Ciena Corporation (CIEN) Earnings Call Transcript & Summary

Good morning, everyone. I hope you had a good start to the day, fully caffeinated. Day 2 here at the Goldman Sachs Communacopia and Technology Conference. I'm still trying to learn the name of it. I'm Rod Hall. I'm the hardware analyst at Goldman Sachs, done networking and telecom infrastructure for a long time. My team and I also do consumer systems, consumer electronic stuff, so we do a broad gamut of things. We're very pleased to have Ciena with us today, one of our favorite infrastructure companies. Got David…

One of the favorites?

Yes, we have a couple. Actually, it turns out you're not the only one there. But David Rothenstein, the SVP and Chief Strategy Officer and General Counsel. So you have a lot of jobs it seems. You must be pretty busy.

Yes, I get to regulate myself.

So welcome. Thanks for coming.

Thanks for having us.

So I guess before I jump into -- I mean, I've got a bunch of specific questions. Maybe it would be good for you to give us and the investors -- because you don't necessarily do these things all that often -- a little bit of an overview of what you do, what do you focus on day-to-day yourself in your role? And then I'll jump into some specific Ciena questions.

Thanks for having us, and nice to see everyone. So I've been at Ciena for 21.5 years through, I guess, 2.5 different financial downturns. And my role has shifted over the course of time. As Rod said, I have been the General Counsel of Ciena since 2008. But I've been taking on additional roles over the course of time. I also am in charge of corporate real estate, sustainability, government relations. I ran human resources for a year, and I'm now the Chief Strategy Officer, in charge, obviously, of looking at our strategic planning processes as well as corporate development with respect to strategic partnerships, M&A, venture capital investments, things like that.

A lot of roles there over the years, too. So the biggest question that we're getting now, and of course, you probably are getting as well. I'm sure you've had this question 10 times in the last week, if not 20, is related to supply. And one of the things that we're trying to puzzle out, I think, is we know that consumer electronics, certain areas are very weak. TVs are very weak. PCs are extremely weak. We hear that although Dell yesterday said that this isn't really happening with them, but we hear that at least some components are becoming more available out in the market somewhere, but they don't seem to be finding their way to companies like Ciena. And so that's kind of a puzzle for us. We're wondering, A, are those components fungible into companies like Ciena or maybe they aren't? And B, if they are fungible, where are they? And is it just a matter of time before they show up and things get better? So I'm just curious kind of what's your take on all the supply situation right here, as we said?

Yes. A lot to unpack there, but let me try to kind of take it piece by piece. So in general, you're right. I mean, supply is the first, second and third question that we and I think all tech manufacturers are getting these days. From our standpoint, and I think it's really important to contextualize it, is that the vast majority of our supply chain are performing in accordance with their commitments, right? And yes, there are lead times that have been extended because of semi shortages and things like that. That's certainly well documented. But the vast majority of them, I'd say 95% of our suppliers are delivering in accordance with their commitments, which is good. And I think echoes what a lot of others have been saying about some degrees of green shoots coming out of the supply chain. What's happened to us in particular over the past few months is we've seen some more acute challenges beyond just semi, and that we've had a handful of suppliers who make very specific integrated circuits that go into our modem technology. And they have effectively not delivered on their commitments that we were even hearing from them a few months ago, both in terms of volume and in terms of timing. And the problem with that for us is because these are what we call them, as you may have heard the term golden screws. These are very low-value parts, $1, $5, $20 parts, largely in terms of power management and control and that gets to part of your question, Rod, in terms of why others are seeing a glut. These are very specific power management and control integrated circuits that go into our high-performing optics systems. And because they gate their incidental to the functionality of our modem technology, but they're essential to their manufacturer, meaning we can't get a modem out without these golden screws. And just to give you a data point to kind of play out that disproportionate impact, of the, say, top 10 biggest problem integrated circuit components that we're seeing today, that would be about $5 million in cost if we could get our hands on them today. That $5 million of cost of those circuits would gate about $1 billion of modem revenue for us, right? So $5 million equates to $1 billion. And that really kind of sums up the nut of the problem, which is why we fell short of our Q3 expectations and why the guide for Q4 was also below our and the Street's expectations. In terms of why perhaps others are seeing, particularly you mentioned consumer electronics. Part of the challenge is that these are not necessarily commoditized. They're low value, they're low cost, but they're not necessarily commoditized. So while some of these parts are used in other industries like consumer electronics, also automotive, and then we think in many cases, we've been allocated away from. In some of the cases that you're seeing the glut, they're just simply not -- they're not fungible. They don't translate into that. And just even within our own portfolio, right, we've seen the problem with our modems on the optical portfolio. We have not seen the problem in our routing and switching portfolio. Those use a different set of power management and control integrated circuits. So we don't have the same supply chain issues in that portfolio.

Right. So how long do you think some of these -- presumably, some of the capacity, well, I know for sure, some of the capacity that makes the specialized stuff for you is also making -- have been making chips for the PC industry and so on. So that capacity can be repurposed. How long do you think before some of that free capacity starts to help things in terms of the how soon before the $5 million starts getting served, I guess is you know...

Yes. It's a fair question. Believe me, we're asking it ourselves. What we are doing is we're spending a lot of time with these suppliers, asking those same questions, having those types of discussions. And what we've said is we do expect an improvement in the volume and predictability of these integrated circuits as we move into our fiscal '23, which is a month or 2 away from now. So we do expect to see some improvement in that regard. But the reality is we purchase -- we procure over 10,000 different components and trying to play a little bit of whack-a-mole in terms of what is available, when is a bit difficult. But we do see some improvement in that regard, which should help address some of our issues in terms of the size of our backlog, which is great, but it also means that we're not servicing some of the customer demand. And frankly, that's not good, right? We're pretty upset about the current situation. We've had those discussions with our suppliers because by definition, it means we're not meeting our customers' needs and that's not okay. So we do expect some improvement on those integrated circuits as we move into '23. In terms of the broader semiconductor issues, make your own assumption, right? The conventional wisdom is some degree of improvement back to normalcy with more regular lead times in sometime in the middle or second half of 2023. We think that's probably a reasonable assumption. Yes. It seems like that's the circular reference consensus view, but then things are breaking loose a lot faster than I think when that view was originally formed, and I think that view is informed by the fact that things have kept happening that nobody expected to happen in China, particularly with regards to lockdowns and stuff. So we'll see whether that feels conservative now as we sit here today, but I guess we'll wait and see. Trying to make predictions in the current uncertain environment, there's a bit of a cool there.

Yes. Is there -- how close to this are you yourself in your role? I wouldn't think too much of -- you're doing too much on this or is it just all hands to the pump, everybody is working on supply?

Well, I'm certainly not the one who is going in and having the discussion about procurement team and the different suppliers. But our CEO, our CFO, myself and a lot of the executive leadership's team's time is being spent on these issues with customer escalations, supplier discussions and trying to answer the exact same questions that you're asking right now.

One of the things that we're, I think, not great at in the investment community is -- we always ask the same questions about these kinds of things, but we really don't understand all the details underlying it and that you're having to deal with to execute. Is there anything that we should be asking that were not any second, third order effects we should be thinking about or looking into that maybe we -- you're not hearing as much from investors or do you think people are covering it pretty well as a subject?

I think people by and large have a read on the situation. If there's one thing that I would call out is that I do think there is sometimes a tendency in the current environment to focus really on the short-term nature of the supply chain challenges and what that means for the business, and that's entirely understandable. At the same time, context is king. And I think one cannot and should not lose sight of the broader industry demand drivers, the secular demand drivers that we believe are incredibly strong, are incredibly durable. We can talk about what they are. But I do see a little bit of that being lost in terms of a longer-term view because, yes, there are uncertainties right now, whether it be supply chain shortages, a land war in Europe, a pandemic, highest inflation for decades here in the U.S., inflation in Europe for totally different reasons, does not lead to a recession, if so when and how deep, labor shortages, war for talent, any number of different uncertainties that it's easy to sit here and say, "Well, why would you invest in anything right now?" But our view is that these demand drivers, the cloud adoption across the customer segments, whether it be service providers, cloud providers or enterprises is so strong and will continue to persist that our strategic thesis is we're going to invest and continue to invest into the uncertainty, both with our organic growth initiatives and possibly inorganic accelerants to that growth.

Yes. I mean that's one of the -- we have more interest in infrastructure technology companies like yours than we do just about anything else we cover. So investors, I think, appreciate that. Wall Street tends to work on the things that it's a rate of change business, so they work on the things that they don't understand. Even -- I think most people do understand the growth drivers longer term, maybe with minor differences.

Glad to hear that.

Yes. So cost, let's talk about costs a little bit. Everybody has seen higher costs in components logistics. Could you just recap the margin impact that those have had on Ciena specifically and talk a little bit about the trajectory of margins from here as this hopefully improves over time?

Yes, absolutely. So what we've said historically is that our target margin range is in the mid-40s. It was meaningfully higher during the pandemic, and that was entirely a function of mix where network or operators simply were not doing any new deployments or build-outs. It was just pure capacity ads, which carry proportionately higher margins. So we were seeing margins in the high 40s in 2020 and 2021. What we've seen in the past year or so is obviously a degradation in the margin profile, and that's really for 2 reasons. One is these semi shortages, which have directly resulted in what we call higher exception costs. Think broker fees, when we have to go into the open market and procure available components, we're paying fees to the brokers to be able to access those materials. There's also higher freight and delivery costs, expedite fees, just getting your hands on anything right now in the semi space and delivering it to our customers cost us more. So our cost on that line would be about 4x typical this year, and that's resulting in about a 400 basis point impact on our margins right now. Additive to that is these acute integrated circuit challenges right now because they gate so disproportionately the modems that I talked about, that's an additional about 200 basis points of impact that we saw in Q3, and we expect to see at least until things normalize in that regard. So roughly, you add the 2 together, roughly a 600 basis point impact. We do think things will improve, as I said, certainly on the latter issue, that will improve into '23. The semi issue, again, make your own assumptions about that. Our view is that we absolutely see line of sight to getting back to our target mid-40s gross margin range. However, we don't think that's going to occur in fiscal '23. The reality is on semis, it's not as though a light switch is going to flip and all of a sudden, everything will be good. It's going to take some time for this capacity to come online to burn down backlog and inventory and then to normalize back. So it will take a little bit of time.

What about workarounds? You guys have talked about redesigns. How is that going? When does it start to impact? Is it already impacting your ability to service backlog?

Yes. So it's a good question. So what you're talking about is we're not just sitting on our hands and waiting for these suppliers to deliver these integrated circuits. We have, right at the beginning of our Q2 in earnest, done a number of different things to mitigate the risk. You mentioned engineering alternatives. One is to qualify alternative sources of supply and another is to redesign certain of our products to accept different alternative components. Those, by definition, take time. These are not jellybean parts. These are highly sophisticated pieces of equipment, as I said, thousands of parts on a chip that fits on the tip of your finger. So that does take some quarters to instantiate. We do expect to start seeing the benefit of some of those efforts as we move into '23. And while we're waiting for that, we're continuing to per equipment on the -- like procure equipment rather on the open market, placing large, advanced POs and of course, working with these suppliers.

Great. So backlog has continued to grow, and you guys have recently -- the latest disclosure of $4.4 billion number, which is kind of mind-boggling in terms of its size. But I think our revenue estimate next year is $4 billion or something like -- it's about just over $4 billion. So your backlog is bigger than a revenue estimate. So I wonder how that's affecting customer perceptions? And are you concerned that it puts you in any kind of competitive disadvantage or create any kind of friction that way with customers? Just curious how those conversations are going, given people are having to wait so long for things?

Yes. No, I mean, obviously topical. You're right. I mean, the backlog has grown exponentially. Just to dimension it for you, we were at $2.1 billion coming out of our last fiscal year. And in 9 months, it's now at $4.4 billion, which is a mind-blowing number. And that's great, and it's not great for different reasons, right? It's great, obviously, because we have, as you said, a significant line of sight for the next period of time in terms of being able to service that backlog and turn it into revenue. It's not great because, by definition, as I said, we're not servicing that demand for our customers. And so that is a problem. We do think that growth rate of the backlog is, of course, going to have to normalize. That's just not sustainable. When it normalizes exactly, how, we don't know. What I think has been nice is that we've not seen as we've said -- we've not seen any cancellations of orders. And we've heard questions about double ordering and forward ordering. And the reality is our customers would take all this equipment tomorrow, if they could get their hands on it. So I think those are kind of a little bit of red herrings, to be honest. So the demand environment in the 2-plus decades I've been at Ciena has never been stronger. Now in terms of how customers feel about that, well, those are hard conversations. We're obviously in allocation mode. The good news is, because we have more absolute and relative demand because of our size, we are shipping more equipment than anybody else, right? I don't want you to think that we're just sitting here not doing anything, right? We're shipping a lot, but we're not shipping enough to meet our customers' needs. So those are conversations that we are having on calls on a daily basis with our customers having to make difficult allocation calls because by definition, they're not meeting their in-service customer requirements and their own internal objectives in terms of rolling out network services. So at the end of the day, we've not seen any cancellations. The demand continues to grow. So we feel good about that from a customer relationship standpoint. But we'd be tone-deaf to think that there is no potential impact. And so 10 competitors in any particular account or route or opportunity, rifle shot their limited allocation to provide equipment more quickly than we in a corner case that certainly is happening. But overall, we think to whatever extent that is happening once these more acute issues start ameliorating and we're able to get those modems out the door, that share will be recovered and frankly, increased relatively quickly.

Yes. And you have control of that situation, you can decide where you want to take the risk that happens, or you don't…

We do, and we are being thoughtful and selective about that. We operate primarily on a first-in, first-out basis; first come, first served. And the reality is that primarily the North America providers, the Tier 1 service providers and the cloud providers got in queue earlier than perhaps some of the ones did internationally.

Well, let's talk about more fun things. Now let's put our line under supply. Network Edge, there's been a lot of momentum that way, particularly for Ciena. We've seen you emerge in Verizon small cell deployments. We didn't think that would happen. We thought that'd be passive optical and ended up active switch to AT&T win recently. It just seems like there's an awful lot of active optical stuff going on in access that wasn't going on before. And I wonder how big of an opportunity do you think this is purely a U.S. phenomenon? Do you see this emerging in Europe as well kind of strategically, how important is that for the company?

Yes. So the way we think about our strategic priorities, Rod, is obviously, from an optical standpoint, we are the hunted, right? We have the world's leading technology innovation, best-of-breed, largest market share and everyone is gunning for us. Conversely, in the area that you're talking about, which I'll call kind of broadly next-generation metro and edge, we are the hunter, right? It's not as that we're new to the space. We've been doing Layer 2 access and aggregation for a very long time. But some of the examples you're raising, I think, are good ones because we have deliberately leaned into next-gen metro and edge as a growth opportunity for us over the past several years. And in fact, we see that as far and away the biggest CAGR of our addressable market over the next several years. Optical Transport will grow. Software automation will grow, but next-gen metro and edge is going to go significantly higher. And the largest source of what we said is going to be an increase in our addressable market from approximately $13 billion to $22 billion over the next several years. It's metro and edge, think cloud services, think cross-haul applications, think Ethernet data services and then what you mentioned residential PON. And we have been leaning in organically and with partnerships. We did -- the win at AT&T, you spoke of, was a very significant residential PON win for us, taking away incumbency from one of our competitors, and we now have significant momentum in that space with over 40 different customers. And that we expect will continue to increase, right? I mean you look at the rate of fiber broadband deployments going on in the U.S. right now. We expect more fiber to be deployed in the next 5 years in the U.S. than in the past 15 years combined. And then you've got, obviously, the rural broadband subsidies, with be it under the Infrastructure and Jobs Act and RDOF and things like that. So there's some significant catalysts there, and that's why we're leaning into that space. And then you mentioned as well  Cell-Site Router, which is nice win at Tier 1 service provider in North America, very happy with that win as well. So we are the challenger. But make no mistake, we're not naive. We're going into a very competitive space with very significantly large and well resourced-competitors in this space. But overall, we think that the combination of the fact that we're leading with our optics and the reality is IP and optical are converging, I think that's proven now. The fact that we've got existing customer relationships and the fact that we come at it with a very nice cohesive closed-loop software automation story, I think, provides us with significant competitive differentiation to succeed in the space.

Yes, that's what I wondered about is whether it's the software automation story that's making these edge sales for you? Or how would -- is that the way you would characterize your advantage, or do you think there are other things also driving people to purchase it in a place that they haven't historically used you?

Yes, I think it really depends upon the customer and the application, Rod, to be honest. But it's certainly absolutely a factor. In some cases, it could be a leading factor, right? In some cases, it's a factor along with the things I mentioned, like the optics and the capacity and the programmability that we bring along with it. In some cases, it's just the customer relationships. So on as that we're talking to the customer for the first time, we've been selling to them, or in some cases, 20 or 30 years. But make no mistake, I mean, software automation is key, right? Service providers need to adopt cloud principles to make their networks more operationally efficient. 5G, we're at the -- I know you don't want to talk about 5G, but we're at the very early stages of 5G, and 5G fuels the need for modernization of OSS/BSS. And that is all about these new innovative services that need end-to-end services, life cycle automation and that's what Blue Planet is all about.

Yes. You've got the advantage of your colleague, Steve Alexander, who's leading a lot of that next-gen [indiscernible] of the future kind of work and thinking, I think. Let's talk about 800 gig a little bit. Last quarter, you mentioned that WaveLogic 5 Extreme modem cumulative shipments were over 44,000 up to that point anyway. Can you talk about revenue contribution of 800 gig at this point or give us any kind of hints as to how big that might be from a revenue point of view?

Yes. So we don't disclose, Rod, the separate revenue proportion from 800 gig. We do disclose port shipments and they're continuing to increase. The way I would think about 800 gig is less about 800 gig as a cycle itself and more about the fact that Ciena has the highest-performing optical engine in the world, right? What might increase -- to your question, what might increase the take rate of deployments of 800 gig? Really, it's 2 things that drive those decisions, right? One is client rate, right? And the reality is the majority of networks are still on 100 and, to some cases, 400 gig. But the ability to carry a client anywhere without having to regenerate the signal is significant. It's significant both in terms of service performance and in terms of the cost and the economic value position. So that's one is the client rate. The second is spectral efficiency, right? And our ability to drive more bits on the fiber than anyone else goes straight to the cost per bit analysis. So our view is that 800 gig is a proxy for WaveLogic 5 for us, which is the world's leading both in terms of technology and market share, 800-gig engine. But for us, if you're a network operator and you've got growing bandwidth traffic and fiber constraints, we're the obvious, of course.

Right. Yes, I was explained that 800 gig and things like that are more brands than they are actual speeds and the speed is how the thing performs in the field and those results vary…

Anybody can announce a hero experiment with 800 gig or we've seen some recently with 1.2 terabits. And I think history speaks for itself in terms of others overhanging the market and what actually gets delivered and when, I think has been proven out quite clearly over time.

So let's talk about web-scale a little bit. You've been a leader there in DCI. Revenue growth slowed a little bit in the last couple of years there. And I don't know if you want to talk a little bit about what's going on with demand in web-scale more broadly, maybe not just for you but for everyone? And then what you would kind of expect going forward in terms of growth there, not necessarily for Ciena, but just for that whole market?

Yes. So the web-scale provider and by the way, we are guilty of this as well. We talk about it in terms of web-scale or GCNs or ICPs. I'll just refer to them as cloud providers, right, Google with the Facebooks, the Amazons, Microsoft. The way we think about that customer segment for us is unlike the service providers who have a different need, right? They need to -- we talked about. They need to transform their networks from these distributed decentralized approaches. The webs, the cloud providers really need to do something different. They need to transform their networks at massive scale, right? So they've got their existing data center infrastructure that they're refreshing and they've got new data center build-outs. Why? Because they need to host the massive demand for cloud services and cloud applications, anything as a service and then take that and interconnect into the service provider regions to create this kind of globally distributed cloud edge. And so we don't see that diminishing at all over the next several years. Now in terms of the demand, what we've seen over the past few years is hyper growth in CapEx from the cloud providers. That growth rate is going to have to moderate. That's not sustainable. However, we expect to see a meaningful growth rate on CapEx for these guys in terms of low double digits over the next several years. So we don't see that demand going away at all. They are driving the lion's share of decision-making and traffic around the world, and we expect that to continue.

Great. Okay. And then another thing on the cloud provider side of things. Google just published a white paper not too long ago, talking about optical switching as a replacement of the aggregation layer, which is usually handled by switching Ethernet switching products, routing switching products. So curious what you think about that? Have you guys been looking at that, talking to people about that kind of opportunity? I know direct attack is not really what Ciena does, but maybe does that eventually move into a coherent sort of an application for you?

Look, I think it's interesting. Maybe it's just interesting to me because I've been doing this for so long. That higher data rates in the data center are driving the same kind of architectural discussions shift that we saw 10 years ago in the WAN, right? In this case, what we're talking about is the need for taking coherent optical technology in the form of an optical switch into the data center. And I think the answer is, yes, we do see that as an opportunity for us over time. It's a logical extension of what we do. How big, how soon? I don't know. My guess is right now, just given where cloud providers are at, I think that's probably an opportunity that you can see for Ciena the 2025 range and beyond.

Right. Yes, it's funny to me when I talk to hyperscale engineers, they've just invented some incredible thing from their point of view and having been around telecoms my entire life and started as a network engineer, I usually tell them that was done in telecoms 50 years ago. I mean many, many of these things are being -- the wheels are being recreated. So it's just kind of funny to watch.

Yes.

Let's talk about international growth a little bit. Can you talk a little bit about what you see going on there, particularly India is kind of exciting opportunities. We look out into the next year. But any international comments you might have in terms of growth expectations or other areas or projects that you think are investors ought to be focused on?

So I think to meet our longer-term objectives, and we said pretty clearly, our longer-term top line growth is 6% to 8% supply chain issues and resulting impacts notwithstanding. To meet that over time, we are going to need to continue to grow our share internationally. That is a fact, right? And we have operations around the world. We count outside of the China PTTs, virtually all of the largest service providers and cloud providers as our customers, and that's great. But it needs to continue, and there are a variety of different opportunities. You mentioned one in terms of India. India is interesting because in '18 and '19, the big remaining 3, there used to be 20, they're rating 3 in terms of Reliance Jio, Bharti Airtel and Vodafone Idea, spent a lot of money investing in building out their networks. And we were the backbone for all of those and took significant market share. That was before the Indian government effectively excluded Huawei from future deployments. Now we saw in India and India, I think, is frankly the best, not the only, but the best use case for what happens when you run your networks hot. They run their networks extremely hot during the pandemic. And so what we're seeing right now in India is really the beginning stages of what we think will be a significant investment cycle by all 3 service providers. They're one because of the need for the capacity infill to meet their demand and 2, because they're just starting their 5G public rollouts there. So that we see as a meaningful opportunity for us over the next several years. And the fact that Huawei is not there, we were doing well there before and that will only present additional opportunities for us, but make no mistake, our competitors are looking at it the same way. And elsewhere, certainly, I think Western Europe in general, I think, is a very interesting battleground. I'm sure Huawei is on the tip of your tongue in terms of what that means for us there. And I think unlike the U.S. and Canada and maybe parts like Japan, where decisions have been made to exclude Huawei, Western Europe is still pretty much a battleground for those Huawei displacement opportunities. And I think that's also a meaningful share gain opportunity for us over the course of time.

Yes, I would think Nokia is fighting hard to try to get in there as well, although it doesn't seem like their road map is quite as robust as yours from an optical point of view. Pluggables, well actually, let me -- before I do that, let me see if anybody has got any questions because we're down to 8 minutes left here. Does anybody in the audience have any questions for David. Anybody? Okay. All right. I'm going to jump into pluggables. And then if you have a question, just raise your hand and we'll get a mic to you, and you can ask it. So pluggables is a pretty interesting strategic challenge opportunity for Ciena. I wonder if you could talk a little bit about that. The TAM, the company has talked about $500 million of TAM in the past. A lot of investors worry about disruption and what eventually pluggables mean for systems manufacturers like Ciena. So I wonder if you could just kind of encapsulate your thinking on pluggables a little bit for us?

And we'll try to encapsulate. It could be...

Especially the strategic side of it, what does it look like longer term?

Yes. I do remember back -- and you remember this, Rod, back in 2018, the conventional wisdom was that plugs presented an existential threat to the optical system solution business. We're sitting here in 2022, and that has not instantiated at all. Here's our view on plugs. There's a lot of talk about it and a lot written about it. But our view is that we are incredibly well-positioned to take advantage of the pluggable consumption model with our customers. And the nearest and best analogy is what happened when cloud providers looked on a substitutional basis, to take optical technology inside their data center shells, right? And what happened there is that led to a disaggregation in offline systems into open line systems. And we were able to serve that need very well with our Waveserver and reconfigurable line system products. So we know what we're doing in this regard when things get disaggregated. And really, that's what plugs are. It's just an entirely different form factor. Will there be substitutional on the optical systems? Yes, there will be. But we feel that we're incredibly well-positioned to serve that need. I would also point out that this is really for the next several years, a cloud provider opportunity. Yes, there are service providers who do have an interest in plugs, but the take rate, just like with open line systems is significantly longer with service providers just given the nature of their network needs and their applications than cloud providers. And so service provider is always going to lag in this regard. But overall, in terms of plugs, we believe that we have and will continue to have the best available plugs out there, and we'll compete very effectively in that market.

What I -- the thing I worry about with plugs is disintermediation. So I was worried that maybe in Arista or somebody is going to sell a switch router box, plugs go into that, that's facing the WAN and -- but it's their automation software then is utilized. But is that correct? Is that something you worry about or is it still on an optical layer of automation that goes on there. I worry about everything.

Of course, you're a lawyer.

It's my job.

Look, I think there have been some pretty notable examples where certain cloud providers try to do things differently. I'm not going to name them, but to build their own white box for which to take their own software on top of it and to disaggregate even further. And for a lot of different reasons, those haven't worked out. And the reality is it's because coherent optics are really hard, right? It's complicated. The investment capacity is significant and it takes a long time. Now that said, are we continuing to engage with our cloud provider customers are on different methods of disaggregating the procurement of optical technology? The answer is, of course. And we'll continue to listen to them and be in the best position to serve their needs. But in terms of some existential disintermediation threat, we don't see it that way.

I've always wondered whether there's any layer of automation that even it's a pluggable as going into an active box, maybe it's a switch router or something like that, there could still be a Ciena automation layer on top of that. Is that a possibility with pluggables or not really? I just don't really understand whether it is...

The answer is, absolutely. That is the beauty of an open architecture approach, and a multi-vendor approach is that you can -- whether it's the line system, whether it's the pluggable optics, whether it's the software that wraps around it, you can play at all or any of those particular elements of the overall solution.

Right, right. Let's talk a little bit about competition and then I want to get into maybe your -- get you to give me a macro forecast. Just kidding. But could you talk a little bit about what you're seeing competitively out there right now? Who is your toughest competitor? What's going on with competition? It's kind of hard to judge. I mean, it doesn't -- it looks like if anything, competition has gotten less aggressive, especially with Huawei's exit, but I don't know how you would see things out there.

Yes. I'd go back to where I kind of broke down our strategic priorities. On optical, the competition is well known, right? I mean, in terms -- we believe that we have, far and away, the best coherent optical technology on the market. I think that's been proven. You've got Infinera out there with a single wavelength 800-gig solution. You've got Cisco with Acacia and then some smaller players. Overall, we feel very good about our competitive positioning in optical and continue to do so. As I mentioned on metro and edge or routing and switching, it's -- we're more of the disruptor in that regard, but feel that the competition is strong, but we're capable of differentiating ourselves. In PON, for example, Nokia is the big dog in that space. And then overall, when you talk about the convergence of IP and optical, certainly Cisco and the routed optical networking infrastructure is the biggest competitive threat. But if I had to pick it over the next 3 to 5 years, I -- personally, we see it blowing down to the big 3: ourselves, Nokia and Cisco with Acacia.

And then software asset-wise, you bought Vyatta, Blue Planet. Anything else -- do you feel like your software portfolio now is pretty complete or do you think there's a lot more opportunity to expand that either organically or inorganically? Is there anything missing from that portfolio?

I think the answer is yes to both of those questions. I'm not being glib. We have, as you said, built out the portfolio, so that -- we believe it's a comprehensive closed-loop automation portfolio with the acquisition of Packet Design for routing optimization insurance with DonRiver for federated inventory, the Centina for unified analysis and assurance. We believe -- and then with, obviously, the virtual routing capabilities from Vyatta, we believe that we have a very nice comprehensive, differentiated solution. That said, we are always looking at opportunities to build out gaps in the portfolio, to meet particular customer needs or a particular application set. And so that's an area, along with others, that we continue to look at inorganic opportunities to accelerate growth.

And then lastly, I'll ask you a macro question, but not on the overall economy, more on Europe specifically. I think to put it mildly, things don't seem like they're going all that well there. So I'm curious whether you've gotten any signal back from service providers, customers there may be indicating that they're looking to reduce investment as they look out into next year? Or do people seem like they're just kind of par for the course continuing forward?

It's -- there are smarter economists around this conference than me in terms of trying to predict what's going to happen. We are not seeing any of that from our customers, whether in Europe or frankly, anywhere else. Going back to my point about the demand drivers. I'm not going to say that any industry is recession-proof because that would be foolhardy. At the same time, the needs for network connectivity over the next several years are only going to increase. For all the trends and the reasons that we've talked about, it's going to result in the need for more optical bandwidth. It's going to need -- the need for more data transfers into the cloud and the need for more cloud CapEx. And so we don't see anybody indicating to us a slowdown in demand.

Great. Okay. Well, we're out of time. Unfortunately, I've got 10 more questions, but I really appreciate it. Thank you, David.

It's my pleasure. Thanks for having me here.

Thanks, everyone.