Techstrong Gang – October 14, 2024

Hello, everybody. We're gonna be talking about cyber cabs today from Elon Musk, followed by, well, just what is going on with women in stem and how come we don't have enough of them in this business? And then finally, we're gonna revisit DevSecOps once again, because there was a whole all day DevOps event that kind of got into this.

And boy, there sure are a lot of vulnerabilities in our software these days. You're watching Techron Gang. We'll be back in a minute.

All right, folks, here's the gang. Once again, I'm gonna make some introductions, but of course, um, we have some faces that we maybe had I've not seen in a little while. We'll start with Tracy Bannon.

How are you, Tracy? Oh, I am doing really well, just right. Excellent.

Nothing like, I, you know, I, as I understand it, you just got off a plate around midnight, so thanks for coming up. I did. So, I may be a little punchy today, but who knows?

It might add to the, uh, double trouble of the two TRAs. All right. Speaking of the two TRAs, we have Tracy Ragan joining us once again from the Southwest.

How you doing, Tracy? I'm doing great. How about you?

And I'm glad that Tracy Bannon on, and it's fun to have two women, and sometimes people confuse us, so now people can see that there's two, actually, two different Tracy's. There you go. Some people might say, well, it's double trouble, but we'll see how it goes.

All right. Then we have John Willis, who's joining us once again. John, are you home, or where are you these days?

I am Actually home right now, so, yeah. Excellent. That's a hard moment for you as well.

I think so, yeah. And then finally, Jon Swartz is once again out in Silicon Valley, and we're gonna start with John, because he was monitoring this whole, uh, cyber cab launch from Elon Musk and his squad. And I guess they unveiled something that looked like a car, but it didn't, from the article, it didn't sound to me like it was actually operational.

But John, jump in and what did you Yes, the, you're very perceptive, Mike. This is a very strange press conference. First of all, it started at 7:00 PM on the West Coast, and it was at a back lot in Hollywood.

So that kind of set, set me to, okay, these, your private roads that only these cars can travel on in a very, very planned, uh, design. So first of all, they started with something from, I don't remember, 2001 Space Odyssey, the Stargate sequence, that kind of psychedelic trip outs. They're going through a time continuum.

They did that for the first 45 minutes of this event. So at first I thought there was a technical glitch just being on x It, it turns out this was the preamble for a very short presentation, which was kind of telling in a certain way. So they brought out this cab.

Basically the cab picked up, Elon Musk traveled maybe several hundred feet, dropped him off at the events. Uh, he described it as something with, there's no steering wheel, two front seats, there's a butterfly wing doors, inductive charging, and a full service driving assistance software. So he, he introduces, he, people are screaming at him asking him, how much does it cost?

When is it available? He, in a sense, this is not prepared. These were remarks are not prepared.

He is basically kind of winging it. And he says, well, it's gonna be cost less than $30,000, and in most likelihood, it'll be available in 2026, if not 2027. He then goes on to show something, which is a, it's a huge van.

It's a robo van, which is seats up to 20 people or can haul cargo. There's no timeframe on that. Then he shows the robots, they're humanoid robots, and it's basically a guy in his suit, right?

So that we, we go, we go through all this. Then I, I, I, I check this, the stock price the next morning, and it's down 6% immediately, because the timeframe is, is off. We've been hearing about this for a decade.

It was very interesting. I will give him his props. This was a, almost a kind of futuristic type of press conference.

Very, very light on substance. It, it is almost like an Apple event where, um, it, it reeled out of control. I mean, I, that's what the, the intent was.

But in the end, it was, uh, some pretty interesting things, very vaguely presented, and it will spur a lot of controversy, a lot of talk. But, you know, it makes life interesting, I guess, Geez, I, no, Michael J. Fox and a DeLorean and back to the future, none Of that.

It looked like a DeLorean, by the way, you know, by the way, the other thing is, this was at night, so it was very dimly lit, so you couldn't really see the car that well. And I'm not sure if that was by design, but, uh, maybe to hide some sort of flaws or something. But it was, it, it's almost, I don't know.

I I, I'm not gonna go there. I was gonna say something. It kind of reminded me of like kind of a bait and switch type of, uh, political stump speech.

But it, there were some elements of that. But then again, it so Musk, well, he, I Have a question that he did call himself dark maga recently. So maybe that's the reason why you didn't Yeah.

Let's, you wrote my Mind. Let's, let's steer away from that. Let's steer away from that.

Yeah. But let me ask you a question. Just thinking about it broadly right now.

Don't we need somebody who is just thinking crazy good thoughts? Yes. Right?

Think about generative ai, and it's, people say it has hallucinations, and it's, that's actually a feature, right? It has this vivid imagination. You gotta throw a lot of stuff away.

Don't we need Yes. Somebody just talking about stuff, putting things out there. Like, I'm not, I'm not an Elon hater, because I do like that he comes up with, he and his people come up with some, you make a pretty awesome ideas.

Yeah. Teresa, I'm just going to just interject this one thing. That's a very, very good idea.

And that's funny that you say that, because I used to go to all the Apple events when jobs would make the presentation. Now, there's a difference between jobs and Musk, but in, in a sense, they're showman. They're, they're opening your minds to different things.

They, they, um, in Musk's case, it can be polarizing. But, you know, one thing I noticed when he came on the scene, there was a lot of resentment towards him, especially from Apple. People with Apple really resented him.

And I thought, because he kind of took the mantle, and whatever you think about him personally, this guy does have a vision, and he thinks big. And when I was watching this, I was thinking, there's no other tech events that I, that I go to or have seen that even come close to this in terms of at least storing your imagination. Well, remember when he sent the, the car into orbit?

Well, right, right. Just think about the showmanship and the fun of that. Like, I, I wanted to watch that just because it was interesting.

There's A dichotomy on that, man. Right? And I'll just say this right now, we don't have to go for a light.

I despise him, but I, yeah, I gotta tour a SpaceX. And one of the things they do when they give you the SpaceX tour is they show you three different worlds, right? And they show you the Earth, then they show you Mars, and then they show you this weird version of a planet.

And it's like, his vision is to actually have people living on Mars, right? Like that, that is the, sort of the true north of that company is to get the first human to Mars, right? Like, and like, yeah.

I mean, it's that kind of thinking. And even the opening, I debate of whether the lawsuit is, you know, you know, his, his version of the lawsuit is that they sold, um, the ip, or they took the investor from Microsoft, the billion dollar original Microsoft into OpenAI with the caveat, unfortunately, I can't understand how billionaires can't find legal documents. But, um, the, uh, with the caveat is that the, when it went to a GI, which is another whole, what is a GI, but like, if you believe that narrative, and even though I dislike him, I actually believe that narrative that Yeah, like that, that like that, that he accepted the ability to, to close the model.

And the caveat, 'cause the original, uh, and I'll shut up in second. The original tent was for OpenAI, is they, they, there was a group of people beyond Musk and others that didn't believe that felt that the world would be harmed if Google sort of owned ai. And I Well, if Any one organization owned ai, right?

Yeah. Yeah. That's what we're trying to get after, is there should be no singularity already With Larry Page and surgery and all that, that they were scared.

So, so, so Jon Swartz, do you believe or buy into this conversation, and I think Tracy was kind of alluding to, but you know, is Elon Musk, the Thomas Edison and the Thomas Edison of our times? He, he may be, that is a good question. It's very debatable.

You know, the, the other I was gonna mention, and John kind of spurred my thinking about this, I wasn't thinking about this before John just was just talking, was in the last several years, there are only about three individuals who I've seen these types of visions from, or heard things like. So there's Musk. And I remember in a couple of interviews, I did an interview with Steve Wozniak, where I, I talked to him about the future.

And one of the things he envisioned was these kind of dome like facilities that out in the desert, where these cities would exist out in the desert and be self-sufficient, almost like a, like a large scale or smaller scale version of a Phoenix or Las Vegas, but enter a dome. And I remember Peter Thiel talking to him when he first had that weird idea of the underwater, you know, Atlantis type environments. So it's Silicon Valley's become, and, and there's no real big ideas.

There's no Hoover Dam project, there's no moonshot. You know, there hasn't been, except for some of these individuals, especially Musk. And I think that's been one of the big criticism is like, everything here is based on iterations or convenience, or here's an app that does something for you while you're running an errand.

And I think we kind of got stuck into that ruts. And I think we really do need big thinkers. They can aim for the sky, they can aim for Mars if they fall short.

At least they made an attempt in what what they end up doing is something that will benefit us and maybe benefit society in the long term, knows. It's, it's true. We do need big thinkers, but, but You don't, you don't have to like, somebody come, you don't have to like them, them, you don't, I I get that for them to be a big Thinker.

I bought a Tesla. I loved the car, beautiful car. One of the best cars I ever drove when I've had a Porsche and I've had a Mercedes, but it had the very worst customer service I have ever seen in a, in a car, in an auto company.

At one point, it took me six months to get my car back from being fixed, and I had to spend the money, which cost me $1,200 to ship it to Denver. And this is when Elon Musk was running the company. So I don't, I despise him because he sucks at customer service.

He sucks at customer service, I Think, I think prohibit after Thoughts, never be anywhere near that. Oh, yeah. Um, so I'm not One of his cars.

He's a, I'm not gonna do anything with him, but I'm off of X permanently. Yeah. I mean, he ruined X, but, uh, among other things, but in a sense, I don't think from somebody like him, I think customer service is almost like a kind of a, a, a nuisance, right?

It, it's all about the idea. It's a new, it was a nuisance. How dare I tell them the car broke and I needed, needed service, even if I'm willing to pay for it.

Right? How dare I say that Consideration? Anyway, so don't worry about it.

That's probably More, I'm glad that this is an opinion show and not a news show. You Got it. Big opinion.

Well, The news was Tracy didn't get her car fixed for six months, and I was spending quite a bit every month to pay for that car. Oh, we are hoot, you know, tracking back to where, where John was going the very beginning. Let's, let's let John get Willis get, Say the radar.

A big thinker that gets very little credit these days. Bill Gates 'cause the amount of money he's putting into, say, solving some real world problems Yes. In other parts of the world that we don't get to see that often.

So, um, anyway, there's some, I'm not, I, I agree everything you said, John, but like, no, I think you're right, John. That's, that's a really good point. Because he doesn't pursue, I think he, to his credit, to Gates' credit, he doesn't pursue the publicity for a lot of the things he works on.

And you know what? All pros take him for doing that. Yeah.

I, sorry. I wanted, and A lot of big thinkers don't pursue the publicity Yeah. And are solving big problems.

And it just happens to be he's in the tech industry. Mm-Hmm. He likes the attention, he likes the, the, you know, the, the crazy, uh, presentations and, you know, and I feel like it's more than that than anything else, right?

Yeah. After my, after my Tesla experience, I'm just like, oh God, That wasn't bad. Oh, yeah.

You know, you know, Tracy, you know, I, I had a friend who was an investor in Tesla, this is years ago. And, and this is when they were having the autopilot issues. She was actually in an accident.

She was come traveling from Los Angeles to San Francisco on a freeway, and it didn't break. And this started all in, you talk about customer so speak, you talk about flaws. So there is a downside.

But, um, in, in a sense, you know, and I also was gonna say that, that that that 45 minute sequence, which never ended by the way, it was just this psychedelic show. It almost to me seemed like it was like kind of going inside his mind, right? It was like, so kind of, it was beautifully yet disturbing at the same time.

And I always, when I think of this guy, and there's that dichotomy that we discussed earlier, you know, you have to separate the ideas from the personality because the ideas are really interesting. The personality leaves much to be desired. All right.

I'm gonna say the last thing on this, 'cause we gotta move on to our next topic. But the one thing that was in your story that kind of screamed out, you know, this is smoke and mirrors, is, you know, he's saying that the other self-driving cars are based on lidar, and he's gonna be using computer vision ai. Well, we already have computer vision ai, so it sounded to me like he was just waiting for enough hardware to drive the algorithms that are gonna deliver at enough capability to make the car go reliably somewhere in two years.

Well, you know, God bless. But anybody can come to that same conclusion. And I don't need a light show and a psychedelic show and, and, and all this other crap to kinda, 'cause you know, if we've been following anything around the computer vision AI stuff, John.

Yeah. Say the same thing. Well, you know, it was interesting.

I just did a, um, I've been really trying to promote my deving book, by the way. I have a Deming book now, uh, the, um, at, in the lean conference. So I went to a conference that was based, um, in Columbus, Indi, Indiana, and we visited a Toyota forklift factory.

And what was interesting, they were bragging about the next gen forklifts. I'm not gonna use lidar. And, um, they, they're basically using sensors and they're finding like all these, frankly in material handling, like the, the, the complexity, the lidar is over expensive.

You've gotta have a distributed sort of system that, you know, there's just so many sort of heavyweight things with lidar. You know, we think of like with autonomous vehicle, by the way. They're not really good at boxes and obstacles and colors, lidar.

And so it's just interesting to see, I don't know if this is a trend or not, but I, you know, just hearing this first time about what you were talking about with, um, you know, sort of Tesla, um, you know, that, but they were very proud of the fact of how much money, how much more efficiency, and just sort of with the computing power, the, you know, the computer vision, power of just sensors and cameras. Uh, they felt that the, it's sort of the future of, um, factory and material handling. You know, even picking out a particular box.

You know, and again, I'm no expert here, but Lidar was supposedly, you know, would do really good with people and cars, things like that. But like, how do you sort of go up into a shelf in the fourth corner quadrant, pull out the smaller versus the bigger box, or the brand box or large box according to the, uh, engineers, we got a great overview of the, their preview next gen technology for forklifts, which would seem like a terribly boring subject, but it was actually fascinating. Um, anyway, that's, it was, when you asked me about that, Mike, I, I was like, yeah, I think then it normally have a lot to talk about.

Alright folks, that's where we're, I think the one thing we can agree on is that Elon Musk is a fabulous showman. We'll be right In a world where every line of code powers the future, every keystroke can introduce new threats. A 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.

All right, folks. We're back in the annual MetLife study on Women in STEM is Out once again. And I just once again found that there's issues here that we're just not addressing.

And I kind of feel like there's two parts of this conversation, and I'm gonna dare to say these things. And then Tracy will probably set me straight. But it seems like one is we have an education issue and we need to go deal with that at the grammar school level.

'cause we're not getting enough women into the STEM sciences early enough. But at the same time, we can't wait 20 years to solve this problem. So nobody seems to have a good answer for what to do in the meantime.

But Tracy Rainin, you have a show that kind of talks, some touches on some of these issues with Techstrong tv. So what was your take of this? Not surprising.

Um, you know, I do believe that there are some interesting insights in that article. Uh, and in particular, the interest in younger women in stem. But it's not, those, those numbers are just still not high enough.

Uh, yes, uh, Jody Ashley and I do Techstrong Women, so happy to be doing that program. And we've had some really, um, interesting discussions on this topic, even though we try to stick with technical issues, technical topics, because most women really don't wanna talk about this. But we did have, uh, Paula Bratcher Rad Ratliff on, uh, in one of our shows.

She used to be, she used to be one of the high level execs at Manpower, and she talked a lot about how difficult it is for younger women to become interested in stem. Now, when we think about stem, a lot of women go into medicine. There is increase in biotech, biotech and women getting funded in these areas.

And women in, in leadership positions, but not necessarily in all of stem, you know, uh, technology. We struggle horribly. And Paula, you know, uh, talked about the effects Covid had on women in technology in particular.

And she told us that they lost a huge number of women because they had to stay home and take and make sure the kids were going to school. And, and, you know, it was covid. They had kids at home, they couldn't do both jobs.

So we have, we have a long way to go in this area. And I just wanna say something about, and it will probably be controversial, but I wanna say something about DNI, um, in companies today, it is a checkbox that's all companies are doing. It's a checkbox.

I, you know, if we go look at Palo Alto Networks, they had a huge, huge in program in DNI, yet they still stuck in lampshades on two women in, in Las Vegas. So we have a problem. And part of the problem, and I'm going to, this is the other problem that I know, is that oftentimes women who have done well in techno in any of these areas, in, in business in general, believe that between two women, it's a zero sum game.

It's not, I have been in situations where the worst manager I ever had was a woman. So women are, we ourselves have to realize that mentorship programs are really critical. We need to be reaching down to younger women and reaching for their hand and pulling them up.

And that if this is not a zero sum game, we can both win. Even if it's a room full of men and you're, there's only two women in the room, we're not competing against each other. And oftentimes that happens.

So there's a lot of cultural problems that go along with this particular topic. Some of it come from women ourselves, and we have to look at ourselves and figure out how to solve the problem. We can't look at d and i programs.

We have to see what we need to do to make the workplace better for women. And there are some interesting, there's a, a Perona, um, which is a little DB company that we interviewed, both the CEO and the CTO, who both happened to be women talked about this. And they said that where their success came from was their co-founder, who was really, he, he really didn't think about, you know, diversity and inclusion.

He wasn't thinking about bringing women up the up through the, the ranks, what he was looking for, or people who were really good at what they did and didn't look at the, at the, you know, how they re what, you know, what their sex was or how, how they referred to themselves or their pronouns. He looked at who was gonna do a good job, and that's how he started bringing women up in the company. So we need both of that.

We need, we need upper management to recognize that women have a lot to contribute in STEM and in technology in particular. And we also need women to see and understand that this is not a zero sum game, and we need to help each other through the process. You know, um, several years ago, about 10 years ago when I was at USA today, we had this project that looked at this topic.

So there's, there's an incredibly talented, uh, reporter. We still lit the paper named Jessica Gwen. And so she and I worked on this together, and we got Jesse Jackson for some reason, got involved in it.

And we did a presentation at, at Stanford with representatives. This was really hard pull off, but representatives from Google, Facebook, and Apple agreed to do this as well as Twitter. And it was a very sensitive topic.

But what what was frustrating to me is that a lot of the things that we're talking about now, we were talking about back then, and the progress is, is glacial. I mean, there is progress, but it is glacial. I think mentorship is, as Tracy said, is incredibly important as well as education.

But, you know, the, the, what was telling was, um, it was, I believe in, in hindsight, a checkbox, uh, initiative for some of these companies because all of the chief diversity officers of those companies that were on that panel were gone within 18 months. And they were replaced. And I don't even know now if, if these companies, I'm sure these companies still have that title, but it, to me, at the time, it felt a little bit like a show on the part of the tech companies because they're so dominated by white males.

It's performative. It's definitely performative. Yeah.

Really was. That's a good word. Yeah.

Tracy, Tracy Bannon. I know you were involved early on in this topic, including working on that show on Techstrong tv. But, um, what is it about mentorship that doesn't seem to be happening among women in the field?

Is there's Mentorship happens. Let me, let me cut to the chase. The problem is not mentorship.

I'm over mentored. We're over mentored because mentoring has become the new mansplaining nerd, splaining, womansplaining to one another. So mentoring is actually a challenge.

What we need to do is more sponsoring sponsorship is different than mentoring. Mentoring and sponsorship. You know, have a them, but sponsorship means that I make an opportunity for you that I believe in you and I make sure that you have what you need to succeed.

And I'm there to catch you when you fall. Very different than the current mentoring programs, which couple people together. And you have lunch every so often.

You talk about your careers. And was it helpful? Sure.

I had mentors and should we go do away with mentoring? No, but we need a foot stomp, foot stomp on sponsorship. And I wanna talk a little bit about a quality that I have found.

So I have a podcast called Real Technologists. And the reason that it says, real technologists started from a rant when I was asked to put my pronouns into my signature block, and I said, no, you're a leader. Well, it's okay, but I don't want to out myself.

And I'm not first and foremost known as a woman technologist. A woman architect. I've been fighting for decades to be an architect, to be an engineer, right?

To be a researcher. Now, is the same true for the Jins? Maybe not.

But my rant was that, hey, I control my adjectives. I control my adverbs, I control it all. Let me control that all.

And that's where real came from. But as I interview individuals, and it's all about getting to their origin stories, what helped them to be who they are, whether they have different gender identities, all kinds of different humans. So it's not specifically female, but one of the qualities was that they dared to say yes when something was scary.

They dared to say yes when there was an opportunity afforded to them. And I think one of the things that we can really help all of the different diversity organizations with is to give people sponsorship, but also help them to say yes, women in specific, as Tracy said, right? We have to help them at the earlier grades to stop saying no.

Right? There comes a point, and there have been studies about this. Some not.

This is not just opinion. When you hit puberty and the brain starts to change a little bit where girls will stop taking those chances where they will throttle back from those things. So I believe we can really help, but we, what we can't do is stop talking about it.

Um, someone said to me, boy, you know, you've been dealing with women in technology. There've been the women's groups. Can we, can we throttle back on it now that the younger generation is doing better?

And I would say, well, hell no. We have to continue to talk about bringing the right people with the right skills to the table. And that expands us, right?

That takes us past the women in technology. It takes us to humans. And yes, there is a dominant group right now, but we can get past that.

We can get past that. So there, my rant over, I really, I really like the idea of a sponsorship of, as opposed to mentorship, because it really is, it's somebody who is re reaching down and saying, I am going to sponsor you. We are gonna, you know, we are gonna make sure you're lifted up.

Uh, and that is an area that women really need to work on. J John will, you know, wait, wait. I want to get John Willis here on this, 'cause I know he is passionate, subject, but let me, I wanna ask you this one question, John.

We both know a lot of guys who look like us, and they have quiet biases, right? They have, um, you know, and sometimes those biases, you know, become less quiet after they had a couple of beers, but during the day, yeah, No, it's, it's, it's, we're still a pretty pathetic industry. And, and, and, uh, you know, I mean, like, I mean the, the Palo Alto is the sort of the, the visual part of the, I mean, I stand on stage.

We all do. And you look in the room and you say, what, what the hell is going on here? Uhhuh?

You know, I'll tell you this, you know, white old, you're not fat. I am, I'm kind of less fat than I was white old fat privileged men. We, we grew up in this real, real world way.

Now, I was never, I never, I always try not to be a good person. I know, but like, you kind of don't know what you don't know. And like, this is the cliche part of me.

DevOps helped me so much here because I got to meet younger people, younger women. And I remember one of the, you know, all of a sudden I'm on the band working, how can we get more women, you know, in some of the earliest DevOps days, we, we, we went for diversity. We, you know, we, we literally, you know, tried to build a, you know, we, even with Jean, you know, we really went strategic about how we're gonna do this.

You know, we're not just gonna go 50 50, but we're gonna go out and find amazing, like the two TRAs, you know, different people who can come in and be like, incredible day one valuable. But, and, and so I was like, okay, well now I'm a good citizen. And then, you know, uh, a young woman, uh, you know, I could talk about the, uh, you know, what happened with, uh, what's her name and Docker, which was a, just a terrible tragedy.

You can take too long. Um, but, but somebody came up to me and said, you know, John, it's great that you're doing this. And every once in a while you're doing a talk about diversity.

And, and he said, but like, can we just focus more on how to keep women in stem? They're like, oh my God. Yeah.

I mean like, like, okay, we're all like, like again old, you know, sort of like, we've seen the light white men, privileged white men. Like, okay. And, but then we're not like yelling at our peers.

Like, to your point, that silent, you know, when you're in, in the bar and a couple of guys get a couple of drinks in 'em and a sudden saying, I don't know why they care so much about he or she, or, and like, that's our time to like, yes, speak up and, and say, you know what? That's b******t. And it's hard.

Wish you're having the, anyway, long story short, I mean, I, I, I, you know, Oh, kudos to you, John, as another friend of mine, Dr. John Cuffed, um, calls it out, calls it out on, on LinkedIn, and in his other social feed, when there's a mantle, he calls them mans, right? When you see a group of experts together, and visually, you're immediately caught by, wow, there's a homogenization that's going on there.

So he's very at, and it's, it doesn't mean that those are not experts, but can't we look at other experts, right? Is it don't always go to your comfort zone. And I think that that's a message for all of us, is not always going to our comfort zone.

I just always go to John Willis 'cause he's in my comfort zone. Oh, stop It. You know, I, I think STEM is becoming unavoidable in this sense.

And I was having this chat with my daughter, and she works in this fashion house and has steer stayed away from STEM her whole life, no interest. And we retire. And I was say, you know, so how's the economy?

You worried about your job? She says, yeah, we're laying some people off. And, um, you know, and I said, so are you concerned about that?

And she said, no, not really. I'm the only one here who knows how the software works. So go figure.

Yeah. Well, one last put, and then we're probably gonna have to close this topic out, but I'm an advocate of steam over stem. Oh, mm-Hmm.

So the, you know, the arts make a huge difference for, for stem, right? Better at math, if you expose children to music, for example. So there's a beauty that comes from Steam, not just stem.

And that tends to merge those worlds that you're talking about there, Mike. It tends to merge, uh, across the aisle, so to speak. It does, and AI is gonna drive it further because the people they're looking for to recruit for AI are all liberal arts people, more so than they are hardcore.

Mm-Hmm. Little bit. Yeah.

I just wanna tell one story real quick here. Um, please. This is about CloudFlare.

When, uh, CloudFlare was started by, um, Michelle Zalin, and I'm not, can't think of her, uh, partner, um, ma maybe Matthew Prince. They came from Harvard. She was the technologist behind it.

She created CloudFlare. She had thought about it for quite some time, worked on it through her college career. He was the sales pitch.

He was a good at selling. So the two of them marched themselves off to, uh, Silicon Valley to look for funding, and they got a big chance, but they took her partner to the side and said, we'll fund you as, as long as you get rid of the little girl. Oh, wow.

So We Have a, you know, if you're, if you're an investor listening, don't pass up CloudFlare because there's a little girl in the room. 'cause that little girl may have been the one who wrote the technology and the, and, and we, we recognize their biases. But you have to understand that you need to get past those biases, because there are other brilliant people solving really hard problems, not just e the Elon Musk and people who look like him.

Yeah. Tracy, that's, that's interesting you said that because I've had five, five women, uh, who I know, who started companies who told me the same type of situations, the same scenario when they were asking for funding, and they, they were met with the same type of answer, same type of attitude. You have no idea what they say to women.

Yes, yes. I had one Investors that, that I robbed product, I product Wish I could tell you what they said. Yeah, but I can wish I could tell what they Said, can't I?

And I said, why couldn't you invest in me if you love my product? And they said, because what if your child, what if your kid gets sick? Oh, Geez.

Yeah. Oh. And I said, let's not down this path.

Okay? I said, well, I'm so thankful that you believe I've been childbearing years. A and it was a very nice co conversation, and I just hung up.

Oh My goodness here, right? Like, you know, like, you don't hear her now recently, you getting a little more noise about her, but like, I've been doing all this crazy research on the history of AI and neural networks, right? She is as important as Koon Hinton.

I mean, because without what she did, you read her book, if you, if she connected the dots and you ne, you know, you hear the Godfather Koon, you hear Hiton, you hear Gio, you very really, I mean, every once in a while you'll see a little bit of a sort of slight of the godmother, you know? So we like continue to perpetuate this nonsense, you know? And she is in my, you know, from my history of neural networks, she's equally as important as Koon and Hiton because she connects the dots between the two of All.

Right? I think we gotta end this conversation here, but I will just end it with this one point, regardless of race, Cree color or gender, it won't be an insult. We'll be back in a minute.

All right, folks. And then earlier this week, there was an all day DevOps event. It's a virtual event.

com. And the main theme was, once again, DevSecOps, not surprisingly, and this is not a topic that everybody's in love with, I'll be honest. It's one of those topics that we all need to talk about it, but it doesn't drive the most amount of traffic on our sites because, well, a lot of folks just shy away from it.

And it's painful is at the end of this. John Willis, I'm gonna start with you here a little bit, but I know you were talking to Alan Chimble about this very issue also at an open text of virtual event. And what is your sense of what's going on with DevSecOps?

Because if I look at the Sonatype report, at least, you know, everybody's still using the wrong version of software, and there's tons and tons of vulnerabilities, and it doesn't feel like we're making any progress. And then the next survey will come out and say, including some stuff that we've done on, uh, text strong research, suggesting that, you know, half the folks at least are doing the right thing, but I don't know if they're getting the right answers, or at least, or if it just means they're trying. Yeah.

You know, shocker, people are writing and consuming more software. Holy Mac, who would've thunk? Uh, you know, wa Andreessen warned us in 2011, right?

He said software leading the world, right? And, and like, and now it's eating Thero a cloud. Like put it to another whole layer, you know, be just the consumability.

It's, you know, I'm not just called J Paradox, more abstractions you create the more you use, right? But, but now ai it's even sort of like, you know, now it's sort of steroid based, right? Like, because the generation of code and, and you know, we can stay away from the efficacy of the code.

But yeah, I mean, so the numbers, uh, you know, the sonotype, bless their heart, they've been doing God's work for 10 years now. I mean, those reports I think have been incredibly, the early years were amazing for us, especially when we were trying to build the DevSecOps narrative. Like it was great to have this, you know, credible amount of like data, you know, and they own made Incent Sonotype had made in Central.

So it was a great sort of grounding place. Um, you know, I think I, you know, like, again, I I'm a big fan of what Sonotype has done for the 10 years. I think, you know, I read the report recently and, um, recently, as in the last day and a half.

But, uh, um, you know, so the growth, the 150, all that stuff, right? But here's the, the one thing I was sort of disappointed at is that, um, they didn't hone in on ai. And AI is like the real vector problem right now because the abstraction of all the code, if we think about all the problems, and, and I know Tracy is an ex, you know, sort of the, the both traces are experts in this, but, um, you know, the dependency map I've always said is beyond human comprehension of just code, right?

Like the, what goes on, you write 10 lines of code, you know, it's could be a million lines of code with that 10 code, lines of code that you create, right? And then therein lies all the dragons. Well, that just up the game, just up two and a half, three orders of magnitude of complexity now with the abstraction and the code that's being built.

So I just would've thought, you know, like a little more drill in, and I know they have the data because, you know, I've seen, uh, what's to say, Steven, give presentations at ETLS. Um, you know, like, you know, sort of some of the other reports now are doing a better job about supply chain. And, uh, you know, like, like for example, um, you know, to get a report, they talked about an increase in PI high in p pip, you know, repository for Python.

Of course, there's an increase there. Like more people are writing Jupyter Notebooks right now, prefer AI models than, you know, people who had never coded before. And again, herein lies this abstraction problem, because now that now we've gotten more code, you know, at least at Java libraries were, were reasonably vetted.

You know, like your Capital One at one point had 20,000 Java developers, right? You know, NIST and all. I mean, there, there's a lot of work that looks the, the, the hard job of figuring out where the dragons are on its code.

But guess a place that hasn't been looked at is academia Python code, right? And like, it's now getting thrown into banks and retail insurance at a level. And so the real story about not is just that, you know, PIP libraries, you know, the pipi is now like 80% increase.

It's like the y and the impact of that. And then I, I think they just, I think they should have honed in on some of the new, the new vulnerabilities like hallucinations and code libraries or, or, um, some of the interesting stuff Wiz is doing like with pickle, uh, a bike code in model inference models, right? And, and escaping out, and the fact that there's a high increase in configuration misconfigure that's not really misconfiguration, but it's configuration vulnerabilities.

We saw this in cloud with Docker Kubernetes, right? Um, so, uh, again, I think the little bit of like disappointment that, 'cause I kind of know a little bit of inside baseball that they do have a lot of this knowledge. I think it should have been blasted or at least referenced more significantly in this report at this time.

Because I think our CIOs, and I'll go one last thing and I'll turn over the mic, but I've been working on this dear CIO narrative, like the CEOs are hiring. We wrote a sort of fictional story. CEOs are hiring Chief AI officers.

I think we've talked about this on this show. The chief AI officers are told to stay away from the ciso, stay away from the CIO, um, they're bringing in Stanford. I've heard the other day that they're only hiring from Stanford and CMU.

Um, so they're bringing these kids and they're, when you talk to them, they're like, why are you guys going on about how hard it is to install Kubernetes? It was easy for me. You know, and so they're building these stacks with Kubernetes and Kafka and Redis, and they're, they don't know anything about hardening and they don't do the research.

They don't, so like, this is like, this is an, like, I can't it, you know, I call it shadow ai, right? Like shadow AI is going to be at least three orders of more complex than shadow it, and we're still cleaning up from shadow it. So I, you know, I, like, I would've liked to see Sonatype spend a little more time.

I think, you know, not that we have to compare Sonatype and, and J Rog, which Jfr is doing a much better job of exposing all these sort of, um, macro level, um, vulnerable opportunities. Tracy Bannon, I know you were at a conference recently where some of this content came up. I mean, what's your take glass half full or half empty?

Um, it depends on who you are and what your skill sets are, whether it's half empty or half full, quite frankly. Um, so I agree with everything that, that John has said. We, there are orders of magnitude of people who don't understand how to operationalize the goodness, right?

Data scientists or data scientists, they're not intended to create something that's production worthy. Um, quite frankly, that's why I spend a lot of time, my Mitre spends a lot of time with our data scientists to say, how do we operationalize this? What does assurance look like so that we can actually take it forward.

The, the number of, um, a dish thought leadership that's made available by these folks is fantastic. But goodness gracious, what happens when you go from experiment right to field of production? There's an entire life cycle that should be there.

One of the things that we're trying to help folks with, and this just broadly, is to say, even when you're experimenting, unless it's just a one-off experiment on your desktop, and you're the only person who's gonna see it, everything means to have a base level of maturity to it, a base level of hardening to it. Like these are the fundamentals. John and I were both in Boca Raton, actually, Textron and Alan brought us together.

And even just working on some, um, LLM libraries, we were just working on some Jupiter notebooks. We realized that we had some really poor practices right there in front of us in terms of having our keys embedded and our secrets. And, and thank heavens that, that Shannon Lee raised her hand and said, yelling at it.

Hold on peeps. Hold on. Look what you're doing.

Yes, it was great. It was great. So the point there is that there's a, there's a, uh, we need to be embracing of the data scientists.

We need to be embracing of the folks who are on the AI side of the house, the mathematicians. We need to help them because they just don't know. They truly don't know, and it's not in their wheelhouse.

So let's give them guardrails on the other side of it. Let's make sure that when we are consuming those things, that we're smart about it. We always were, we were, well, always were, we always were, we were smarter about it when it wasn't coming at us quite as fast.

We were smarter about it. We evaluated, we didn't let our developers just bring down the latest package. In every organization I've ever worked with, we've had a two step hop.

You could bring something into a sandbox and evaluate it, but to actually be part of our corporate repository that was leveraged. And there were two, there's a second step, which was the evaluation and us approving it, and we hosted it, right? We didn't let you just simply go down out to, out to the great old internet and pull down what the latest and greatest was.

So there are lots of different approaches to it, but where John is going is right, whether you call it shadow or not, we got a whole lot of people doing a whole lot of stuff. And we need to make sure that they understand whether risk areas are and what are the small things that they can do to help make us all just a little bit safer. Tracy Ragan, you have been on this show multiple times saying that DevOps is broken, and now we're talking about extending that very broken DevOps processes out to more junior developers using higher level abstraction tools that hide all these nuances that Tracy's talking about, so they don't even know they probably exist.

So how is all this gonna turn out? Well, first of all, we're kind of, I always say DevOps is broken, uh, for many reasons. Uh, most of our DevOps pipelines are built around monolith, not a decoupled architecture.

And if you think about what we've done with decoupled is now, instead of having one big statically linked set of, instead of binaries that get released out together, and we, we check the, you know, we, we run static code analysis against all of them. We're doing it against every single container. And we hope that everybody does the same process, but many times they don't.

And just because you have done static code analysis or scanning code at the point in time of the container build, doesn't mean that there's not gonna be a vulnerability found five minutes after you just did that. So this is a real time problem, and we've u many companies have used binary repositories. They use all of the tools to try to keep them safe, but we still are not winning the, the, the battle.

And that's because our tools are just old. We are not making investment in this space any longer. I know that I'm in this space and I'm knocking on investor doors, and they wanna see what we do with ai.

And I'm like, I don't need AI to solve this problem. And part of this is when you have a, when you have a, when you have one vulnerability that could be now distributed across literally thousands of containers, what you need to know is where, what version of that vulnerability, um, it is, is the problem. It's not just every single, not it, the package may be fine, it's the version of the package that's the problem.

So you need to either put the older version in or upgrade to the new version where the, where the, the patches fixed, and they talk about that in that report that it could be a year before somebody consumes the correct package. So a lot of this has to do with scripting. Things are, you know, they, it, it's very, um, obfuscated where this data is coming from.

And now we have it in thousands of Russian dolls that's running it as containers across our, our, our clusters. And in that Russian doll, you have a package, you have multiple components, and then you have on top of that what we would call the solution that's put inside a container. And every time we do that, we pull in more those vulnerabilities.

So we could have duplicate vulnerabilities across the entire Kubernetes cluster, and we can't find where it's running. So what do we do? We take a, a sought off shotgun, and we try to hit the target.

We say, okay, we're just going to reinstall everybody's gonna get a new version of this package the next time they build. And does that always happen? No, it doesn't.

So it's a hard problem to solve. We need new tooling specific to this area. We need new, fresh, fresh ideas.

And by the way, that might be coming from a woman. Oh, let me, uh, my first question is, how do you say it? Is it maka dolls?

So that was a fantastic, the Russian doll analogy. That, that I think that that's a, that's a beautiful one. Um, a second thing that I'd like to get some opinions on is something that I was talking about with Mike earlier, and that is being at some different DevOps events recently, in addition to needing new tools, how do we re-energize the professionals?

Because there is a certain exhaustion, uh, in our, in our industry, in our field, right? We've got a little ton of information that's coming at us. But the folks that I see at these different events, they're not necessarily the new in career, they're not younger, they are well established, and they're going there to get a little bit of, uh, you know, just a little bit of extra energy or somebody that says, yep, you're right.

You're going about this in the right way. So how do we take not just the tools, but how do we reinvigorate because we keep just hearing about sir, some AI on it. Just, just sprinkle some AI on it.

And if you've got challenges with your SDLC sprinkling AI on it, isn't I immediately gonna get rid of your, your friction, right? It isn't gonna make your value stream any more effective. How do we help our, how do we help the humans right now in all of this, Right?

We only got a couple, we only got a couple of minutes left, so I'm gonna let John Willis answer that question because, well, you were at the start of this whole thing, so I'm glad I was gonna duck out. Man, that's a No man, that, that would maybe you Willis, that's for you, right? You know, maybe they get out going to put my cliche hat on, like, and back to Tracy.

Uh, the other Tracy's comment, like, all I agree with everything, and I just, I've been concerned about this. The, the sort of, the, the complexity of it gets worse. But to my point earlier, now, we have people who don't even know what they don't know.

When you look at an AI stack, when they mature away from notebooks to stacks, they're using Kafka, they're using Kubernetes, they're using all these tools. So I'm, I've been saying to people, you think you've written an AI application, you've written 85% of what you've written is basically how native infrastructure. And so now it's like the, the, the now more than ever, maybe that's the dear CIO let's scream for the Tom Anderson.

Like, this is as bad as it is. It's going to get a lot worse when we have like HR groups bringing in Stanford kids to build co-pilots that think that like, what's so hard about installing Kubernetes? I just hit a couple of buttons, right?

And, and like, like, maybe we need to scare everybody and get people our, what we do excited, you know? All right. I'll say one last thing.

I know Mike, we're gonna shut this down, but the problem we have right now is I talk to a, a woman who was a chief AI officer or a large, large organization that are billing stacks, and I asked her specifically, have you reviewed any of this with the ciso? And you could have heard a pin drop in the state where that headquarters was because she knew she had to answer it. And the answer was no.

Right? Um, this should be, and, and so the, this fictional story that we wrote is where the CIO is savvy enough to go, I'm not gonna let this happen. I'm going right into the Chief AI's officer and I'm gonna plant myself there, and I'm gonna make sure this doesn't happen.

So may, you know, like that's my sort of like tunnel vision answer in general. I mean, I, you know, I don't know. It's a hard problem because the people are less concerned about the hard people wanna do the easy and the abstraction just keeping getting better.

So it's keeps burying all the ugliness. All right, folks, you heard it here. Hey, look, the professionals are scared, so maybe you should be tuned.

All right. Hey, I wanna thank everybody for being on the show. It was awesome.

We ran a little bit long, but hey, no problem. We were having fun. Um, everybody else watching, please stay tuned for the next few episodes of the text on TV broadcast for the day.

Until then, we'll see you next time.