Techstrong Gang – September 19, 2024

Hey, happy Thursday to you. We've got an interesting, uh, show. Is generative AI a productivity mirage?

Is it all just smoke and mirrors our at autonomous agents the future, and minding the incident management gap? That's what we've got in store for you. You're watching Textron Gang.

Hey, happy Thursday everyone. Alan Shimmel here for Techstrong, and I am holding down the fort in our studio here at Boca Raton. I, I think it's gonna be this way a bit, as everyone is kinda migrating, uh, seasons are changing, clocks are falling back, and, uh, it is what it is.

But nevertheless, our gang is assembled and, you know, they're all over the place. But let's introduce them to you here on Textron Gang today. Um, first of all, we'll go to the czar of Silicon Valley, our editor in charge there, John Schwartz.

Hey, John, how are you? Hi, Ellen. It's good to see you.

I am living the dream as in Dreamforce this week. That's, that's Big. Uh, Salesforce show, Dreamforce and everywhere.

Have you ever been called before, John? No. I, I don't know whether to take that as a compliment or is it, It's the motion deck around you should be Okay.

Or political appointments. So those usually don't work on you either, or politic Political appointments. Don't answer your pager.

All good? Hey, Josh. But let, let's move on.

Let's move on from John over to Mitchell. Mitchell, you're home in, in Colorado? Yeah, we, uh, pagers don't work above a mile high, so we're safe here.

Oh, yeah. It's not on our agenda today, but, oh my God. You gotta ask yourself.

What the heck? Um, It's a James Bond movie. It is.

That really, I mean, that, that was, it's Insane. Insane. Anyway, Mitchell, of course, is our CTO and Futurum Group, CTA.

It's a pleasure to have him on. Good To be here. And then, uh, joining us from his new, well, he's right outside Yankee Stadium where he wants to be.

Actually, Harrison, New York. Don't kid yourself. That's not the Bronx.

It's much tonier. Mike, I've been, I'm in the great state of the Bolshevik. So here we go.

Bolshevik State. Okay. When you don't have beepers.

And then join joining us in, in the Charlotte Raleigh Durham Triangle area. It's our, uh, developer, uh, advocate extraordinaire, hope Lynch, and hope great to have you here. Thanks for joining us.

Oh, thanks. Thanks For being here. Uh, having me here and, and talking about pagers.

You don't know what a rabbit hole. I went down last night trying to discover how they did that, and I did. But, Um, well, no, they did, but it turned out it what the company that supposedly manufactured them in Taiwan didn't manufacturing.

They licensed the name to another company that's supposedly in Budapest, but doesn't have a working phone number and points back to the Taiwan address. It seems to me that that Budapest company is just a front. Yep, Yep.

Yes, yes. And, and I mean, this is, this is something we'll see in the movies one day or something. I had this feeling.

I mean, They, they've just handed, you know, a big part of a plot line to someone truly. I mean, Stuxnet was a great story. This is, that's nothing compared to this.

Oh, yeah. It's crazy. I, I don't want it to take, we have a lot to go on to.

We Have, we, we, we will be, we will be coming back to this topic on this cybersecurity issue. Well, because at the end of the day, look, this is all about supply chains. Yeah.

This was the ultimate supply chain attack. Anyway, let, let's, we'll save it. We're gonna save that.

Uh, talk about SBOs though. I just had to throw that out there. com, generative AI and the DevOps Productivity Mirage.

I don't know if I actually believe, I think this is one of those studies that maybe, Uh, I don't, I don't know. I just don't Agree. We will see, let me explain to you what these folks at Uplevel did.

They actually created, it's not a survey, it's a study. And they had two groups that control group and a test group. And they monitored their productivity through a bunch of metrics, including pull requests and things like that, that are pretty much standard door econom of metrics things.

And what they found was that while we have seen survey after survey where developers say that they are more productive, um, it's not actually translating into something that organizations are benefiting from by seeing more software actually delivered faster. And they're also in this survey, encountering more vulnerabilities. And I was talking to Tracy Bannon about this very issue, and the Mitre group has a similar survey that they are study that they're launching, and they're asking people to participate in because they have similar suspicions.

And other folks are starting to talk about this issue where it's kinda like there's a, the developers have a perception bias where, because something was easier for them, but somehow or other that benefits the organization they work for. And well, turns out it may not be the case. Now, this may not always be the case, but right now it seems like there's significant amount of, uh, well, to your phrase, smoke and mirror.

So, hope, I know you looked at this when you're in DevOps. So What's, you know, what's your feeling? Uh, I'm having a distinct feeling of deja vu.

This really reminds me of way back ish when there was this metric around lines of code. If you produce so many lines of code today, you're doing a great job and you're a great developer. Uh, so quantity of output, equating it with productivity, but ignoring quality, how easy is it to maintain and is there any actual value to it?

And with ai, uh, anecdotally and by studies, um, we know there are more bugs. So developers are getting more code faster, but then they're turning and spending more time stitching it together, debugging it, uh, ensuring it meets quality standards, making sure it meets compliance standards. So I think this is, um, an old problem just with, with a new face.

You know what, Alan, we did a, uh, shameless plug, uh, Textron research did a study on ai augmented DevOps just released about two months ago. And, uh, we titled it AI augmented because that was really what the study was telling us is it's not about generating code and people don't have write code anymore or something to that effect. Um, it's really about all parts of, so, uh, SDLC software development lifecycle, and is AI playing a role in whatever form, generative ai, machine learning, et cetera.

And, and about 25% of the respondents who had over 500 respondents, um, Sid, yes, they're using AI in water multiple places, um, across the SDLC, the top areas were development and testing, but most of it is augmenting, like code completion. Small sections of code maybe are written. Um, but you can see, kind of tell where we are because only 14% of people just trusted the output of any ai.

Everyone else, it has to go through pretty serious review process before they're willing to let it kind of flow down whatever part in the process. The other interesting, I think the most interesting stat out of it was the people who rated themselves more in the higher end of the maturity and using DevOps. They've been doing it a while.

Their organizations are, are DevOps organizations. They tended to be using, uh, more AI in, in DevOps across the SDLC and also seeing more results. Again, it's modest now, but that, that also impacted how much they got ai.

So to hope's point, it's not just about spitting out code and let's measure, you know, that as output. I think software is a much more complex process. So I, I'd like to weigh in here.

Yeah. I thought as much. Yeah.

You know, Upworks, what do they do for a living? They're, uh, jobs Platform. Yeah.

No, they, they provide people with jobs, right? You don't, you hire people to judge. I think they have a, I think they have a platform.

I feel like I walked into the back room at Rick's American Cafe and I'm shocked that there's gambling going on. Mm-Hmm. What did you expect them to say?

Now that being said, I, I agree wholeheartedly with with hope's comments and with, and with the findings that Mitchell has. You know, how do we, how we, how do we measure productivity gains in the DevOps life cycle, you know, in the, in the SDLC and the CICD process? I think it's a question of augmentation still at this point.

But yet, you know, I, I was out in, in Austin last week at Jfr Swamp up, and I had a chance to go to dinner with my friends, John Willis and Patrick dubois. You know, Patrick Dubois's, the guy who started DevOps, right? He coined the term, he said, you know, Alan, I was so down on DevOps over the last couple years 'cause I really felt it kind of made me ride its course a little bit.

I wasn't juiced up about it anymore. What, what new was there to explore and this whole AI thing and how it's gonna help DevOps has got me so juiced up that I've been on the road ever since doing workshops, speaking, evangelizing how AI is making life better for DevOps folk, how AI through the use of ai, so many companies, products that we see out there right now, including some of these platforms are gonna be so easily duplicated or, or not cloned, but replaceable at a much less level that, you know, it's gonna create havoc in, in the software valuation space. These are two smart guys who know a thing or two about DevOps.

I think I agree with them. I think there's a, we're we're on the proverbial journey, and I think writing code and running ahead on code is nice. It helps developers.

I don't think we really have seen the part of gen AI that really helps the DevOps team. And I think what we're gonna see is more advanced reasoning engines in these LLMs, and we'll be able to create agents assigned for specific tasks. And that's when we start to see the productivity gains.

I think we're kind of, the cow is over there, the proverbial half over the fence, as they say, and that the front end of this thing is helping the developers. The backend is for the engineers is, is still being written as a story to come, Because I think for the developers, it's still a chat interface. Write, write me, write me some code that does this, that and the other thing.

Write an application in this language that does that, that's a chat bot kind of thing. And that's what, that's the current state of art with a lot of generative ai. But you're right, this is Co-pilots is the other, you know, right.

In my id. Right. But, um, generative ai, not generative agent ai.

And, and just, you know, I, I downloaded iOS 18 yesterday. I don't know how many of you out there Did it? Raise your hands.

I have an iPhone 15. My 16 comes Friday, and supposedly I have Apple Intelligence in this IO s 18 Apple 15 phone. And I, I, I did a search for an Apple Intelligence app.

I went online and wrote, what's the deal with Apple Intelligence? And, you know, search that there's nothing you could point to per se, that that interfaces that it's, it's built into the environment to make it better able and more responsive to your needs. I think that's the AI we're going to see that is gonna help DevOps teams.

You're talking about is AI a feature or a product? It may not even be a feature. Maybe just so behind the scenes we don't realize what's AI and what's not.

And that's, that's sort of like what we want with security. You want it to be transparent as much as possible, and not in your face. I don't want it to be dealing with 25 chat bots either.

But to what Alan was talking about in terms of the, what these agents will do to platforms, that's a very interesting conversation because, you know, I was talking to the folks at Docker about this in their interleukin, if I start applying AI to these workflows, I'm not necessarily gonna wanna license separate testing tools and separate things for every function. It's all gonna be one motion, it's all gonna be integrated, and it's all gonna be orchestrated. And if that's the case, then to Alan's point, yeah.

A lot of these platforms that we today invoke separately are gonna become features something else. Yep. John, I heard Agen AI is huge at Dreamforce.

They're talking about agents for everything. Exactly. So Exactly.

I was just gonna weigh in on that. Um, last week, uh, agent force, which is something that Salesforce just announced, basically autonomous agents for very specific tasks across vertical markets was announced. And no, we'll go into it in the next segment, but Benioff is talking about this as kind of a way to redefine where Salesforce sits in the enterprise market.

In a sense, Salesforce and I, I, I'll just key on them because of what they, what they announced and what they are intend to do. Um, they see this as their next evolution, their next wave. Uh, Betty off Ashley last night was Ashley Tuesday, excuse me, was talking to Jensen Wang and that, and, and he talked about how he would wake up in a cold sweat thinking about, hey, uh, AI had had completely upended what they were doing.

So in a sense, they want to create billions of these agents. We'll see what happens. I mean, it just, to me, it's just, this is kinda this classic conundrum here.

Uh, and I, I feel for the developers, in a sense, you're thrust into increasing productivity while evading all the vulnerabilities based on this study. And it'll see, it'll be interesting how it shakes out, especially with something as ambitious as Salesforce is trying to do. Yeah.

You Know, I, I hear that thing about agents and I shake my head sometimes and I just go, tell me some task that you really enjoy doing, that you're really upset that an AI agent's gonna do for you. Because I have yet to see one. It's, it's, uh, to me it's like, um, for finance, when when they learned, ooh, RPA, we can get it to do these lower level tasks.

Some people were threatened by it, but other people jumped for joy. Uh, I, I really think it's the same thing. If, if there is a little agent who can come in and knows I wrote code, and it can say, oh, would you like me to put together your, your blurb for the release notes?

Who would not say yes to that? Absolutely. Little agent Curring around writing.

Hey, I call that whole story the Revenge of Clippy. But, uh, anyway, that is a horror film, not a with the, anyway, Howard, the one thing that Jensen Wang said that was interesting, he says, when we, when we have these agents, it's gonna be like welcoming a new team member. Um, it's like another employee.

It's like onboarding employees, but the work for many cases go, is going to be done before you even get to work, before you even arrive in the, the next morning. So there's a positive to that, getting rid of the unnecessary stuff that people, most people don't want to do. Absolutely.

Hey, we gotta take a break. We're taking a break here on Text Drug Gang. We got a quick word for you, and we're gonna come back and talk more about autonomous agents.

Um, Salesforce and Nvidia. John just previewed it for you. You're watching Textron Gang.

In a world where every line of code powers the future, every keystroke can introduce new threats. As software evolves, so must security, it's time to rethink how we protect our digital world. Join the leaders in DevSecOps and AI at the OpenText DevSecOps Virtual Summit on September 24th.

Discover how innovation is transforming software delivery faster, more secure and smarter. From AI driven security to the truth behind cloud security. Get the insights that will keep you ahead of the curve.

Don't just watch the future unfold. Be part of it. Register now and secure your place in tomorrow's world.

Hi, I am John Swartz here at Textron Gang. Welcome back. Salesforce and everyone's favorite AI partner, Nvidia, on Tuesday night announced the strategic partnership to develop AI features for large customers.

Specifically, they're gonna develop these autonomous agents and interactive avatar experiences that should be interesting. The resulting products would optimize predictive and generative AI workflows for things like crisis management in the event of a product recall or service outage, realtime weather impact for travel and logistics for the supply chain, realtime customer support resolution. So what these companies Salesforce and Nvidia are trying to do is create, they say billions and billions of agents that will do all these tests, many of which we don't want to do during, uh, sit down chat later Tuesday night.

Between Benioff and Wang, they talked about the opportunity, and in, in a sense, they want to create, as I said, billions of these agents. And, uh, Salesforce in, in this essence, thinks this, this is going to create this huge next wave for them. Because if you haven't already noticed, the last few months have been pretty brutal for Salesforce.

They've been, um, looked upon as a laggard for a AI development. They think this is their big breakthrough. And in, and Jensen agrees with them.

We'll see how it, how it can, how it fold unfolds. It's gonna take several years. Um, I'm gonna throw this out to the, to the panel.

Um, what were your thoughts on this and what type of impact do you think it might have? Not just on Salesforce, but in in general? I Always liked the Salesforce story because they were basically saying that they were gonna provide the trusted governance and the framework for using multiple LLMs.

And I can ask, you know, so what if I'm behind on development of LLMs? There's plenty of other people doing that. So it's a question of, you know, to me it's not necessarily the leader of AI isn't the company that builds the biggest and greatest LLM or provides the GPU.

It seems to me it's gonna be the company that figures out how to harness all this stuff. And they might be in a better position to do that than the other guy. So I wouldn't count 'em out on the AI side, that sort, they wouldn't start throwing around words like lager.

It would reminds me Of, well, mark Benioff kind of that his, his, the behavior of Mark, mark Benioff kind of betrays that because he is in a semi, how can I say this? Not a panic, but he's concerned because he keeps reaching out to reporters. I know, because he's texted me twice in the last couple weeks reminding me how important this is.

Right. So Mark, is he protest a little bit? Mark does I Nv right?

No, but John's the czar. Of course, mark would reach out to him. Uh, no, he wanted, he wanted something.

Um, but Mike, just, I'll briefly just interject. It's interesting what you said, because in a sense, what Salesforce is doing here is kind of like what Apple does in a sense. There's an established market, but in, in the case of Salesforce, they really have doubled down on security and governance and compliant and compliance.

And it's something that Mark to his credit, has talked quite a bit about to me and to a lot of other people. So in a sense, they see this basically as, as kind of a, not just something that would help Salesforce and these ordinary tasks, but something that would have significant impact on science, science, education, et cetera. Yeah.

It kind of takes me back to my, I used to manage Human Factors group had two times in my career before we called it customer user experience or customer experience. And the secret sauce is in using AI secret sauce is, is figuring out the experience or interaction models for a company like Salesforce. I mean, they're not in, they're not in the background processing insurance claims.

Uh, so much of what they do is in upfront working with humans, you know, internal processes that have been automated, working with sales team, uh, communicating with customers. If that, to me is the, the key. If they can figure out the models, avatars, not avatars, whatever.

It, it may not be the right idea. We don't, may not need the next clippy to refer back to Clippy. Um, you know, it, it's, it's, that may or may not be the right idea, but it's who figures that out?

Because I don't really want a better chat bot from the chatbots we've got on websites now. They're all pretty useless to me anyway. I, I found 'em really frustrating.

You're so limited. But I also, there's times I want to talk to a human. So it's, I think the way they figure this out will determine there's win or loss, not how much AI they use.

You know what, to that point, Mitch, I only want to say representative once and get to a real human. How many of us scream into the phone? Representative, representative?

Well, would I could get you a representative more quickly if you gimme representative or representative. Can't they just make the zero button to get, you know? Well, I'm, I, I don't know what you, you're doing it all wrong 'cause it's gonna be my digital representation of me calling a digital human on the other side.

And the two of them are gonna argue about it. One, and then let me know how it turns out. So Your, your bot's gonna say representative 25 times.

Well, I'm gonna teach it to say really loud. Let me ask you another question once the over under on how many fireside chats with other CEOs as Jensen Wong gonna do this year? Oh, One two weeks.

That's a great, It's at least I lost track Adobe. Yeah. It's at least two a month, I think right Now.

Two a month, I think just two a week. Yeah. Seems to me like it's a lot.

That's, you know, that's like, you know, Where it's weird. They have, they have to Go ahead, Jack. Well, I was gonna mention the far saint jet they had was interesting because I, I would say more than half of that was a glorified psychiatric session.

I mean, Jensen on, on multiple occasions was talking about his career and what this meant to him. You know, there were years at Dreamforce where Jensen would pop into the, the press room trying to get anyone to write about his company. Yeah.

And he's, You know, did running that company 30 years, that company that does game processes or something. Right. Right.

That's Great. For 18 to 24-year-old white boys. I mean, I'm all for Jensen seems like an earnest and nice guy.

But I have to ask the question. Yes. Is anybody else doing anything interesting in ai?

'cause it doesn't seem that way. Well, every time Sam Hartman opens his mouth, he gets in trouble. Yeah, That's be careful.

I just call as I see him out here because of, of course, I'm not in the rarefied air of Silicon Valley, like our royalty, but down here at Bo Hotel, that's what it looks like to us. Yeah. Um, in the, I think the phrase you're looking for is over exposure, right?

Uhhuh? Well, they may come a voice Yes. Where there's maybe over exposure, but we'll see.

Anyway, anything else on this? Otherwise, let's take a break and let's come back to, I'll put it in my best British, uh, voice mind. The incident gap, please.

Or the incident management gap, please. Uh, no. Okay.

You are watching Textron Gang. All right, folks. And we're back.

And as promised, we're talking about, well, a survey that Atlassian did where they talked to folks who spanned everything from software engineers to your typical IT administrator. And they're asking folks about how incident management gets handled. And the interesting thing about it is a lot of the folks seemed to think that they were doing a pretty good job and they had it under control despite all these outages that we seem to hear about.

But what was interesting, Mitch, was the number of organizations where the DevOps team was actively involved in incident management was relatively low. And I can't make up my mind if that's a good thing or a bad thing, but from your experience, you know, how aggressively should the DevOps team be involved in these incident managements? Or is are they like the second or third tier?

And we only call when, you know, the proverbial s**t really hits the been, Well, you know what it made me think of thanks for lasting for doing this survey is if compared this to what it was maybe five years ago, you know, is it better or is it worse? Is 'cause in theory, we, it's like Growing a presidential debate, Mitchell. Yes, I Yes, exactly.

Boy, rerun the tape, right? Well Make may make it grade again. Is that right?

Make It grade Again. Are you better off? Let's not even go here For sure.

You know what? I'm liable to have someone camped out at the, uh, office here for 12 hours looking for me. So I think they're On their way.

A think they're just going around in Florida. I think they're on their way. I I'd start packing up, but anyway, you know, it it, you can, is the glass half full or is the glass has half empty?

Or is it just half-assed? You know, um, To me it, it is always, it is always, you know, who is the developer who owns that application? That was always the question for operations.

And hey, you have that list. You know, who's pager pagers, you know whose cell phone, you know who to call, who, who to get ahold of when there's an issue with this app. That's a little more complicated in the world of microservices and cloud native and different cloud providers and platform engineers and all of that.

To me, that's it. It's a pretty complex web. And anybody that's got that figured out I think is, is well a step ahead.

So I think in the survey, Mike, it was like, uh, 60% of organizations where developers are part of the it, it team managing incidents. I don't know. I, that seems like not too bad to me.

I'm not, I'm not disappointed by that number. You know, does every developer want to be working on incidents? No.

You want the right developer working on incidents, right? That's my my view of it. I've got a few thoughts on this one.

So again, shock shocking gambling is going on. There's a gap between DevOps and ITSMs. Who knew, right?

com. But I I think fundamentally here when we talk about incident management, incident management is generally post deployment incidents, right? Yeah.

Because that's, that's when incident management kind of rears its head, right? And, and so because it's post-deployment, the DevOps team is not usually frontline though one of the principles of DevOps was having your developers and your DevOps team more involved at that frontline, that tier one, as we used to call it, right? Tier one.

But I think by and large DevOps team members are tier two support folks. And, and so, you know, with the advent of autonomous agents and ai, more things theoretically get handled at tier one. And, and so there's not just a a, a a, an incident management gap.

The, the, the gap goes to not just the volume, but to the quality, if you will, of, of, you know, if it gets up to the DevOps team, it's usually a, a bigger deal, a harder problem, a more acute incident than, than gets handled down here, uh, by your, you know, normal ITSM kind of, uh, responses. Alright. Hey, hope, um, in theory I'm investing in de ops to prevent the incident in the first place.

So are we getting any better at preventing the incident or you know, and maybe the ITSM guys are feeling better about things 'cause we are Preventing it? Well that, that really, that is a big question and that really does depend. I have been in some organizations where they still support the hero mentality, right?

So there is little incentive to reduce incidents because the person who gets promoted, the person who gets recognized is the person who has saved the day. But in other organizations that I've worked where they've, they've gotten rid of that and it's more, you know, the quieter and calmer you can make this, the more things just run, the better we know you are at your job. So the rewards matter.

But also how are organizations integrating ITSM with DevOps? I have worked with some teams where you rotated someone who is actually, uh, an engineer on the team with being one of the people who respond to an incident, right? So they always would have skin in the game because they knew at some point I may have to be the person on the resolution side.

I am not just the person, you know, pitching it over the wall, getting away from that, um, to the point made earlier was, was one of the original hopes of DevOps, right? You don't just develop and throw it over the wall. You're also there, um, when it's in use and when something goes wrong.

So I think overall it's a cultural shift. It's a need to probably integrate tools because right now tooling for DevOps and service management often are very separate. Um, but I think organizations are drawing closer together.

But again, culture change is going to be a big part of that. I know you're long enough hope to that when you start out a response, well, yes, we're in for some good wisdom. So, because That is a hot button though.

I mean, that Is, I know that's a noted question because it it gets to the heart. Yeah. And, and, and I'll give you something else.

Not only similar kind of issue is security. Are we more secure now than we were four years ago On the old issues? I, I gotta tell you, I was talking to a friend of mine yesterday, actually, it wasn't a friend, it was my brother.

He's my friend, but he's my brother. And he spent a good chunk of his day yesterday locking up all his credit reports and putting fraud alerts in because he got a batch of letters. I got home in my mailbox, and I'm not kidding you, I had this many envelopes between my wife and my children and I from all bunch of different folks that, you know, um, we're breached.

Uh, our information is available on the dark web, everything, social security numbers, all kinds of stuff. And, you know, giving me all this free credit monitoring, like that's gonna help and a million dollars worth of coverage. You gotta go sign up for all these things.

They give it to you. Right? And, and, and, and the four days that took to send you the letter was really helpful.

Yeah, but well that's assumed they did it in four days. This is the USPS we're talking about. But anyway, um, you know, but yet I know that our security is better.

I know we're doing a better job generally on security. However, the concentration of this kind of data is such a, a irresistible honeypot to so many different people for so many, you know, to so many different reasons that I, I venture to say all of our stuff is out there already and has been breached, You know, to to, to take another look at it. What is interesting, uh, not interesting actually, that Atlassian did this study.

'cause they're one of the few DevOps players that has a really strong presence, maybe stronger presence in operations. ITSM. Yes.

Yeah. Um, and DevOps and, and early in design, I mean, they have a designer tool that of course Jira kind spans the whole thing. And they've got a number of tools now that span that.

And then specific, you know, I started a lot of my DevOps tools when I started were Atlassian tools. So they, they are in that place where things can intersect and they've done a lot of work of building an, an AI platform underneath their products that leverages is leveraged in different ways and different products, but also a database that's shared across those, those products. They've done a lot to bring things together and integrate even multiple versions of Jira into one.

So they're, they're in an interesting position to, to make that more seamless for customers to bring ops and development together to bring more ITSM into the development or development into ITSM. So I think from a positioning standpoint, this survey, not surprisingly, but this survey does a lot to validate that. Yeah.

I think, you know, in general we keep talking about whether or not we're better at security. Um, the point is the bad guys are a lot better at what they do. Right?

Let's give some credit to the other team here. Uh, you know, there's a an old thing. I mean, if you ever followed the Civil War when they asked Pickett, why is it you think you lost the Battle of Gettysburg?

He said, well, I think the Yankees had something to do with it. Um, you know, the point here is the bad guys are getting better at this stuff. Yeah, no, they're more sophisticated and you know, we talk about things like ai, they're using AI pretty well too, right?

It goes and theyre very cuts both ways. Yeah. They're consistently pioneers and Right, and, and, and crime, right?

I mean, early adoption, just like the porn industry was in, in, in adopting technology and showing others how to Yes. I, I could tell you stories about that. Uh, I do, I do love the idea of an embedded DevOps engineer in an ITSM workflow for a couple of weeks at a time.

'cause honestly, you know, they should spend more time out on the frontline where their software is because they'll take those lessons home and make better stop. Well, You know what, when Mitchell and I, you know, we were two of the three co-founders of a company called Still Secure Mitchell ran vp. He was VP CTO and VP of Engineering.

He ran the development team, but he also ran the the help desk team. And that was something Mitchell remember he used to do. You used to rotate in developers Absolutely.

To be frontline, uh, answering the phones to just hear what real customers were Saying. A lot of ideas were generated by developers sitting on that support line going, you know what? We could solve this problem or make it easier to figure out what the problem is.

Or, or, or, so it's just like in the telephone days, back in the mabell days, I dunno, they still do it. New employees oftentimes went on ride alongs for installers, pole climbers, installers, things like that, and saw what the people in the field do. There's nothing better than experiencing in it.

Maybe don't, you don't have to climb that pole. But watching the person do that in the work that they do and, you know, it's, it's hard work that pop folks do in the field and the issues that they deal with. So there's, I think some I-T-M-S-M experience sitting in ops for a while.

It's good for everybody. I'm not sure they do that because they wanna have a better customer experience or they just wanna eliminate the whole function altogether. 'cause they aid it.

Well, some do. Yeah. Depends on whose book you read, I think.

Whoa. All right. Hey, I gotta pull the plug on today's gang.

It's been a, a heck of a Thursday, but we gotta get ready for Friday. And, uh, we've got a full day of, of, uh, of things to do. John, enjoy.

Uh, is Dreamforce over now, or? Oh, It, it goes through Thursday. So yeah, pretty Much ending, ending today.

Okay. Yes. And um, Mitch and Mike, it's good to see you guys not on the road.

Hope we're looking for a big announcement from you maybe in the upcoming next week. Hope so. Yes.

Good luck. But until then, hey, this is Alice Sch on behalf of Textron Group. You just watched Textron Gang.

Have a great day. Stay tuned for our full Textron TV schedule immediately following this program. bye-bye everyone.

Discover Textron Group, the epicenter of tech innovation. We are your go-to for reaching IT leaders and practitioners worldwide. Our secret impactful content that sparks awareness, engagement, and top quality leads with us.

You'll access editorial websites, streaming videos, virtual events, custom content analyst research, and more. Join our satisfied clients. Let's revolutionize your tech journey.

Contact us today and tell your story to the world in the most powerful way with Textron Group.