In this episode, we chat with Job Snijders, a principal engineer at Fastly and a developer and art director for the Open BSD project.
Job has over 10 years of experience in the Internet routing and security field, working for various organizations and communities such as RIPE, NTT, and NLNOG. He’ll share his story of how he got interested in Internet routing security and how he created and contributed to various tools and standards to improve the global routing system.
We also talk about his creative side and how he used a visualization tool to draw the Nyan Cat on the RIPE statmon interface by manipulating the visibility of prefixes in the global routing system.
We’ll learn Job’s perspective on innovation, collaboration, and feedback in the network industry, and how he manages his busy schedule and avoids imposter syndrome.
Join us for this fascinating and inspiring conversation with Job Snijders.
Every time I draw away the curtains and unpeel yet another layer of how the Internet really works, I discover another room filled with people just like me who are trying their best with the information available to them at that time.
Thanks for being an imposter - a part of the Imposter Syndrome Network (ISN)!
We'd love it if you connected with us at the links below:
Make it a great day.
Machines made this, mistakes and all...
[00:00:00] Chris: Hello, and welcome to the Imposter Syndrome Network podcast, where everyone belongs, especially if you think you don't. My name is Chris Grundemann, and I'm here with the always amazing Zoe Rose. Hey! Hey, this is the Job Sniders episode, and I think you're going to love it. Job is a principal engineer at Fastly, where he analyzes and architects global networks for future growth.
[00:00:34] Chris: Job has been actively involved in the internet community in both operational, engineering, and architectural capacities, as a frequent presenter at network operator events, such as NANOG, ITNOG, DKNOG, RIPE, NLNOG, and APRICOT, and in a number of community projects for over 15 years. Job is the co chair of the IETF Grow working group, a director of PeeringDB, a director of the Route Server Support Foundation, and a member of the RIPE NCC executive board. As if that's not enough He's also a developer and art director for the OpenBSD project.
[00:01:15] Chris: Hey Job, would you like to introduce yourself a bit further to the Impostor Syndrome Network?
[00:01:20] Job: Hey Chris, hey Zoe, thank you for having me here. Excited to have a conversation, uh, I guess about me. Which is something I'm not super used to because usually I very much focus on technical stories, problems, and solutions.
[00:01:38] Job: But yeah, I'm a Dutch Netherlands based engineer, and I really like internet routing security. And that's where I direct virtually all of my life's energy towards, but I also have some hobbies not to, you know, take away from just being a geek all of the time, like hiking and running. So this is me.
[00:01:59] Chris: Fantastic. Well, we'll dive right in here. I personally consider myself a creative person and I often call myself a creative technologist. But you, Job, as I mentioned just now, are the art director for the OpenBSD project. Can you tell us a little bit about what that means?
[00:02:16] Job: It means, it is my responsibility to ensure that every half year when OpenBSD publishes a new release of the operating system, There's artwork that goes along with it.
[00:02:27] Job: So I, uh, commission artists to create a poster and a banner and, uh, something to, uh, you know, cheer up the celebration of yet another release. So every six months I work with a new artist, you know, put a little bug in there here on what the theme or concept of the art could be, and then let them do their thing and I pay them for it because I am not of the type, you know, this is great for your resume.
[00:02:52] Job: Um, so the trick here is, uh, uh, we, we print and sell t shirts and the profits of selling merch are used towards funding the next artist.
[00:03:04] Zoe: Oh, nice. So you don't pay them in exposure. That's good. Yeah. I, I, there's one part I love about security is the constant. Development of stickers and, uh, all of those fun things.
[00:03:20] Zoe: No, that's really interesting. I have a question that is not specifically job related, but more, how in the bloody hell do you measure, manage such a seemingly busy schedule? I mean, you've got like all of the directors, all of the chairs, and, and you have a job, and you do this. How do you manage that?
[00:03:39] Job: Am I managing it?
[00:03:40] Job: I don't know if I'm doing the best job possible, but I think the trick is to my, my employer is super supportive of the community work I do. So some of the time budget definitely comes from that space. And I think another good ingredients is to build small teams that are efficient in working together in a way that the activity doesn't cost everybody more time than is needed.
[00:04:05] Job: So for example, for the, the route server support foundation. This is a foundation I set up, I think, three years ago to, uh, basically convert money into the production of open source software. And the developers we hire in this foundation are very autonomous. So they, I don't need to do daily handholding to, to check in if they're being productive and are on the right track, but it's a very hands off approach.
[00:04:32] Job: My fellow board members each have excellent skills that save me a lot of time. So the treasurer is actually a financial professional. So, yeah, I think, uh, it's, it's, uh, in part to be surrounded with people that, that work well together and in part, all of these activities very much aligned with my, my desire for innovation and internet routing security.
[00:04:58] Job: So it energizes me to some degree to see progress in the individual projects that I contribute to.
[00:05:05] Chris: Yeah. So, I mean, I would assume anyway that the, the route server support foundation, since that was kind of your creation, you've been able to kind of build that team around you of great folks who can support the organization and not take up all of your time.
[00:05:18] Chris: I wonder in a lot of the other cases, you probably had less control of building those teams. Do you? Do you seek out projects that already have great teams? Do you get involved and then, and then build a great team around you? Or have you just been lucky in some cases to have that kind of support structure so that you can do so many things based on, you know, sharing the load across some other great individuals like that?
[00:05:38] Job: Yeah. I, I'm not sure there, there is a conscious strategy because especially for the elected positions, you're not in charge about who else is elected. I mean, the closest thing you can do is to, to encourage people you admire, to make themselves available as candidates. But I guess so far, yeah, luck definitely is an ingredient.
[00:06:01] Job: And for instance, uh, chairing an ITF working group, uh, for me doesn't feel like an an a burden on my time because I, I am as an, an author of various internet draft proposals, quite familiar with, with the process and how things should work. So I can very quickly move based on, on that experience and having been part of that community.
[00:06:25] Job: So yeah, it's, you gotta be conscious about the activities that you pick and make sure that they are well within what you like doing and have some experience with.
[00:06:38] Zoe: Yeah, well, I recently saw, I can't remember which social media platform it was, but it was a picture of learning to say no. And it had like a bunch of different tasks and it said.
[00:06:48] Zoe: Uh, yes, yes, no, no, yes, yes. And then you could see the yeses were 100%, the nos were obviously zero. And then it had the other side as if you say yes to everything and it's like percentages of progress and nothing is complete. And so I think that just goes really to what you're saying is. You can say yes to everything, but if it's not something you're necessarily that good at or enjoy, it's going to take longer and actually impact the rest of your areas as well.
[00:07:17] Zoe: So I think that's, that's one thing that I've had to learn. And I'm still learning that saying no is good too.
[00:07:24] Job: Yeah, definitely. I have to say no to things as well. I know that at this point in time with the obligations I have to various institutions, there, there is no additional time to take on more volunteer roles.
[00:07:37] Job: So something somewhere needs to free up before I can change positions. Definitely. And that, that means saying no.
[00:07:44] Zoe: Yeah, definitely. When we prepare for these talks, we get like some quotes or LinkedIn history and that, and put it in a chart. And one quote that we had was internet routing system hacker. What does that mean to you?
[00:08:00] Zoe: And, uh, and. What would be your definition of a hacker as well? I mean, obviously I work in security, but I always like to hear everybody's definitions.
[00:08:09] Job: I'm a hacker in the sense that the OpenBSD project uses. I like to tinker with things, look for their weak points, their strong points, understand the system, and then come up with, uh, innovations to, to further strengthen the system.
[00:08:25] Job: And in a sort of playful way, I envision that there is this game between on the one hand, uh, myself as a player. And on the other hand, unknown entities that maybe know more than I do or are looking at things differently and to try and be ahead of everybody and building defenses, uh, to help protect the global internet routing system.
[00:08:48] Job: So a lot of the time I look at programs from the perspective of how can I break it, what error conditions could exist in the future that, that negatively impacts just components in the whole machine, and then try to fix that before someone else finds the same issue and exploits it.
[00:09:08] Chris: Another form of hacking, other than kind of taking things apart, putting them back together and looking for vulnerabilities.
[00:09:15] Chris: As you mentioned is potentially just really playful or doing things that are, you know, use a tool or a system for something that it wasn't intended to do for, and you've done that as well. I believe one of the most fun things I've seen you do anyway is drawing the Nyan Cat on RIPE's StatMon interface.
[00:09:34] Chris: Folks who don't know or aren't following internet routing may not know what any of those words mean other than maybe Nyan Cat. Um, so, so maybe you could tell us a little bit about, about what that, what that project looked like and, and, and what it was and, and maybe even why you did it.
[00:09:49] Job: Sure. Yeah. Yeah. So some time ago I stumbled upon a very interesting visualization tool that RIPE NCC had developed and you could input a prefix. And then it would show a graph where it showed the stability of the more specific prefixes over time. And it, it used a color gradient, uh, where red meant the prefix was not visible and green means the prefix is visible on many vantage points. And I don't know exactly how or why, but at some point it popped into my head that this graph, which was intended as a debugging tool for operators to offer a certain perspective, the stability of prefixes in the global routing system also be manipulated to some degree and used to draw to purposefully draw an image. In this tool.
[00:10:46] Job: So I happen to have access to a giant IP four block, my, my previous, uh, employer NTT made, made a slash 16 available to me and this tool worked on, on the, the slash 24 boundary. So this effectively gave me 256. rows, uh, and by flapping the individual slash 24, I could create pixels. So I did a small test run and, uh, printed the phrase NTT and a heart, uh, in the system.
[00:11:21] Job: And when that succeeded, I was like, all right, I gotta go big. What is something that can, what is like an endless image and the Nyan Cats, that moving GIF, where it just keeps moving forward in a wavy fashion. Uh, felt very appropriate for this type of visual outputs display. So I put a small Python script to work and it took like months during which I kept absolutely silence about the effort.
[00:11:49] Job: And at some point I was like, okay, I think it's, it's ready to, to share with the world. And I think the world, uh, thought it was very humorous to see that, that very serious debugging tool used in such a way.
[00:12:02] Chris: I love it. Really fun. And definitely great. I think a lot of folks got a kick out of it. And I think it also served to potentially promote.
[00:12:11] Chris: This tool that ripe had out there. I think, uh, I think there's definitely at least dozens of people who maybe didn't even know that this interface was available until people started passing around, uh, you know, links and images to the, to the cat. So I think it, uh, it worked out great on many fronts. And it also kind of reminds me of, you know, obviously in a more serious fashion, the time you've put into a bunch of other tools, right?
[00:12:35] Chris: So IRRD, the OpenBSD RPKI validator, there's some RPKI clients and debuggers, the BGP Q4 kind of policy builder, OpenBGP D. I think I can probably go on and on and on with just a bunch of words that won't make any sense to anyone who hasn't used these tools, they can find them all, but, you know, back to Zoe's point about, you know, not only do you have all these kind of director level positions that you're, that you're doing in addition to your full time job, but there's a lot of software you've developed over, over the, over time, which often I'm sure takes months or years of, of work to, to get it right and, and move it forward.
[00:13:12] Chris: So I don't want to ask the same question again. I'm like, how do you do all this? But but I I do see a theme here. We we talked to Dean nelson a while back on the show and then he talked about his ability to do so many things was in large part because All of the things were focused kind of on, on one piece, right?
[00:13:30] Chris: And then he kind of had his, his kind of personal mission statement, I guess you would say, around what he was working on. And although he had a bunch of projects, they all kind of, you know, locked together and, and move forward. And it seems like you've done the same thing, right? I mean, everything kind of revolves around internet routing, internet routing security and pushing those things forward.
[00:13:48] Chris: I mean, and does that resonate with you? Is that, is that kind of part of how this works and being able to get so much done? Is that it's focused in a way?
[00:13:55] Job: Yeah, absolutely. Earlier in my career, I was a network operator, the ISP I worked for used Cisco and Juniper gear and, and you, you had to work with the tools that you had as my understanding of the internet routing system grew, I started to recognize that there were.
[00:14:12] Job: These giant holes in the system ripe for abuse and abuse in this context isn't even evil adversaries trying to ruin somebody's day, but also people that just make a typo. And if you look at your keyboard that, you know, the numbers are super close to each other. So it's very easy to punch in a two when you intended to press the three button.
[00:14:38] Job: Um, and then. Suddenly someone on the other side of the planet is like, holy crap. My prefix is being hijacked and it's causing me downtime. And what is going on? So looking at the tools that the internet community had available to improve the quality of the data that is passed around in this system, uh, I started to recognize that there were gaps that needed addressing and that perhaps nobody else other than me was going to do it.
[00:15:07] Job: I think to some degree, there is, uh, always this, this bystander effect that people acknowledge. Yes. This, this is not optimal or, or even worse, this thing is broken and that's it. So, um, I didn't want things to stop there. I felt that there was ample opportunity to move the status quo in a better direction and that it, that I was positioned in, in a way to, to make that change, that there were people around me or, or employers that were supportive of such initiatives so that.
[00:15:42] Job: If there's time and budget and willingness, then yeah, let's, let's make it happen. So in the case of IRRD, the original IRRD was developed by merits associated with the university of Michigan, if I'm not mistaken. And it was this program written in C and organically developed over the course of, of multiple decades.
[00:16:03] Job: And I realized at some point that that type of database would benefit from a cleanup mechanism based on RPKI. And RPKI in this context is a cryptographically verifiable, higher source of truth compared to what the IRRD daemon contains. And I recognize that, that, that extending the C version of IRRD to integrate in a certain way with the RPKI was, was not feasible.
[00:16:34] Job: Every time I changed something in it. Program in some other random part of the program, things started falling apart. And at some point I concluded like, I lack the oversight of this program and we need to start from scratch. Uh, this is the only path forward. We, this program is a dead ends. So I went to my employer with a pitch and I said, Hey.
[00:16:58] Job: I know this developer and she is very experienced with this type of database and she would do a great job if we allow her to rewrite it. And this, this was years ago. And I also recognize that the time in this sense is not so relevant because as long as things happen at some point in time, then The decades after that, we'll enjoy the benefits of that program.
[00:17:25] Job: So yes, it would have been cool to have it yesterday. The next best day to plant the tree is tomorrow, right? So, so in the case of the IRRD program, because the development itself was outsourced to, to a super knowledgeable person required from my side to focus more on, on testing and architectural questions.
[00:17:47] Job: And significantly reduced the workload for me. And now we've come to the point where merits themselves in October deploy IRRD version four, the full rewrites. So we've come full circle that, that the original organization. Has now latched onto to the new version, um, that was developed outside that organization and IRRDB be being the world's biggest database of IRR and now receiving RPKI based filtering is a massive step forward in the global routing security posture because by them deploying the software, everybody benefits.
[00:18:28] Job: And at that point, I don't care that it took five years, it, it happens. So goal achieved.
[00:18:36] Zoe: Well, two things stood out to me there is years ago, I took an entrepreneur class because I had to, and I at the time was like, this is silly, but actually a lot of things I learned to there were really beneficial. But the, the first thing that you mentioned is, you know, you were inspired by your normal working day and the challenges you found, the limitations you found, because I think a lot of people.
[00:18:56] Zoe: get so overwhelmed, they're like, I'm not creative, I can't think of things, I don't know what to do. It could be as simple as, well, what issues do you see and what solutions can you propose? Because actually, if you're having that issue, somebody else is as well. And that's, it's like super niche, but even that's still cool.
[00:19:18] Zoe: Um, and then the second thing that, uh, stood out to me is how you started that, you know, you propose to the company to make use of their resources because, you know, our time costs money, money costs money, you know, so there is limitations to what we can do as individuals. But we can also, as you say, you know, make use of those resources that already exist in there to create for the community and lead by example, you clearly been very successful in that approach as well.
[00:19:49] Zoe: So those are really those for me were really two takeaways that were really interesting. I've never taken that approach. I've always been. Oh, I'll do it myself. But I really like the idea, especially now that I'm about to be a mom of two, I really like the idea of making use of existing resources, even if there's grants out there, for example, because some governments do certain grants.
[00:20:10] Zoe: So that's really, I like that point that you made. Another question that came up through listening to you talk about this is, These are quite complex topics and often when we go into things that are quite complex and take a long time to solve, we feel a little bit like, are we doing the right thing? Am I smart enough to do this?
[00:20:30] Zoe: Am I achieving what I want? So has there ever been a time that you actually felt like an imposter and maybe got a little bit demotivated and how are you able to kind of overcome that?
[00:20:40] Job: Yeah, that is a super fun question. I mailed that over before we started this session. It's like, is, is imposter syndrome something I, I have experienced?
[00:20:51] Job: And I think the answer is probably no, because years ago in a moment of, of insecurity, I. I reflected on what is it I do? Can I do the things I want to do? And from time to time, confirming that things are real. So for example, I use my own software. I have my own router at home. I use OpenBGP. I use RPKI validator.
[00:21:20] Job: I use IRRD. I think this almost physical connection to those projects and not. Being in a position where I give guidance over a project where I do not experience all the outcomes myself, I can, I have this feedback loop that, that I am in fact doing something that is meaningful because if where I would feel insecurity is, is when I, when I am the one that has to make a decision, like we, we implement this feature or we don't, or we, we do design a instead of B.
[00:21:53] Job: And then will not be the one that suffers from that decision if it was bad. And as long as I suffer the consequences of my own decisions, I can confirm that I'm on the, on the right track. So really, really eating my own dog foods is, is I think helpful. And I, I also. Consider it sort of exercises. Like I, when I propose something in the IETF, I oftentimes write the implementation myself so that can confirm that I can program the thing I'm proposing.
[00:22:29] Job: So it's not imaginary. It's not a theoretical exercise. I can confirm to myself that it is real and that helps myself to understand whether the proposal has merits, but also helps the community understand that this is not that the proposal is real. So, yeah, I think that those type of activities help, help me stay grounded and to some extent avoid the pitfall of imposter syndrome.
[00:22:59] Chris: Uh, that makes a lot of sense. We had another guest on not too long ago, Rosemary Wang, who talked about something kind of similar, which was that, you know, she wrote a book and we asked her kind of the same feeling, right? I mean, how scary is that? And she had a similar approach, which was, you know, I'm not trying to be an expert in all things or in everything or, or even just general, but I'm talking about my experiences.
[00:23:23] Chris: And so I know that the thing that I did worked for me at that time, and that's all I'm telling you is that the thing I did worked for me at that time. And as long as I stick to that context, then there really isn't too much fear there because it's just a true statement at that point, it's, it's not, um, I'm not kind of overreaching or, or.
[00:23:40] Chris: Making any grandiose claims or anything like that. It sounds like that's been similar for you because you're actually working on the thing you're working on and then suffering the pains if it's wrong and knowing that it's right when it works. Yeah, that makes a lot of sense.
[00:23:52] Job: And another key ingredient is having a good support network.
[00:23:56] Job: So both inside the OpenBSD project and the IETF, there are reviewers that, that will confirm your understanding, that will compare that the natural language you use to describe the change matches up with the code that you've wrote and in the OpenBSD project, you cannot commit code to the code base.
[00:24:17] Job: Unless at least one other person knowledgeable in that area also looks at it and is willing to attach their name to it. And the same goes for IETF. I can go to a working group and pitch an idea, but if, and if the idea gains some, some traction, there will be, there are multiple review points where, where other people take a look at what the proposal exactly entails and whether it is implementable for them.
[00:24:42] Job: And these multiple rounds of review at various stages of proposals also help me stay based because I, I'm not flying solo this, this, all of this as a team efforts and the participation of other people is key to that.
[00:24:58] Zoe: I was wondering, cause you seem to have quite excellent experiences with your community and with your places you've worked and the support, as you just mentioned, the support of your community.
[00:25:12] Zoe: I am curious if you were ever in a situation where either maybe you didn't have as supportive of an environment or maybe there was a person that you didn't work that well with, and that? One, you identified that, because that's one thing in my career I struggle with, is identifying where maybe it's not the right environment for me.
[00:25:31] Zoe: And two, how did you get past that? Did you immediately leave, or did you think, oh, okay, I need to restructure the way I'm working, for example? Then again, you could say, no, I've always had the best people ever. I don't know.
[00:25:44] Job: Yeah, I think, uh, like years ago, more than a decade ago, I worked at a small ISP and the relationship with my manager was challenging at times.
[00:25:56] Job: The manager was somewhere on the spectrum and yeah, this impact of communication and I am skeptical that I am truly normal myself, but yeah, at some point I started to recognize that, that being part of the company cost stress, sleepless nights. Um, I think it was for feedback from friends where they pointed out like, Hey, this, this isn't necessarily normal.
[00:26:23] Job: You know, you have a choice that I realized that it might be good to change my environment and look for other opportunities. And at that point I started my, uh, one person entrepreneurship, uh, as a consultant and even later returned to that company because we, we managed to restore our social connection, but also joined other companies.
[00:26:45] Job: And all in all that, that was super cool because being a consultant meant that my hourly wage was decent enough that I only needed to work two or three days a week for, for money. And that gave me two or three days a week to, to work on volunteer things, because I recognize that I don't need to work full time for a salary, but I can work full time for everything that interests me.
[00:27:13] Job: So for instance, the analog ring debugging project is a direct result of me having that time available back then. And especially in the beginning, it was a lot of work, which I happily put into the project because I was the boss of my own time. And I, you know, I had my, my basics, uh, covered in terms of food and shelter.
[00:27:33] Job: So there was opportunity to, to build things that. No, made no commercial sense whatsoever. Yeah. I think looking back, I I've, I've worked at wonderful companies with wonderful people, and I hope that will stay that way the remainder of my, uh, the career.
[00:27:51] Chris: Awesome. Yeah, we do too. And unfortunately we do not have any extra time right now.
[00:27:56] Chris: We are all out. That's often the case. So we run out of time before we run out of topics. Job. Do you have any, any current projects or causes that you'd like to highlight to the imposter syndrome network before we close out?
[00:28:09] Job: I am spending all my time on improving the RPKI. I think, uh, there are so many routing security incidents happening, uh, that the only path forward is to, to continue investments in a cryptographically supported infrastructure that is the RPKI.
[00:28:28] Job: And, and up until this point, it's to some degree, a little bit underutilized. Uh, the RPKI is this framework on which we can build multiple applications. But so far we've only really used one application, which is routes origin authorization and routes origin validation, but that's just one part of the puzzle.
[00:28:50] Job: So there is also. AS path spoofing and route leaks, uh, and other types of incidents that are not yet fully addressed. So putting time and effort into really making a dent in, in how routing has been going for the last 30 years and seeing that improve over the years. Seeing the likes of merits, uh, embrace RPKI integration, even the IRR level is a very exciting field to be working in.
[00:29:19] Job: And, uh, how do I wrap this up? I think not accepting the status quo is, is how you, you get there. You, if you see a problem and it bothers you. It probably bothers other people. If you can find those other people, you can band together and maybe come to some consensus on what the solution could be longterm or midterm for, for that given problem and environments like the North American network operator group to find like minded people that are maybe suffering from, from the same challenges to use the infrastructure that the IETF gifts us where there is a formal process to submit ideas and a process to follow to receive feedback and review and go through the motions to eventually publish things in a place where other people can find them. Yeah, there's a wealth of resources out there and it's fun to make use of those resources and to encourage other people to use those resources.
[00:30:20] Chris: Absolutely. Yeah, that's great. Job, thank you so much for sharing your story on your journey. Uh, with the imposter syndrome network and thank you to all of our listeners, uh, for your time, your attention and your support. Please do remember that if you found this episode insightful or interesting We'd love for you to consider paying it forward by letting others know about the show and the great guests we have on.
[00:30:43] Chris: We only have a few moments left. We've probably gone a little longer than we normally do Job, but I I would like to know If you could quickly tell us what the most valuable lesson you've learned in your career is.
[00:30:56] Job: Um, all we do was built by other humans that are as fallible as, as we ourselves are. Very early on in my career, I viewed the IETF as this far away entity, and I have no idea how any of that works and how Cisco or Juniper or others decided.
[00:31:19] Job: What features would be part of their products? I was like a passive sufferer of other people's inventions. And then through the years, I learned that, that there is ways to, to fix problems that you, you can, for instance, open a case with the technical assistance center of the vendor and point out that there is a defect and that they fix it.
[00:31:42] Job: That if there is issues in internet standards, that those internet standards are not set in stone, but can be updated and amended over time and that there is a process to do so. And every time I draw away the curtains and unpeel yet another layer of how the internet really works, I discover another room filled with people just like me who are trying their best with the information available to them at that time.
[00:32:09] Job: And, and that on the one hand, that there is no perfect oversight of the internet, that it's, it's not perfect. Can be scary. On the other hand. It is encouraging that it, it was built by people just like me. And that means that, that over time things can become worse, but more importantly, improve and that I can be a part of that.
[00:32:34] Job: So yeah, there is no magic in all of this. Uh, or the magic is the fact that it's a human built system as a human I can contribute to it.
[00:32:46] Chris: Awesome. I love that. That's a great way to end it. Thanks again, and we will be back next week.