Google Fashion Shopping Site Makes Debut

You know how remote and strange the fashion world is when you come to Google’s headquarters in Mountain View, Calif. For one thing, employees are zipping around the sprawling campus on scooters and bicycles, so that pretty much eliminates platform shoes and minis. And for another, there are way too many snack stations at Google. Fashionistas are funny about food.

But go a couple of blocks from the main building, and the mood and the desk décor are conspicuously more invested in style. One employee, Abigail Holtz, has on an ivory silk mini dress with a plunging neckline and a pair of high heels. A colleague, Marissa Goodman, is more casual—but no less savvy. She used to design women’s clothes for Old Navy and Esprit.
In a deliberate collision between nerds and fashion mavens, Google has created a new e-commerce site that significantly improves how fashion is presented and sold online. The site, Boutiques.com, which is expected to go up Wednesday morning, may also change how people shop for clothes.
Boutiques.com has so many capabilities and components that even Google engineers have a hard time qualifying it. It is a collection of hundreds of virtual boutiques merchandised — or, in the new parlance, “curated” — by designers, retailers, bloggers, celebrities and regular folks. You can shop in the style of, say, the actress Carey Mulligan or Mary-Kate and Ashley Olsen — among the celebrities who signed up for the launch — or you can build your own boutique and amass followers who can comment on your taste.
It is a place, then, to show off your fashion acumen, much as millions of Polyvore users already do with their picture collages.
It is also a source of inspiration. In every boutique on the site, there are dozens of additional choices inspired by a designer’s or celebrity’s style — generated by algorithms — with product photos that are much larger and sharper than on other shopping sites. And if you don’t know how to wear the leopard pumps you just bought, there’s a panel of street-style photos on the right side of the site that visualizes the shoes in more expressive modes. Indeed, whatever your style preference — classic, romantic, casual — the inspiration panel automatically adjusts for them, like a support group that can read your mind with surprising precision.
And that may be Boutiques.com’s ultimate game-changer — how precisely it analyzes your preferences to give you what you requested. As many online shoppers know, search engines tend to give you stuff you don’t really want. A request for fern-colored shoes might yield fern shoes, plus fern-print blouses. But, as two experienced online shoppers found when they tested the site earlier this week at Google’s New York office, if you ask for cobalt blue shoes, you get them. And if you refine your preferences with a click or two, you get even more specific styles.
The process at Boutiques.com is accomplished through visual search technology, and what style experts like Ms. Goodman and Ms. Holtz conveyed to Google code writers about the nuances of fashion — from color and pattern to silhouette and what looks good together and what does not.
The technology was actually developed by Like.com, a Silicon Valley company co-founded by Munjal Shah, which Google acquired last summer for a reported $100 million. Before the purchase, Like.com had created a number of fashion e-commerce sites, including Covet.com, and the styling tool Couturious.com.
“I’ve always been impressed with Like.com,” said Sucharita Mulpuru, a vice president and retail analyst at Forrester Research, who is familiar with the work on Boutiques.com. “I was just floored by the technology back then and it’s evolved since. They’ve just honed the algorithm.”
Fueled in part by new gadgets like the iPad and more dollars spent by retailers on technology, online sales have generally outpaced brick-and-mortar sales. “I feel e-commerce in the last 12 months has caught a second wind,” Ms. Mulpuru said. According to Forrester, Internet sales of apparel and accessories this year will account for 14 percent, or $25 billion, of the $173 billion that Americans will spend online.
Mr. Shah is the team leader for Boutiques.com, with a left-brain-right-brain group of technicians and tastemakers. As he said in interviews conducted over the past week, “Online fashion shopping has to be universal and curatorial at the same time. This is an answer.”

A number of big companies, most notably Amazon and eBay, have been trying to get a bigger slice of the online apparel pie. But while they have improved the stylishness of their fashion pages, they may be ultimately constrained by their somewhat static platforms. It’s hard to mix DVD players and $900 Christian Louboutin peep-toe pumps. 

Meanwhile dedicated fashion sites like Shopstyle.com have gathered fans.
“Shopstyle’s done one of the best jobs in my opinion of creating the right high fashion experience,” Mr. Shah said. “But we think of it as layer one. It’s kind of broken things down, but they didn’t go for a detailed categorization and they didn’t personalize.”
At the time that Mr. Shah and his team at Like.com created Covet.com, a fashion personalization site, they didn’t really have the full picture of all that was possible on the web. “We captured your preferences but we couldn’t analyze the items to see if they your met your preferences,” he said. “We did one half but you need both halves. We achieved the other half only by rebuilding the technology with a whole new way of analyzing patterns and silhouettes.”
In simple terms, what the style experts did was come up with about 500 words for color, shape and pattern — robin’s egg blue, for instance, and gingham — and then the engineers trained the algorithm to know what each was. They would have pictures of what gingham was and what gingham wasn’t. “We did that word by word by word,” Mr. Shah said. The trouble is a lot of sites don’t have, or use, vision technology. They end up stuffing in a bunch of key words and the search engine gets confused. So you get fern-print blouses when all you really want are fern pumps.
Despite the amount of products that a search on Boutiques.com kicks out, the download time is very fast, and choices appear on extra-long pages so you don’t have to keep clicking. Virtually every kind of information is analyzed — price, brand, color and so on. The site also includes a system called “Complete the Look,” for which Ms. Goodman wrote “a ton of rules,” Mr. Shah said, “and our computer vision and machine learning guys implemented them.”
Additionally, there is a good sense of discovery on Boutiques.com; items come to your attention, almost as they do in stores, that you didn’t necessary plan to buy. Seasonal trends, like fall’s military looks, can be boosted on the site. Again, to Ms. Malpuru, “that’s where Munjal gets it—fashion is about discovery.”
Users of the site will have the option to take a personal style quiz, which ranks a broad spectrum of loves and hates, but Mr. Shah is convinced that most people will prefer to find their “style twin” and shop in that individual’s boutique.
Among the designers who signed up are Tory Burch, Oscar de la Renta, and Isaac Mizrahi, who for one plans to offer signature pieces like a plaid cocktail dress and a military coat.
Bloggers include Bryan Boy and Rumi Neely of Fashiontoast. Other celebrities, which Google said it has paid to host boutiques, are Claire Danes, Ashlee Simpson and Nicole Richie. (By the way, Google authenticates the celebrity boutiques, so imposters be warned.) Retailers include Neiman Marcus, Barneys, Shopbop, Net-a-Porter, and Scoop NYC. Potentially, fashion magazines could have boutiques. So could a character from a television show.
“Somebody may come in and built 20 goth boutiques in the first day,” Mr. Shah said.
On Monday, Simone B. Oliver, a Web producer for The New York Times, and Jane Son, a publicist, gave the site a test drive for this article.
As the two women, both avid online shoppers, slowly got used to the site — which they later admitted could be overwhelming — Ms. Oliver said, glancing at the product categories, “I’m curious why you put shoes first.”
Ms. Goodman replied, “Shoes are always one of the most popular categories.”
Ms. Oliver laughed. “Good answer.”
As Ms. Son typed in “duck boots” on her laptop, yielding a bunch of choices, Ms. Oliver searched for cobalt platform pumps and a black leather shift dress, quickly finding things she liked. She skipped over the celebrity boutiques, preferring the blogger and trend boutiques: “I get more inspiration from girls on the street than celebrities.”But, she added, “I think the celebrity choices are useful for body shape. There are some people in the limelight that have a similar body shape to mine, and a certain silhouette would look good on me. And shades of skin tones.”
Mr. Shah nodded. “We could add skin tones to the preferences.”
Both women liked the inspiration panel, and also how refined the search was.“On other sites, you can’t edit your choices as much,” Ms. Son said. “And I need an edited selection.”Ms. Oliver rated the sense of discovery “A-plus,” adding “I just found out about a shoe brand that I had never heard of—Velvet Angels. They’re more in my price range than, say, Yves Saint Laurent. That was fun. I was looking for something else and it popped up.”
She said to members of the team, “I love that you guys have so many options but you also have the options that make sense.”
Yet, after the meeting, both women identified an obvious shortcoming of Boutiques.com: as curated as it is, a lot still comes up in a search. Suggesting that too much information may be a turnoff to inexperienced Web shoppers, Ms. Son said, “It’s going to take some getting used to, that’s for sure.”
Nodding, Ms. Oliver said, “I feel it’s an amazing site but there are a few aspects that are not very intuitive. Some people might go back to the regular Google search and look for their boots.”
Fortunately for them, the site has that option, too.

http://www.nytimes.com/2010/11/18/fashion/18googlefashion.html?

Chi chít thủy điện ở miền Trung

TT - Chỉ với hai tỉnh miền Trung là Quảng Nam, Thừa Thiên - Huế cùng hai tỉnh ở Tây nguyên là Kontum và Đắc Nông mà có gần 150 dự án thủy điện lớn nhỏ đã, đang và sẽ triển khai. Dân khoa học ví von việc triển khai tràn lan các dự án thủy điện ở miền Trung và Tây nguyên là đánh cược với thiên nhiên, mà phần thua chắc chắn thuộc về con người.

Tan hoang rừng già

Ngày 18 -10, chúng tôi vào công trường thủy điện A Lưới (huyện A Lưới, Thừa Thiên - Huế) bằng tuyến đường công vụ còn nguyên dấu đất bùn đỏ. Con đường mới mở nhỏ và rất dốc men theo sườn núi cao với những cánh rừng già bạt ngàn, bên dưới là vực sâu hun hút. Để có được con đường này, đơn vị thi công đã tìm mọi cách đốn bỏ hàng chục hecta rừng già mà dấu tích để lại là những gốc cây to tươm nhựa đỏ bầm như máu.

Khoảng 2km trước khi đến công trường nhà máy là một cảnh tượng hùng vĩ hiện ra gồm ba tầng bậc taluy đỏ thẫm, loang lổ, xé toạc những thảm xanh nối tiếp của rừng. Trên suốt chiều dài 10km của tuyến đường công vụ có rất nhiều thân gỗ lớn đã bị đốn hạ còn trơ gốc ven đường. Cạnh đó, nhiều phách gỗ đã được xẻ chờ cơ hội chuyển đi.

Càng đi sâu vào trong, những phách gỗ đỏ tươi nằm ngổn ngang càng nhiều... Ông Lê Viết Ngọc Vinh, hạt trưởng Hạt kiểm lâm A Lưới, cho biết chưa thống kê được chính xác bao nhiêu hecta rừng bị triệt hạ để phục vụ công trình này, nhưng nó là một mối nguy lớn cho tương lai.

Trên trục đường Hồ Chí Minh đoạn qua xã Mà Cooih (huyện Đông Giang, Quảng Nam), dù lòng hồ đã được dọn dẹp, khai thác, tận thu gỗ trước khi tích nước đúng kế hoạch, thế nhưng vào những ngày này trên con đường công vụ dẫn vào đập chính của thủy điện A Vương, cảnh những khoảnh rừng xanh thẳm đang bị chìm dần trong lòng hồ khiến nhiều người xót xa.

Sau trận bão số 9 vừa qua, hàng ngàn thân gỗ lớn từ khắp các thượng nguồn theo nước trôi về nằm lềnh bềnh trên mặt hồ. Ngay tại khu vực cửa nhận nước nhà máy, nhân viên vận hành phải vất vả dùng những thùng phuy rỗng kết lại với nhau tạo nên dây phao nhằm ngăn thân gỗ, củi tràn vào cửa nhận nước phá hỏng nhà máy.

Trận lũ vừa rồi gỗ trôi về lòng hồ nhiều đến mức ông Nguyễn Văn Lê - chủ tịch HĐQT Công ty cổ phần Thủy điện A Vương - phải thốt lên: “Nếu cảnh này còn tái diễn thì chừng mười năm sau hồ A Vương sẽ không còn nước để phát điện”.

Theo ông Lê, mới sau một cơn lũ mà lòng hồ thủy điện đã bị bồi lắng một lượng bùn dày ước đến 10m. Như vậy không lâu nữa lòng hồ A Vương sẽ bị bồi lắng lấp đầy, do vậy không còn chỗ chứa nước để phát điện. Theo ông Lê, nguyên nhân chính do rừng đầu nguồn mất quá nhiều nên không giữ được nước dẫn đến xói lở đất rừng.

Một giọt nước qua bốn cửa tuôcbin

Với chiều dài trải dọc từ Kontum đến Quảng Nam trước khi hợp vào sông Vu Gia đổ ra biển, sông Đăk Mi với nhiều tên gọi như sông Cái, sông Nước Mỹ hay sông Bung được đánh giá là hệ sông có tiềm năng thủy điện lớn nhất Quảng Nam.

Trên con sông Đăk Mi dài chưa đến 100km dự kiến ban đầu chỉ có hai nhà máy thủy điện là Đăk Mi 1 (225MW) và Đăk Mi 4 (210MW). Tuy nhiên một thời gian sau, dự án Đăk Mi 1 đã bị “xẻ” ra thành ba dự án theo bậc thang gồm Đăk Mi 1 (58MW, nằm trên địa phận tỉnh Kontum), Đăk Mi 2 (90MW) và Đăk Mi 3 (45MW).

Riêng thủy điện Đăk Mi 4 vẫn giữ nguyên quy hoạch. Cả ba dự án này đều nằm trên địa phận huyện Phước Sơn (Quảng Nam). “Với quy hoạch theo kiểu dày đặc như vậy thì một giọt nước từ nguồn chảy về đến biển phải qua bốn cửa tuôcbin” - ông Dương Chí Công, giám đốc Sở Tài nguyên - môi trường Quảng Nam, ví von. Theo báo cáo, đến thời điểm này toàn tỉnh Quảng Nam đã có tổng cộng 62 dự án thủy điện được phê duyệt, trong đó riêng trên sông Vu Gia - Thu Bồn là 10 dự án. 47 dự án thủy điện vừa và nhỏ khác cũng đã được địa phương này cho phép lập nghiên cứu, đầu tư.

Trong khi đó tại Thừa Thiên - Huế, một loạt nhà máy thủy điện nằm trên các nhánh của sông Hương đang được thi công ồ ạt nhằm kịp tiến độ phát điện vào cuối năm 2011 như thủy điện Bình Điền (44MW), A Lưới (170MW), Hương Điền (81MW), Thượng Nhật (6,5MW). Theo ông Nguyễn Duy Thành - phó giám đốc Sở Công thương Thừa Thiên - Huế, tỉnh đã thông qua quy hoạch phát triển hệ thống thủy điện vừa và nhỏ của địa phương từ nay đến năm 2015 bao gồm 12 nhà máy, trong đó riêng trên sông Bồ có bảy dự án.

Đầu tư thủy điện: ngon ăn, ít rủi ro

Hơn hai năm trở lại đây, phong trào đầu tư vào các dự án thủy điện đang trở nên rất “hot” khi hàng loạt doanh nghiệp trong và ngoài ngành ồ ạt nhảy vào mà không cần chờ địa phương tiếp thị.  

Theo một cán bộ tài chính trong ngành điện, việc đầu tư thủy điện hiện đang rất “ngon ăn” bởi hiệu quả kinh tế cao, thu hồi vốn nhanh. Thủy điện công suất càng lớn, địa hình tốt thì suất đầu tư thấp. Cũng theo vị này, với suất đầu tư bình quân 25 tỉ đồng/MW thì một dự án chỉ từ 8-10 năm là thu hồi vốn.

Một chuyên gia ngân hàng cho biết phần lớn dự án thủy điện luôn được các ngân hàng ưu ái tài trợ vốn. Hợp đồng giải ngân nhanh và thuận lợi hơn các dự án khác. Lý do là đầu tư thủy điện có lợi nhuận trên vốn cao và ít rủi ro, lãi suất thu được ổn định.

Bà Trần Thị Oanh, tổng giám đốc Công ty cổ phần Sông Côn, cho hay mới gần hai tháng phát điện nhưng thủy điện Sông Côn 2 đã đưa lên lưới hơn 60 triệu kWh. Với điện lượng hơn 210 triệu kWh/năm thì sau chín năm thủy điện Sông Côn 2 sẽ thu hồi 1.050 tỉ đồng đã đầu tư trước đó. Trong khi đó một cán bộ của thủy điện A Lưới cho rằng nếu kịp phát điện trong quý 1-2011 thì chỉ sau chín năm nhà máy sẽ thu hồi đủ 3.234 tỉ đồng đã đầu tư.

Điều đáng nói, theo cán bộ thủy điện A Lưới, trong khi các nhà máy thủy điện lớn như Hòa Bình, Trị An... đều phải giảm công suất vào mùa khô kiệt (từ tháng 9 đến tháng 12) do hết nước, thì ngược lại thời điểm này miền Trung đang vào mùa mưa, các nhà máy điện đều chạy hết công suất nên rất kinh tế, lợi nhuận mang lại rất cao.

Sức hấp dẫn từ các dự án thủy điện cũng đã thu hút nhiều doanh nghiệp ngoài ngành điện quan tâm. Mới đây, Công ty Phú Thạnh Mỹ đã khởi công thủy điện Sông Bung 4A với tổng vốn đầu tư 1.200 tỉ đồng.

Tương tự, Công ty xây dựng tổng hợp Trường Thịnh (một đơn vị tư nhân ở Quảng Bình) tham gia 308 tỉ đồng đầu tư thủy điện La Trọng (thượng nguồn sông Gianh). Tại Bắc Trà My (Quảng Nam), Công ty cơ khí áp lực Mạnh Nam đầu tư thủy điện Tà Vi. Trong khi đó tại huyện Phước Sơn, Tổng công ty Đầu tư phát triển đô thị và khu công nghiệp VN làm chủ hai dự án thủy điện Đăk Mi 4 và Đăk Mi 4C.

Qua huyện miền núi Nam Giang, Tổng công ty cổ phần Xây dựng điện VN làm chủ đầu tư các dự án thủy điện Đăc Pring, Sông Bung 4...Tổng công ty Cơ điện nông nghiệp và thủy lợi đầu tư thủy điện Đăk Mi 2, Công ty cổ phần Đạt Phương đầu tư thủy điện Sông Bung 6, Công ty cổ phần Hoàng Anh Quảng Nam thì thủy điện Sông Cùng hay Công ty cổ phần Xây dựng 699 là dự án Trà Linh 3... 

NHÓM PV MIỀN TRUNG

http://tuoitre.vn/Chinh-tri-xa-hoi/Moi-truong/343699/Chi-chit-thuy-dien-o-mien-Trung.html

 

The IT component in insurance industry performance

Top companies in our survey grind down costs by consolidating IT and rationalizing processes. They’re growing faster as well.

The Great Recession has caused plenty of pain in the European insurance industry. Rock-bottom interest rates have reduced returns on the insurers’ holdings, while economic disruption continues to increase the volume of claims payouts.
Not all companies are suffering, of course; indeed, many are thriving. For insights into the factors that distinguish the top performers from the rest of the pack, we surveyed more than 80 life insurers and property-and-casualty (P&C) insurers across Europe.1 We were particularly interested in identifying the factors that most directly contributed to their superior cost performance, including product development, marketing, sales support, operations, and IT, as well as other support functions. Despite the strong economic headwinds, we found not only that companies in the top quartile of our survey outperformed their peers on costs but also that this operational strength—contrary to conventional wisdom—produced more robust revenue growth.
A yawning gap in costs

One of the more startling survey findings was the size of the cost difference between top- and bottom-ranked players. For top-tier insurers, the total operating cost per policy, across virtually every business function, is at least half that of companies at the bottom of the stack (Exhibit 1). Bottom-quartile players’ unit costs can be twice as high as those of top-quartile ones.
Some of the biggest differences result from the top companies’ much better cost management in sales support, operations, and IT. The consistency of this gap also runs contrary to the industry’s common wisdom: companies may squeeze costs in one area but must compensate elsewhere, with higher costs, to maintain revenues or buttress customer satisfaction. Instead, we found that top-tier insurers sustained their low-cost position without sacrificing top-line revenue growth. Leaders in our survey increased their gross written premiums—the industry’s standard revenue measure—two percentage points faster than the broader market index did.
The survey also calls into question other industry credos (Exhibit 2). The first is that larger insurers have an implicit cost advantage. Our survey data show that while size provides some economies of scale, it often creates administrative and operational complexities that impede even larger efficiency gains. Second, it is often assumed that variations in the wages and regulatory regimes of different countries strongly influence the performance of the insurance companies that do business in them—in other words, that high wages and stiff regulations depress profits. Yet our findings indicate that top performers use operational excellence to dampen the effects of high wages and regulatory constraints. By moving back-office functions to cheaper near- and offshore locales, these companies deftly use labor cost arbitrage to create a lower average cost base. As a result, their cost performance is superior across all geographies.
Finally, we found that the best operators, despite their vigilance on costs, don’t allow that frugality to damage vital functions, such as customer service and due diligence for issuing new policies. When one large insurer found that its cost base was significantly higher than that of industry leaders, for example, its managers attributed the difference to the additional spending needed to ensure longer-term growth and profitability. The company assumed that its lower-cost competitors cut corners in areas such as risk assessment and undertook fewer medical checks to authenticate claims or confirm eligibility. Yet our analysis shows that low-cost leaders did not cut back in these areas and actually achieved higher levels of profitability than the company in question.
Fragmented IT and operations raise complexity and costs. The top-tier companies did more for less thanks to their operational excellence. In fact, among P&C insurers alone, top-quartile players’ profit margins, based on the industry standard combined ratio, were four percentage points higher than the average of the companies we surveyed.
How low-cost companies do it

Behind the low scores of our survey’s poor performers, one factor predominates: operational weakness. These companies are more likely to have an ungainly operational footprint, with limited cohesion among operating centers and fragmented IT systems, so they find it harder to streamline processes, share common tasks, and achieve economies of scale. Such deficiencies, which often disrupt a range of processes, lead to high backlogs, as well as frequent work-arounds and manual interventions—factors that slow down response times, raise costs, and hurt customer service.
We have identified three core operating principles that differentiate top-performing companies. These offer a roadmap for insurers seeking to improve their competitive cost position and growth prospects.
Create a centralized structure
Top performers are more likely to centralize their IT and back-office processes and to consolidate corresponding reporting lines under a single point of contact, often the COO, improving both productivity and governance. They are more likely, for instance, to pool medical experts instead of segregating them by location and channel. This approach reduces the number of resources needed—in some cases by as much as 40 percent—while increasing the quality of the guidance.
Taking this approach, one insurer—ranked in the bottom quartile back in 2007—began a three-year transformation to consolidate its back-office support functions and associated IT platforms and to shrink a sprawling network of over two dozen operational centers. These moves shaved 27 percent and 25 percent from the cost of operations and IT, respectively, and bumped the company from the bottom to the upper-middle quartile in cost performance. The savings were reinvested in several new product initiatives, which eventually raised the company’s market share in several key areas.
Ramp up automation
Top performers make wider use of automated processes, often deploying self-service systems that help customers, insurance agents, brokers, and independent financial advisers process simple transactions (such as low-risk claims) with minimal resort to adjusters.
One P&C insurer cut its processing times by 30 to 40 percent when it automated high-volume, low-risk claims for the replacement of damaged vehicle windshields and the like. Using computer-based, straight-through processing techniques that combine separate steps into an integrated work flow, the company’s process is now completed electronically, without manual intervention. Another insurer, faced with rising claims losses, decided to automate its loss-management process. By standardizing decision making through rules-based programs about eligibility requirements and new compliance guidelines, the company reduced its leakage from payment errors by 4 percent of revenues. These savings, which allowed the insurer to keep down prices for its most popular products, gave it a leg up on the competition and helped raise its revenues.
Use lean process design
Many low-cost leaders have applied lean-manufacturing techniques to alleviate backlogs and improve response times. That design focus allowed these companies to streamline their core activities across the business, to eliminate redundancy, and to route tasks and manage work flows more effectively.
One life insurer, for example, seeking to help its brokers, revamped the redemption process to speed up payout times, a key driver of customer satisfaction. A diagnostic revealed several problems, including frequent back-and-forth steps needed to complete documentation and a system that often jumbled together urgent and routine requests, making it hard for adjusters to prioritize tasks. The result was often weeks-long delays in the vetting and approval process.
In response, the insurer created two separate work streams, one for administrative or routine requests that were less time sensitive, the other for payouts. The company also established strict performance targets governing the payout timetable. For each step in the process, it set time limits, such as the number of days a given piece of documentation could rest with an individual agent or process owner. These changes shortened the payment process to 8 days, from 18, while improving overall productivity by 25 percent. In addition, the company now measures customer satisfaction, obtaining feedback directly from the broker network and using that information to further refine the process.
As our survey shows, organizations that take the steps we recommend stand a strong chance of outperforming their peers. While these findings center on the insurance industry, we believe that the same thinking can be applied to other service industries in which IT and operations can play a key role in speeding time to market and improving the way companies meet their customers’ needs.
https://www.mckinseyquarterly.com/Financial_Services/Insurance/The_IT_component_in_insurance_industry_performance_2700?gp=1