how often they can use that. If you don’t have the compute power, if you don’t have the budget to spend on the compute power, you may not be using AI as much as perhaps someone at your business wants you to. I can see that being a more realistic prediction.
Sascha Giese:
There is actually something happening. So most of the gear is Nvidia. They sell some specific chips for AI. They have a ban on selling those chips to China. You can Google it, but there’s no ban selling Nvidia 4090s in China. So China ordered like hundred thousands of them and repurposed them into their, I don’t know, Alibaba or whatever their data centers. I read about it last week. It’s quite new story. It’s interesting.
Sean Sebring:
We’ll have to ask AI to help us think of an alternative solution to silicon so we can continue to manufacture these things en masse. Not too similar, but similarly just talking about, I wouldn’t call it a limitation, but a change is the energy efficiency. We talked about this one pretty briefly when we were preparing for this and I thought this one was incredibly fascinating and could lead to more, but energy efficiency, it’s talking about using data centers kind of in a follow the sun model so to speak. So that where the, and Sascha, you can probably expand on this for us, I think you had talked about it a little bit while we were preparing, but having the load follow the sun so that it’s leveraging solar energy while it’s performing its job.
Sascha Giese:
That is a session I attended a couple of weeks ago and it was really fascinating idea. So if you have a very light workload that sits in the cloud and feeds its data into a central location, this workload could potentially and automatically follow the sun just like you explained it. So go from data center to data center, wherever the energy is sourced based on renewable sources. So probably Texas a lot, I guess there’s a lot of sun, but there’s a light in Texas, so where will it move? Will it move to Europe, will it move to Australia? I don’t know. And if we just talk about the script a few megabytes or so, the cost of transferring or running is neglectable compared to the energy consumption based on renewable resources.
Sascha Giese:
I thought that’s a brilliant idea. It’s obviously something that won’t be able to… Won’t work with any work node. If you have a big VM somewhere, the cost of moving that thing around wouldn’t justify the idea. But in general it’s a good one and I think the whole situation with being sustainable and energy efficient is not limited to AI. I would say that is one of the bigger predictions for next year, which is not solely focused on AI. Something has to happen, and I think data centers are the second-worst element when it comes to sustainability right behind flying around the world.
Chrystal Taylor:
Sticking with the kind of green initiatives. We had talked about this, I want to say last year you made a prediction around green initiatives and the green code I think is what we talked about last year, but Sascha, so you guys can look that up. But I think that it is fascinating that there are more predictions around improving our footprint in tech like company-, industry-wide. There’s a prediction that I found really interesting around data centers exploring heat reuse. And this prediction came out of the fact that there is currently legislation being discussed in Germany… It was in Germany… to require them to reuse heat in their data centers as energy, a certain percentage as a renewable source of energy in their companies. And I found that really interesting and the predictions around basically the rest of Europe and then the rest of the world potentially following suit if that regulation becomes law, becomes a required thing. So I found that really interesting. I don’t even know how you would go about it, I guess any other geothermal or something like that, things that rely on heat to convert it into energy, but why wouldn’t we try and do that?
Chrystal Taylor:
I don’t know what the cost behind that is but one of the predictions or the things that I’ve been listening to recently was talking about how businesses are not going to invest that much in green initiatives basically if the cost is higher than the result. If they can be proved that it’s going to save them money or something like that, or if it’s going to save face for them, that maybe is potentially enough of an incentive. And I think that is A, sad, but B, probably realistic.
Chrystal Taylor:
But yeah, I find it really interesting to explore the heat reuse in data centers. I mean, everyone knows they get hot. There’s a lot of computers running in one room and even though we have improved over the last several years, have improved cooling systems and all of that for data centers, I think that there’s still quite a bit to be done in that area.
Sean Sebring:
I don’t know how scalable this one is, but my best friend works for LCRA, which is… What is it? Lower Colorado River…
Chrystal Taylor:
Authority.
Sean Sebring:
… Authority. Yeah, there you go. LCRA. And they actually have a really, really cool way that they leverage the Colorado River to cool their systems. They actually use it. It comes from the Colorado River and so they kind of have their data center being cooled by a natural resource. And so it’s similar to wind and solar energies, just leveraging an existing source to take care of a job without having to burn more energy. I’m sure that at scale that could be a problem. We don’t want to heat the ocean. Not that that would happen, but you got to be cognizant of what’s downstream of the heat that we’re adding to the Colorado River. Is it going to affect anything negatively? But it’s just another kind of cool use case of, hey, let’s just let this take care of itself. We run the Colorado River, let’s use the Colorado River to cool our systems.
Sascha Giese:
Of course, I mean all these things are baby steps, but every little helps.
Chrystal Taylor:
Yeah. Speaking of baby steps, we talked about this a little bit in our pre-talks, but I do want to bring up the other major sustainability prediction that was in there, it was in Gartner. And their prediction was, “To support long-term ecological balance and human rights by 2027, 25% of CIOs will have compensation linked to their sustainable tech impact.” And Sascha’s laughing now, but I remember when we had this conversation, we were all a bit skeptical, so I don’t know where they decided that this was a thing. I’ve never heard of anyone compensation plan being tied to it yet. So if there’s any out there that would be interesting to find out. That’s where they got this prediction from. It’d be nice if it was.
Sean Sebring:
I love this prediction. Yes, but have no idea who, well, obviously Gartner said it, but based off of what data are you creating this prediction, what legislation. I mean I love the idea of more accountability like that. I think that it would be amazing. But yeah, I would love to see more of why that is a prediction and maybe we could look into that more. But I agree. Skepticism it was visible in Sascha’s laugh, your smile. Yeah, it’s an interesting one. That’s definitely probably the most interesting one on here because where did that come from? I never would’ve expected to see that in a prediction.
Sascha Giese:
It is probably coming from an AI.
Sean Sebring:
They’re trying to influence us with some accountability. That would be great.
Sascha Giese:
No, I mean look, there’s different C-levels who are more accountable for directions or things that happen to a company and yeah, I’m totally with you. Accountability should be a bigger thing for C-levels, but 25% of CIOs will have compensation linked to the sustainable tech impact. Sorry, that is not going to happen.
Sean Sebring:
So here’s an interesting one. Platform engineering seems to suggest moving back to internal platforms and this is actually data-driven. You can see this happening. So there is a trend to it and from Gartner it is a prediction that it will continue to suggest moving back to internal platforms. What do you guys think?
Chrystal Taylor:
Well, I found this interesting because they have the flip side, they have the flip side prediction as well where use of cloud platforms were increased.
Chrystal Taylor:
“We’re increasing cloud, but we’re also going to increase our moves back to internal platforms.”
Chrystal Taylor:
I think this is realistic though. VMblog had similar prediction there as well where they’re starting to shift back to data centers for some things. And I think that this is an interesting response. For me, it’s similar to the reaction we had to the pandemic. At the beginning of the pandemic, we all had to make a choice of how we were going to support everyone at home and we put in plans and we made quick decisions of this is how we’re going to support it. And then after people started going back to work or things like that, they had to revise all of that stuff.
Chrystal Taylor:
You make these quick decisions of like, “Oh, this is the hot thing, this is the thing we need to be doing.” Some executive has made a mandate that says we must be X percent in the cloud or whatever, and you have to follow through on that. Well then you can start to see diminishing returns on some things. It doesn’t make sense for everything to be in the cloud necessarily. It’s a tool just like any other, it’s a place to host things, but it doesn’t necessarily mean it’s the right tool for the job every time. So I think that we have come as an industry to the realization that it doesn’t solve our problems and sometimes it costs us more, and that I think is where this shift is coming from. It costs us more or it is not as accessible as we want it to be or we don’t have as much control over it as we want to have.
Chrystal Taylor:
And I think that that’s where these predictions are sort of coming from is that developers, especially when they’re developing smaller applications, they want to be able to touch all the pieces of their code and everything, especially during the initial development process, they need better access to it. You eventually you transition into a higher scale thing as you’re developing it, but in the beginning especially it makes sense maybe to have that all more local where you can have better access, easier access. You don’t have to worry about as much controls or the cost of transferring data to and from the cloud and things like that. So I think that this prediction makes sense in a lot of ways. We’re coming to the realization that it’s not the answer to every problem.
Sean Sebring:
I think like most things that trend which cloud was trending and for good reason too, it can be a good solution. And so it trended not just because of its capability but because of what it said about you as an organization. It was the more modern look, the more modern feel, the more mature, more capable. So there was a lot associated with moving to the cloud from the stigma perspective. And so now that the stigma is kind of a little over, people are actually taking a more realistic look at things and saying, “Is this better for us right now that it’s not just a trend then we should be going there. Everyone else is doing it.” They’re having a more realistic look at is this the right move for us as an organization too.
Sean Sebring:
And I think the second thing is, like you mentioned Chrystal, this was probably the first thing to me, but the second thing is cost. I am a video streaming subscriber to many platforms now and they all continue to go up so at some point you realize that the subscription model is starting to do a lot more than you realized to your bank account. The same thing could be true from a cloud perspective. You just have to weigh what makes the most sense. This could be a move that’ll be a stick-it-to-them from a subscription perspective that hey, if you want cloud and SaaS to continue to be the number one choice, it has to be affordable too.
Sascha Giese:
I was actually talking to a very big company on a conference who are in exactly that situation. I tried to leave all the names out of it, so they got a huge bill from their cloud provider month by month by month, by month by month and they bit the bullet, okay. But at some point they were like, it’s just not feasible anymore. So they checked what old infrastructure they still had on site, which was basically collecting dust but was still okay. They bought like five, six more servers, modern servers to give it a little bit more nice touch and put their development system back on prem. Now they got a huge bill for the software which they’re using, I give it the name, this is OpenShift, okay, so it’s not exactly something that’s really cheap, but it was a calculation for them and they have this with a maintenance contract for an X amount of years and it is significantly cheaper than running this in the cloud. But Sean, as you mentioned, some workloads like a mail server or collaboration platform they’re perfect in the cloud. There’s no need to keep those things on prem, particularly in the times where we live, where people work from all over the world and just even from a cafe or whatever, it’s just much more simple to keep such things in the cloud.
Chrystal Taylor:
Yeah, I like the idea, Sean. You were talking about subscription services earlier and I like the idea that we’re coming back around to the realization that maybe basic cable pricing wasn’t so bad.
Sean Sebring:
No, and it’s a convenience thing too. So even as now that Netflix streaming or Hulu, we’ve gotten so used to having what we want on demand right there, whereas the cable approach was “Is it on TV right now?” Thinking back to the channels that you would flip through. So it’s just, again, it’s a cost-benefit analysis and I think that people are taking it a little more seriously now because it’s not just the trending thing of this is where we’re headed, the cloud. It’s not that it’s going away, it’s just that people are looking at it I think more realistically now.
Chrystal Taylor:
Yeah.
Chrystal Taylor:
Switching gears a little bit. We talked about supermarkets earlier, which is why it’s driving me to this next prediction that I saw, which was that we’re going to have more machine customers, which I found really, really interesting as a concept which were for online shops and things like that, each shops you have essentially a bot that can behave like a real customer. It’s not just a RUM, real user monitoring, or a bot that is prescribed to do a set thing, but they use AI to generate a personality. So you don’t know what it’s going to buy, you don’t know what it’s going to do. And I found that really fascinating. I think that would be really interesting, especially for, I mean I worked retail for many years, I think that it was a really fascinating concept to try and improve your stores by getting data from… I mean effectively it’s a script or whatever that’s running and if this, then that and if not, then this other thing.
Chrystal Taylor:
So I think it’s really interesting that we can walk through the whole steps even better than synthetic user monitoring or real user monitoring. It’s kind of the next evolution of synthetic user monitoring I think is what this machine customer sort of resembles to me is that right now we have the ability to do synthetic monitoring where you can pretend to be a user, but you tell it what steps to take. You tell it to go download a file or go put thing in a cart and whatever, but you can’t complete the transaction. You can’t go all the way through the entire process. There are limitations to that and of course it is prescribed, so at a certain point you may run into things like caching and things like that that are happening. So you’re getting an artificially lowered sense of how long the process would take for a new user for instance because if you’re not turning off that caching and things like that. There’s all those limitations to those things which are great and useful, but I think this is the next level of it. So I find it really intriguing to see where that’s going to go.
Sean Sebring:
I’m curious, I don’t quite get the use case for this year. Is it while developing, let’s say a website or an application so that they get feedback from a bunch of fake customers too. Okay.
Chrystal Taylor:
QA.
Sascha Giese:
It’s testing and the AI is probably creative, can do unpredictable stuff like humans would do, but obviously cheaper than humans. But now I’m just wondering what the AI is doing with those 20 boxes of toilet paper that arrive.
Chrystal Taylor:
I’m sure they have a developer customer ID that they can then remove all that stuff from orders or things like that. There’s got to be some backend management that’s going to be required for these things but use case scenarios, I think it’s really quality assurance. You’re not only able to test how long it takes them to get things in their cart and get through the transactional process and whatever. You’re able to see, depending on how the AI is built, maybe it’s looking at trends and things like that so that you can see what people are buying more of based on what this AI bot is doing.
Chrystal Taylor:
I can see lots of use cases, but I was in retail for quite a while, like I said, so I can understand how this would be really useful to an electronic storefront especially to go through the process of if it’s collecting data from elsewhere or maybe it’s fronted with collected, these are the things that people have bought the most in this time period in the last whatever. And so it goes through the process. I don’t really know, but I find the whole idea of it very interesting. I think it could be useful in the end.
Sean Sebring:
I have actually, that triggered something for me, where it’s similar I think exists already for… Okay, I’m a gamer and in fact I think all three of us are, we are all three gamers, but I play a mobile game on my phone and it’s a 5v5 for the most part. And oftentimes I’m like, “I swear this guy is a bot.” I’m like, “I swear the person I’m playing with is a bot.” And so I talked some trash and then guess what? The bot talked trash back and I’m still not convinced that he’s not a bot just because he did that and I’m like, and there’s a reason that they would want to put bots in there because they want to inflate it so that matchmaking doesn’t take so long. You can always play and if they don’t have enough players, they’d need these machine customers kind of like the prediction, but they’d be players instead. And so I believe AI is not just playing the character but also communicating with the team based off of how they’ve seen the other players communicating, which means they’re really good at trash talk too.
Chrystal Taylor:
See that just makes me think what’s the next level of that. This makes me go into the malicious side of things like they’re going to start using and maybe they already do, I don’t know, maybe they’re going to start using bots to catfish people and all these other things. Obviously there’s two sides to every coin. There’s the benefit that we can see, but we should never forget about the other side of this where people can use these things for ill.
Sascha Giese:
Isn’t there when you are a chess player, they have this Elo rating I think. Have you ever heard of this?
Sascha Giese:
And I mean if you use machines to push your ratings, I could see that as a consequence.
Sean Sebring:
So going back to cloud, another prediction from VMblog is cloud infrastructure security, probably not just security in my opinion, but here it says security will shift left away from monitoring to more proactive. I feel like that is and has been happening so I’m wondering if they just mean more of it now that AI is becoming more capable. I feel like monitoring is not a thing of the past, but if we can tell our tools exactly what we’re looking for and then ask the tool to do exactly what I’m going to be doing, right, it’s very scripted for the person who’s reacting to the monitoring. So this totally makes sense to me and it’s already been happening and I agree it will continue in 2024 and continue to grow and get bigger and bigger.
Chrystal Taylor:
Now Sean, I’m a very experienced monitoring professional and I don’t like you saying that monitoring might be a thing of the past, I think that it is…
Sean Sebring:
Your job is only changing, Chrystal, not going away.
Chrystal Taylor:
No, of course not.
Sascha Giese:
That’s why we call it observability these days.
Chrystal Taylor:
Yeah, but it’s evolving. It’s evolving. I don’t think that it is ever going to be a thing that isn’t used, but the fact of the matter is that we do have to get to a more proactive state. And I think you’re right, that is already starting to happen. We’ve definitely seen it, but I think the idea behind the prediction is that it’s going to become more common, that observability and other proactive forms of monitoring whatever they happen to be. And proactive resolutions, which I think is the key here. It’s talking proactive in the monitoring and everything like that but I think resolutions is where the key to all of this is.
Chrystal Taylor:
If you can get the knowledge ahead of time to take preventative action, so like forecasting, we’ve been doing that for years. That’s very mild understanding thereof. We can forecast in six months you’re going to run out of space based on the usage that you’ve had over the past six months or whatever.
Chrystal Taylor:
Those things you can then predict and say, okay, well to be proactive in the two months before I’m just going to add more storage or I’m going to offload X amount of data somewhere else or whatever. You can make these decisions ahead of time, which prevents downtime. And I’ve talked about this before on TechPod where we have reached a real level of impatience as a global society where we’re not willing to wait for things at all. So page load times can’t be… I mean if it’s taken more than five seconds, people are leaving.
Chrystal Taylor:
These are real problems that businesses are facing that they have to deal with. And the proactivity, which is becoming more and more common is helping them to get ahead of those problems before they become a problem. If your e-commerce site goes down for five minutes, maybe that costs you $600,000 or whatever. It’s tied to real dollars. So that means that they’re spending more attention and more time into investing on the front end so that those things don’t happen.
Sean Sebring:
I have an interesting example of the proactivity from a security standpoint and a much more practical application. Also, just quick note, talking about us demanding consumers. Moving from the U.S., I now live in Ireland, moving to Cork, Amazon’s just not the same. I can’t do same day deliveries and so I’ve reset my expectations when it comes to my demanding consumerism.
Sean Sebring:
But anyway, another thing about the travel and hats off, I had been a Google Fi customer and when I went to France last spring and when I moved here to Ireland again proactively my phone just said, “Hey, I noticed you’re here, don’t worry, we’ve got you covered based on your plan.”
Sean Sebring:
And kind of similarly again is banks. Since you can install your mobile banking, it actually knows where you are and it will see that you’re next to an ATM. It can say you’re in a different country. Do you need assistance from a security perspective? So those are really proactive ways rather than, whoops, I got locked out of my card because I traveled and forgot to update something and then I swiped in a different country and now it’s fraud, and so they locked it.
Sean Sebring:
So they’re adding it to all sorts of different things. Again, it’s already happening. It’s just continuing to trend and I’m really happy with it. I’m also one of the people who doesn’t really care if Google gives me ads because I’m like, “You know what? I actually did want that.” You’re listening to my whole house and conversations all of a sudden stuff’s showing up. I’m like, “Yeah, but actually I was talking about that I was interested, so thank you for sending me an article.”
Chrystal Taylor:
Don’t get me started, Sean. It’s different conversation. Conversation for another day.
Sascha Giese:
The Amazon thing actually, there’s a different reason and that is the Brexit. So previously Amazon stuff came from the UK. Now it usually doesn’t because you have to pay extra so it’s being shipped from the continent.
Sean Sebring:
Well, Sascha, that’s why I use the German Amazon store more than the UK one being here in Ireland.
Sascha Giese:
Yeah, it’s cheaper. Even the shipping costs are more expensive.
Sean Sebring:
And the time too.
Sascha Giese:
But what I probably will see, and this is the last time I’m talking AI for today, I promise, it’ll remarkably change the way we access and process data. So there’s no more going to DBA and “Hey, can you run that query for me and put it in a nice chart?” Or we use super clunky tools like… We use Tableau here. It’s a nightmare. At least for me it’s a nightmare. Wouldn’t it be totally awesome if it could go just like, “Hey Siri, how much toilet paper did we sell in Texas in the last quarter?” And get those results instantly. I could see that happening more and more.
Sean Sebring:
I love that idea.
Sascha Giese:
Yeah, my Siri is just replying. I don’t know what she told me.
Chrystal Taylor:
She’s trying to answer your question.
Sean Sebring:
There’s your AI for you.
Sean Sebring:
No, I love that. I like the accessibility to that kind of data, which is again why I’ve been such a fan of the availability of ChatGPT. It’s just like having a handy assistant that can answer a bunch of little questions for me and I don’t have to put a note for myself, “Hey, remember to go research this,” and then I forget because I didn’t actually make the note. I can just do it on demand and get some answers and feedback live right there. It’s super cool, super handy.
Chrystal Taylor:
Well, I’ll switch gears again because the last one that I want to cover of the predictions that I found really interesting is of course around learning, which if anyone has ever heard me talk about anything before, I do like to talk about keeping your skills up. I mentioned it earlier.
Chrystal Taylor:
This one is that we’ll see an increase in the use of intelligent technology for skill augmentation and accelerated time to competency. And I found that super interesting. So the prediction is the time to competency thing is for onboarding new employees. So using intelligent technology instead of using an individual person and walking them through and holding their hand for six weeks or whatever to use our technology to increase the time to them actually being able to return results on whatever their job responsibilities are. And I do think that there is a good place for technology in learning and skilling up.
Chrystal Taylor:
I mean there are so many opportunities now that we didn’t have even 10 years ago to learn stuff. I don’t feel the need to go and get a master’s degree in a regular institution and go to classes and all of those things and no shade to anyone who does that. I think there are different paths for everyone, but there is so much available able online and through classes and through the use of AI that you can learn things. And as I say that I said earlier, be cautious because it’s only as good as the data that backs it. The same thing is true of any learning classes that you take or any online free classes. There’s a lot of materials out there, there’s a lot of really great things available. So many people are writing blogs and all of those things to help you learn new things because we all learn different ways, but make sure your source is trustworthy.
Chrystal Taylor:
Obviously there’s still not great things out there too, just like there always were. I mean I can definitely remember learning things in school from school textbooks that are not right. So you learn later that those things are not the whole story or they’re not telling you everything you need to learn. So there’s definitely an element of that behind it. But I like the idea of using intelligent technology, whether that be in the form of AI or whether that be in the form of a machine customer. Maybe that machine customer you have to be on the other end and you watch how it goes through the process and that’s part of your learning experience or anything, whatever that happens to be. I think that there’s room for it, and in my experience, a lot of times companies do not budget time appropriately for onboarding new employees in the sense that the person that they use for the onboarding that already works there has to do their full-time job on top of also onboarding the new employee.
Chrystal Taylor:
And they don’t give them additional time for that. They don’t offload their workload for that time period or anything like that. And that’s not true for everyone, but it is often the case where you’re just expected to do your regular job and also let me teach you.
Chrystal Taylor:
Even back when I was a server. You are a server, you go to a restaurant and you have a trainee server that’s going to be there. They’re just walking around with the full-time server doing everything that they’re doing following them around like a little puppy. They go everywhere they go, they are teaching them the whole time that they’re actually doing their job and it’s all walks of life. It’s not just technology that does this, but I find it really intriguing. Maybe that’ll help offset some of that workload of training a new employee to do your job or the junior level of your job or whatever. If you can have a little bit more assistance from technology to help them onboard.
Sascha Giese:
I’m stumbling about accelerated time to competency. That sounds amazing to me. But time to competency should be measurable then, isn’t it? I mean, when I’m not traveling, I try to learn Kubernetes, which is a nightmare. It’s taken weeks already. It would probably take months before I completely understand it. If a machine could help me with that, I would totally buy into it. I would love it, yeah.
Chrystal Taylor:
Well, and I know part of that is people learn in different ways and they typically have one way of learning a thing, especially if you’re onboarding a new employee. This is the set way we have of training a new employee. We’re not going to do anything differently. So the time to competency I think is different for every person, but if you can reduce it in any way, I think those should always be considered.
Sascha Giese:
I think we have a good example for ourselves. Look, I’m with SolarWinds for almost 10 years and when I started I’ve been told, “Here’s the demo website, there’s the documentation. Good luck.” That’s it. And nowadays, when someone starts, we have all these learning portals. We created our own. We actually have a department for that, so it’s manual labor, but I could see very well that in the future when such solutions exist and are affordable, it will be hot stuff. Everyone’s going to use it.
Chrystal Taylor:
I just was going to ask Sascha if he has any predictions of his own for next year.
Sascha Giese:
My prediction for myself is I will go to Bali for a business and I actually do, and I’ll attach a full week for pleasure just to enjoy the sun. That is the biggest thing I planned so far for next year.
Chrystal Taylor:
That’s amazing. But do you have any tech predictions? I know in the past we have, when we were Head Geeks, we had come up with tech predictions. Do you have any tech predictions that we didn’t cover in this episode so far of your own?
Sascha Giese:
Well, we talked about the green thing, which has to happen. I don’t want people glue themselves in front of data centers instead of the roads. I don’t know if they do this in the US, they do this here frequently. Maybe we can find proper technology to help us fighting climate change. That is definitely something interesting.
Sascha Giese:
What else will we see? We will see an over saturation in the market for AI frameworks, I guess because everyone comes around with the next best thing. And keep in mind when a company is evaluating like a normal piece of software, like one of our tools, maybe, the evaluation process can very well take a couple of months. Now imagine that with an AI system that will take ages. And actually there’s one more thing. So when a software vendor creates software that has AI built in, there’s two ways where it’s coming from.
Sascha Giese:
One is A, the framework. So I buy a framework and just adjust it or I create my own. Now remember what happened two years ago with Log4j, with a small piece of open source software, which everyone used. Suddenly there was like a zero-day in a piece of software and everyone had to change their tools.
Sascha Giese:
Now imagine that happens with one of the AI frameworks. Oh boy. The consequences could be really disastrous if you consider how powerful these things are and how very well-connected they are.
Sascha Giese:
At some point we’ll see AIs fighting AIs. Isn’t that almost a Terminator scenario? Almost? That was machines against humans but we would see AIs fighting AIs and we just sit there and eat popcorn and see what’s the outcome.
Chrystal Taylor:
I’m sure that we’ll have no repercussions on humanity.
Sean Sebring:
We’ll just have to fight Skynet with some solar winds.
Sascha Giese:
Yeah.
Sean Sebring:
What a fun way to end today’s topic on trend predictions, thank you for that Sascha, and thank you Chrystal.
Sean Sebring:
Let’s roll into our rapid fire questions round now. And since we’ve had Sascha as a guest before, we’re going to try and throw some different stuff his way.
Chrystal Taylor:
Sascha, what is your favorite video game of all time?
Sascha Giese:
All the time. Well, we are not recording this as video. Oh, my tattoo’s on the back. We wouldn’t see it anyway. I have the tattoo of Quake on my back, the very old game from the nineties with that round thingy and the nail through it because it was the first time I saw one of those, what was the name? Those 3D cards. Voodoo, I think was the name. First time I saw actual 3D, and I will never forget this moment in my life. Obviously this is like 30 years ago. I don’t play it anymore. Actually, there was a re-release. You can get it for $5 or something, but it’s showing its age. But the impact it had on my life was pretty amazing.
Sascha Giese:
These days I’m still amazed by Cyberpunk, the add-on, pretty great. One of the best games these days, and I’m still playing it once a year or so in a different way. So you can play it in different ways.
Sean Sebring:
That’s awesome. And also Sascha, perfect description of the Quake logo. I hope you all were able to visualize that the way I was, that I don’t know. I’m actually kind of curious now since we’re all gamers. Chrystal, can you share yours? I know it’s not really rapid fire at this point, but go ahead.
Chrystal Taylor:
My favorite video game of all time is The Legend of Zelda: Ocarina of Time, which is the most basic answer, but it is true. I have played that game, I want to say upwards of 15 times. And it doesn’t change like Cyberpunk does. It is the same story and nothing about it is different. And yet I have played it many times over. I love that game. I will say when the newer games came out, Breath of the Wild and now Tears of the Kingdom, that those are very close, like second place for me. But because of nostalgia, Ocarina of Time is going to be my favorite forever.
Sean Sebring:
That was my answer too. And there’s no shame in it. Just the orchestration, the music in those just so good.
Chrystal Taylor:
So good.
Sean Sebring:
In fact, they came out with them available for a Nintendo Switch online and I played through them with my daughter and that was great memories for us. She’s of course also now into Breath of the Wild, and we’ve done some Tears of the Kingdom, not as much as I’d like, but moving was a lot for us. But yes, agreed. Fantastic video game. Love it. And it’s not a basic answer, there’s a reason it’s a lot of people’s favorites.
Chrystal Taylor:
My current favorite though, that I’m playing right now is Baldur’s Gate 3, which is also probably not surprising me.
Sean Sebring:
I have yet to pick that up, I really want to.
Sascha Giese:
I played it, but I didn’t finish it. I don’t know, it didn’t really hit me. Didn’t work for me.
Chrystal Taylor:
I’ll tell you, my best experience has been playing with other people. So it’s good by itself, but it is great with other people.
Sascha Giese:
I don’t like people.
Sean Sebring:
Sascha, what talent would you most like to have? If you could just snap your fingers and give yourself a talent, what would it be?
Sascha Giese:
I have a keyboard in front of me. I would love to actually be able to play like piano. All I can do is very simple stuff. I would love to be able to play the piano.
Sean Sebring:
We get musical answers a lot for that, and I agree it’s something that you can admire, but you can’t just pick up the way you’d like to. It takes a special part of the brain and years of mastery for most.
Sascha Giese:
And it takes musical theory, which is like, “Yawn.”
Chrystal Taylor:
So Sean asked me this question before I was a host and I was just guesting, but I like this question. So if you could be one of the first people to go live on Mars, would you?
Sascha Giese:
In theory, yes, but I’m not sure how Amazon is going to deliver. But yes, why not? Why not? Interesting experience. Well, you have to wear those weird suits when you go from home A to home B. Maybe not so great, but I would totally do it, yes.
Sean Sebring:
Awesome, awesome. Yeah, it can’t be that much different than living in Ireland, right? It’s raining outside, you got to put on a big suit, walk to your neighbor’s house. So you’ve already practiced that.
Sascha Giese:
So you haven’t been to Ireland long. There’s the saying, “It never rains in a pub.”
Sean Sebring:
Okay. Let’s see. This is good insight about you, Sascha. What would you consider as your greatest achievement?
Sascha Giese:
My greatest achievement. Seeing how proud and happy my mother was when I was on the title of a magazine for the first time, of a print mag. That happened twice to me. I sent copies both of them to my mother and she was absolutely happy, and that was an amazing moment actually.
Chrystal Taylor:
That’s so wholesome.
Sean Sebring:
That’s a beautiful moment to even just picture. Awesome. If you could choose what to come back as, what would it be? If you were to reincarnate, that’s a fun one.
Sascha Giese:
Come back as like an animal or…?
Sean Sebring:
Yeah, you could be an animal, a rock, a tree. Yeah.
Sean Sebring:
What would you come back as?
Sascha Giese:
I would like to be my cat.
Sean Sebring:
Your cat.
Sascha Giese:
My cat is allowed to do everything, is totally spoiled.
Sean Sebring:
Okay, let’s take this further. Are you still the owner of yourself? Would you be the cat of Sascha and Sascha is still there?
Chrystal Taylor:
That sounds confusing.
Sascha Giese:
If I could swap with my cat instantly, I probably would.
Chrystal Taylor:
Freaky Friday-style.
Sascha Giese:
I would. I mean, I don’t see him right now.
Sean Sebring:
This is wonderful.
Chrystal Taylor:
I’m imagining a cat Sascha, but is on a call just meowing. It’s like, “What’s wrong with Sascha today?”
Sean Sebring:
Would it meow with a German accent though?
Sascha Giese:
No, it is actually a Turkish cat. We got him from a Turkish family and his name is Chicken, which means dirty dude, for whatever reason.
Sean Sebring:
Super fun, awesome answers, Sascha. And thank you so much for joining us today.
Sascha Giese:
Thanks for having me.
Sean Sebring:
And thank you listeners for joining us on another episode of SolarWinds TechPod. I’m your host, Sean Sebring, joined by fellow host Chrystal Taylor. If you haven’t yet, make sure to subscribe and follow for more TechPod content.
Sean Sebring:
Thanks for tuning in.
Sascha Giese:
Auf Wiedersehen.
