# Security tools and the companies that make them

URL: https://securitytable.ai/security-tools-and-the-companies-that-make-them/
Published: 2022-12-11
Duration: 3179 seconds
Season: Not supplied
Episode: Not supplied
Hosts: Chris Romeo, Izar Tarandach, Matt Coles

## Show notes and chapters

In the inaugural episode of the Security Table, the gang discusses Mark Curphey's article, "[A Security Tools Crash Is Coming](https://www.linkedin.com/pulse/security-tools-crash-coming-mark-curphey/)." We consider the four conditions Mark describes, and then we riff on what it means for the security world in 2023. We also uncover several debates that will resurface in upcoming episodes, such as SBOM: what is it really for?

Follow AI Security Table:

➜ Home: [https://securitytable.ai/](https://securitytable.ai/)

➜ X: [https://x.com/AISecurityTable](https://x.com/AISecurityTable)

➜ LinkedIn: [https://www.linkedin.com/company/ai-security-table/](https://www.linkedin.com/company/ai-security-table/)

➜ YouTube: [https://www.youtube.com/@AISecurityTable](https://www.youtube.com/@AISecurityTable)

## Transcript

**Chris Romeo (0:01)**: Welcome to the first episode of The Security Table. We're so glad that you joined us to pull up a chair, I guess, around our virtual security table. My name is Chris Romeo, and I'm joined by a few of my friends who are passionate cybersecurity people who like to argue. Well, that's not exactly why we invited them to the table here, but we do want to explore the world of software security, application security. This, this is a thing that all we, we all love. And what brought us all together. And so we've got Izar Tarandash, Matt Coles, Tonya Ward. This is the folks that pulled up a chair to the table. And so I'm going to dive us right into our first issue and an article that was published in the last week or so. Mark Kerfee, the guy who was behind OWASP from the very early days, put this article out there. It started as a small LinkedIn post and then grew into an article. A security tools crash is coming. Wow, what a big thing for us to think about here. We're all AppSec practitioners. We all interact with tools in a lot of different ways. And I'm not even gonna tell you what Mark said. I want to hear what some of the other folks think though, just about that idea in general. Yezar, coming to you first, like what do you think about this idea that there's a security tool crash coming?

**Izar Tarandach (1:21)**: I'm running around in circles with my head on fire 'cause I just joined a security company creating security tools. So the way that I look at it is, you know, there's a general tightening of the screws everywhere. Don't have to tell anybody about it. Everybody already knows that all indicators and indexes are going down and unfortunately people are getting laid off and things like that. But I think that together with that closing of the purse and people not throwing as much money into security as they used to, buying anything that's shiny and that will solve all their problems as a silver bullet, which of course we know is totally a thing. Uh, on the other side, they will have to replace these— some of these people who are being laid off with automation. How good is it going to be? How effective is it going to be? I don't know. But in the grand calculus of things, it could be that that's even an incentive to go for some other tools. On the other hand, it is a very big space, but it is very saturated.

**Tanya Ward (2:36)**: So can I ask, how many possible tools can you have? Like, do we even know what that is? And like, every time we put tools out there, do we think about the processes that we need to put in place, the people that need to run them? I mean— Wait, well—

**Izar Tarandach (2:52)**: Sorry, Matt, did you just go into a kindergarten and ask a kid how many toys they can have?

**Matt Coles (2:58)**: Lots of Legos, lots of Legos.

**Izar Tarandach (3:00)**: Lots of Legos.

**Tanya Ward (3:01)**: But then you lose the Legos and then you don't know, you know, I mean, at some point there's oversaturation.

**Matt Coles (3:08)**: So just keep in mind that we've been working for a number of years towards getting very specialized tools to do one thing very well, and you need, of course, then a lot of those tools to be able to manage the the environment and the, the, the space here where we were talking about a lot of different things that can be, can be focused on and test and tested with these tools, things that we rely on, on them for. And having one monolithic tool or a small set of monolithic tools that do a lot of things, that's, that's been happening, that's happened in the past. And, you know, driving towards single single-function specialized tools, you're going to need more of them.

**Izar Tarandach (3:48)**: So, right, but if we look at the market the way it is right now, and correct me if I'm wrong, in that saturation we have a large set of tools that either overlap or do exactly the same thing. And in order for the best of each class to, to come up, people have to declare that Right, so if we are tightening the market, it means that the weak ones are going to fall along and the strong ones are going to surface and lead the market.

**Matt Coles (4:21)**: Well, if I can interject there for just a sec, so, uh, one of the things that is also a, uh, is a related topic to this is vendor lock-in, right? If you rely on, on a particular tool or set of tools from a, from a single vendor, you know, it's really difficult to get out of that and learn about or take advantage of innovations from other vendors because your infrastructure and your tooling and your processes and everything have been set along that particular path. So I think that's a challenge that we have to be very careful of here.

**Tanya Ward (4:54)**: Yeah.

**Izar Tarandach (4:55)**: Not only that, but who's going to decide what the best tool is? Because the best tool for me is not the best tool for you.

**Tanya Ward (5:01)**: Yeah, absolutely.

**Izar Tarandach (5:02)**: So of all those that are going—

**Tanya Ward (5:04)**: Yes, I was just going to jump in. So, you know, we're talking a lot about tools, but you see, like, when you look across everything that we do within security, is there a particular— it's like we, we try to solve a problem by automating something, by adding tools, but sometimes as we keep going in, we find that there's gaps. So that's, that's why more tools are being developed, and then the lack of integration. So between these tools, and then somebody has to solve that problem. In your, in your experience, are you seeing a particular gap where we need to start building more tools? 'Cause we know there's a lot of tools for scanning and threat intelligence and logging, all that good stuff.

**Izar Tarandach (5:42)**: That goes back to the problem of which one is the best. We are in a domain where we have difficulty saying we are doing a good job of something because we don't have the right metrics. We don't really know if everything that we see is everything that we are looking for. So, if you can't measure that, if you don't have metrics that you can compare from one to the other, how exactly do you say if you have enough or if you have coverage or if you have the best?

**Chris Romeo (6:12)**: But I mean, think about the SaaS market, right? Like, when was the first SaaS tool invented? Probably, I mean, it's got to be in the '90s, right?

**Tanya Ward (6:20)**: '90s.

**Chris Romeo (6:20)**: Yeah, it was in the late '90s or so. And, but then I also think about like what's been the innovation in SaaS? Since the beginning of it right now.

**Matt Coles (6:29)**: Well, actually, I think that there's been sort of a— I'm sorry, Chris, I didn't mean to interrupt there. I think that there's been sort of this, this wave pattern of innovation, right? There was a lot of— first it was basically grep, right? Really, it was keyword searching. And then it was, you know, control flow and data flow analysis and looking at control graphs and, you know, translating to some, you know, natural language and whatnot. And then And now we're back to, again, looking at the single function, we're back to things like semgrep, right? It's grep, but with semantics added. And so, I think it's gone in waves. We've, you know, it's, what's the word I'm looking for?

**Izar Tarandach (7:15)**: I think it's not quite the wave. Yeah, go ahead, please. More like, what's the name of the race? A weapons race.

**Chris Romeo (7:23)**: The arms race.

**Izar Tarandach (7:24)**: Yeah, it's an arms race. We had these two big names, and it was funny, like, every time that you went to a conference or to an exposition or something, you would see that one of them inched forward, and then the other one pulled, and then the other one pulled. And it was always those two. And I think that you guys know who I'm talking about. But then we had the jump to, as Matt says, back to the the beginning, looking with grep, but a more smarter grep. And now we are seeing this ML and AI and whatnot thing coming up, and that's the next weapons race, right?

**Tanya Ward (8:00)**: Yeah.

**Izar Tarandach (8:00)**: Who's going to have the best model?

**Tanya Ward (8:02)**: It's— sorry, I think you've actually nailed it on the head. Tools are dancing with problems. So think about maybe several years ago, hardware. Like, did we have tools for code signing on hardware? You know, how did we necessarily scan hardware? Now we're into supply chain. Oh, so now we have all these tools coming up to figure out how do we actually manage the interdependencies in our products, transitive dependencies. So you're right, I actually think then AI/ML, there's going to be more tools. So going back to the question, I think we're always going to need tools because the problems that we try to solve today are not going to be the problems that we need to solve or automate for tomorrow.

**Matt Coles (8:39)**: And we also need to keep in mind that as the technologies and the technology stack changes, yeah, right, the, the The technologies lend themselves well to different types of analysis or different, they have different capabilities. And so a tool that did exist either can't function against this new, the new tech, the new technology stack and you need something else or it changes how they, how let's use the SAS example. It changes how analysis is done and the false positive rate that you get from it. And so There's those aspects in addition to the value-add service, right? How do I as a tool vendor manage integration into an enterprise? What's my license model that's changed over time? How do we get better monetization out of a tool? Then you have the influx of open-source tools, right? Everyone wants to create something.

**Tanya Ward (9:35)**: Totally.

**Matt Coles (9:36)**: Yeah. How do you figure out best of breed when that best of breed is a moving target?

**Chris Romeo (9:41)**: Yeah, I mean, you think open source though, right? Like open source in the enterprise from a security tool perspective is, I mean, with the lack of support can be a blocker to using something. I mean, Semgrep is not really a good example because it's an open source, but there's a company wrapped around it that helps you. And I happen to like Semgrep. I think it's a cool tool. I think they've done some really good things. So I'm not throwing anything bad in their direction. I'm just saying like, If I get a, let's take Bandit. Okay. Bandit is a SaaS tool for Python, fully open source. I don't think there's a company behind it.

**Tanya Ward (10:15)**: No.

**Chris Romeo (10:15)**: Like, can you put Bandit into your enterprise though? Like, does it, it without the support that comes from a commercial tool. So is open source even something open source tools? Are they even part of the conversation here? Yes. Are we just talking about commercial tools?

**Matt Coles (10:29)**: No, I think that they have to be. And, and, and specifically because, and let's, let's talk a little bit about who, who's running these tools, right? Security teams versus developers, right? Developers need to stay ahead of, of, they need to stay ahead of what security is asking them to do or, or stay ahead of what their customers are asking them to do. And so they're always looking for bleeding edge and that, that introduces some challenges, right? Shadow IT is always a challenge and, and we can talk about that at length, I'm sure. Uh, versus commercial tools. I mean, you need to do that at an organizational level to really be effective. And so now you have security teams generally driving those purchases, whether those are central security teams or local within development orgs. But those are— I think we need to consider both. Maybe the questions are different we need to ask.

**Tanya Ward (11:19)**: Matt, would you consider it— No, I was just going to ask a question like if you're going to bring open source in and it's not supported, I mean, this is just advocating more that developers, security teams need to be more heavily involved in the open source market to be able to, you know, provide that level of support.

**Chris Romeo (11:41)**: That's one of the things Mark was saying. That's one of the things Mark was saying like in the context, not necessarily in the context of this article, but I've heard him talking about that from the OWASP mission. The fact that, you know, we've got these open source, we've got these other companies that are part of that are taking advantage of what OWASP provides, but they're not giving back. That's part of the challenge here, is that you do have some people that are using these tools, but they're using them for commercial means because the open source license allows it, but they're not giving back to the bigger community to help move the whole industry forward.

**Matt Coles (12:17)**: That's right.

**Izar Tarandach (12:18)**: Yeah, but the problem is that as soon as they start giving back, then that's when we start looking at certain open source tools as having a company wrapped around them. The thing, one of the things that we ignore about open source is that it suffers from a huge problem in terms of especially going towards the enterprise use. People want to write the stuff that's interesting, that's shiny, that's sexy, right? Nobody wants you to, to be writing the exporter for JSON because it's going to go into some big system and Workday and whatnot and something that the enterprise needs. For that thing to work, right? So that's where the enterprise should go into, into open source, in, in my opinion. Provide that envelope to the Shiny stuff that you need for that thing to operate into an enterprise, uh, environment. Leave the rest alone. Don't, don't embrace the whole thing. Don't try to own the whole thing. Don't try to direct the whole thing. But give that cocoon where the, the open source developer knows that if they put their time in, they will have the satisfaction of working on something that's of interest to them, but at the same time, eventually can end up in an enterprise.

**Matt Coles (13:29)**: Can I ask a quick question here? In ISAR, and obviously the rest of you as well, have we moved beyond the barrier of GPLv3 as a blocker for enterprise developers to get involved in open source projects?

**Izar Tarandach (13:48)**: I have seen at least 3 or 4 big places in the past that are very reluctant to let their developers go back to something that's GPL v3. I think that there is still a view of this is a virus of a license. I'm not a legal person.

**Tanya Ward (14:05)**: Yeah.

**Izar Tarandach (14:05)**: I'm not a lawyer, but that's what I heard.

**Chris Romeo (14:08)**: Yeah. I was gonna say Matt just offered the gut punch of the whole giving back to open source. Like, Oh no, I shouldn't have.

**Tanya Ward (14:18)**: What I think is really good about open source, the tools that come out of, you know, all these different initiatives, is there are a lot of companies out there that don't necessarily have the level of funding as big enterprise. And so it does become available to everyone. It's like everyone has access to these tools. And that only can make security better, you know, across the industry. Whether it's small businesses to enterprise businesses.

**Chris Romeo (14:42)**: Yeah.

**Izar Tarandach (14:43)**: So, and I think that you touch on a very important point too, uh, Tanya, perhaps even without going there, but the interoperability. Like you, you talked about the number of tools and Matt talked about the ubiquity of them, but sometimes we do want these things to pass results from one to the other. And I think that it's very rare the tool that considers putting things out in a consumable way. But even more rare than that is the tool that's ready to import somebody else's results and translate things and agglomerate things in a way that I am looking at, for the lack of a better word, finding from the angles of all the different tools that I have. And I don't see n findings, I see one with all the information from all the different tools.

**Matt Coles (15:33)**: The challenge is, I think, that that lends itself to yet another tool to provide that bridge.

**Izar Tarandach (15:40)**: Or yet another standard format. So I wanted to point out that I think that we are 30 minutes in, and this is the first time that this is going to be said in this setting. Take, for example, the Open Threat Modeling format.

**Matt Coles (15:56)**: Hmm.

**Izar Tarandach (15:58)**: I guess. That's, that's been adopted now by OWASP Threat Dragon, of course by Aries Brisk, by PyTM. And these are already 3, 3 tools, one of them commercial, 2 of them open source but with different approaches, and all of them using this open format and approaching it of course from different sides and different views. But it is something that came from an enterprise setting with the commercial offering starting it. And in my view, it's a positive example of what enterprises could be doing with open source more.

**Matt Coles (16:41)**: What you're calling out is standardization as a means of providing— well, I guess it's a double-edged sword. If you standardize you can get both a collection of tools that now are competing on different aspects, or they all become the same, in which case it doesn't really matter which one you choose.

**Izar Tarandach (17:09)**: I see what you, what you mean, but again, take the OTM as an example. The format is open in terms that every tool can add its own thing, and there are very, very little required attributes in the whole format.

**Chris Romeo (17:22)**: Yeah, but let's pick on OTM. Let me pick on this for a minute because Ken, when I think of things like this, it's like, what really is the value prop, right? Like, a company starts something and then they put it out there and, you know, a company always has something. A company doesn't create an open standard because they're good people. And I'm not picking on OPM or I'm not picking on OTM, whatever it is, or anybody else along the way. Is it really something of value though is my question to you. What is the value proposition for me of this open threat modeling format?

**Izar Tarandach (18:02)**: I can only give you my personal view and I have that. It's worth exactly what you paid for it. And I think that in the case of threat modeling, whenever I look at the commercial offerings out there, they already tend to be enterprise-grade level with lots of fields, lots of dialogues, lots of workflows and this and that and the other one. But as a threat model practitioner, if I can offer someone a simpler experience like the OWASP tools or any number of other open source tools, and they can experience that and they don't lose their work when they move from the simpler to the more complicated because they need to follow that progression, right? The enterprise wants to be enterprise and wants to integrate with all kinds of stuff. If they have that format that will take them along with the results they already have and they they don't have to go through the look at one window, go to the other window, create the same thing. That for me is already a win. So that's why I like the fact that they came out with something that's so open that we could use.

**Tanya Ward (19:10)**: So I've got 3 points on that. So the first one is when we think about tools and what you've just been talking about, can it scale? Like when we build out, we start off with an idea and incubate it. Open source, and we think about enterprise, like, are we really thinking about scalability? Number one. Number two, are we also thinking about how usable are these tools? Like, at the end of the day, is it product managers, people who understand how users use the tool, building it? Or is it security professionals thinking, oh, well, this is what I think would be a good idea, and obviously taking feedback in? And the third point is, I always think about the end users at the end of the day. When you look at the market of all of these tools out there, Where do I begin? Like, I mean, where do I begin? Is there a guide? Is there a guide on where you start? Like, of all of these different scanners, you talk to your vendor, you talk to your vendor, they tell you where to begin and they take you for dinner and it's the, it's the one, you know, but, but 10 lists on BuzzFeed, you know.

**Izar Tarandach (20:11)**: Yeah. It's interesting cuz the, the first point, the scale, will it scale for me to everybody that I hear somebody say, "But will it scale?" It sounds like that guy on YouTube that had the thing with the mixer. What was his catchphrase? "Will it—" I don't remember, but you guys know who I mean. And people just scream, "Will it scale?" You know what? It might not be the tool for you. Just start using it. Check it out. See if it gives you what you want. It's probably— rather than learning how to use the tool, It's probably part of a process of how you're going to put that capability into your pipeline. So fine, it's great. I think that the, the forward-looking question, which is will it scale, is awesome. But if you are looking at a tool that nobody's selling to you as it scales, then you're— Blend. Blend, yeah. You're probably in a place in your journey where You should be more concerned with does this do what I need it to do and how is this going to get into my process. Once you figure those two, then you probably can start looking around and talking to your vendor and going for dinner. Yeah.

**Chris Romeo (21:25)**: I want to kind of bring us back around towards a couple things Mark had in this article because I do want to get y'all's impression of this. I think the first one, if anybody wants to argue on this one, I would love to hear it because security teams want less and not more tools. Tools was one of Mark's first. Okay, you want less? You want less or more tools? You want more? More work?

**Izar Tarandach (21:45)**: Okay, I'm gonna do the cop-out of the century. I want the right tools. Yeah, difficult. What do I mean by the right tools? Coming from a point where 20 minutes ago I said that we can't measure this stuff and all that kind of stuff. I'm coming from the side of Anybody who has heard me rant about stuff before knows that I like the idea that shift left shouldn't even be a term. And Matt has explained that to me a lot of times. Shift left is something that shouldn't exist. You should begin left, stay left. Left is where you live. Go left. Because I think that it's safe to say that today we live in a situation where we have long-hanging fruit and all the rest of security. And most people have not yet dealt with the low-hanging fruit. So the end of that rant is that I want the right tool that I can give to a developer and they will take the low-hanging fruit, and it can give it to a security team and they will get the higher end of the stuff. I love those tools that have, uh, the little button on the corner advanced that opens that crazy dialogue with 1,000 more options and anything. And the not so advanced is, do you want it in dark mode or not? Of course, of course.

**Tanya Ward (23:10)**: It seems that you're looking for Nirvana. Like, can you really have a tool that, that kind of does all of those things?

**Chris Romeo (23:17)**: Yes, that's an announcement of Izar's new startup. It's called Nirvana. It'll do all things for all people in all places having to do with security.

**Matt Coles (23:26)**: You can't afford it. It's priceless.

**Tanya Ward (23:31)**: And the catchphrase is security really is a silver bullet.

**Chris Romeo (23:34)**: There you go, I like it. We're making t-shirts. I got the t-shirt order right now.

**Izar Tarandach (23:39)**: The catchphrase is you've got a lot of problems, will it blend? Oh, that's good, that's good. I like it.

**Chris Romeo (23:46)**: That's good. All right, so Matt, how about the market right now? How about the market though? Are there— so Mark's other thing he's saying here is there are way more vendors than the size of the market can support. Are there more vendors in the space right now than the market can support?

**Matt Coles (23:59)**: Matt Coles. I think the answer is going to be yes. And, and, well, I think that there always has been because we innovate, right? And there's— there we have— I mean, here in the US at least, we have a capitalist system, so People want to compete. They want to continue adding features that they one-up them. As Izar said, there's this arms race, so to speak. I think that there are so many options. Finding, not only understanding what you need or want, But finding all of the— figuring out which ones of the options that you have do what you want or need is a challenge, right? I mean, Tanya raised it, you know, is there a list somewhere? Well, sort of. And so you either go with what you know, or you're asking your peers. And actually, that's a— maybe that's a good question. Maybe I'll— maybe that's something, I don't know, maybe Tanya might be able to help us with. Is there a network? Is it the network that allows you to make the right tool choices from the plethora of vendors that are out there?

**Tanya Ward (25:18)**: Oh, I don't know. I mean, I'm just thinking, yeah, no, I, you would think there is. I think it's more how do all of these things converge together and when you have all these different tools, how disruptive can it be to jump from one to the other for a company? I also think about performance, developer performance. We ask them to go here, we ask them to go there. We go here, there, everywhere. And so each of those is a learning curve. There's a disruption the more we ask them to jump to different places. So, Isar, I can tell you've got thoughts here. I want to hear them.

**Izar Tarandach (25:57)**: I want advice! Sorry, okay, the three of us are in a Slack where a couple of times we have already seen somebody pop up and say, hey, does anybody know these two? Does anybody, has anybody worked with that one? So yeah, there is a network in there. We do talk about this stuff, and the way that we do, others I'm sure also do, right? But I, I think that's two very clear signs for me that the market is totally saturated. My byline on LinkedIn is, I do not have purchasing power, okay? And I do not answer any phone call from a number that I don't recognize that comes from Boston, Maryland, or a couple of other places, right? People, stop pushing this stuff down their throats. Stop with the, the, the 1,000 invitations for webinars and for this and for that and for anyone. Everybody's got Zoom allergy right now, so one more webinar is not what's going to make me buy your tool. On the other hand, write a nice article that solves the problem that I have, I'm all ears, right? But it's not me.

**Chris Romeo (27:11)**: Like Stranger Things is about to—

**Tanya Ward (27:13)**: Why didn't I— Sorry, sorry, that's me. I've just turned it off. I apologize. I thought we were in a Stranger Things episode.

**Chris Romeo (27:20)**: I was like, oh, that's funny.

**Matt Coles (27:21)**: My phone started ringing at the same time.

**Izar Tarandach (27:23)**: Helicopter outside just because I was talking about the—

**Matt Coles (27:27)**: it was the tool vendor calling.

**Chris Romeo (27:28)**: That's right, it was somebody wanted a demo. Like, can you introduce me to Izar Terendash? Because I'm trying to get him on a demo and he's not coming to my webinar, so I don't know what's happening here.

**Izar Tarandach (27:45)**: Where was I going with that? Yeah. So in short, the market is saturated. The good news of the market being saturated is that then we have some really, really cool solutions out there.

**Matt Coles (27:58)**: Agreed. And by the way, I would say we're saturated, not from, not necessarily just from a company standpoint, but from a commercial offering standpoint. But I mean, how many, How many tools on GitHub, for instance, are there? Yeah. I mean, maybe not for threat modeling just yet, but for static code analysis or for web security test, especially web security testing, right? Or something that works on, you know, cloud, you know, Kubernetes in the cloud, right? And so, I mean, there's probably hundreds, if not thousands of offerings out there. And so, You know what the thing with some of those is?

**Izar Tarandach (28:40)**: Most of them have that awful catchphrase, "single pane of glass." Yes.

**Chris Romeo (28:44)**: Oh, nobody says that anymore. Come on, no one says that. They said that in the '90s.

**Izar Tarandach (28:49)**: My monitor looks like the cockpit of an F-35.

**Tanya Ward (28:53)**: Chris, Chris, Chris, Chris's first company is called Nirvana. He's still stuck in the '90s. That's true.

**Chris Romeo (29:01)**: Do you have a heads-up display on your single pane of glass? That was supposed to be an innovative feature that was going to be released later. All right, let's, let's react to something else that Mark said here as one of his 4 conditions. Um, we can skip over the venture market, uh, you know, about startups are going to run out of cash in 2023. He's making a statement there. 2023 will, will tell that story. But what about— so he says that the global economic climate means security budgets are being frozen, and in many cases declining. Yes. So across the board, are we seeing, you know, is that, is that what we're seeing when we look across the industries? I guess these are, you mentioned kind of layoffs and stuff like we're seeing, we're seeing people pull back. We're seeing people reserve cash, which means there, there is, you know, you're not gonna have as much money in your security budget just cause you're not gonna have as much money in any of your budgets because we're pulling back and we're, we're trying to, to project past 2023. Is that consistent with what everybody is, is seeing across the industry?

**Izar Tarandach (30:00)**: I am starting to see it. Yes. I'm seeing hiring getting frozen, of course, but the signs are there.

**Tanya Ward (30:09)**: So I'm, well, I'm curious to understand what's been the trigger for that. Do you think that during COVID a lot of the remote work, you know, working from home, there was an increase in, like, was there like a major increase in purchasing tools and now we're seeing the crash of it?

**Izar Tarandach (30:26)**: Look, I, I don't think that any of us is Ray Dalio or something that can explain exactly to you what all the factors are, but I think that it's fair to say that there is a, a general tightening of everything. Yeah. Unfortunately, I don't think that security has been hit as hard as other parts of the tech industry. Yeah. Although as those taper down exactly because some of the factors that you mentioned, the need for security personnel is probably going to go down as well. All of a sudden, you don't need 20 people to monitor the single pane of glass of 5,000 engineers working, and you're going to need only 10.

**Matt Coles (31:04)**: Well, so let me ask a quick question here, because— and I'll admit that I maybe didn't dive into Mark's article as well as I, as I should have— is there a difference between security tools to enable the enterprise to be secure, you know, zero trust, network defense. Oh, Matt swore.

**Chris Romeo (31:25)**: He swore right in the middle of the podcast. He said zero trust.

**Matt Coles (31:28)**: Yeah. Versus development tools that help companies make sure that the things that they build are secure. And I think that they are different. They're different classes of tools, certainly. They're managed differently. They're operated differently. They do different things. Operate, yeah. And does that change the discussion around purchasing and budgets?

**Chris Romeo (31:53)**: I think to Tanya's point, right, you're seeing a bit of a COVID hangover as some of those VPNs get decommissioned at the end of the day, right? You remember in those first 3, you know, months of COVID like the VPN providers 10x'd their product sales or whatever, because everybody's like, we gotta have remote, we gotta remote, we gotta work from home, we gotta have this. Get this thing going. And so, yeah, I think that's a great, that's a great observation though, Matt, that there's probably some of that, there's some pullback gonna happen as a result of people going back to the office and not needing the giant infrastructures that have been created as a separator there. But yeah, the development tools, that's a good way to think about this. Is it really where all of us play on a daily basis that's gonna be hit, or is it other parts of security?

**Tanya Ward (32:40)**: Well, I think, I mean, think about, I mean, ESPOM, for example, right? The generation. I mean, there's tools that are having to come out to be able to support that, validating it. So I think there's still going to be a need for tools to solve problems, such as, you know, the generation of these artifacts from the bill of material, making sure the machine readable.

**Matt Coles (33:02)**: Can I just jump in here for just a sec though before— and, I know you're waiting for a point, but, but It's interesting to note that the tools, the class of tools that you're talking about are not security tools. They're data manipulation, data management tools, and they're not security tools. And so they have other functions or other uses as well, right? There's a lot of tools that work on XML and JSON data that are not specifically looking at security. The fact that SBOM exists as a security thing is interesting, but it's structured data.

**Izar Tarandach (33:37)**: So yeah. And in conjunction with that, let's not forget this is not exactly a case of, uh, plug and play, or that you can just reach into a switch cabinet and just start creeping cables. It takes a long time to take some of these, these tools offline, right? And you're going to have to keep paying while that, that's out, and you're going to need people to make the switch. So I would say that 2023 may be the perhaps the year that we won't see renewals, but probably not the year that we're going to see companies going down saying nobody's paying me this year. That should be probably 2024.

**Chris Romeo (34:16)**: Yeah, man, I'm gonna institute us— I'm gonna institute a swear word list here on the security table. Where's the jar? Zero trust is one, and I'm thinking I need to put SBOM on the list I think that we need a, uh, uh, those are bingo for next episode. We definitely do. Tanya, you were gonna say something.

**Tanya Ward (34:32)**: No, I was just gonna say, so going back to your point, Isara, is it because you feel like enterprises, developers, whoever it may be, have no longer see the value in these tools? Like, have they— like, no, I'm asking, it's a question.

**Izar Tarandach (34:51)**: No, no, no, no, no. I, I think that it's a matter of, yes, there is value, but if we put everything that we've been talking about so far together, the different interfaces, the lack of interoperability, the how hard it is to put in and put out, how many people do you need to manage something like that? People are saying, are going to start looking at those tools and, and say, is this really giving me the, the ROI that I need?

**Tanya Ward (35:17)**: The ROI. Yeah. Right.

**Chris Romeo (35:19)**: Ooh, there's a, there's an important term. ROI. Yep.

**Izar Tarandach (35:22)**: Did one of us not think of that? Okay.

**Matt Coles (35:23)**: I was gonna say that.

**Tanya Ward (35:24)**: Is that, is that the same as value? Is that the same as value?

**Izar Tarandach (35:28)**: Single pane of glass.

**Tanya Ward (35:33)**: Return on investment. I think it's about, you know, you wanna make sure cost versus benefits, right?

**Izar Tarandach (35:39)**: So next Halloween, I'm coming out with like this huge back of my face. As a single pane of glass. Yeah.

**Chris Romeo (35:48)**: It'll be available on Etsy, ladies and gentlemen. You can buy it and ease our face on there. Well, let's react to a few more of— so Mark kind of had some conclusions in this article too, and I'm curious to get thoughts on, you know, what, how we react to what Mark is kind of predicting is gonna happen in 2023. Let's see. Well, here's one. Unless you're a top 5 priority, you'll either not get a look or you'll be deferred to a later indeterminate date. Does that seem like—

**Matt Coles (36:20)**: Can you explain that one? Can you explain that one? What do you mean by top 5 candidate?

**Chris Romeo (36:25)**: Well, I think this is, this is his language, so I'm putting words into the article, but, you know, top 5 priority being like, hey, the security team has 5 things that we decided in 2022 that were things we have to wrestle with and solve in 2023. So if you're not on that list, if you're not, you know, when you think about the enterprise, right, the average enterprise has a list of how many? 50? Does the CISO have a list of 50 things?

**Izar Tarandach (36:50)**: Maybe more, maybe less. If any team out there is managing to get to the 6th of their top 5, I want to know who they are because they probably got like amazing people and, um, a great amount of money, but I don't think that anybody manages to go beyond their 5 anyway every year.

**Matt Coles (37:11)**: So it's interesting, I'll jump in here, I think that, you know, to talk about, to say if you're not on the list of the top 5, you're not going to get a look, that suggests that the tools have to be, have to have those capabilities or have to have the, that has to have that offering. And so it goes back to the one tool that can do a lot of things so that can be, it can be on the list. Okay. As opposed to a company that does one thing really well, you know, let's take static analysis and, and I don't know if that's gonna be on the top 5 list for a lot of people, right? But if somebody does static analysis really well, are they gonna want to dilute that by doing something else just to get onto the top 5 priority list?

**Izar Tarandach (37:56)**: Yeah. On the other hand, since we have been surfing like the crest of this wave for a long time and now things are going to start, as predicted, going down, there is a lot of companies with a lot of cash at hand. We could end up seeing some consolidation happening here with bigger companies getting smaller ones and incorporating that into their products. We could see some products that are already by themselves big, start offering smaller versions of other functionality as a side effect.

**Chris Romeo (38:31)**: Now, when I think of every time I've seen that happen in our industry though, uh, never ends well. I can't, I can't think of anybody that I'm like, that is an example of how you put together a bunch of small tools into a single pane of glass.

**Matt Coles (38:46)**: I know, I know one.

**Chris Romeo (38:47)**: I can, I could offer one, but, uh, I don't know if we're not gonna name names. This is a, this is a no-name name. Podcast.

**Matt Coles (38:52)**: So there is, I know, I know of one that I, if I had to put on the list, there would definitely be one that would be, but, but, but you're right. It doesn't happen either. It doesn't happen well or it doesn't end well.

**Chris Romeo (39:05)**: Oh yeah. True. Yeah. I mean, some products have gone, you know, I can remember some that were acquired in the, in, you know, 5, 7 years ago. And then it's like, where is that? Oh, it was bought by this company. Oh, well, where's the product? Uh, it doesn't exist anymore. Oh, so you bought it and then just killed it.

**Matt Coles (39:21)**: And all the engineers that were working on it left for greener pastures.

**Tanya Ward (39:26)**: Do you know, Chris, you actually bring up a really interesting point. I mean, this goes back to basic economics, doesn't it, right? When the market starts drying up, right, competitors start— don't they start, you know, kind of merging and acquiring different, you know, products in order to kind of shorten the market down? Yeah, no, I think that's normal.

**Izar Tarandach (39:47)**: You circle the wagons.

**Tanya Ward (39:50)**: Yeah.

**Chris Romeo (39:51)**: Yeah. And yeah, I mean, I think it's, I think it's a good point. Like there's, there, we're probably gonna see it happen, whether it's gonna create super companies or not of 10, 10 different tools brought together. And, and you're like, wow, that integration of these two, of these 3 or 5 tools together created a really super value proposition. I'm not gonna hold my breath. Nope. Maybe other people out there will, but I've seen it happen 3 or 4 times in our industry and I've just watched it go. I got another thing that's jumping out at me from Mark's analysis here. And I want us to wrestle with this one. He says, "We will see good enough baked into developer tools and cloud platforms like GitHub and AWS. Good enough will become common, replace existing tools, and be the preferred new tool." So, imagine your provider like a GitHub is— and they've already done it with some stuff. They've already brought some tools into their fold. Um, so what he's suggesting is you'll start to see some of the people you're already doing business with provide you with some side kind of products that may not be as good as the thing that's the top of the market, but it'll get you 75% of the way. What's your reaction to that?

**Izar Tarandach (40:57)**: Look at the example that he used, GitHub. Okay. Yeah. I, I, I think that it's important to, to latch into that because it goes back to what we were saying before at the start. Yeah. It's a start, but it may not be big on capabilities, but it is big on integration. So GitHub today already offers you a pipeline that's populated. You can have a bit of a little bit of this, a little bit of that, and you get results. So it's not like you, you're, you're blind in the middle of the, the storm, right? You are doing something. You are catching again that low-hanging fruit, which sometimes is all that people can deal with. But the important thing here is that, that good enough comes from a place that you are already paying for something, so you already have all that process going on. You already trust the relationships that you have with these people, some amazing sales engineers out there giving awesome support, and it's something that's already the, the flavor that you know. You don't have to change your environments to actually use that, or God forbid, train people.

**Chris Romeo (42:00)**: I think Matt's about to disagree with you here by that look on his face.

**Matt Coles (42:03)**: Yeah, I'm having a hard time with this whole thread of conversation because what basically you're saying is the lowest secured, the lowest commercial vendor produces something that's good enough. Where does that leave the open source community? First off. And second, let's talk about vendor lock-in again. So if you get used to using, you build your entire environment, your entire processes and your personnel training and et cetera around a set of stuff that's good enough because they're there, you know, you're not getting out of that anytime soon, I imagine. So I think there's two challenges there that's not favorable. True.

**Izar Tarandach (42:48)**: And let's throw a bit more in the mix that may play one side or the other. Look at most enterprises' security teams when they are part of the, the, of the process of bringing in new stuff. So the kind of vetting that they have to do on that, yeah, they much prefer to go to someone that they are actually paying, which means that somebody can get sued at some point if needed, than going to the open source and having to read all that code and running the risk of we missed something and we are bringing this sticking Bombit here, right? So there is that, uh, um, I think, uh, um, easiest path that things at some point tend to go through when you have 50 things to review and you know this guy, you worked with him before, you paid them before, you just go more of the same. Yay. Right. Yeah.

**Chris Romeo (43:41)**: But is this issue really open? Like it isn't really, this isn't really an open source issue though, right?

**Izar Tarandach (43:45)**: No, it's not. But open source will give you a bit more difficulty into vetting that thing. Unless you want to say, well, what's the whole thing with the lots of eyes, these shallow holes and whatnot? Many eyes make, no, light work. Many hands make light work, I don't know what it was. Yeah, but the problem is that that's also an illusion because let's be honest, not every security people out there are reading line by line of whatever comes from GitHub and saying, Yes, we vetted that and we are 100% sure of what came in. And next week we're going to do the same thing for whatever update comes in.

**Chris Romeo (44:24)**: But look at Dependabot. Let's use Dependabot as an example. I don't think— I guess they might— maybe GitHub charges something extra for it or whatever. I don't know if they do. Maybe they don't now. I think they might have done in the beginning. But if you look at Dependabot, Dependabot kind of revolutionized a problem in our space. I would argue that Dependabot was ahead of a lot of the commercial vendors for a time as a value-add. And so in that case, I would say that's not even like a cheaper version that's bundled in but isn't best in class. I don't know, I haven't looked at Dependabot in a while to know if it's still hanging with the top software composition analysis tools, but it was for a time.

**Izar Tarandach (45:08)**: A couple of months ago, I— Did participate in a comparison of, and it did pretty well. Was it the best? I don't know. We, again, we don't know how to measure our success, but, uh, yes, it was good enough. You see, we, we ended up getting there in the good enough. Yeah. Because it plugs a hole that I need. It costs me next to nothing. It's already in the tools that I use.

**Chris Romeo (45:34)**: So yeah, why not? Isar is arguing for good enough. I know Matt is not loving it.

**Izar Tarandach (45:39)**: I'm not arguing for good enough. I'm saying that by default we grow into good enough.

**Tanya Ward (45:46)**: I think good enough is exactly what it is. It's good enough. I like what you had brought up about the integration. I think it always comes back to the problems again that people are trying to solve within the businesses as well. And integration, do you favor integration over additional capabilities? There's multiple products that we use on a day-to-day basis that aren't even security in nature that used to not be integrated. They're now integrated together. We see a value-add because things are just seamlessly moving across. Productivity goes up. When things are kind of isolated and it's excellent, it fulfills that niche or that's one small thing, you still— it goes back to my point. You know, I have different tools and you're spending more time and productivity because you're jumping across. So I think it comes down to a balance between the two things.

**Matt Coles (46:36)**: And by the way, I was not arguing against, I was arguing that how do we avoid getting stuck in good enough? Right.

**Chris Romeo (46:44)**: I'm just, I'm trying to create some controversy here, Matt.

**Matt Coles (46:46)**: Oh yeah, you're doing a good job. That's all I'm doing.

**Chris Romeo (46:48)**: Stir the pot.

**Izar Tarandach (46:49)**: That's what I'm doing. That's a great point. That's a great point because again, if we can agree on that, again, sorry, 80/20, the vision of the low-hanging fruit and the advanced stuff. The people who are not going to get stuck are those for whom the good enough covers the, the low-hanging fruit well enough that they find themselves with time and budget to go after the more complicated stuff, right? So some people are going to look at that as a learning curve and actually walk the curve and then improve their tools, just the way that we talked about the tools with open, uh, formats there in the beginning. and all the people are going to say, hey, you know what, it's the illusion of action. I am doing something. My CISO, my CSO, they can't scream that I'm not doing anything because here, here it is, I'm doing something, it's good enough.

**Tanya Ward (47:40)**: So if, if, so going back to the two points, it all still comes down to ROI, like the cost versus like what are you getting out. Like you have to still show that when you have a tool that you're, you're solving that problem, like you have to be able to measure it as well. Yeah.

**Chris Romeo (47:59)**: But if I'm not paying for it, if I'm getting it through my another platform provider, what is the ROI?

**Tanya Ward (48:04)**: No, no, no. I'm not, but Chris, nothing is for free, right? People, you, you deploy it, you still have, you know, you have to train on it. You know, there's still a hidden cost to things. You maybe write some scripts to support it. Yeah.

**Izar Tarandach (48:16)**: You know what's the, the image that just came through to my mind? Somebody working with an open source tool, let's say a detector of some sort, and it doesn't detect something and there is an incident and horrible things happen. And I want to see that people come to GitHub and say, I would like to talk to your manager about your tool. You know, it's, it's at the end of the day, at this, again, that word scale and at the value of things that most of us work with. It's very hard to, to trust. No, I'm sorry, not to trust is not the word. It's very hard to justify trust in open source just because of what could happen on the bad side, on the probability that something really bad happens. Especially now that we are starting to see this whole liability thing happening with CISOs and hey, you are culpable for stuff that happens and whatnot. That's going to make, I think, the case for open source tools that would sit in very, uh, not delicate but very, uh, um, central decision points and would be responsible for a lot of security. It would make the enterprise case for them even harder.

**Chris Romeo (49:38)**: I noticed that Izar keeps opening Pandora's box after Pandora's box. Let me just quickly say something about, you know, the general general trust of open source. And then let me go to the next thing. It's like, you know, that's, that's a, that's a 10-hour debate right there.

**Tanya Ward (49:52)**: At the end of the day, we're all security professionals. So if we're going to have security tools coming into our environments, don't we want to make sure that they follow good security practices and also vulnerability response practices? Like, if you check for Log4j and Log4j happened to be in a security tool and they don't patch it for 12 months, Do you think that's a problem?

**Izar Tarandach (50:12)**: Wait, wait, wait, wait, wait, wait, you're telling me that security tools cannot be secure?

**Tanya Ward (50:17)**: No, I'm saying we're security professionals that— okay, okay. No, I'm not.

**Izar Tarandach (50:25)**: I'm sorry. I mean, I just—

**Chris Romeo (50:29)**: Ezzar just is, he's the pain in the ass.

**Tanya Ward (50:31)**: He's on fire today.

**Chris Romeo (50:33)**: In the group, he's just poking every bear. Is that what you do when you go to the zoo? Do you just poke the wild animals, the killers, and just be like, hey, look what happens when you poke the tiger.

**Izar Tarandach (50:44)**: Only the slow ones.

**Chris Romeo (50:45)**: Only the slow ones. All right. I want to get your, I want to get, I want to get, here's our final issue I'm taking out of Kurphy's report here. I want to get your take on this, cause I think this one could be fun. He says, in order to stand out from the crowd, marketing teams will be making increasingly salacious claims. I'm going to have to look that word up or someone can tell me what it is. No, you'll see an ROI of 500%. That one's real today. Zero risk is a reality. Our AI-powered system always has your back when you are asleep. Struggling or worried vendors will all be shouting about why their area of focus is the most important to secure and why their product is critical. Is this any different than what we've experienced for the last 5 years? No, it sounds louder.

**Matt Coles (51:24)**: Maybe they're going to get bigger claims. You know, it's going to—

**Izar Tarandach (51:28)**: Yeah. Their budget to go down as well so they can't play more.

**Chris Romeo (51:31)**: I mean, if I hear one more time that our solution's powered by AI and ML. Oh my, that's insane. I'm like, what does that actually mean? Does that mean? Nobody ever knows. Nobody can ever, they, it sounds so good though. Like I wanna say like the Security Table Podcast is powered by AI and ML. Yeah. What does it mean?

**Izar Tarandach (51:53)**: You know what's the thing with AI and ML? Once upon a time, the magic dust of security was crypto, as in cryptography, not in funny money. And some very smart people could actually explain that to us. Nowadays, the magic dust of security is ML, AI, and nobody can explain that to anyone. Because you know what we need?

**Chris Romeo (52:15)**: We need an AI that we build that can explain it to us. Right stuff. There's got to be, there's got to be.

**Matt Coles (52:23)**: You're trying to get it in, aren't you? Don't say no names.

**Chris Romeo (52:27)**: I think we're coming to the end of our first entry in the security table here, and I know I can say it's been a ball of laughs and a lot of fun just kind of going through this. And I know we had 3 issues we were going to talk about. We got through 1 in just a little bit under an hour, but you know what? We had some fun along the way. So Tanya, Matt, Izar, thanks for sitting at the security table here, and we'll do this again with the next episode with our next issue. Thanks for putting it together.

**Tanya Ward (52:55)**: Yeah, take care. Awesome.
