Techstrong Gang – September 26, 2024

Hello, everybody. I'm Mike Azar. Today we'll be talking about AI once again, because well, they're kind of like transformers.

There's more than meets the eye, and there's a lot of debate about who's good and who's bad. And then we're also gonna be talking about, well, CrowdStrike executives got called to Capitol Hill, which was kind of like going in the principal's office. And finally, the CNCF is after those patent trolls.

You're watching Textron. All right, folks, we're back and we got an awesome lineup as usual, starting with Mark Kinkle, who we haven't seen in a while. Mark, welcome back to the show.

How you doing? Great. Thanks for having me, Mike.

All right, of course. Then we have John Schwartz hanging out in Silicon Valley. John, always good to see you once again, So it's good to see you all.

Happy To be here. And then our representatives from the Futurum Group today are Kimberly Bates, who's out in Colorado with also Mitch Ashley, who happens to be in Colorado. And I have a question.

Have you two ever actually met in person yet? That would be too easy, Kendall. We have not yet.

We, we actually have in Paris, we met at, uh, Kup Con earlier this Year. You know, I, I find it's just easier to fly. It's kind of like I have all these friends from NetApp that live here, or HP live here, and it's a whole lot easier to go to the conference and see them there than it is to have us together here.

Better world, better world. So, so, so we gotta just travel halfway around the world to meet somebody who lives across town. Isn't that the way of it?

All right, well, let's jump in. There's this fierce debate going on around ai, and if you've seen anything, uh, the coverage of this is all over the map. We have those who are the true believers that say that this is gonna transform productivity in the economy.

And more and more, there's skeptics starting to emerge, including folks who are analysts at Goldman Sachs, who are also turns out arguing with other analysts at Goldman Sachs about the value of ai. And this conversation is starting to get a little strange because, well, we've seen a couple of things happen. First, philanthropic is saying that they will, uh, generate a billion dollars in revenue, mostly through API calls, and other companies are starting to make some money off of ai.

But, uh, the folks that use it seem to be questioning what exactly to do with Gen ai and where does it fit into this whole thing. So, mark, I know you follow this whole area closely, but is there, is there a nuance in these conversations that we're not appreciating? Uh, I think there's a couple things.

I mean, first off, when you have a hammer, everything looks like a nail. And that's what AI is today. It's everybody, and there is potential.

But right now, there's a lot of nails that we shouldn't be hitting. And so I think there's definitely a lot of value in certain areas, you know, helping to streamline customer service. There's really good in the image recognition and healthcare.

There's, there's little pockets. But just because AI has the potential to be so transformative today, doesn't mean it won't be there tomorrow. I think, you know, the Internet's agreed example, like we over invested early on on the internet, and you got companies like Global Crossing, which put, um, fiber across the Atlantic Ocean to start connecting the internet, but that company tanked.

com, all those big, uh, e-commerce companies, you know, they drove data center build outs, but it was Amazon and eBay and others that actually benefited more down the line. So I think that's where we're at today is, you know, it's, it was just so transformative to see Chat GBT. It was like a magic trick to even people like me who understand technology.

And then as you start to like, scrape away the layers, you're like, yeah, I sort of see, you know, what The Magician has up the sleeve. And I think that, that we're making this investment. They're definitely companies like Scaled DOIs, making about half a billion dollars a year open, ais making over a billion, you know, Nvidia is killing it, but they're all, you know, at the sort of lower levels, those, those companies that are like going, you know, concierge healthcare, they'll, they'll get there.

There's, you know, a million other sort of use cases. But right now, I think we're overinvesting and we will reap those benefits in, you know, a couple years in more and more verticals as, as time goes on. John, what's the mood in the valley?

I mean, to Mark's point, are people getting a little bored with the party trick and now it's kind of like, well, what are we doing with this thing? And there's some real, um, you know, hard conversations to be had? Well, you know, when, when the market leader Nvidia gets dinged for reporting less than ideal ROI in otherwise incredible, exceptional quarter, you know, there's something up, there's a level of impatience, angst.

Uh, actually, I recently, I talked with a couple of executives from HubSpot who are on the product side about the state of AI and the, this roiling impatience, uh, they both told me, and they made this comparison that Mark had alluded to earlier. They think that the current state of AI wave is somewhere to where the internet was in 1992 or so. So it's fairly early.

They still believe that there's this gestation development period. They think it's a few more years before we eventually things take off. And we have this inevitable fallout, or maybe even an implosion where companies that are generating $10 million in revenue are valued in the billions.

So that's, that's years away. Um, and you're right, the, these, his Goldman Sachs reports are like as clear as mud, right? There's no one knows, right?

No one knows. Even even Sheridan in his own report says much more is unknown than known, which basically means I don't have a clue. And I think that's the fear out here.

Especially as, uh, companies like Salesforce, I always pick on Salesforce, but they, they scramble into this space thinking everything's gonna, is gonna rise, the tide's gonna rise with all the boats, and it, it, it's, it's, it's early. It's just still, it's just too early. I Dunno, Mitch, is it too early for analysts to delve into what's going on in here?

Because, you know, I think John just gave the whole analyst community a back of his hand there for a second. Well, the, and the analyst community, I didn't mean that. I'm sorry.

I, I, I took it. Well, I'll talk to you later, John. Anyway.

I don't, I don't, I don't throw you into that group, Mitch. You're above and beyond. Okay.

I want one of the good ones. All right. Kimberly was too, by the way.

Um, I, I'm sorry for Kimberly too. Okay. I'm talking about Goldman Sachs in particular.

Sorry about that. In, uh, in the analyst community, we have this very sophisticated term we like to use called fomo. And that's what AI is all about.

And, and it's, I I like Mark's example, uh, around the internet because, you know, comparing this to something like blockchain, where there was a lot of hype around it, every security problem and a lot of other things can be solved by blockchain. And, you know, everything, everybody in the security world thought, you know, it's the next second coming of whatever, and it found its place, but it was way overhyped in this case. It requires hardware, it requires, you know, all the Nvidia chips.

It requires new kinds of data centers, especially with generative, generative ai and all the investment it takes in training and the power to run those thing. So, and, and the network to run it at, uh, 800, 800 gigabit per second, inter interconnecting, all these GPUs. So we are, again, over rotating on the investment.

I'm not saying we don't need an investment, but there's a bit of a land grab of, you know, I better buy some land. I better at least put, you know, a foundation on there. So I've got my claim, I'm not sure what I'm gonna do with it yet.

And so we're, we're in this learning period of what's the best use besides chatbots, what's the best use of generative ai? And some people are figuring out, a lot of people are still working on it. At the same time, we have, uh, machine learning, um, algorithms and models and AI models that some good work is happening in that too.

For example, I, uh, just did some work with a company called Feature Byte who is working on the problem of AI models are much different than software. They're very iterative between data engineers, uh, uh, a AI specialists and, and domain experts to get the model accurate. 'cause it actually has to produce a, an accurate result that you're intending, um, when you roll it into production.

And they, where they're working on sort of the slow parts of that process, how to speed that up. So there's some good things happening across the board. So we have to be careful not to whitewash everything as, as AI is, is way overhyped.

It's gen AI that's way over hyped. Yeah. Kimberly, you know, using Mitch's metaphor around, um, staking claims in the gold Rush, it seems to me that this is playing out all over again in a similar way to what we might've seen in the 1840s when, you know, we had all the Gold Rush people and the people who made the money sold shovels and jeans.

So are the cloud service providers, the makers of jeans and shovels in, in this whole thing, and they're laughing all the way to the bank? Well, I kind of think the team here is focusing very much so in the technology companies and commenting on where the technology companies are. And so let's, you know, move over to what, what we're really talking about in terms of ai, generative AI in particular, and, and what the impact is to companies and the, the rollout of the, if you will, the applications, the use cases, et cetera.

And in order to do that, yes, I need GPUs, I need all the technology, thank you very much for the servers and the infrastructure. But what really needs to be done is to create the models as you were talking about Mitch. Um, that can be rolled out.

Those models take time. And, you know, one of the things that you can look at the big enterprises that are cautiously moving into this direction, you know, first thing that they needed to do, not only were they, they early on acquired all the technology and they're trying to figure out how to implement it. But the other big thing that they had to do was address all of the legal, you know, priorities in terms of compliance.

How are they gonna approach it? How are they gonna keep this thing from doing things that, that would, you know, harm the brand They needed to get that done. Then they needed to set this up.

And, and you have two different types of companies. This was really well said. Um, I was, this week I was with NetApp and George Curian who was talking about, there's two kinds of companies.

There's ones that have well ordered systems and data, and they can train because they've got well ordered systems and data that can then other companies that he is talking to. He says they're a year or more away from being able to do this because they don't have that discipline. We heard the same thing from, from, uh, Jeff Clark at Dell, you know, with the companies they're dealing with.

So you have companies that can move rapidly and start delivering, but they still have, you know, they're looking at things like risk. They're looking at things about how hard, you know, what will I do with the company? And that means things are gonna move a little bit slower to make sure they are right.

You know, we've seen the ugliness that this can do. So they're gonna go, whoops, let me hold back. And I think that, you know, I'm sorry, but Wall Street doesn't understand this.

Wall Street thinks you can flip a switch and run with the technology and, um, you know, and so they're, they're hyping it, they are certainly hyping this. Can, can I just mention, oh, can I mention something really quickly? Yeah, yeah.

I, I was just wanted Tomia Kopa, when I was talking about finance, when I was talking about analysts, I was talking about financial analysts ultimately still and about Wall. I know I, I can say the Catholic nature of me, but no, I, I actually think this is a trap. And this is something as a financial journalist, and I've been absolved of that you see this overreaction Yes.

By the day, not just by the week, by the day, in terms of how stocks move. So I just wanted to interject that, because this is a hyperscale and for these guys, their, their opinions might change every 20 minutes and let alone every day, A couple, couple of our fathers there. And you're fine.

Okay? You're go, we, we, I mean, they do have, and those guys have a f and I, you know, they should actually, I think they should be ashamed of themselves for what they're doing, you know? Yes, I am benefiting personally.

I think we all have, in terms of our portfolio somewhere, you know, that, you know, the stocks have raised, and, and I, I kind of feel like the gen AI craze has saved us from a recession. So there's goodness in that. Mm-Hmm.

But, but at the same time, it's like, guys, go do your homework and learn what it takes to make this stuff happen. Um, and the amount of work these guys have to do to make it happen and, and quit being so, like, I'm Goldman Sachs. Come on, gimme a break.

So, so, so Mark, uh, mark, I, I, let me ask you something here. 'cause I, I'm, in my mind, I'm starting to bifurcate a couple of different things here in these trend lines. One is I will look at the investments that an open AI is making, and some of the companies that's part to train these massive LLMs, and it's billions upon billions of dollars.

And there's been rumors that, you know, heck, they're so far underwater that there might not be financially viable. And I wonder about that side of the game because those things require these massive levels of ROI that I'm not sure are gonna get generated. On the other hand, it seems like we're seeing the second wave of smaller LLMs that are targeted for domain specific use cases, and it's still early for that stuff.

But that stuff looks kind of awesome when all this stuff around magenta ai, this to me, looks to me where the magic really is gonna be. And this other stuff is kind of like, um, bad theater. Yeah.

So I think there's sort of these sort of grandiose moonshot. So let's just talk about LOM. So there's, the whole idea of a GI is artificial general intelligence.

That's human grade intelligence, that's lays off what's here today. And what you're mentioning is, what I'm looking at is more of genic workflows, and per specifically, persona based. So sales development reps, customer service reps, you know, scheduling agents, things that have very narrow scope.

Um, it can do that pretty well. And over time we'll have agents that can solve general problems. But that's a way off 'cause of the reasoning problem.

Um, the other thing that you mentioned is, is really the, the, the massive investment in AI right now, let's just look at incentives. And if we look at who's investing in it, we got Microsoft, who's a public company who invested in OpenAI, and they're making real revenue, even though that is very, very, you know, they're making a billion plus dollars on a valuation of one 50 billion. That's not great public company, but, but they're demonstrating, you know, revenue at scale.

What isn't demonstrating revenue at scale is this a hundred thousand, you know, different startups and they're being invested in by venture capitalists who aren't gonna make money on the actual appreciation. In general, they're making money on their management fees and for them to scale their businesses, they want to invest more money, which is good for us in general as an economy and society, but probably not so good for the people that put money into those funds because we know in these large funds that they simply don't return what the s and p 500 does. So I think what, what it all boils down to is, you're right, the, the, the narrowly focused LLMs are really solving a lot of the tasks that are making us not nearly as productive as we could be today.

I think in the long term, a lot of this investment is gonna be for things that may trickle down into other more successful companies, but it is overhyped in a lot of areas today. And, you know, the hype is, is I think it's called, uh, there's a name for it, a Maya's Law or Maris Law, where we gr vastly underestimate technology in the, or overestimate technology in the short term and underestimate it in the long term. And that's AI for us today.

But, You know, if I can add into that, you mean last, I think it was last March, I met with, um, the folks from Swift, the, the transaction people, the international international transaction people, and they were presenting on the, on foundational model that they were building, um, that had to do with, um, looking at international fraud transactions. And their ROI that they were painting was in the multi, multi-billions of dollars of the return on investment because how big the fraud is internationally. And, um, it, and out of that, I went, wow, you know, and exactly what you were talking about, John, it's this WOWness about saying, okay, so we're gonna see every industry coming out and in sub industries coming out some level of foundational model.

And those people are probably gonna use the LLM, the large language models to build those things. I mean, you guys are smarter than I am on the technical side, but I would assume one of those pieces, whether it's chat or chat, GPT or open, I, you know, whatever it is that's gonna come in, and they'll use that as some sort of process to treat their, these foundational models, build that foundational model, and then bring that out to the market and sell it, um, to the, the different people within their industry. So it's, that's another market that's gonna come up.

You know, Kimberly, I I hosted a panel on Tuesday this week, um, and we had a gentleman from MasterCard talking to exactly about that. They're using AI for fraud. And it's one of those, you have to play catch up, right?

You're continually, uh, uh, modifying and adjusting the models to, uh, chase after whatever the lead techniques are. But one of the things, it re it re it really stood out for me, uh, to quote the great prophet c Cuba Gooding Jr. Show me the data.

It's about the data. It's, it's not about having the chips, it's not about having the model. It's, it's the data and the data governance, the what it, what it contains, and what regulations and governance you have to ma manage.

Think of the SWIFT people. Think of MasterCard, visa, uh, what PII is in that information. What has to be anonymized?

Where is that data located? Where is it being processed? In what model, in what data center, in what country?

Right? And I have different requirements of how it has to be managed, what security requirements or what security, uh, guidelines, uh, have been implemented there. So without those, those mature processes in place, uh, you're left with generalized models, generalized LLMs, generalized, uh, small language models, which now you're up to the, okay, is it okay to use this?

'cause what's the copyright? Has it been secured? Is there any stuff in here I shouldn't have access to?

That's why people are so interested in rag for front ending, uh, those, those models. So, but you know, I follow, I follow, uh, uba, good Gooding Jr. Show me the data, And it is difficult.

But anyway, I'll, I won't wax on anymore. We, we can go on about this forever and we're outta time on this particular subject, but I, for one, I'm looking forward to that congressional hearing someday on this topic because, well, there's a lot to cover in here that goes everywhere from issues that the SCC might wanna look at to, you know, how this tech is actually being implemented and who's governing this stuff. We'll be back in a minute.

In a world where every line of code powers the future, every keystroke can introduce new threads 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. All right, folks, we're back in. 5 million devices around the world when all was said and done.

Um, the first thing, of course was they apologized. But John, you and I both watched this hearing. Um, what was your take on what was going on here?

I mean, it seemed a little, um, cordial in times. Yeah, yeah. It was surprisingly muted.

Uh, Adam Myers is the, uh, senior vice president for counter adversary operations at CrowdStrike. To start things off, he read a long apologetic contrite statement, kind of in line with what George Kurtz, the CEO has been saying. George Kurtz also met with one of the committee, uh, chairs, Adam, excuse me, Eric Swalwell, who's a congressman here.

And so there was kind of a very, very contrite, conciliatory nature to it all, which is in stark contrast, Mike, as, as we've experienced with these other hearings where you have a bunch of cow tech leaders being grilled by Congress and making promises and saying they would defer on decisions. In this case, actually, uh, CrowdStrike laid out some minor changes they've made so far in terms of, uh, the speed at which you would adopt your software updates. And, um, they, they were mentioning they had just met with Microsoft on moving forward on this.

Uh, what's interesting was that I think, given all the hearings that are going on, and, and in this case, there was a basic fundamental understanding of what happened for the most part among the congressional members. Of course, there were some cringey moments is when a representative from Tennessee earnestly asked if AI was to blame for the faulty software updates. And other members insisted on calling it a breach.

But, but for the most part, it, it was, it was kind of a healthy interaction, almost a mature conversation, which is something we're not used to in these hearings. Um, so I, I thought of it as a positive going forward, Mitch, I watched this thing and I kept looking at it, and I'm going, DevOps say DevOps say DevSecOps, no one. And it did not happen, but never, clearly what they're doing over at CrowdStrike is they're gonna manage the content updates the same way they manage the code.

And, um, arguably that should have been done from the get go, and hopefully more people are doing that. But maybe that's the lesson we're all learning from this, is that, you know, what content is go, It, it, this could have been monitoring software on computers. It didn't have to be security software.

This was not a security breach, actually had nothing to do with security other than that's what the CrowdStrike Falcon agent does. This is a software release process. Yes, there may be some QA issues also before you get to that point, but the fact that it was, is pushed out so rapidly without recognition that there were issues.

And so usually stage releases, not just in your environment, but also in customers, especially someone that can have that kind of impact. So that, that to me is where the gap is. Um, I don't know what penalty box you were in that you had to watch the congressional hearings.

I wasn't in that penalty box, but I doubt they talked about software releases. That's, that's what we need to be addressing. And not just CrowdStrike, we have a lot of other people running agents and software on our computers that potentially could have some catastrophic damage.

Maybe not as large, but let's get to the real problem here. Yeah, my, uh, that Mitch, it was interesting because Mars was trying to explain to them how the process works. You was mentioned there are 10 to 12 updates a day.

And, and again, it, it would always ve veer over towards it, how is this a security risk? And it, you can sense his frustration, but there's really very little he could do. But I think he did a, i I think he did a good job of explaining this is how this works.

Um, and, and again, there was the amount of clueless questions, and I don't, I don't mean to be harsh, but the amount of clueless questions were a minimum this time at least. Yeah. There was two congress people who understood the difference between pushing out everything all at once and putting, pushing code out in stages and understanding that that gave you an opportunity to roll things back.

And, um, so, you know, two outta, I don't know how many people were on that committee, 10 or so. Um, There's hope. You're saying there's hope.

There's hope there was there, there was some hope in there. I I was not completely, uh, chagrined to see the level of questions, but yeah, there are other people there who, you know, couldn't buy a clue and how they got on the committee, I don't know. But, um, mark, I mean, is this all theater, do you think?

Or are we gonna pull something out of this? So I've, I've found it all sort of humorous. You know, I feel like politicians trying to manage tech is like, I don't know, having someone who's never played chess coach a professional to chess team, you know, they, they understand how to move the pieces.

They have no idea how to win the game. Um, you know, they, it's all good intentions, but at the end of the day, I think the market needs to, to decide, you know, CrowdStrike and Delta, just 'cause Delta is a regulated, um, industry more so than others with the, uh, FFAI guess that's why they took this on. But, you know, the bottom line was it wasn't, it was just incompetence.

And it wasn't gross safety incompetence, it was growth gross, you know, operational incompetence from CrowdStrike and the way that they put out, you know, releases Delta and the way that they handled, you know, upstream releases from their vendor. And then the thing that made it a little bit, you know, more complicated than other application releases is that the CrowdStrike software and Falcon has access to Ring zero. And what it did was it bricked computers and they couldn't roll back because it required, um, physical intervention.

Like if it was a website scheduling update, they could have rolled it back and probably have that in place. But because they're sort of, you know, highly secure application, and if you, you know, don't understand how the chess game is played, that seems really scary. But the really scary thing is that somebody's CIO didn't have s sos that allowed for, you know, secure rollout and rollback.

And like you said, all they needed to say was DevOps platform engineering. Um, welcome to 2024, get your stuff together. And I'm sure after you drop, you know, the market is actually probably doing more to enforce this than, like you said, the very cordial government hearing.

That's all theater At the end of the day, you know, when your stock drops, you gotta reinvent yourself and update your products so that your customers stay there. 'cause I know my, my brother has an MSP and they don't sell CrowdStrike, not because there's anything wrong with CrowdStrike. They're like, it was a premium brand and it was overpriced, and the margins were, you know, really good for CrowdStrike not good for him.

I think that the market will either CrowdStrike will react and offer better options, better education for your customers, or people like Sentinel One and you know, Microsoft and everybody else that does endpoint protection will, you know, benefit from it. John, there was one lovely awkward moment during the hearing when, uh, one of the congressmen asked the representative from CrowdStrike. Um, so, uh, we think, you know, our estimates are that this thing costs about $50 billion in, uh, damages and, uh, what are we gonna do to make everybody whole?

And that was followed by a long pause of silence and a little, he and he and he and, and then the subject got changed. So, um, Yeah. Sound like a debate answer.

Right? Let's, let's talk about immigration. Sorry.

Anyway, um, let's, you know, sorry Bob, I couldn't resist. But you know, the one thing that that, that always plays out in these, these hearings now and plays out in, in something our, even our President Biden would talk about the profound risks associated with AI hears what I think has been going on, and people in the industry know this is what's the, the lawmaker's point of view. There's a backdrop in of overcompensation overcompensation for not letting what happened with social media affect ai.

So they're trying to get ahead of the problem. So when they're kind of the earnest attempt to do so, they're, they're scrambling, they're, it marks it, they're playing checkers while the industry is playing 3D chess, right? They're just try, they're just trying to maintain any type of, uh, tangential, um, holding of, of, of people accountable.

So I think that's what's happening. So they're overcompensating. Um, again, they, they, they're kicking themselves over all the damage that's been done by social media, and they don't want this to happen.

But I, and I think in, in so doing, they've overcompensated by looking at every conceivable thing they could fight wrong with, with, with the technology that's gonna do incredible things. I don't know, Kimberly, should we take any implications from the fact that the CEO of CrowdStrike didn't show up for this event and that they sent a vice president? I mean, 'cause in other industries, you know, you usually see the litany of CEOs get called up, you know, like it's, there's a mess in the airline industry and suddenly there's four of them sitting at a table with their, uh, hands up like this.

Um, but we didn't see that here. And in fact, we didn't see that with the Microsoft case either. They sent over some other vice p you know, they sent over the president, I think, who's their legal counsel and the head of the lobbying effort.

But, um, you know, it's interesting to me that we're not seeing CEOs in the tech sector showing up for these congressional hearings. I hadn't thought about that, but that's an interesting kind of play because, you know, when you have a CEO show up, it's more newsy possibly. So if I send, if I don't send the big guy the big face, I mean, if, if, if the Facebook situation wasn't Zuckerberg, would it had been covered the way it was covered, you know, if it wasn't, you know, or Dorsey or whatever.

Those kind, those guys, you know, they're, they're welded out and well seat. So, you know, okay, so let's say they do something about, you know, Tesla, you know, if Musk showed up there, he would just be all over the front page of the New York Times. Maybe if he sent his chief scientist on to talk about self-driving, you know, or um, auto autopilot and that kind of thing.

You're just not gonna have that, the press, press numbers. So, and then I could probably justify it by saying, these people probably know more than, you know, whatever. So I think it's kind of maybe a PR move.

I don't know. I have to go ask them. I'm just speculating and I know nothing, You know, you know, ly that is, that's a good point because when the CEOs were, were showing up like Zuckerberg, even Nadella Cook at one point Mm-Hmm.

They would often answer, I really don't have that information. I'm gonna defer, we're gonna have our staff work on it. Yeah.

And they would kick the can down the road, in this case with Myers. And, and in other examples, I think Brad Smith, they do know a fair amount more of what's going on at a granular level. So it is a little bit more useful.

Otherwise you just have a bunch of non-answers. At least you have some knowledgeable, insightful denials or not answers in this case. Alright.

You know, what's gonna happen next time, you know what's gonna happen next time, next time the company's gonna send an aa an ai, I Really, you're gonna say that they're bot Right need to testify for them on behalf of the company because, well, let's be real. If you looked at the prepared statement that they sent out, and AI could have read that just as easily Yep. And probably answered the questions maybe even a little better.

I don't know. All right. I know I'm delving into the land of cynicism here, But, well, you know, you could type that into chat GPT and say, here is all the information about what they're asking about chat.

What should I tell them? We'll, still amongst we assume answer Let, should be talking to, For the US House of Representatives about How do you think Congress got to figure out what questions to ask? We should have thoughts, we should have genic agent, we should have genic AI for both the questions and the answers.

Answers and just view it virtually. Right? Right.

Save a lots. My agent will talk to your agent. I think to your point about the fact that the CEOs didn't show up is really, these congressional committees have no teeth.

If it was Lena Khan from the FTC calling them in. Yeah. But like, yeah, it, it really, I didn't, I just watched enough clips to get the same impression that Mike has, has been, uh, giving is it's, you know, it's a bad cocktail party.

And, and they probably should have broken it up before they start. All right. You know, in the future though, that AI agent is gonna have to confer with an AI agent lawyer to answer the questions that are answered by the AI agents that the Congress people have.

So, you know, you can see how this whole thing will just be like, Hey, let's skip the whole meeting all together and just publish the transcripts and call it even. I Think that's what we're headed. Yeah.

All right. Well folks, um, I guess I'll be sad when that happens 'cause there just won't be as much drama in the world, but it is kind of where it's going. All right.

We'll be back in a minute. All right, folks. We're back and we have more illegal eagle chatter to have a conversation about.

But the Cloud native computing foundation is after, or helping the defendant at least open source projects who are being hit by patent trolls. These are folks that kinda buy up patents and then go looking to see who actually might be abusing these patents. And then they send them a little letter saying, you owe us some money.

And if you don't, we'll see you in court. Kind of thing, mark. Um, this has been going on forever in a day, but the open source community anomaly is supposed to be, you know, I wrote some code, I gave it to everybody.

I don't have any money, but that's making from it. So, you know, you can have 10% of nothing. Yeah.

So I, I actually was sort of surprised this was a news story. So I went into Google Trends just to see if, like, I am not producing tons of IP that I'm worried about software patents on. And the trend, it sort of peaked in 2013 and has not been a thing, uh, since that has at least covered the news.

And I think, you know, um, the Linux Foundation, and I used to work with the Lin work for, and I was a member of Linux Foundation. I'm not sure what the news is here. 'cause I don't see this as a, as a trend other than it's probably a good benefit for these organizations to pull their patents and have, you know, a unified front to make it more difficult.

And that's, you know, open Invention Network, which came around I think 2012 originally funded by IBM and IS works with the Linux Foundation along with, um, the new guys. Well, they're not new. They've been around for about anytime unified.

I think it's a good benefit for the industry to say, listen, this is not a profitable endeavor for nmps, which is, uh, um, non-practicing patent something or it patent trolls. NPE is the term of art. But, um, I don't see that there's a huge story there other than it's just one more step that says you're gonna have to invest a lot more to be a PAT patrol when you have more of these organizations out there with more members who, you know, license to the main, main, uh, or at least have an agreement not to, um, you know, sue each other and keep it from what basically is the bigger problem, which is software patents in general are, you know, it's, it's a pretty, if you look at what, what required is required to get a patent, it's sort of sketchy.

That's, you know, it has to be novel to someone skilled in the art. Well, 85% of the patents I've ever seen seem pretty obvious to me. So I guess I've considered myself partially Skilled.

You know, mark, I, I suspect that why would, why would CNCF do this? It's probably then, this is just my suspicion, this has become a big enough nuisance to the projects and to the contributing companies that you don't want it to become such a big problem with trolls, patent trolls, that people don't want to contribute software anymore. They don't wanna work on open source projects.

They don't wanna sign up as a sponsor of them. So maybe it's more preemptive, maybe it's gotten to a point where it's starting to become an issue. But that would be the kind of, okay, so they, they've got someone who's going to manage this whole, uh, NPE for us, should that happen on our projects.

So we're not exposing ourselves to a whole bunch of legal work or whatever and just kind of say, forget about it and walk away. You don't want that to drive the, uh, contributions to open source down. You're such a glass half bull guy, Mitch.

I got, I think I, I think there's, I think there's something a little more Nuanced to foot here. Okay. I think if you look at the Lennox Foundation and the CNCF, they operate on, you know, what I call the, the kindness of strangers business model, right?

They wait for somebody to go develop something and then they get handed this open source project that somebody gives them, and then they try to get other people to validate it and away they go. But if the thing that they were originally given has all these patent infringement dependencies attached to it, then they'll get sucked into this bigger conversation. 'cause those NPE fellows are just waiting for enough people to use this thing to make it interesting for them to sue somebody.

So I think that that's part of the issue, is they really need to vet more of that code to make sure that it's not violating some patent that some company didn't think through, or somebody created as a, a rival to a different commercial offering and then decided, well, let's just do 'em one better and we'll give it to the CNCF to destroy that market altogether. You know, there's this malicious behavior among the vendors sometimes that, um, isn't necessarily, uh, in the most, uh, shall we say, uh, egalitarian sense of open source. But, Okay.

So I think one of the other trending things that's happening here is that up until we're, we're now seeing the popularity of open source, particularly in the container market. And if that is, you know, as we've looked at it, the next platform that we're gonna be sitting on. So, and that is also where a lot of the AI development is going, is on the container platform.

So I, you know, as we look at companies like, I don't bring Broadcom in the situation, the VMware kind of thing, you know, one of the solutions to the VMware pricing problem is to start developing all your applications on containers. And that is all open source. So if ubiquitously, you start to see companies, and I'm not talking about Linux.

I'm talking about CNCF items here. If ubiquitously, you see that every enterprise company out there has a container environment in there and it's using that CNCF code, now all of a sudden these nmps is that what you call them, um, have got a bigger impact, potential impact across the board on companies and how things are deployed. That becomes more interesting because more money to be had, I have a little Argue, wait, argue against me.

Here you go. I think you're touch, you're touching on something important. And it's, it's, this is my combination of, you know, cynicism and practicality all in one.

What you're bringing up is that Kubernetes and container platforms are mainstream technology. Now when, and this is a trade organization that wants to make sure they're like, they're, they're the guardian of these assets. And in the early days when you joined CNCF, you joined because Kubernetes was hot.

It was mainly, you know, vendor dominated by people that wanted to sell stuff to end users. Now it's getting into a maturity phase where you start shoring up and doing all the boring, mundane things like protecting from practice. And this, in my opinion, is really just a new feature to make sure that they keep growing the foundation, that there's member benefits so that Kubernetes becomes welling, you know, is, is a stable place for people to continue to deploy it.

When it, as Mike said, when, when these vendors like Google gave Kubernetes to the Linux Foundation or specifically CNCF, it was, you know, it probably came with a lot of baggage that, and I'm sure it did 'cause I ran one of those projects. It is, you're shoring up a lot of holes and then you're taking into account a whole industry. And I think it's, I think that this is not a new trend or anything.

I think it's just, let's make sure that we're continuing to add value as a trade organization and making sure that Kubernetes is well maintained for a long time. And that's what the Linux Foundation does, is they, they keep the, the kids from, you know, their referee on that soccer field of technology, along with they're a guardian for this technology to have a long life that, you know, everyone from CrowdStrike to Delta to, you know, eBay to Amazon is using in their infrastructure. And we wanna make sure that, that that upstream provider, that center of gravity is, you know, well maintained.

So, you know, it's the, the problem, the commons, it is our commons for IT infrastructure, especially cloud, and they're just making sure that there's no, uh, you know, disasters. So another piece to that, and I'd love to hear your thoughts on this, is, you know, I'm not a developer, but with AI in one of the core applications for AI is code development. And I'm an open source developer, and I go in there and I go throw something up there to say, Hey, write this for me.

And if, if somebody has dropped in some of this patented software, all of a sudden it pops out. And that is a risk I would think. Yeah, it's always a risk.

I mean, like, like I think Mike or Mitch said early on is you can have 10% of nothing because these folks are normally doing, you know, open source, it's on GitHub, there is no direct sales in the early days. A lot of it comes out of industry. So, you know, is there patent enforcement when you're using it for your own own usage?

But patents are such a, one thing that's easy to, to look at is copyright. So we can look at all this code that's out there, make a comparison, and it's apples to apples. There's a set of tests for the US PPO that, you know, like, and I made allusion to it earlier, is it's a matter of opinion.

Does this, is this novel? Is it novel to someone skilled in the art? Is it, you know, does it have some kind of uniqueness that deserves patent protection?

And when it comes to things and software, you know, there was a big deal about Amazon's one click PO purchasing back in the day. And I think that that ended up lib being litigated and everybody has one click purchasing from Shopify to Amazon to, you know, apple. Um, so I, I think it's, it's a matter of the, the patent thing is very hard for a developer to understand, because what happens with these nmps is they go and find something that they think as a really good chance of being infringed on, and they go shopping.

AI is actually gonna make that easier. They're gonna be able to, to search patents and then cross reference and identify targets better. And hopefully people like, uh, United Patents and Open Invention Network are using the same, you know, the good guys and the bad guys are using these tools to sort of, you know, come to a standoff.

So let's come to the heart of the conversation, mark, we'll start with you. Is the patent system broken? Hell yeah.

Have you, I mean, this is the whole problem is that there's, it's too subjective. And if you look at things, I, I had a worked for a company that had a patent on making changes to multiple systems at the same time using Java. Now, any developer out there would go, yeah, that's obvious.

It's multi-threaded system. It's gonna do that. We had the patent, we probably should have sold it, we would've made more money.

But, um, it's just not, there's certain things that are patentable. And the problem is that you're at the mercy of how much money you have to spend putting into the patent and how much time and how good the patent inspector is, whether they issue a valid patent, and then whether or not somebody wants to challenge that. So when there's patents for something with not a lot of competition, great.

When there's a lot of competition and there's money there, then you'll, I I would tend to guess that it's, it's just not a good system. And Germany doesn't have it for, for that reason, plus many others. But, um, I think our software patents are not a great great thing for our industry.

I'm gonna jump in and say I think it's run its course. Um, we, we are at a point where companies do patents for defensive reasons purely in some cases. So when and if they get sued, they, they can trot out whatever things they may have that may, uh, argue against what the claim is against them.

On the other side of it, uh, their startups oftentimes that won't do patents. They don't want to go through the process. It's not worth it.

Their valuation is not gonna be based on how many patents they have. Other cases they are, you know, maybe in pharmaceuticals or something, but not in the software business. I'm not saying people don't ever do it, but it's not a given.

You gotta gotta protect all of this with a patent. So at some point you ask what's the value of it, other than in, in, in certain cases where it does influence valuation of a company or the defensive part of it is just because we're all just in legal battle with each other and, uh, the patents are the weapon. So let's, let's take the weapons away and see what the legal battles look like, You know, who has a lot of patents and then regularly takes people to court over it.

IBM probably at the forefront of this thing, right? A friend of mine wrote a lot of those. So yeah.

And I, IBM is the one that originally funded the open and best, uh, invention Network, if I recall. I Think that's right. And I think you're right.

They own those patents were more profitable, and now I'm sure that they're probably less profitable and more expensive. And also we just don't have the same r and d costs to recoup from software that we do in pharmacy. So that makes it a different, you know, game.

All right. So should we at this point fund an effort to go and find, I don't know, all the patents that are owned by some entity that's not actually building software and review those patents and challenge them and at the same time maybe have a, a, a moment in time where we all agree to like, Hey, we're gonna give up our patents for all these silly things so that they don't fall into the hands of these folks and we can just have an a moment as an industry and say, let's put this to bed once and for all. Is that kind of like bringing all our guns into the middle of the town square?

Kind of, yeah, and actually I think that is what the intent of these companies are. I think you license those patents to those, um, OIN and uh, United Patents, and then they have 'em in the portfolio, even though you maintain membership, but with other conditions. And I think you hit it already on the head, Mike, is everybody should be bringing their guns to the Count square, not just some of the people.

That was Kimberly's idea, but yeah. Sorry, What's, what's the movie where the, The American revolutionary people to do so? There we go.

I, that didn't work very well for the Brits. That was both, I think there's A movie, I can't think of the title. It's, it's, um, A Night of Anarchy.

Basically all all laws are are taken off the books and you know, people go to the street and wrestle One night a year Where all hell breaks in. Yeah, exactly. Maybe we do that with patents and see what happens.

Patent That, that does, that doesn't end well. But maybe we do need some patent control legislation. It might be more successful than certain other control legislation that we're putting out there.

Well, more, it was called The Purge, Mitch Purge and thing that we should do with Software patents. So I agree out loud. I thought you might sign up for that.

That was pretty ugly. Oh my gosh. All right, folks.

Well, I enjoyed the conversation today. I wanna thank everybody for participating, and I wanna thank you all for watching. Um, coming up next, of course, we have Ira Winkler has a few opinions about this whole return to the Office thing that Amazon is pushing through.

And it may surprise you that, you know, he, he has issues with all those, uh, what does he call 'em, uh, keyboard warriors who were giving Amazon a hard time about, well, it's called Get a Job, go To Work. And there's no such thing as a perfect job. So grow up.

Anyway, folks, Put your big boy and girl pants on, go to work. There you go. Go to the auction and After that, we have a great lineup of Textron TV content.

We invite you to check that all out. Until then, we'll see you next time. Hi, this is Ira Winkler with this edition of Bite Me Today.

I wanna address, and there's so many things that actually come together on this one. Um, fundamentally, let me, the first thing I'll start with is everybody now criticizing Amazon for forcing five days back in the office. No more work from home.

If you work at Amazon or whatever the exact rule is. And I'm sitting there watching and listening and reading about people say, ah, everybody's gonna quit Amazon. Amazon's gonna lose all their best people, blah, blah, blah, blah, blah.

And it's like, are you actually kidding me? I almost dare you to quit just because Amazon is doing that. Why do I say that today?

In the same breath, I see something about IBM laying off 7% of their workforce or something like that. Yesterday I saw that Cisco was laying off 17% or something, some outrageous numbers. Thousands of people are being laid off.

And I'm not saying, gee, everybody thank Amazon for keeping me employed in a sarcastic way. But the reality of the situation is the job market's tough. And yes, if you don't like it, you are not here to second guess an Amazon corporate decision.

Maybe it'll work against them. It's frankly none of your business. The reality of the situation that you have to consider is that if people don't wanna work there, they won't.

Most people unfortunately, do not have the option. So you standing there being a keyboard warrior saying, aha, now everybody's gonna quit Amazon again, I dare you to find it. You don't even work there in any of the cases I've seen where people have been criticizing Amazon.

So, you know, to that extent. And at the same time, you know, this is kind of also related because I periodically see lots of people complaining about job advertisements, and they'll say, how could these people look at, they're only paying $50,000 for what's claimed to be an entry level person that needs A-C-I-S-S-P, whatever, whatever, whatever. And again, I'm sitting there thinking, you don't like the job?

Don't take it. Somebody might need that, given all the layoffs that that job, that underpaid job could be a godsend to them. Now, am I advocating predatory behavior by employers?

No. I unfortunately have been in a position where I could not pay people as much as I want, but just for example, who's gonna underpay people? Frankly, there are a lot of nonprofit organizations that need cybersecurity people.

And what are you gonna say if you can't go ahead and pay people the salary that IBM or Accenture or somebody else might, that therefore you shouldn't be hiring people and offering them a job at what you can afford to pay them. You know, again, let the market decide. I see so many people out there saying, so, oh, entry level position, that requires a prerequisite.

Yeah, a lot of them do. I address that in a previous edition, invite me, but at the same time, what are you, I mean, all these people are out there expressing outrage, telling people that, no, you should not have this job on the market. Well, what are you sitting there doing to any of you that are criticizing anybody's job vacancy?

I challenge you to actually put out your own job vacancy and pay people what you think. The reality of these situations is all these people who are outraged and shaking their fist into the clouds saying what bad people they are for offering a low salary, none of them are offering any damn jobs to anyone. So if you are gonna listen to all these influencers, what you need to understand is they're not giving you any jobs at all.

And the people who are, frankly, again, like I said, there were positions where I was paying entry level sock people. I actually raised rates by 10 to 20% for any of the SOC people we wanted to keep. Fundamentally though, they were hired, many of the better ones were hired away.

Actually, not even the better ones, but just about anyone I trained was being hired away by Rapid seven for roughly twice what we were paying them do. I wish I could have paid them more. Yeah, I wish I could have paid them more, but it wasn't in my budget to do so.

And again, like I said, I got them 10 to 20% raises to begin with, which was a lot for them. But either way, don't listen to all these people who are shaking their head fists at the cloud saying, how dare they offer a job? How dare Amazon bring them back?

None of these people is doing a damn thing to give people the right salary to give people alternatives to Amazon. And unfortunately, the reality is there's a lot fewer options to Amazon today than there was even a week ago. Tens of thousands of jobs have been lost in the last week, and you are complaining, oh gee, now everybody's gonna quit because they have to go into work.

Yeah, it kind of sucks. I admit I haven't been inside a job or in a full-time position at a desk in an office for God, more than 20 years now, give or take. But you shut up and let people make their own decisions.

Don't start telling people they should be quitting when they don't have alternatives. Anyway, that's Ira Winkler for this edition of Bite Me 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.