Why haven't we seen full adoption of automation yet in the networking industry? I mean, it's been years since we started down this road, over a decade, maybe more.
So with me today is Chris Grundemann, cofounder of the Network Automation Forum and the AutoCon events held twice a year focused on, you guessed it, network automation, and even more specifically, trying to answer that very question of why it hasn't been fully adopted yet.
Chris's response to that question is interesting. It's less technical than you'd think.
This is a great episode today and really top of mind for me just a couple weeks after AutoCon 4 in October 2025. My name is Philip Gervasi, and this is Telemetry Now.
Chris, thanks so much for joining today's show. So I've known you for a long time, but we never have had much of an opportunity to sit down and just chat for a while. I was on your podcast once a while back, a few years ago now. So I am really looking forward to talking today.
Yes. About, you know, the network automation forum and AutoCon and but also just talk tech and where, you know, where we are in the industry right now. Now I know I'm gonna guess at least that most of our audience, being networking nerds, probably know probably know who you are. But for those that might not, would you give a little bit of a background, a little intro for yourself or whatever you're comfortable sharing?
Yeah, sure.
Think I guess my I'm a network engineer by trade and have definitely veered off onto many adjacent paths over the course of my career since then, since starting out. I got my real start about twenty years ago at the turn of the century I just like finding ways to weave in turn of the century into conversations these days.
The first network I built, we actually built with software because we didn't know any better. This was early 2000s and so things like Zebra and Quagga and Bird and some of the virtual routing engines that are open source available now were either just getting started or weren't quite available yet and definitely I didn't know about them. And so we ended up writing our own router OS which sounds much more grandiose than it is. It was essentially a bunch of Perl scripts that called IP tables and stood up static routes and NAT rules.
But it's definitely been very interesting that from that kind of beginning, then I dove into more kind of real service writer networks in in kind of jobs at bigger companies and and learned about, know, real routers and switches and and then, you know, networking and that kind of stuff. And to see it kind of come back full circle where, you know, I've been watching for the last ten plus years, ever since OpenFlow came out and kinda really made waves. Been watching the network automation space and how that's evolved. And then definitely in the last two years since Scott Roban and I started the network automation forum, I've had front row seats to this kind of revolution of network automation of really applying software to networking, which in a lot of ways for me kind of feels like coming back to the beginning and starting over, which which is great.
So, yes, that's me. Mean, in the middle there, I I worked for the Internet Society. I worked for several service providers.
Spent a lot of time flying around the world talking about technology, I've built some networks, I got to write some patents, wrote a couple books, and an ITF RFC. And yeah, so you may have seen my name around there, as Phil mentioned, but also a lot of the behind the scenes stuff too. Yeah.
You've been busy for the past twenty years.
And it is interesting that you started off with software and not with, you know, just racking and stacking gear. And then after the fact, you know, like, maybe I can do this a little bit more efficiently. You know, you you you flip that whole thing on its head. So that's really neat. I didn't know that about you.
You know, it's it's interesting. I'm thinking about, like, twenty fifteen, twenty fourteen, twenty sixteen. Right? In there, like, those couple years.
That's when everybody in the networking industry as far as, you know, my my memory, that's when everybody started talking about network automation. And, you know, you started hearing about engineers learning network engineers learning learning Python. That's the first time I ever heard of Ansible. I didn't know what that was prior to that and things like that.
So really, really interesting how it's been ten years now that we've been having this conversation. Now I know that the conversation started much, much earlier than that. You talked about Pearl scripts literally decades ago, the turn of the century, if you will, which, yeah, I when you said that, I'm like, yeah. Nineteen o three.
How old are you, man?
That's that's crazy. I did see a meme the other day that said thirty years ago. Yeah. The early seventies. And then you do the math, and you're like, no. No. That was the mid nineties.
Right.
So I I can't compute that yet. Tell tell me a little bit about the network automation forum. We have had, Scott, you mentioned, you know, your your partner at the Network Automation Forum on once or twice now to talk about it, but I'd like to hear your perspective. Tell me how that started, why that started, what the goal is.
Yeah. Absolutely. So, essentially, I mean, the the real mechanics of how it started was was Scott and I both saw this need. I think we both had a common client.
We do we both do consulting and and contracting work, and we both had a common client who was looking to do like a user forum for their product. And and we both independently were were talking to this person and and said, what makes more sense is to build a big tent, do an automation conference and be one of the sponsors and bring in other vendors and make it vendor neutral instead of trying to, you know, start a user forum from scratch. It seems more likely to be successful if you really look at this holistically. And and that company and that individual weren't really interested in that.
But having heard it from both of us said, hey, you two guys should talk. And we did. I think Scott had seen the first version of the state of network automation survey. I had known people who know Scott.
Scott knows so many people through all his training and SE time and and all that Yeah.
So I definitely known of Scott and and and then we met for the first time at a Nanog meeting actually and kinda sat down and said, hey, you know, can we do this? Can we pull this off?
Luckily, we were both entrepreneurial enough or dumb enough to say, yeah, let's let's let's do it. Let's roll the dice. And, you know, I think this was July or so of twenty twenty three.
And we went ahead and pulled the trigger, signed a contract with a hotel for a conference in November of twenty twenty three, and decided to see if if there was a community there that wanted to get together. And lucky for everyone involved, I think there was. Right? It it it definitely played out bigger than I would have ever expected.
The way I think about it is essentially, we lit a fire looked around and it turned out there was a bunch of people standing around the dark. And I think people have now come to gather around this campfire and talk about network automation and things definitely for me have advanced very rapidly. I I don't know if if the industry has advanced as rapidly as it seems like it from from my seat. But I definitely understand what modern network operations using automation should look like so much clearer than I did two years ago.
Right? I had a kind of an inkling of what automation was. I I had my own opinions about it. But now having, again, this front row seat to the the experts in the field really coming and talking about what they're doing, seeing the conversations on Slack.
Like, I I feel like I now really get it. And I think that's happened for a lot of people. So I'm really excited about kind of the success of this community coming together and and sharing information so openly.
Yeah. Yeah. So you're saying that even you as the event organizer, who you know, I've seen you at these events. You're running around like crazy, you know, coordinating and putting out fires and whatever you have to do. Proverbial fires. Right?
Yeah. Luckily, no real fires so far.
Yeah. Luckily. You know? But but what you're saying is that you've actually, like, learned a tremendous amount specifically from these events and advanced your own knowledge and understanding of network automation just from attending these AutoCon events.
A ton. A ton. Yeah. And I would say it's definitely split. I mean, definitely at the events themselves, there's a huge kind of high bandwidth information download from both the presentations and then the people in the hallways and in the bars afterwards talking about the presentations.
But also from the Slack channel, there's a few thousand people who are on there and some of them are quite active and there's debates that rage from time to time. And between the two, I've gotten a really clear picture I think of of how we should be doing this.
So then what like, you know, we're talking about network automation. What what's the actual goal of the conference? Is it just to get around and talk about automation and, like, what are you using? What are you doing?
Or what what's the underlying impetus for starting the conference in the first place? Yeah. You know, obviously, you talked about community a couple of times. So getting folks together to to talk and professional development and learn.
I get that. But what what is the overarching goal for network automation form?
Yeah. So we we kinda founded it with a question of why aren't we Well, I think what we said is why haven't we seen full adoption of network automation yet?
Exactly as you kind of open this up, right? We've been talking about this for at least ten years, probably longer. If you look at service writers, right? Service writers you know, have been doing some kind of scripting in the network for a long time. Right? It just was required for big networks to get stuff done.
Enterprises have have have come around to it a little more recently, I think. And it depends on how you classify hyperscalers. Are they enterprises? Are they service writers? What are they? I guess they're hyperscalers. They've been doing this for a while as well.
But but it's not everyone. It's not everywhere. And and so that question kind of bubbled up of like, well, why not? You know, when when OpenFlow came out, I think it was two thousand twelve, it made a huge splash across the industry and and and people kind of were screaming that everything was gonna change.
It was a big kind of hype around that. And then it kind of petered out. There was definitely some companies, some vendors who put OpenFlow into their products and things. My realization at the time being so totally confused by it at first and then kind of digging into it and understanding it.
I I was lucky enough to be working at a think tank at the time. And so got some to spend spend some really time like digging into like, what is this SDN thing? And my conclusion was people want to automate their I don't think that people really cared too much about the separation of control plane and data plane, which was kind of the SDN claim to fame. But what I realized was people did want to automate.
So you're you're really talking about more of an operational focus then. Automate as in the sense of, like, do things better, run the networks better.
Is that is that right? Absolutely. I think so. Yeah. Yeah. Yeah. And it does bleed into, like, design and and architecture and things like that to be able to build a well automated network.
It also blends into things like organizational culture and and organizational structure, which which is pretty cool.
So I I wanna I wanna ask you what the answer to that question is, that that hypothetical question or maybe it was a literal question that you, you know, asked folks at AutoCon Zero. I remember I was there in Denver a couple years ago.
But before we do, we just had AutoCon 4 we.
I was there, so I say we, but you just had AutoCon 4 in Austin, Texas.
I don't want to get too much into a recap. I know the videos are going be on YouTube soon, but tell me a little bit about your perspective there. How have things changed? How is AutoCon 4 for you? But also how have things progressed now? We're looking at AutoCon we started with zero, so that's five events now, right?
Yeah. Yeah, absolutely. I mean, one kind of obvious fact, guess, is just the growth of the events. Think, the first AutoCon Zero, about three forty people showed up, which was mind boggling at the time that that many people showed up when we had announced this thing, know, a few months before. But AutoCon 4 in Austin, as you said just a few weeks ago, had over seven hundred folks show up for the conference.
In that time, we've also added workshops. So now there's two days before the event where we do hands on workshops.
There's now sixteen different half day workshops. So four at a time during four time blocks.
This AutoCon 4, we also added in a separate track. So we we did an advanced track, which honestly, I don't think hit the mark and we're we're gonna kinda go back to the drawing board on on that and try and find some better ideas around around content for for that piece of it. We also had a leadership track, which I think did hit the mark and really provided some good insights into how to manage up and and manage teams and and that kind of stuff and how to look at this from a more business perspective. So so just overall, you know, the the community, the people who show up has grown and the event itself has grown quite a bit.
Also, we've gotten a lot better at it, I think. Definitely, this was not only the biggest event, but it felt the smoothest. That's maybe very, very selfish and personal because I think hopefully, no one else sees all the rough edges. But I think Scott and I were way less stressed out this time than we have been ever before, so that was great.
Yeah. Yeah.
Yeah. I mean, I was there as an attendee and and a and a workshop presenter, so I did not notice any of those rough edges. I thought it was great. I will say that I discovered probably the best tacos in my life around the corner. And that was I mean, yes, all of the presentations were amazing, and meeting up with folks was amazing. But as somebody who enjoys really good food, you know, that was that was really special.
And Fantastic. Think in the Slack, I mentioned it a couple times. And by the end of that week, I mean, I don't know how many people ended up going to visit that taco place around the corner. So, you know, I say that in jest because, you know, one of the things that I love about conferences in general, autocon or not, is connecting with folks, the hallway conversations and all that kind of thing. And what I noticed about autocon is that, sure, there's this focus on network automation, on tools and resources and architecture and all these kind of things. You just mentioned a leadership track. So even bringing it out of the pure technical conversation into some of the the how does this relate to the business and to management and the people conversation.
But also, there's like the birds of a feather meetups that happen every single morning. There's the folks that are all getting together to go to that NAF sponsored social event that's happening afterwards. So those are all very, very important, very special to me just in making those connections that, you know, selfishly that help advance my career because I get to know people and and build my own network people network, not, you know, IP network.
And and also, you know, connect other people. I'm always looking for, like, can I record a podcast with this person? Is there a blog post hidden in this content right now? So I think that's been a huge benefit for me attending personally.
Absolutely. And I think that definitely gets to the root of why AutoCon has been so successful in general is, yes, it really is a community event. And so I cut my teeth, I grew up kind of in the industry going to Nanogs, the North American Network Operators Group meetings, also Aaron meetings. And then when I worked for the Internet Society, I ended up getting to go to NOGs all over the world to help start some NOGs around the world.
And we really tried to kind of model that, you know, very community focused, very practitioner focused approach to conferences for AutoCon. And I I think we've done pretty well. And luckily, there's a great community there that that really has kinda jumped in and and and known all the right things to do and and how to how to socialize and how to network and how to take advantage of these things. Because I do think that those, you know, going out for breakfast burritos, going out to grab tacos at lunch, you know, grabbing food at the social event or the happy hour, maybe some drinks even, you know, after after hours is is just as important as the presentations, right?
I think, know, there's something that somebody probably said something very smart about, you know, learning only happens with reflection. Right? So you you can you can watch a presentation. If you don't then sit down and think about it and and work through it, I don't know that you've learned as much as as with that discussion.
Right?
And so if you can see a presentation and then talk to five other really smart people about what they saw in the presentation and see these different perspectives and different understandings, know, they're coming from different backgrounds, they have different networks Oh, yeah.
You really learn a lot more and I think you take a lot more back with you because of those conversations on top of the presentations.
Okay. You know, let's stick on the concept that not the concept, but this topic of learning.
Just was it yesterday or was it this morning that I saw a LinkedIn post from the network automation forum about looking into doing some sort of actual formal training and instruction at future events. What is what's that all about?
Yeah. So we so I was actually Scott's idea. It was his brainchild was to have these workshops before the event. And so we did it as early as AutoCon one.
So the second event, it was in Amsterdam. We had two days of workshops before the event. It was a single track, so we could only have, like, fifty people in there because that's about about the size of a a classroom where it starts to break down if you have more people than that, especially for hands on labs and that kind of stuff. Yeah.
Right. Right. And then we we quickly expanded that to have, again, these, you know, four simultaneous sessions in four blocks, AM PM, Monday, AM PM, Tuesday. So we have sixteen potential workshops and it's if you buy a workshop ticket, it's it's choose your own adventure.
You get to pick which of the four and each time block you wanna go to.
And the way we're doing that is in partnership mostly with vendors. There's there's been some other who weren't vendors who came and volunteered their time to put together some workshops. And it's been great because in Austin, we limited it so that each vendor could only do one workshop. So you had sixteen very different perspectives on network automation.
That's been great.
It is inherently because the folks who have the resources to put on a workshop for sixty people with lab resources for free are mostly vendors. Does skew that direction, right? And so there's a little bit of you're looking at products and over tools in some ways, which wasn't the end of the world. But what we wanna do is try and add into that mix.
Maybe we could get somebody to come do a Python for network engineers training, one of the vendors isn't really going to do. Not really what they want to be showcasing. And so there's maybe Yang, maybe Jinja, there's a bunch of kind of foundational topics that I think fall outside of what a vendor is going to talk about that we could provide more training to give, especially newcomers, a leg up on network automation and kinda get into the mix. So right now, as you said, we put out a request for information.
We're really just looking at is this feasible or not. Obviously, if we have to pay trainers, the cost of the workshops will probably go up. So that's something we're taking into consideration. We do try to keep ticket prices as low as possible so the most people can can attend.
But we do also think that it, you know, maybe worth having even a full two day journey of of kind of introduction to network automation and really get some foundational topics in there in addition to what we're already seeing from from from from various vendors and and different folks who have have ideas to to share. Right. So at this point, we're just gathering information to see what's possible, and hopefully, we can continue to enhance the workshops as we go forward.
Yeah. But it does speak to the heart of what you're trying to do, which is, of course, to build community, but now talking about education and getting folks to advance in their technical knowledge, but also in their ability to advance their project with their manager and get budget and things like that. Because I did hear talk, I've heard those conversations as well from the main stage too. One of the of the talks was all about how to secure budget and how to phrase things or put things into terms of ROI and things like that and the cost benefit.
So that's all very real in an operational context. Sure, we want to understand what strings are and variables and all these things and and what an Ansible playbook is, but when it comes down to it, it's all just interesting tools and toys that we don't get to use because the budget wasn't approved to run the project.
Exactly right. Yeah. And I think you're right. I think there is several facets to that initial question, right? Why haven't we seen full adoption of network automation yet? There's a lot of answers.
Many, know, and I don't know if I want call it a failure or not, but like many failures, there's almost never one cause, there's a lot of causes.
The education piece is addressing the potential skills gap. I think in general, my understanding, experience so far is that there's a lot more network engineers that are really comfortable with network engineering topics than there are that are comfortable with automation topics.
And just like on the other side, there's a lot of software developers that have no idea how networking works. Right? And so sharing that information to kind of bridge that gap, I think, you know, the education piece really fills that.
But to your point, if if you're not getting staffed right, if the organization is not structured properly to break down those silos and allow people to work together, if the culture is not right to allow experimentation and failure, if you can't get budget, then those skills don't really matter. And so there's that aspect of it as well. So I I think, you know, we're really trying to to, you know, answer that question or or maybe even, you know, fight back against the answers, right, the challenges that have popped up is that how do we address those and help the industry move forward so that we can have bigger, better, more reliable networks? Because I think that does require some form of automation and observability and now maybe even AI. Yeah.
Yeah. And, oh, AI, is that something that we're going to talk about today on this podcast? Yes, are. It's in the notes.
But before we do, you mentioned observability just now, you mentioned orchestrate. I don't know if you mentioned orchestration, I was thinking it when you said it. But ultimately, you're talking about education and I was a network engineer for many years, so often I wanted to be educated about what the vendors were doing because I was implementing brand X switch or brand y overlay, whatever it happened to me. And so it was important to me to learn what vendors were doing because, again, they had a lot of the resources and were pushing a lot of the innovation of the day.
So it never bothered me because I know other conferences are a little bit different where they really wanna downplay any kind of vendor participation. It never bothered me, you know, starting with AutoCon Zero, that there were vendors involved, and I think it was handled well in that it wasn't, you know, just these constant vendor sales pitches. It's, hey. This is the technology that we're putting out there right now. Let's show you what it does and how it relates to automation, how you can run your network better.
And so that kind of married with what you're doing now with that independent educational voice, I think, is really great. Now how that's gonna flesh out? I know you're still working on that. But another thing that I was thinking is how you know, when I was walking around just here in AutoCon 4 a couple weeks ago, I noticed it prior, but especially this past one. You know, I noticed vendors that were talking about things that weren't exactly network automation, that they were visibility or observability or, you know, those other things.
AI was one of them as well that, you know, sort of fit into this network automation ecosystem. So, you know, to you're kinda zooming out a little bit, you know, going to an AutoCon and understanding that this all fits into an entire, you know, technology landscape where you're feeding data into your, you know, event driven automation platform that you built or you bought or whatever it happens to be.
So that's something that I noticed is that there are, I don't know if you want to call them like automation adjacent technologies that are still very relevant to understand, to learn, know, both from a technical perspective and also like how does this fit into an architectural landscape as a business owner.
Absolutely. And I think there's definitely a little bit of I mean, we could have a semantic debate here for sure. Luckily, has never really broken out. But we called it the Network Automation Forum, which is a forum to talk about network automation. The event's called AutoCon, which is the automation conference.
But what does network automation mean? And I think part of what we've done in the last two years as a community is defined that. And I definitely think we could have called this the advancing network operations group or or or conference. Right? I mean, I think this is more holistic and and maybe a hat tip to to Scott and his kind of total network operations podcast and and and series of events that he's doing that really kind of looks at that kind of more holistically. How do we kind of combine all of this together and and run things in a more similar fashion?
This is really about, you know, how do you build and operate the best network possible. And I think that, you know, automation is an umbrella term for orchestration observability. I think artificial intelligence in this context is mostly about either gathering data or automating. And so network automation is definitely a bit of a shorthand for all the things you need to know, that are beyond just the protocols to really run a network well.
Yeah. So do you believe that AI does fit into I mean, you just said so. So I guess you do believe.
I think so.
To to me, I see AI as another form of automation in this context anyway.
Right? I mean, that that's not like a great definition of AI more broadly, but I think in the context of network operations, AI is a another form of of automation. Again, in that context of automation being a really broad term, that's little bit nebulous. Yeah.
Well, you know, doing programmatically performing actions, whether that's deriving meaning from data, so data analysis, that's an action. That's doing something programmatically. Whether it's doing that or pushing config or doing other kinds of advanced things, whether there's a large language model involved or not, you know, using the umbrella term, the classic traditional term of AI. I think it does certainly fit into that, especially as you're using those tools as part of that greater workflow, that greater operational kind of ecosystem of how data feeds, you know, an alert, which feeds an action, which feeds a visibility, you know, dashboard and and, you know, a ticketing system, all these things that link together in real life.
Right? So network automation is you know, we've been defining it as more than just a collection of Python scripts in a folder somewhere. It really is this entire thing. So tell me about upcoming events.
I know AutoCon five is coming up in was it May or June of next year?
Oh, June. Yeah. I think it's the week of, is it the sixth of June? It's basically the the the the second week in June. I think it's the Yeah.
I should probably hold on.
Should actually the dates. Date.
I think so. Yeah.
So We're gonna edit this part out.
We're gonna leave this part in.
Alright. That's great. I love it. It is the eighth through the twelfth is the week.
The actual dates of the week. June eight through twelfth. It is in Munich, Germany. So that's what we've kinda done is if you notice the even numbered AutoCons are in the US, the odd numbered AutoCons are in Europe, and and so the next one and the in the springtime is also when those happen.
So, yeah, June eighth and ninth will be the workshops, and the tenth through the twelfth will be the conference In in Munich. Yeah.
We're really excited. So are the conferences pretty much the same? I mean, I'm gonna assume that they're somewhat different, culturally. Obviously, it's a different location, things like that. But are they are they are you do you intend for them to be the same?
Yeah. So far, they've been very much the same. So far, we've kind of seen it as one continuous event series, and we're learning things at one event and taking them to the next event regardless of where that event is.
We have started to see a little bit of difference between, you know, the the the type of content that folks in Europe want versus the type of content that folks in the US They're fairly minor, but it does seem that the appetite for more vendor solutions, it seems to be slightly higher in the US. I mean, this is a broad generalization. And there's it seems to be a slightly higher appetite for open source and build it yourself in in Europe. And I don't know if that's just a scale of companies question or a cultural thing.
I, you know, I don't know. Also, Europe is a lot of countries. So again, this is gross oversimplification. But there does seem to be a little bit of difference in in what people wanna see.
So we're we're starting to take that into account and and there may be slightly more tailored choices. Because so far we haven't been really considering where the event is. We just choose the best speakers based on what we know versus tailoring to the audience, which which we may do a little more of going forward.
Okay. Yeah. So how how are speakers chosen? I already know the answer to this, Chris.
Yeah. But I want you to yeah. I want you to tell us. How how are speakers chosen?
How are workshops chosen? How does that all come together?
Yeah. So both the both the speakers and the workshops go through a a call for we have a call for speakers and a call for workshops that we put out. People throw in their ideas. We collect those up.
And then we have an advisory board. So we have twelve folks from the industry. These are all, names you would probably know many of them. Alright.
If we if you if you go to the website and look at the about page, you can see who's on the current advisory board. We're about to kinda refresh that for for twenty twenty six. But we try to get a mix of folks that are, you know, folks who work at vendors, folks who work at service writers, folks who work at enterprises, folks who work for consultants. Again, this is really kind of this big tent approach we've taken to to all the things and we do that in a microcosm on the advisory board.
And then those twelve folks go over the submissions and and essentially rank the submissions from, you know, what they think they would like to see and and what they would what they think the audience would like to see. Right? What what's most beneficial to to answering these questions and to moving the industry forward. And then we take all those ranks and and then from that, it it shakes out pretty clearly who the top, you know, contenders are.
And and then that's where I take my kind of creative liberty to to actually build the agenda and try and order them in a way that that makes cognitive sense. I try to build an experience that you're gonna kind of journey through as as the conference unfolds, but based on that material of the talks that the advisory board has selected and the workshops as well.
And and the workshops as well. Right. Right. And then I noticed that you're partnered with several other, like, media companies. And, you know, last year, I saw the USNUA was involved. Pack and Bushers has a presence there. Is that's just, you know, a mechanism to to reach more folks to stay out there in the community, that kind of thing?
Absolutely. Yeah. Yeah. Exactly. So be a hundred percent. Yes. So at USNUA, actually, last year in in twenty twenty four, they actually hosted the social event at the Denver AutoCon AutoCon two, it was Two.
Yep. Which was great. And and and we've done some cost promotion there. Obviously, what they're doing is really cool, kinda pulling these local user groups together to talk about networking where folks don't really have to travel.
They can just pop over for a happy hour. I think that's amazing. And so we definitely were trying to kinda promote them when they were promoting us. Packet pushers, definitely, you know, I've known Ethan and Drew for a long time and having them involved just made a ton of sense, I think.
So there's two pieces to that. One, they show up and and cover the event. And so if you're watching on LinkedIn, they do these awesome summaries of talks kind of in real time so you can you can follow along or catch up afterwards.
In addition to that, they're also on-site recording podcasts and things like that because, again, we've got all these bright minds in one place. So capitalizing on that makes a ton of sense. There's some other podcasters who come and do the same thing, some other industry analysts who come and and talk to people there.
And then also, we actually provide some packet pushers benefits for our sponsors. So so so the higher tier sponsors get to do a video with packet pushers and things like that. So Right. That also helps kinda boost the value to the sponsors. In addition to the event, they also have these kind of long tail media assets, which I think works out really well.
Yeah. Absolutely. So, Chris, I wanna I wanna stop here, though, because earlier, you know, we focused on education as one of the reasons not education, but a lack of education, I would say, is one of the reasons that we haven't reached full adoption for network automation yet.
Folks needed to learn about these other tools and mechanisms, and even that they just exist, let alone how to do it.
Is that the only thing?
Is there any other conclusion that you've come to as to why we haven't reached full adoption of network automation yet?
Absolutely. Yeah, there's several things. Do think that there is skills is one thing, and individual learning is one. I do think that in addition to that, some of the lack of skills isn't that there's not information out there.
We actually live in a very information rich time. I mean, when I first learned the networking, I think, you know, Google as a search engine was a brand new tool that was kind of exciting. There definitely wasn't a YouTube. There definitely wasn't TikTok.
There wasn't, you know, all all these courses and things you can get to, all the books that have been written since then. So, you know, we're definitely you know, if you wanna learn this, you can. Right? There there's people to learn from.
There's there's materials and things like that.
Of the things that comes into play there is what I would call an identity crisis. And I've written a little bit about this. We've talked a bit about this in the But, you know, there is definitely this, I think, connection of a lot of network engineers to their CLI of choice.
And I think, you know, while, you know, Cisco certifications have done great things for for the industry and for teaching a lot of people how to do things. If you if you look at it really clearly, the CCIE is a race to hand jam, you know, manually type configs into a bunch of machines. And so getting to CCIE is great and it does represent a lot of learning, but it's also a test of like marathon pounding the keyboard. And I think because of that in a lot of ways, we have adopted that as, that is what it means to be a network engineer.
It is knowing the CLI commands by heart, knowing, you know, the BGP protocol, you know, steps so that I can troubleshoot things, like, really, really being kind of that one man machine, almost a cowboy that can go in there and just ride around and and fix things. And and that I if you're if you attach to that as your identity, the network automation can be very, very scary. And so I think there's a lot of people who are or have anyway, and I think this is loosening it up in a lot of areas, but have refused to learn these skills because it it really challenges that sense of identity of what it means to be a network engineer.
So I think that's definitely part of what we've done is is kinda socializing that at these events and and in these communities.
Beyond that, I think that the idea that one person is going to know all the things and be able to automate a, you know, enterprise or service provider scale network is a crazy myth and a crazy idea. Right? There's been talk at at some presentations about yetis riding unicorns and and and this is kind of the expectation there of of a one man show or one woman show who's gonna go in and and know Python and Jinja and YAML and BGP and OSPF and, you know you know, Cisco and Nokia and Arista and MicroTik or, you know, the the list is just way too long. And so organizations coming around to the idea that what you really need here is a multi cross functional team to be able to run a network.
And you don't want these silos, these firewalls between system administrators and security folks and network folks. Right? You need to kind of really, like, break that down. And and so I think that there's some cultural stuff there Yeah.
Organizational structure things there that come into it as well. So those are kind of the big three, I think. I think it's, you know, that this personal identity stuff that leads to the skills gap. I think there's a lot of cultural things and organizational things.
You know, businesses looking at the network as a cost center and not as something that they should invest in is definitely part of the problem, and that's why we've talked about some of these, like, TCO and ROI conversations. Yeah.
Do you think any of that's starting to change finally? Because, you know, what you're talking about for the most part, there's these are all people issues.
It's not like there's a protocol that's broken or too slow or something like that, or we need a better programming language. I mean, everything that you mentioned there involved people learning stuff and system level, system people, system level changes, which is incredibly difficult to do.
So so have you have you seen any improvement?
I think so. Absolutely. I I and, again, my my view is a little bit skewed because, you know, a lot of where a lot of the time I spend now is in and around the the NAF community. And as a community, I think we've definitely all kind of leveled each other up a little bit and thinking about this. The fact that the community continues to grow is a good sign and it gets referenced outside of this. So I do think there is is change happening. There's change afoot for sure.
Maybe still slower than we'd like, but that's again part of our mission is to is to spread this word and have folks involved and have folks know where the resources are to be able to move these things forward. Because I do think, yeah, it is and and this is where, you know, the reality of call it net DevOps comes into play. Right? Because DevOps was never about tools.
It's about methodology. And and that's what we need to understand here for network automation as well is the tools are there. Right? We don't need to invent new protocols.
We don't need to invent new tools. Sure. Cool new tools can help make this all simpler. But right now, everything's available to build a fully automated network, and it's really about the sociotechnical system Yeah.
More than it is about, you know, individual technology.
You know, I have always prided myself on coming up with interesting terms.
Sociotechnical system is one that I've never heard or used before, and I like it. I like it. It makes a lot of sense for the context of what we're talking about today. There is this whole technical world, and then we got this people issue to deal with and bringing them together. That seems to be what's causing things to be so slow. I mean, it's almost inevitable if you think about it, like trying to change people's hearts and minds.
But I also wonder if we're going to start to see a little uptick in that pace of change and that rate of change as folks literally age out and retire, and you have new folks coming in. And I talk to engineers that are in their mid-20s, a couple of years out of college. It's a very different mindset than what I had for a long time. Like you said, my identity was very much entrenched.
For me, specifically, it was Cisco because I went through that whole certification track. But it was very much entrenched in my ability to memorize syntax and know how to do stuff on a router and make things work. And to be fair, I needed to know that stuff. I was a VAR engineer for most of my career.
I needed to go out there and plug stuff in and very quickly make things work and then get out and go to the next customer.
So for sure, I really do think it's inevitable, and it's going be a slow change because of that. But like you said, we're starting to see change. And I think some evidence for that, me personally, is the difference between AutoCon Zero and AutoCon 4. Now, this might be intentional or this is organic.
I don't know. But at AutoCon Zero, it was, I think, a little bit more maybe a lot more technical in nature with regard to the presentations and the conversations that I heard and things like that. And then at AutoCon 4, that was all there. That element was absolutely there.
But I saw the addition of a leadership track, as you mentioned. And also, some of the talks from the main stage were about wrapping this into how do I marry or how do I align these technical KPIs of my network and of automation of reducing MTTR and being more efficient, all these things with business objectives and with money from the main stage. And so I see this, or I observe this evolution from autocon zero to AutoCon 4, which to me is evidence that it is really much more explicitly trying to address that sociotechnical aspect of the problem.
Absolutely. Yeah. I mean, the nice thing is we do have decade or more of DevOps to kinda steal from. Right?
There they like, these aren't Yep. Inherently new problems because we're apply like, we're just applying kind of you know, these methodologies and solutions that have been discovered in other areas to networking. So we have a leg up, I think, right, where a lot of the kind of the DevOps stuff that for, you know, happens started happening really had to build some of this stuff from scratch. Of course, they didn't build from scratch either.
They they they stole from lean manufacturing and and from physical manufacturing world and and kinda brought that in. And now we can do that again and kinda steal it again and and reuse it again and look at, you know, how to make networks more efficient.
Yeah. It is funny, like, when you see a network engineer just recently discover GitHub. And you're like, it's been around for a little while. People have been using this for a little while. But it's still cool. Mean, the fact that folks are discovering these technologies and saying, I can do this more efficiently this way, and my version control can be more consistent and effective.
That light bulb goes off, so it's pretty neat. Absolutely. So Chris, how can folks learn more about the Network Automation Forum, specifically AutoCon five coming up, and, you know, any kind of other information? I know that there the at some point, the AutoCon 4 videos will be up on your YouTube channel. Right?
Yes. At some point. We'll get the slides up on our GitHub pretty soon.
The the best kind of hub for everything is network automation dot forum. That's where the website is and that has links out to the Slack and to the GitHub and and to everything else.
Definitely, you know, join the Slack. I think that's where, like, real time conversations are happening whether it's about network automation itself or the meta announcements about what's happening with AutoCon and and and who know, where it's gonna be, what's gonna happen, speakers, tickets going on sale, all that stuff all happens on the Slack. We also have a LinkedIn page that you can follow. There's a YouTube channel that you should subscribe to, and then when those videos drop, you'll get notified. And there's also a mailing list if you if you'd rather be informed that way on the website is where you can find all of this stuff and links to it. So network automation dot forum is definitely the place to start.
Great. Excellent. So, Chris, thanks so much for joining today. It was a real pleasure. Looking forward to talking to you again one day. I'd really like to dig into, what your thoughts are around this entire AI phenomenon that we're dealing with, the networking specifically. So I think that's my invitation to you to come back on in the not too distant future to dig into that with me.
Yeah. Thanks for having me. There's definitely a lot to talk about there.
Yep. Yep. For sure. So for now, thanks so much to our audience for listening. Bye bye.