Kathleen Nemeth - Shaygan Kheradpir - Chief Executive Officer and Director Robyn M. Denholm - Chief Financial & Operations Officer and Executive Vice President Rami Rahim - Executive Vice President and General Manager of Platform Systems Division Vincent J. Molinaro - Chief Cusomer Officer and Executive Vice President

Jess L. Lubert - Wells Fargo Securities, LLC, Research Division Amitabh Passi - UBS Investment Bank, Research Division Pierre Ferragu - Sanford C. Bernstein & Co., LLC., Research Division Ehud A. Gelblum - Citigroup Inc, Research Division Jeffrey Thomas Kvaal - Northland Capital Markets, Research Division Mark Sue - RBC Capital Markets, LLC, Research Division Simona Kiritsov Jankowski - Goldman Sachs Group Inc., Research Division Brian T. Modoff - Deutsche Bank AG, Research Division Benjamin A. Reitzes - Barclays Capital, Research Division Kulbinder Garcha - Crédit Suisse AG, Research Division Ittai Kidron - Oppenheimer & Co. Inc., Research Division

Greetings, and welcome to the Juniper Networks Third Quarter 2014 Financial Results Conference Call. [Operator Instructions] As a reminder, this conference is being recorded. I'd now like to turn the conference over to your host, Kathleen Nemeth, Vice President of Investor Relations. Please go ahead.

Thank you, operator. Good afternoon, and thank you for joining us today. Here on the call are Shaygan Kheradpir, Chief Executive Officer; and Robyn Denholm, Chief Financial and Operations Officer. Vince Molinaro, Executive Vice President, Chief Customer Officer; and Rami Rahim, Executive Vice President, Juniper Development and Innovation, will be available for the Q&A portion of the call. Please remember when listening to today's call that statements concerning Juniper's business outlook, economic and market outlook, strategy, future financial operating results, overall future prospects, planned operating, expensed reductions and capital allocation plans are forward-looking statements that involve a number of risks and uncertainties. Actual results could differ materially from those anticipated in these forward-looking statements as a result of certain factors, including: economic conditions generally or within the networking industry, changes in the overall technology spending and spending by communication service providers and other major customers, the network capacity requirements of service providers, the timing of orders and shipments, manufacturing and supply chain constraints, variation in the mix of products sold, customer perception and acceptance of our products, rapid technological and market change, litigation, the potential impact of activities related to the execution of Juniper's Integrated Operating Plan and other factors listed in our most recent 10-Q and 8-K filed with the SEC. All statements made during this call are made only as of today. Juniper undertakes no obligation to update the information in this conference call in the event facts or circumstances change after the date of the call. Our discussion of the financial results today will include non-GAAP results. For important commentary on why the management team considers non-GAAP information a useful view of the company's financial results, please consult the press release furnished with our 8-K filed with the SEC today. For the detailed reconciliation between GAAP and non-GAAP results, please see the Investor Relations section of our website. In today's call, we will also be providing forward-looking guidance. As a reminder, other than revenue and share count, guidance is provided on a non-GAAP basis. All guidance is forward-looking, and actual results may vary for the reasons I noted earlier. A reconciliation of non-GAAP guidance to GAAP measures is not available on a forward-looking basis due to the high variability and low visibility with respect to the charges which are excluded for non-GAAP guidance. As a reminder, we completed the sale of our Junos Pulse security product line earlier this month. Please ensure your financial models are appropriately adjusted. Further information and our long-term model will be shared during our Investor Day later this month in New York City. Please note that today's call is scheduled to last for 1 hour. [Operator Instructions] With that, I'll turn the call over to Shaygan.

Thanks, Kathleen, and welcome, everyone. I wanted to give you an update on progress at Juniper, including the impact of industry trends on our revenues, our work on our Integrated Operating Plan and steps we are taking to ensure we are well positioned to deliver shareholder value going forward. While we have made significant progress executing beyond our IOP commitments associated with capital return and operating expenses, we are disappointed with our revenue performance this quarter. Our Q3 revenue declined 5% year-over-year and 8% sequentially. The primary driver was lower-than-expected demand and slower ramp-up of new projects from service providers, particularly in the U.S. That said, our traction with Web 2.0 leaders has allowed us to partially offset the decline in carrier spending and increase the diversity of our revenue base. Notwithstanding these industry headwinds, we delivered non-GAAP EPS within our guidance range. We did so by maintaining strong gross margins, diligently managing expenses and executing aggressively against our IOP capital return goals. We have met or exceeded our commitments set forth in our Integrated Operating Plan in February while not losing sight of strategic engagements with key customers. We achieved our original cost reduction target this quarter by getting to a $500 million quarterly OpEx level from $539 million in fourth quarter of 2013. This came 2 quarters ahead of the original commitment. We're now further reducing our operating expense levels to $1,900,000,000 for the fiscal year 2015, a meaningful reduction from current levels. We believe that by continuing to take this kind of surgical approach to our cost reduction efforts, we can still invest sufficiently to grow our business and lead the next wave of IP networking innovation. Our One-Juniper organizational structure is allowing us to get products to market faster and cheaper as our common operating system and coalesced engineering and go-to-market teams continue to operate faster and more efficiently. In fact, we are excited about our product pipeline ahead and pleased with progress in our pivot to strategic vertical markets focused on cloud building and high-IQ networking. We have also exceeded our capital return commitment set forth at the Integrated Operating Plan. Our commitment through Q1 2015 was to return $2 billion of capital. We will return greater than that by Q4 2014. Our commitment through 2016 was to return $3 billion in capital. We have increased the size of our capital return program by $1.1 billion and now plan to return $4.1 billion by the end of 2016. Given market conditions, we will take an aggressive stance and will plan to repurchase $1.5 billion by the end of Q2 2015. We also announced today a $0.10 per share cash dividend to be paid in Q4 with the intent to grow it in line with earnings over time. I'd like to spend a few moments discussing the macro environment and our business. As I've stated previously, I believe the global networking market is at an inflection point, presenting substantial opportunities for growth. The underlying long-term demand trends support the need for increased investment in the network. Based upon historical trends and discussions with customers, we feel investment will return in the second half of next year, and that our integrated portfolio of routing, switching and security products will be well positioned to address critical customer requirements. Despite weaker macro conditions, the market fundamentals for Juniper's routing products continue to be healthy, and our customer base continues to be increasingly diversified. In Security, we saw good demand for high-end SRX products from service providers. Switching revenue declined quarter-over-quarter as we experienced an ebb in the wave of customer build-outs that we saw earlier in the year. Though we continue to see momentum for our switching products, we expect modest improvements until the recent wave of customer design wins ramps in 2015. We continue to see our integrated solutions as a fundamental building block and a differentiating factor for customers of cloud building and high-IQ networking, including the focus on secure routing and switching solutions for active defense. So what should you expect moving forward? We are focused on continuing to drive long-term shareholder value through profitable growth. We assume the carrier spending will remain constrained for several quarters, and we are streamlining our business with the before-mentioned operating expense reductions. We expect this to result in healthy year-over-year operating margin expansion in 2015. I am enthusiastic about our technology roadmap to deliver the next wave of IP networking, and I'm confident in our strong market position. We remain intensely focused on operational excellence, cost discipline and targeted growth initiatives. We appreciate the support of our shareholders and thank our employees for their ongoing commitment and relentless dedication to execution. Now I'll turn it over to Robyn to provide more details on our financial results. Robyn M. Denholm: Thank you, Shaygan, and good afternoon, everyone. The September quarter was challenging from a revenue perspective primarily because the Service Provider demand, particularly in the U.S., is slowing more than expected. Despite this, we delivered non-GAAP earnings within our original guidance range due to our continued focus on cost reduction initiatives. Before I go into detail on the results of the quarter, I'd like to provide an update on our Integrated Operating Plan. We continue to make good progress against our targets. We have achieved our IOP quarterly operating expense target 2 quarters earlier than expected. We also completed the consolidation of our Sunnyvale campus and negotiated an early exit of our lease, which has resulted in a credit of $26 million against our prior restructuring estimate. As announced on October 2, we finalized the sale of the Junos Pulse business. As a result, we anticipate booking a GAAP gain on the sale in Q4 of 2014. As expected, our Q3 results reflect a full quarter of this business. We are executing well on our capital return plan. We paid our inaugural dividend of $0.10 per share in September. We also completed our $1.2 billion ASR in July and repurchased an additional $550 million of shares in Q3. Since inception of our Integrated Operating Plan, we have reduced our fully diluted share count by 10% and 13% over the last 2 years. I'm very pleased at the pace of our cost reductions and our progress towards a more efficient capital structure. In the outlook section, I will provide detail on the additional costs and capital allocation action mentioned by Shaygan. Now let's turn to an analysis of the Q3 results. Demand was relatively weak in the quarter. Our product book-to-bill was approximately 1, and bookings were almost $100 million less than we expected. Total product deferred revenue declined $57 million year-over-year and $29 million versus last quarter. The sequential decrease was mainly the result of the sell-through of channel-related inventory and the reclassification of the Junos Pulse deferred revenue to a held-for-sale asset category. Total revenue for the quarter was $1,126,000,000, down 5% year-over-year and 8% sequentially. Product revenue was $810 million, down 10% year-over-year and 13% over last quarter. Services revenue was $316 million, up 11% year-over-year and 5% sequentially. For the quarter, GAAP diluted earnings per share were $0.23, an increase of $0.04 year-over-year, including a $0.03 benefit from restructuring credit. Non-GAAP diluted earnings per share was $0.36, up $0.03 year-over-year and down $0.04 versus last quarter. The sequential decrease was due to lower revenue, partially offset by improvements in gross margin and operating expenses. Looking at our revenue for Q3. There were 4 primary factors which negatively impacted our results versus our expectations. U.S. carrier spending was worse than we had expected, and Web 2.0 Switching revenue was lower than anticipated due to the timing of deployments and the slower ramp of new projects. To a lesser extent, we saw a softening in the broad-based enterprise market and a greater-than-expected slowdown in demand from Eastern European service providers. Now let's look at the revenue results in detail. Americas revenue was up 3% year-over-year and down 5% sequentially. Americas Service Provider was up 1% year-over-year and down 7% sequentially. The year-over-year increase was driven primarily by Web 2.0 and cable provider growth, partially offset by decline from carrier demand. The sequential decline was driven by weak demand by large carriers and a significant reduction in Web 2.0 Switching revenue in the quarter, partially offset by continued strength in cable providers and Web 2.0 routing. Americas Enterprise was up 5% year-over-year and flat sequentially. Sequentially, we saw continued strength in the U.S. federal market and financial services, which was offset by some softness in the broad enterprise market. EMEA revenue was down 5% over last year and 11% versus last quarter. The year-over-year decrease was due to declines in Western Europe and the Middle East, primarily in Service Providers, whilst the sequential decrease was driven by weakness in Eastern Europe. APAC was down 28% year-over-year and 19% versus last quarter, primarily due to Service Provider weakness in China and Japan. Service Provider revenue for the quarter was $742 million, down 6% year-over-year due to EMEA and APAC and U.S. carriers, partially offset by a slight increase in the Americas from Web 2.0 and cable providers. Sequentially, Service Provider revenue for the quarter was down 11% due to declines across all 3 geographies. Enterprise revenue was $384 million, down 3% year-over-year and sequentially. APAC and EMEA were down in the broader enterprise market year-over-year. Sequentially, the decline was partially offset by continued strength in the U.S. federal market and financial services. By product area, Routing product revenue was $533 million, down 12% year-over-year and 14% sequentially. The declines were driven by weakness in the carrier market in both core and edge. We saw good traction in both the PTX and MX2020 product lines. Enterprise Routing was healthy. Switching product revenue was $155 million, an increase of 5% year-over-year, driven by QFX into service providers, partially offset by a decline in the EX, mainly in the broad enterprise market. Sequentially, Switching product revenues declined 22%, coming off a record revenue in Q2. The decline was the result of lower demand from our Web 2.0 and softness from our broad enterprise customers. Security product revenue was $121 million, down 16% year-over-year but up 9% versus last quarter. Our SRX platform and security software were down 6% year-over-year and up 11% sequentially, mainly as a result of increased demand by service providers. Moving on to gross margins and operating expenses. Non-GAAP gross margins for the quarter was 65.2% compared to 64.4% a year ago and 64.2% last quarter. The improvements were driven by an increase in both product and services gross margins. Non-GAAP product gross margins were 66.1%, up 0.7 points from both a year ago and last quarter, mainly due to favorability in product mix and reductions in costs. Non-GAAP gross margins for services were 63%, up 1.7 points from a year ago and 2.5 points versus last quarter. The increases were due to lower support-related costs from operational improvements and savings from variable costs. We are pleased to report that our non-GAAP operating expenses were $493 million, well below the range we gave last quarter, reflecting our focus on cost as well as the benefit of reduced variable costs in the quarter. Our headcount at the end of the quarter was 9,059, which includes approximately 200 employees who transitioned as a result of the sale of the Junos Pulse business. Adjusting for the exiting Pulse employees, headcount declined 7% versus Q4 of 2013. Non-GAAP operating margin for the quarter was 21.5%, reflecting a year-over-year expansion of 1.7 points due to continued cost reductions. We experienced 0.1 point sequential of decline due to lower revenue, partially offset by improved gross margins and lower operating expenses. The non-GAAP tax rate was 27% compared to 26% last quarter as a result of the increase in the mix of U.S.-based taxable income. The GAAP tax rate was 37.4% compared to 24.9% in Q2 due to a change in the geographic mix of earnings and other onetime items. We ended the third quarter with approximately $2 billion of net cash and investments. The decline was primarily due to the capital return of $594 million. Onshore cash and investments represented approximately 26% of total gross cash balance. For the quarter, we saw a net cash outflow from operations of $79 million, mainly due to timing differences in working capital, specifically an increase in accounts receivable, payments for incentive comp, tax payment and a reduction in deferred revenue. In Q4, we expect to return to our historical pattern of strong positive cash flows. DSO was 49 days, up from 41 days last quarter due to the timing of shipments and invoicing linearity. Capital expenditures were $43 million, and depreciation and amortization expense was $44 million. Now I will provide our outlook for Q4. As a reminder, these metrics are provided on a non-GAAP basis, except for revenue and share count. The demand environment for our largest U.S. carrier customers continues to be challenging. We have good relationships and design wins with these customers. However, the timing of deployments and improved demand is uncertain. At this stage, we are taking a prudent and cautious stance on revenue over the next several quarters. As previously mentioned, we completed the sale of Junos Pulse, and the quarterly revenue impact of approximately $30 million has been factored into the Q4 guidance. For the fourth quarter of 2014, we expect revenues to range from $1,025,000,000 to $1,075,000,000. Gross margins are expected to be 64%, plus or minus 0.5%. Operating expenses are expected to be $480 million, plus or minus $5 million. Operating margins are expected to be 18.5% at the midpoint of guidance. We expect a flat tax rate versus the third quarter. We expect non-GAAP diluted EPS of between $0.28 and $0.32 per share, assuming a weighted average share count of approximately 435 million. I'm also pleased to report that the board has approved a dividend of $0.10 per share for the fourth quarter. For the full year 2014, using the midpoint of Q4 guidance, revenues are expected to be approximately $4,576,000,000. Gross margins will be approximately 64%. Operating expenses will be approximately $2,030,000,000, and operating margins will be just under 20%. EPS is expected to be approximately $1.35. Excluding Junos Pulse revenues of $95 million for the first 3 quarters of 2014, Pulse adjusted full year revenue will be about $4,480,000,000. Now I'd like to move into a discussion of how we are planning the business for 2015. We see the long-term demand drivers as healthy, and we are confident in our innovation pipeline. We continue to earn good design wins across our carrier, web services, cable and enterprise accounts. That said, we are planning for the overall revenue environment to be challenging over the next several quarters as near-term factors are impacting demand from our largest Service Provider customers. Given this market environment, we are targeting 2015 operating expenses of $1,900,000,000, plus or minus $25 million. This is a $130 million reduction for our estimated full year 2014 operating expense levels and translates to a $260 million aggregate IOP commitment. To achieve these additional savings, we will continue to carefully manage headcount, drive efficiency improvements and prioritize revenue-generating projects and resources. As part of our ongoing commitment to drive shareholder value, we also announced earlier today that the board has approved an additional $1.1 billion to the previously authorized capital return program. This brings our total capital return commitment to shareholders, including dividends, to $4.1 billion through 2016. As a reminder, our original IOP commitment was to return $3 billion to shareholders through 2016. We have been opportunistic and aggressive in reducing our share count and have paid our first dividend. This brings the total return to shareholders through the end of Q3 2014 to $1.8 billion. We expect to continue to be opportunistic and aggressive in reducing our share count given the current market conditions, and we are currently expecting to complete a minimum of $1.5 billion in aggregate share repurchases before the end of Q2 2015. Given these additional actions, even with a cautious revenue outlook, we expect significant year-over-year operating margin and earnings per share expansion in 2015. In summary, we remain committed to returning value to shareholders while maintaining the financial flexibility to invest in innovation and growth. I am very proud of our team and thank them for their continued dedication and commitment to Juniper. Now let's open the call for questions.

[Operator Instructions] Our first question comes from Jess Lubert from Wells Fargo Securities. Jess L. Lubert - Wells Fargo Securities, LLC, Research Division: You gave some color on your expected operating expenses for 2015. Can you also let us know if you expect to grow next year? And can you help us understand how you're thinking about Q1 seasonality given the weaker second half? And then despite fairly soft router sales, you suggested fundamentals in the business remained healthy. So I was hoping to understand what's driving confidence here in this market. And given the carrier focus on NFV technologies and reducing costs, what leads you to believe the business will snap back in the second half?

Jess, this is Shaygan. I'll make a few comments and I'll pass it to Robyn. One, I would say that the visibility to revenue in the near term, in the next few quarters, is poor. And this is because we're going through a cycle, as you know, and these cycles take 2 to 4 quarters. We think it started last quarter. Plus, in our last earnings call, we also mentioned that there are some specific dynamics in the carrier market in the U.S. So put them all together, it gives us poor visibility into the near future on revenues. Because we think these cycles typically take 2 to 4 quarters, at this point in time, our planning assumption is that growth will return in the second half of 2015. And we're planning our business accordingly. Robyn, do want to say a few words? Robyn M. Denholm: No, I think that's very good, Shaygan. In terms of the near-term environment, as we mentioned, for the fourth quarter, our guidance -- we obviously put that number out there. And then we do expect that we're in this 2- to 4-quarter cycle, as Shaygan talked about. For the full year next year, we're not putting out specific revenue numbers. As Shaygan mentioned, the environment is challenging. We expect that to continue for the next several quarters. So our planning assumptions at this point is that we expect that challenging environment to continue through the first half of next year.

Our next question comes from Amitabh Passi from UBS. Amitabh Passi - UBS Investment Bank, Research Division: Shaygan, if I look at Juniper over the last 4 years, you had an up year, a down year, an up year, a down year. And it seems like the company is becoming increasingly cyclical. So I guess I wanted to get your thoughts. We talk a lot about a lot of secular underlying drivers, but we're not quite seeing it in your business. And so would love to get your thoughts in terms of can we get a secular growth trajectory for Juniper or is the company increasingly now cyclical?

Amitabh, good question. I mean, if you look at the curves, your statement is actually correct. It looks like sinusoids going back many years. But you also -- the other observation is the magnitude of the sinusoids are coming down in terms of how they swing. What I can say is our business is increasingly diversified. If you look at it this year, Amitabh, most of our growth has come from Web 2.0s, a little bit from cable and so forth. So our strategic vertical that these are selected, we selected them for this specific purpose because we think, be it cable, carrier, Web 2.0, financial services, federal government and so forth, these are the verticals that are going to carry us forward. And we think while some are taller than others in terms of growth capability, we think over the medium to long term, they're going to be within -- I mean, some of them are going to be within the same zone. So we see continued diversity of revenue, but we are -- traditionally has been -- we have a carrier base. And as you know, the carrier CapEx cycles do exist, and it's something that we don't control industry spend. And when they swing down, we swing down with them. And when they go up, we go up with them. But now we have these other 3 engines, and we expect them to grow over time. Robyn, any more comments on that? Robyn M. Denholm: I think -- I think that said it well. I think also the question around the router cycles and whether we expect them to continue, I think Rami can comment on that as well.

Sure, happy to. I mean, when I look at the Routing fundamentals, obviously the first thing you're going to look at is traffic. And all our indication is that traffic continues to grow across all of our customer base, whether it be the Web service providers, telcos, cable operators. I think that's sort of a truth among all of them. There are certainly trends that are happening in the industry and where -- these are obviously trends that we factor into our strategy. But when you take a look at these trends, whether they be Layer 4 to Layer 7, service virtualization or things like the virtualization of the CPE, so low-end routing virtualization or SDN for the purpose of automation, I view these as all largely complementary to our strategy. So this is very much an opportunity for us. It's very, very difficult to virtualize the kind of product, or I'd say impossible, to virtualize the kind of product that solved the performance necessary to deal with capacity growth. So this is why I'm confident in the long term.

Our next question comes from Pierre Felgaru (sic) [Pierre Ferragu] from Bernstein. Pierre Ferragu - Sanford C. Bernstein & Co., LLC., Research Division: So you -- on the Service Provider front, you mentioned specific dynamics that prevent the carriers to spend in the near term. I think we understand well what's happening in the U.S. You have like the debate with Netflix. You have like net neutrality. You have ongoing merger discussions. But what was very surprising this quarter is actually weakness of your revenues in the rest of the world. You mentioned China. You mentioned Eastern Europe. You mentioned Japan as well. Could you maybe review these geographies one by one? And maybe tell us -- it's probably like 2, 3 major clients in each of them. So it would be helpful to get some kind of color on what's happening, what's creating this lack of visibility in these regions as well?

Thanks, Pierre, for the question. Actually, we have Vince Molinaro here, who is our Executive Vice President for go-to-market globally. And he -- Vince, can you respond to Pierre... Vincent J. Molinaro: Sure. Thanks. Thanks for the question, Pierre. Like Robyn discussed and mentioned, we were down in Asia Pacific, primarily because of China and Japan, largely due to Service Provider and a few projects that have come to conclusion. And we have seen a first half with some real good design wins that will carry forward. As I look into Q4, in Asia Pacific in general, look to flat to slightly up. And the reason for that, from my perspective, we're seeing growth in Korea, Malaysia, Indonesia, Southeast Asia in general for opportunities for the broad portfolio. Rami just mentioned traffic growth, and specifically in that part of the world, LTE and mobile build-outs presenting opportunities for us in access and aggregation in mobile backhaul as well as traditional core routing, edge and security. I'll also say in Asia Pacific, the focus in our verticals is beginning to resonate with both our Service Provider and Enterprise customers, particularly in Web and in cloud provider, Web 2.0 companies in India and Japan. So hopefully, that rounds out the picture for you. In EMEA -- in EMEA, I kind of characterize that steady as they go. Good first half. And again, as Robyn mentioned, some geopolitical issues in Eastern Europe and Middle East. We are seeing good traction with new design wins across the entire portfolio. And again, that is in our Enterprise and our Service Provider business across all verticals and even in the strategic verticals like higher education, for example. I think the other area of opportunity, the European providers, they typically invest as the demand presents itself. Traffic is growing in EMEA, 38% to 50% in the public space year-on-year. High network utilization, so we're -- and everything mobile. And that growth is pretty significant. So we're watching the industry dynamic and in the geographies that Robyn mentioned. But in general, we just see a steady trend going forward.

Our next question comes from Ehud Gelblum from Citigroup. Ehud A. Gelblum - Citigroup Inc, Research Division: I have a couple of clarifications first before my question. First of all, Robyn, I just want to make sure that you didn't mention anything about the 25% operating margin target for 2015. If you can comment on that, if that's still on the table. You didn't mention it, so I'll put that out there as a clarification. Lower OpEx, another clarification. How much of that, Robyn, was reversal of prior bonus accruals so not really where the OpEx would've been if we were doing it on a quarter-by-quarter basis? Third, Switching, obviously, fell pretty substantially and you gave some good reasons. I'm assuming within that, was it mainly the Web guys you're talking about? Are there other verticals that we should just be aware of in there? Was U.S. Service Provider in there? And then finally, Kathleen, I know you're going to love this, but AT&T CapEx, if you look at what they did, they're obviously a big part of what you guys do and dictate a lot. They were very strong in the first half of the year. They haven't change their full year number, $21 billion. It's still pretty much set in stone. So if you mapped out AT&T's CapEx, wireless, wireline, however you slice it, the fall-off in Q3 and the further fall-off in Q4 was put in stone back in June, July. So what really did change as you went through this quarter that seemed to be a surprise versus what AT&T was telling you? Was it kind of a mix in that CapEx number? Because the total CapEx number seems to kind of flow with what your new numbers are versus what you had previously thought going into the preannouncement.

Thank you, Ehud. So there were 4 comments there, 2 clarifications, 2 questions. We will address 2 of them, 1 clarification and 1 question. Which would like them to be? Ehud A. Gelblum - Citigroup Inc, Research Division: Okay. How about the lower OpEx bonus accruals for the clarification and then the AT&T CapEx question for the question.

Okay. All right. You got it. Thanks, Ehud. Robyn M. Denholm: So I'll address the OpEx question. So you're right, we did have some variable comp in the quarter in Q3. I actually called it out in the script. We said it was about $10 million in terms of the fourth quarter guidance. So in other words, we did 4 -- we're putting out $480 million, plus or minus $5 million. There's about $8 million to $10 million in there that's variable-comp related.

Yes. And Ehud, on your second question, obviously, we don't comment on any specific customer. So let me make my comments general. And in fact, I read your report. Nice job on the one you've just recently written on another customer on the same issue. And I think it's -- you can project that to my comments. So it's -- as you said, it's the mix and where service providers are spending their money and what part of the network. That's the answer to the question, essentially. And so it is -- if that's in the mix area, that is the answer to the question.

Thanks, Ehud.

The next question comes from Jeff Kvaal from Northland. Jeffrey Thomas Kvaal - Northland Capital Markets, Research Division: My question is, I think, a number of us on this side do worry a little bit about the declining R&D budget and to make sure that the pipeline continues to generate the technical -- technically leading products that you are known for. Could you give us a little bit -- perhaps maybe a bit of an example of how you can be reducing the R&D and yet continuing to not suffer from a thinner pipeline of product or maybe an example of where the efficiency has come into play? I think that would be helpful.

So Jeff, I'll make a few comments and I'll then have Rami go into perhaps examples and give more color. So first of all, in the whole Integrated Operating Plan, the issue has been focus and unleashing the talent of our best R&D and go-to-market people for innovation. And so we said that many, many times we know R&D is the lifeblood of this company. In fact, I am super excited about the product pipeline, and this is not infinity. This is in front of us. So you should rest assured that innovation and engineering and product pipeline is healthy. It's broad. It's deep and it's very exciting. And so you should just rest assured on that. And we are very focused to ensure that our R&D is focused on things that matter to the customers, where the market is going and making sure that we are not wasting our resources in areas that don't matter. Now with that, I'll pass it to Rami.

Sure. So most of, if not all, of the changes that we've done in R&D to reduce the OpEx have been structural in nature. So it's around streamlining the organization in such a way that allows us to deduplicate functions that were, quite frankly, duplicated only because of the organizational design. So that has been eliminated. Beyond that, we've been very, very vigilant with respect to the priorities of the projects that we're working on, making sure that everything that we are working on is aligned 100% with the company's strategy. And as you know, we've said in the past, that has resulted in stopping development of certain projects and products. So there are a number of pre-revenue-generating projects we were working on that we are not anymore. We just divested the Pulse business as you are aware of not because it's a bad business but because of the fact that we could not invest in it ourselves and also all of the other things that we know we need to do for the company's strategy. The last thing I will say is now that all of R&D is in a single organization, one of the fundamental tenets, if you will, of our engineering principles is to leverage, is to make sure that all of our technology components across silicon systems and software are highly leverageable across all of our product lines. And this is something that we're very much executing towards.

Thanks, Rami.

Our next question comes from Mark Sue from RBC Capital Markets. Mark Sue - RBC Capital Markets, LLC, Research Division: If I look at the year-over-year decline, it's actually the worst in the company's history since the IPO. And that's despite the company being more diversified and despite traffic growing every single year. So underneath the cyclical nature, is there a fundamental change that's going on in the business? Is there an increasing structural issue, price deflation, shifts in purchasing power by carriers? And if that's the case, we understand you're addressing the OpEx for the cyclical part, but what are the plans to address the structural part?

Mark, yes. I think if you look at the numbers, the overwhelming majority of this is carriers. It's U.S. carriers. It is the cyclical nature of their buying. Once you open that cover, there are things. There are pluses and minuses, and they always are in the business. But I would say is that is the bulk of the issue that we saw in the third quarter. Robyn, do want to say a few words on that? Robyn M. Denholm: No, I think that covered it, actually.

Next question comes from Simona Jankowski from Goldman Sachs. Simona Kiritsov Jankowski - Goldman Sachs Group Inc., Research Division: I wanted to ask a clarification and a question as well. Just on the clarification, again, I wanted to follow up on the question on 25% operating margin for next year, if that's still on the table as your target. And then my question was just digging in a little bit into the Switching business given the sharp decline there. Can you just characterize for us a little bit how lumpy that business is in terms of customer concentration? Are there are a handful of maybe 10%, 15% customers in there that could have driven the magnitude that we saw in the decline?

Simona, I'll take a crack at that quickly at both, and then on the second one, I'll pass it again to Rami. So on the first one, 25% is our long-term target and very achievable. The visibility to revenue, as you know, has turned quite poor just in the past 6 weeks. If you look at the CapEx numbers from the big carriers, you would see how they have changed and it's just the past 4 to 6 weeks. What we control is our costs, our products, our customer relationships, the design wins. All of those are well in hand. We've got great customer relationships, a lot of design wins. On the costs, we have given you definitive numbers, which is 1.9, minus 25, plus 25. And based on our track record on our costs so far, you can see how we come with costs. We're doing that while preserving ability to invest sufficiently for growth despite the challenging environment. And when you put it all together, we're going to expand operating margins significantly from 2014 to 2015. And this is just arithmetic, this number, depends on revenue. And revenue, we don't control industry spend. The visibility is poor. And these things can change very quickly. They have just in the past 6 weeks. And so whenever -- and we are constructive on the second half of this year. We think we're going to get growth back. That will depend on the revenue number. But we are very focused on the 25%, and we think it's very achievable as a long-term target.

Switching. So should I comment on Switching? Simona, so first, to your question about concentration. Certainly, there is a Service Provider and an Enterprise component to our Switching business. The Service Provider component tends to be fewer larger deals, fewer larger customers than the Enterprise. That's natural. We've always said that the SP portion is going to be lumpy, and this quarter was certainly an example of that. And I would say, I mean, we're certainly disappointed in our results. I think we can and will do better in Switching. So what gives me confidence about the Switching business going forward? First is we've had a number of wins that just haven't yet been -- gone into deployment mode that will kick in, in future quarters. Second is if you take a look at what we've done in Switching in terms of R&D over the last couple of years is we've realigned the organization. We've focused it on a new strategy and product roadmap, and we're now just starting to see the results of that strategy in terms of new products like the QFX5100 being a great example of that. But we're not done yet, and there's more to come. And I have a lot of confidence in what is more to come in terms of our ability to grow this business. And I'll just pass it to Vince for performance. Vincent J. Molinaro: Yes. Now let me just tag on that, Rami, with respect to Shaygan mentioned the vertical focus and the customer differentiation. We're starting to see both in Service Provider and in Enterprise a number of data center wins to leverage the portfolio and positioning forward, especially in the massively scalable data centers. We're seeing a much broader architectural conversation that puts us in a position for continued growth. So I'm excited about that. And then automation and everything else that's going on relative to the environment in the data center, looking forward to again diversity and portfolio of customers where we have applications that position and drive our Switching business. So thanks.

Yes. And I'll just wrap it up, Simona, on -- as Rami and Vince said, I'm actually very encouraged by all of the vertical engagements on Switching and the design wins, and I'm enthusiastic about the ramp in 2015.

Thanks, Simona.

Our next question comes from Brian Modoff from Deutsche Bank. Brian T. Modoff - Deutsche Bank AG, Research Division: I think when you [indiscernible] data, about whether it's a cyclical issue in Routing or if there's something more structural going on, like maybe Metro Optical using that Switching substitution for some of the Routing. The question is really around Switching in terms of -- are you guys planning to do a spine switch? This is clearly an area as you note that you could see some growth in because there's still growth in Switching and you have a small market share. So are you going to do anything architecturally to really get into that market in a bigger way?

Brian, we obviously are not here to unveil our product roadmap. Suffice it to say, as I said, I'm personally very excited about it. And I think it was Brian's report we just recently read, right? We have a lot of comments on that, but that we'll take up online. But Rami, can you sort of take that?

Sure. So look, on the Switching side, I'll just sort of revert back to what I mentioned for Simona, which is that we're not done with our roadmap. We're not going to get into specifics on this call. But I think there's a lot of confidence all around on our roadmap. And in terms of talking to customers, we're getting a lot of very positive feedback on the roadmap. As far as Metro optical and so forth, there are certainly trends and aspirations by all involved in that market. And one can easily talk about IP and packet-based transport and so forth, but actually executing on that, there aren't that many people in the world that can effectively do it and do it competitively and scalably. So for that reason, I'm not 100% sure I agree with the premise that there is sort of this long-drawn-out trend, if you will, in Routing where everything will move to optical. I just don't buy it. That said, there are -- there is an evolution in architectures for the Metro and even for the core where the packet and optical demands will be much closely integrated together. This will be done with things like intelligent software that can balance traffic more effectively across the 2 layers, also done in terms of automation. And if you look at our product roadmap with respect to, for example, NorthStar, this is the controller that does exactly that. It does it with our own packet technology and partner optical technology. And we're seeing, actually, some very good interest from customers. I think it's early, but we're seeing very good interest from customers on such approaches.

Thank you. Thank you, Brian.

Our next question comes from Ben Reitzes from Barclays. Benjamin A. Reitzes - Barclays Capital, Research Division: Back to the Switching question. Could you just -- I think the Enterprise number for Switching might have been below as well what you were thinking at the beginning of the quarter. And what did you notice competitively there with Cisco ramping new products and Arista doing things? And I guess everybody's asking 2 questions so I'll sneak one other in there. Strategically, you guys have done the IOP and the Junos divestiture. But is there any other strategic options on the table? Anything else that we should be thinking about that you have in your ability to enhance shareholder value that we haven't looked at yet in the analyst community?

Ben, I'll make a couple of quick comments and then I'll pass it around. So on Switching, we are actually quite bullish about our Switching products. They really shine most where scale, sophistication, openness, programmability, hardened IP protocols, those things matter. And so we are very bullish about that. As far as your comment, your second comment, we -- as a matter of course, we review our portfolio regularly. And we are in that sort of phase and that's how our process works. So we do that regularly. But Rami, can you take a crack on the Enterprise question specifically?

Yes. I mean, most of the -- most of what affected us in Q3 in Switching was on the SP -- in SPs. And that's, again, the lumpier portion of our business. There was also a broader Enterprise component, but it was a minor component compared to the SP side. As far as strategy is concerned, it's scoring and winning more and more of the opportunities from customers that truly, like Shaygan said, value high capacity and high scale. And those customers tend to be fewer and larger. So there is -- again, there's that lumpiness component to it. And then also, like I said, there is still an element to our product roadmap that has not yet been completed that I think is going to be necessary to make sure that we continue the growth. So that's, again, what it is that we're counting on for continued growth in the future. Vincent J. Molinaro: And I think from a customer perspective, you've seen the announcement on Nike, and that is a complete portfolio but switch-led. And those are the kinds of opportunities, when you take the full portfolio, leveraging the competence and capability of our switch platform, allows us to get into these global data centers, allows us to provide the agility both on a converged campus as well as among data centers to deliver private cloud. And so I think that's a good example, and it's just a ramp and replication of more opportunities like that where we either lead with routing and pull the switch or lead with switch and pull the broader portfolio. And then I -- think about MetaFabric in general, which is the Routing, Switching, Security where we can lead with that enabling switch technology and pull through the entire portfolio and create that end-to-end solution for customers like Nike and others that we're winning designs around that need to move into production. But it is that better together with leveraging the switch and pulling the rest. MetaFabric is, I think, a perfect example of that.

Thank you, Ben.

Our next question comes from Kulbinder from Crédit Suisse. Kulbinder Garcha - Crédit Suisse AG, Research Division: Yes. I guess my question is that you spoke before, Shaygan, about the product cycle, and it normally lasts 2 to 4 quarters. I guess the question is -- this still seems to have materially surprised Juniper. So my question is what is it about the product cycle that surprised you this time and transpired because you -- you have had these cycles before, but this seems to have caught Juniper by surprise again. And I guess just linked to that, I guess the secular concern would be is there something more serious competitive going on, and how would you respond to that kind of concern?

Yes, I mean, it's a good question. It's difficult to call cycles. I think there are a number of analysts on this call right now and I'm not sure how many of us would have called the cycle. And as I said, in addition to that, there is some -- what also complicated it in terms of visibility, Kul, is some specific industry dynamics that is going on. So those 2 together make visibility quite poor, and the industry dynamics isn't one. It's probably, I would say, half a dozen things, and some of them are different on one customer than the other. And then on top of it, you've got the sinusoid with noise fluctuating around it. So I guess that's the answer. And we all wished that we could tell the cycles, but we didn't see it and I don't think many people did either. And in fact, just in the past 6 weeks, we've gotten -- the numbers have moved on CapEx as well on the area that we work. Kulbinder Garcha - Crédit Suisse AG, Research Division: I guess just on the...

Rami, do you have any other comments?

Go ahead, Kulbinder, and then we'll take one more question. Kulbinder Garcha - Crédit Suisse AG, Research Division: Yes. I guess my question and follow-up would be just you mentioned the need for revenue diversity. With the new products, the new projects, is it fair to say that you would have the revenue diversity in a year that you can mitigate some of this going forward beyond that? Is that what it might take to get this diversity of revenue that you think Juniper might need? Robyn M. Denholm: Yes. I think, Kulbinder, on the diversity of revenue, that obviously takes time. And we will have more information next week at the Investor Day on how diverse some of our revenue has gotten over the last 5 years. But to Shaygan's point, we obviously still have some concentration around carriers. And in particular, when those cycles happen, sometimes it's very difficult to get visibility into the depth or breadth of the cycle. And so you're right, the depth of it did exceed our expectations in this quarter. But our view is, at this point, that we're planning for the revenue environment to be challenging. And we've obviously taken appropriate actions in terms of the cost structure and making sure that we're investing in the right things over the right period of time to catch the wave of growth as we come out of this cycle. And that's what the team is very focused on doing.

Thanks, Robyn. Thanks, Kul.

Thank you. We have time for one more question.

Our last question comes from Ittai Kidron from Oppenheimer. Ittai Kidron - Oppenheimer & Co. Inc., Research Division: She's been killing names all night. Ittai. Yes. I guess my question is more on the incremental OpEx cuts. Shaygan, Robyn, I know cuts are never easy. And when you first announced the IOP plan, you had a very heavy OpEx structure where, I would say, on a relative basis, it was easier to identify the cost-cutting. I guess the question from here, how easy or difficult is it to squeeze that extra $100 million out there? Are the things that you need to do a bit more structural and longer in duration in order to squeeze those cuts? Or they can be -- whether either easily identified or quickly extracted? Can you give us some color on the process there? That would be great.

Ittai, yes, we did IOP that we announced in February for $160 million. We are doing $100 million more. On a year-over-year basis, it's $130 million more in OpEx savings that we are doing. It's a variety of actions. I would say the first $160 million we did, it was tilted more heavily towards the structural things we need to do. Robyn will talk to some of that. This one, we've got a little bit of that. But we also, in IOP, said that we are also sharpening that muscle of continuous focus on efficiency and ensuring that all our dollars go to where it matters most. And this is a reflection of the company strengthening that muscle to be able to take yet another 130 million year-over-year out while preserving a very exciting R&D portfolio, and as Robyn said, be in a position to catch the revenue growth that, from a planning perspective, we are announcing in the second half of next year. Robyn? Robyn M. Denholm: Yes, I think Shaygan said it well. What we're trying to do -- what we are doing is continuing to sharpen our focus in terms of productivity and efficiency across the company. And as you could see by the results for the third quarter, we are ahead of plan of the original IOP. And so our view is that we can take these additional costs out. It's never easy to take cost out, as you mentioned in your question. But with the right focus and the right attention, doing that deliberately across the company by managing headcount and continuing to work through discretionary expenses as well as effectiveness and efficiency across the company. And the team's very well aligned around that.

Thanks, Ittai. Okay. That is all the time we have today. We'd like to thank you for participating on the call and your great questions, in some cases clarifications. And we look forward to speaking with you next quarter. Thank you. Bye-bye.

Thank you. This does conclude today's teleconference. You may disconnect your lines at this time. Thank you for your participation.