Better to create than predict - from Silicon Valley

臉書隨筆：柯文哲的荷蘭病

2023-12-29T17:25:00.000-08:00

以下是臉書上想法的轉戴。也是我對高科技業文化的看法。只是遇上了柯文哲的囈語，順便寫下來。

柯文哲真的了解現在世界科技、資訊及資金流動的速度和複雜度嗎？這些流動的東西要能夠沉澱形成有用的效益，很大程度在於所在社會的文化、制度及成員的組成。這更不是「柯學」舌燦蓮花能達成的。

台積電十年以前也沒有今天的影響力。除了台積電自己的努力，還有一半是因為IBM, Intel, Global Foundry 的自滿和怠惰。很有可能十年後有其它的競爭，造成台積電的影響力減弱。這不是什麼荷蘭病。這只是資訊透明及資金充分流動的國際社會一定有的現象。只要文化及制度不變壞，競爭力就會在。

高科技產業沒有永遠的優勢，只有永恆的競爭。政府的工作不是形成什麼了不起的政策。最重要的是協助產業 “stand on their feet, yet stay on their toes”。

沈榮欽老師已經解釋了為什麼「荷蘭病」不適用於此。我的看法有別於此，並不是訴諸權威。

高科技產業的核心是 “Only the paranoid survive”。Intel或是IBM加Global Foundry 的人力及財力與台積電在伯仲之間。現在只是公司文化與半導體代工不甚契合。但是這些公司的人才技術都在。如果這些公司有心改變，三到五年之內，就有可能趕上。與台積電分享最尖端的巿場。即使台灣和台積電不犯任何錯誤，這還是有可能發生的。

高科技的人才及資金是全球流動的。而知識是開放的。即使台積電競競業業，也不能保證它的優勢持續存在。更不能保證沒有替代性技術，在「一夕之間」抹殺過去累積的經驗及資本優勢。

台灣目前在半導體上的優勢，以及與國家整體產業的分布，在長一點的時間尺度上，根本不算什麼。怠惰自滿（complacency）是高科技產業的癌症。那才是高科技業或科技政策規劃者要注意的。

静悄悄的與虎謀皮

2020-11-04T00:00:00.279-08:00

杭特拜登的渤海華美(BHR)到底幫了中共多少忙？我試著不用媒體報導來分析？尤其不用歐美媒體新聞為資料來源？需要的資料都在渤海華美的網站上(http://www.bhrpe.com/)，相關公司的新聞稿和産品資料，以及美國政府的公開文件。

第一個案例是在2015年協助中國航空工業集團公司(AVIC)購買Henniges Automotive (http://www.hennigesautomotive.com/).

中國航空工業集團是個什麼公司？先來看看以下的公司及産品。

成都飛機工業集團

殲-10 - 解放軍空軍數量最多的第四代戰機
殲-20 - 所謂的第五代隱形戰機

瀋陽飛機工業集團

殲-11
殲-15 - 解放軍海軍遼寧號及山東號航空母艦艦載戰機
殲-16 - 四代半重型長程戰機
殲-31 - 研發中。與USA F-35極為相似。可能繼殲-16成為下一代解放軍海軍艦載戰機

陝西飛機工業集團

運-8
運-9 - 最近繞臺的反潛機、預警機、電偵機就是以此機為機身。

西安飛機工業集團

轟-6 - 繞臺及劍指關島的中程轟炸機
轟-20 - 研發中隱形轟炸機
運-20 - 相當於US C-17的運輸機

哈爾濱飛機工業集團

直-9
直-19
直-20 - 仿UH-60 Black Hawk

這些公司的共同點就是他們都是中國航空工業集團的子公司。中國航空工業集團就像是蘇聯時期的米高揚(Mikoyan MiG-25, MiG-29, MiG-31米格機)設計局，或蘇愷(Sukhoi Su-27, Su-30, Su-35)設計局(也稱蘇霍設計局)，及下屬的飛機製造廠。美國是不可能讓米高揚或蘇愷來美國購買軍事科技的相關産業。

Henniges Automotive是一個汽車零件公司。同時也是一個軍民兩用的科技公司。Henniges的軍用科技一是防水(sealing)組件，一是防震(anti-vibration)附件。有一説是因為殲-10座艙下雨漏水。這種農場文傳說無法用來作有效評估。但是防震附件是減少噪音，增加武器發射平台穩定性的重要技術。這對潛艦，坦克，遠程火箭發射車，海軍艦炮設計製造上，有重要影響。所以這項中國航空對Henniges買公司的交易，是必須經過美國聯邦政府Committee on Foreign Investment in the United States(CFIUS)的審核。

CFIUS由九個政府單位派代表組成。其中對此案有影響力的是商務部，國防部，及國務院。從2013年的南海仲裁案起，美中在亞洲對抗的形勢就已經展現。到了2015，歐巴馬政府已經開始重返亞洲。對於中國的軍事發展已經有了戒心。如果美國不能賣軍事科技給像米高揚或蘇愷這樣的公司，渤海華美用什麼理由說服國防部和國務院對這個案子點頭，同意讓中國航空購買美國軍事科技產業？

另一個投資案中廣核(中國廣核集團有限公司 China General Nuclear Power Corporation, CGN)。

世界上有六個擁有核子動力潛艦的國家，美、英、法、俄、中、印。但是現下只有法國及美國擁有核子動力航空母艦。而法國的唯一一艘核子動力航空母艦，戴高樂號，只有美國尼米兹級核子動力航空母艦四成的大小(約莫四萬噸)。輸出動力只有美國尼米兹級航空母艦的三分之一。也就是說現在只有美國有技術開發製造八萬噸以上的核子動力航空母艦。為什麼不把現成的核子動力潛艦的引擎拿來加大，或多放幾個引擎，來造新的航空母艦？現在最大的核子動力潛艦，除了俄國的颳風級，最大也不過兩萬多噸。這種核子動力引擎，拿來核子動力航空母艦上用，引擎放滿了，就沒有空間放飛機了。這就是為什麽法國核子動力航空母艦大不起來。也是為什麼解放軍山東號航空母艦及後續艦隻不是核子動力，實在是力有未逮。

美國核子動力航空母艦的引擎，用的不是商用核能電廠的設計。它的燃料棒用不同的配方及製程，所以濃度更高。反應爐的設計因此也有差異。這種反應爐的單位體積輸出功率大於一般的核子反應爐。這是為什麼美國能作到其他國家作不到的核子動力航空母艦。

渤海華美在2015年11月投資了中廣核。第二年(2016)就爆發了中廣核間諜案(https://www.justice.gov/opa/pr/us-nuclear-engineer-china-general-nuclear-power-company-and-energy-technology-international)。案件在2016年底結案。涉案嫌犯認罪合作。這件案子在辦案過程中，或有一些起伏或爭議。但是單看這案件牽涉的技術，主要是西屋公司(Westinghouse)的燃料棒成份及製程，以及微型化的反應爐設計。而西屋公司就是尼米兹級航空母艦的核子動力供應商。

從公開的資料看不出來渤海華美在中廣核投資案後參與了什麼核能産業的活動。但是事涉國安級的軍事技術，在2016年之後中廣核已經是美國政府盯上的公司。渤海華美並沒有因此從中廣核撤資。現在世界上擁有核子動力艦艇的國家，除了印度，都有商用核能的科技公司。原因在於研發軍用核子動力技術太昂貴，一定要靠民用商業收入來支撑開支和維繫研發人員。這些公司與所在國軍事主管部門多有密切關係。渤海華美投資的中廣核，是一個有解放軍背景的重點公司。渤海華美明知故犯。這個投資決定如果不是為虎作倀，就是與虎謀皮。

其實這些新聞並不是只有保守派媒體報導，自由派媒體過去一樣有相關報導。畢竟媒體不是智庫，要不要作深入分析，在於媒體公司的立場及商業考量。渤海華美還有很多對中國的投資案。如果你對國際石油化工産業有興趣，可以看看中國石化的投資案。如果你對戰略礦産物資有興趣，可以看Tenke Fungurume Mining的案子。對財經投資有興趣的話，那裏也有幾個證券商及銀行的投資案可以研究研究。

杭特拜登在去年十月宣布將辭去了在渤海華美的職務。渤海華美在中國的投資界裏，也不能算是一家大公司。如果一家小公司能在短短數年内，精準投資，協助中國從事一些國家級的大案子，並且通過美國政府審核，這是一個令人訝異的故事。中國航空Henniges的案子展現了渤海華美對美國政府的影響力。中廣核的案子代表渤海華美的態度。那些嗑藥或私生活的照片，跟這種國安級的問題，完全不能相比。西方主流媒體對此默不做聲，不也是共犯。

#西方媒體的沈默

讀汪浩先生「借殼上市」一書有感

2020-08-23T16:51:00.007-07:00

汪浩先生的「借殼上市：蔣介石與中華民國臺灣的形塑」一書，以蔣介石的日記為軸心，加上中華民國國史館史料及美國官方文件，描述了自1948到1971這段「中華民國來臺灣」，「中華民國在臺灣」，而至「中華民國是臺灣」的歷史。

蔣介石的日記在解密之初，只有少數人能够接觸。其中一人是黄仁宇先生。黄仁宇先生的「從大歷史的角度讀蔣介石日記」，只談到約莫1945年二戰結束。也許是黃仁宇先生的個人經歷及學術興趣，對蔣介石日記的最後二十五年，並無著墨。或許當時並不是所有的日記都已公開，黄仁宇先生也礙難從筆。

在這本書出版之前，早已有網路報章陸續登載汪浩先生的一部分研究所得。我對部分內容已略知一二。書中從蔣介石違憲「復行視事」，行軍事政變之實。韓戰救臺，第一及第二次臺海危機，最後一次的反攻大陸，到退出聯合國。歷經杜魯門，艾森豪政府到尼克森及季辛吉的聯中制俄。最引我注意的反而是「蔣介石運來臺灣的黄金去哪了」一章。

蔣介石下令準備將黃金運往臺灣的時機，1948年11月，就在徐蚌會戰之中。蔣介石在那時已作最壞的打算。不知白崇禧是否知情，如何是想。一場投入數百萬陸軍的大戰還未結束，而蔣介石已經在準備錢打下一場戰爭了。我受黃仁宇先生著作影響甚深。寫史論事，若無財税金融、統計數字以輔，非常容易流於個人意見，爾言我語。汪浩先生經由史料估算蔣介石帶來臺灣的黃金，在不到一年的時間，已泰半用盡。在財政上抓襟見肘，企盼美援。蔣介石一生都在籌錢。黃埔建軍跟蘇聯借錢。北伐時，對內跟上海、湖北商會，對外與英美募款借錢。其中細節可見於宋鴻兵先生的「貨幣戰爭」。之後更是一路靠美援。蔣介石的一生革命事業，可以寫成一本革命財政史。汪浩先生是財務金融的專家。只是不知道他有沒有這樣的計劃。

一方面有感於史料解密，對化解社會爭議及分裂的正面貢獻。但是也要感謝汪浩先生這樣有心人，才能讓史料有聲音。這本書呈現了一個由蔣介石本人的視角。與美、日周旋，從反共復國，到救亡圖存，而至1971年的蔣介石版一邊一國，「中華臺灣共和國」。這是現今臺灣處境的時空背景，也是臺灣多年至今，對自身生存經歷的掙扎奮鬥歷程。其實自由民主法治臺灣的存在，就是對中共最大的挑釁。蔣介石甚至蔣經國都了解，唯一不挑釁的方式，只有投降。蔣介石所為，也不過孫文所囑「聯合世界上以平等待我之民族，共同奮鬥。」只是真平等無法及時降臨，政治人物只好隨機應變。而我一介平民，只能解囊購書一讀，以應汪浩先生著書之義。

5G 的世代

2019-02-01T15:12:00.003-08:00

最近5G在臺灣是個熱門的話題。看了一些近似神話傳奇的論述，覺得義不容辭的要談一下。

傳說中5G的神奇功效是速度超快，反應時間超短，而且一旦5G上市，4G/LTE就會人間蒸發。要談5G之前必須先談一些基本概念。5G的兩個主要頻段，一個是現在LTE的頻段。這個頻段能提供的改進不大。主要是讓電信業者能再使用已經擁有的頻段。另一個是在25GHz以上的頻段。這才是傳奇中5G的厲害之處。但是這個頻段有個物理上的限制，不會因為那個頻段叫5G，就變得不一樣。

電磁波在大氣中傳播，頻率越高，衰減越快。這意味著5G需要更多的基地台，構成高密度的無線網格。每個基地台不會因為通電了，就自動連上互聯網。每個基地台都需要一條資料鏈路(back hauling)連上廣域網路(WAN)或城市網路(MAN)。這些連網線路也不能是省油的燈。5G說我讓大家上網速度大增。如果每個人都拿一條消防栓噴水，這後面的水管就得粗的像條河。今天5G佈建的資金和時間的需求問題是在這𥚃。加基地台要錢要時間。牽線也要錢要時間。如果上5G的月費是現在的三到十倍，還有多少人要用。如果不這麼貴，電信業者就得慢慢來。台北好不容易蓋好蓋滿了，新北只有一半的覆蓋率。桃園新竹更少了。拿一個5G的手機，一上高速公路就回到美好的舊時光。傳說中的應用，比如說像是虛擬互動，在大安森林公園也許可以，過了泰山收費站可能就有問題了。4G/LTE還是得頂著先。更不要説自動駕駛，上下三義交流道還是自己開比較「即時」。

5G加IoT也只是一個帶動唱的噱頭。在百貨公司，超級市場，辦公室或是家裏是不須要5G的。因為沒有高速移動，其他的科技產品可以取代的，就不用花大錢用5G設備。因為那些5G設備都有很高的專利成本。

反應速度快這說法也只是半個事實。資料都在數據中心裏，基地台再快也要等資料到了才能送。谷歌的搜尋不會因為是5G就突然快得不得了。資料如果不在臺灣，就要到日本或新加坡的數據中心去拿。如果網站在中國，就還要過電子長城。那也不會因為上了5G網路就忽然變快了。

5G不是一個根本的改變(revolution)。因為高頻寬的用處二、三十年前就知道了。但是物理及技術上的限制，造成龐大的資金需求，進而影響佈建的速度。最近臺灣節目中談論5G的內容，根本是「財經專家談物理，政論名嘴説工程」。然後又把華為的5G技術捧得太高了。我先前轉貼了一個有關5G SEP(Standard Essential Patent)的報告。華為是名列前茅，但是並不具有決定性的影響力。更不要説要以一當十。華為最厲害的技術是作生意可以不用看成本。這個專利是資本主義國家裏的公司沒有的。如果了解這點，就可以了解最近各種媒體的報導，只是一種有計劃的政經宣傳，用來打貿易戰。與事實有一段很大的差距。而臺灣如果想要藉5G帶回電子硬體產業的商機，可能要多花點時間了解3GPP成員的生態，再重新打這個算盤。

A note regarding car-to-car IoT

The best record for transmission from a 5G radio station to a 5G client is between 1 and 2 milli-seconds. Using cell tower as a relay for communication between cars, the transmission is in spoke-hub-spoke mode. One way takes 3 to 4 milli-seconds, round-trip up to 8 milli-seconds. Most communication protocols require handshaking/heartbeat messages on top of a shared communication medium. That is a lot of round trips back and forth. A dedicated connection for peer-to-peer communication using un-licensed spectrum may be more efficient than 5G. Besides, when crossing cell boundary, a cell-to-cell handover is required. A 5G cell is small, that is, a lot of handover's will happen throughout the span of conversation. Two cars in conversation may be in different cells momentarily. I don't know whether this issue has been addressed. It is a shame that the car industry cannot pull their acts together. It may only take one autonomous driving company to put out an open source implementation and turn into the de-facto spec, then leave the 5G car-to-car communication in the dust.

自己的責任自己負

2018-11-24T15:33:00.000-08:00

英文裏有個說法：You want something done right, do it yourself.

LGBT 同性戀團體現在應該有這個覺醒。要爭取自己的權益，要自己站出來。怎麼站出來？大遊行，記者會是遠遠不夠的。遊行只是大點名。記者會要看媒體的取向。要爭取權益就要組織。要組織就要人，要錢，要時間。
捐錢很難嗎？每個月只要少吃一次肉圓，或是少吃一次鹽酥雞，或是少喝一杯咖啡，或是少一次小型消費，就可以開始每個月固定小額捐款。組織經濟來源或許不充裕，但是絕對不會是主要的因素。主要的因素是同志團體的覺醒。這是權益的訴求，也是政治訴求。如果同志團體不能了解過去的訴求手段，己經被證明是完全失敗。不能改弦易轍，採取更積極的作為，婚姻平權在這一代是不會到來的。

Fake it till you are caught red-handed

2018-05-13T16:27:00.001-07:00

Theranos is one of the great cases for studying Silicon Valley company taking risks. But in this case, it is not a risk-taking in the category of luck, or the category of success vs. failure. It is a case of blatant fraud.

https://www.wired.com/story/theranos-and-silicon-valleys-fake-it-till-you-make-it-culture/

http://money.cnn.com/2018/03/14/technology/theranos-fraud-scandal/index.html

Theranos is a health technology startup company, claimed to have revolutionary blood test technology that used very small amounts of blood. Theranos' CEO, Elizabeth Holmes, founded the company in 2003 at the age of 19. It sounded like a billion-dollar idea, a potential unicorn company. Its technology and approach were disruptive, exactly what Silicon Valley venture capitalists were looking for. By 2011, the company has been founded for more than 7 years and received about $89 million of funding. It was not a short period of time in terms of valley's startup mentality. Located in Silicon Valley, it must have gone through the boom of Web 2.0 in 2006-2007 and the recession caused by subprime mortgage crisis from 2008 to 2009.

Yet things took a turn in 2011. Holmes was introduced to former Secretary of State George Shultz. Shultz subsequently joined Theranos board of directors, and gradually other all-star members from both public and private sectors. Since then a slew of big name investors came on board: $150 million by Walton family (heir of Walmart founders), $120 million by Rupert Murdoch (Fox/News Corp, owner of WSJ), and $100 million by Betsy DeVos (current Secretary of Education Departement). Theranos received total $1.4 billion (USD) worth of investment by end of 2017 and had a peak valuation at $9 billion. But its technology had never came close to fulfill its promise. Through a series of whistleblower and external challenges, the company can fake no more.

This story was not an outlier of Silicon Valley's startup jungle adventures. It was the scale of the late stage investment in this company and all the public figures involved and their influence that made this story extraordinary. At the same time, a non-investor such as myself may probably also shoulder some of the investor's loss through the purchase of daily household items, monthly fee of TV cable service, and nonetheless college student loans.

第一次文化大革命

2018-03-26T12:33:00.004-07:00

中國的第一次文化大革命，是漢武帝的罷黜百家，獨尊儒術。

漢武帝即位時，媽媽（王太后）在，阿嬤（竇太皇太后）也在。朝內掌權有影響力的，都是這些外戚。這些人不是與漢武帝看法不同，便是與其他外戚官員互有嫌隙。漢武帝想要有一個方法把整批人換掉，好做自己想做的事。這時就要一個既創造時代潮流，又名正言順的中心思想，來推動人事變更。當時朝中流行的是黃老之術。政策內容多要求「輕徭薄賦、與民休息」。漢武帝於是找來了董仲舒等人上書變法。以孔孟儒學之名，行革命變法之實。把黃老一派的勢力，逐步汰換而出。

為什麼説那個變法是假文化之名，行革命之實？漢宣帝，漢武帝的曾孫，曾經如此訓斥他的太子（後來即位為漢元帝）：「漢家自有制度，本以霸王道雜之，奈何純任德教，用周政乎！」而他對一般儒生的印象，更是低劣。「且俗儒不達時宜，好是古非今，使人眩於名實，不知所守，何足委任！」這時距離漢武帝罷黜百家，最少也施行了六、七十年。這是劉家自己在家裏教育下一代，下一個皇帝的話。如果漢武帝的獨尊儒術，是真正的政治文化革新，那他的曾孫，這位律己甚嚴的漢宣帝，對兒子也許就不會説出這樣有逆先祖訓示的話。

奈何選賢與能

2014-10-01T13:25:00.000-07:00

選賢與能的概念不適用於現代的民主選舉。

賢與能既不是候選人提名的標準，也不是當選後執行付與權責的原始動力。為什麼選民還用選賢與能來作為評薦候選人的第一標準？民主政治的基本假設在於「權力使人腐化，絕對的權力使人絕對的腐化」。選民要傳遞給候選人的信息不應該是期待賢能的想法。賢與能要不是暫時的，不然就是塑造出來的形象。選民有的權力一是讓候選人落選。落選的事實對於候選人事業、心理、及經濟上的壓力，和對政黨組織的震撼，是最有效的溝通。另外一個權力就是抑制公職人員的權責能力，形成當選後競競業業的壓力。

壓力來自於完整的投選機制。如果沒有完整的投選機制，選舉／罷免／創制／複決，人民是沒有真正的權力。沒有權力，卻期盼賢能，那是墮落。這也是一種形式的腐化。

暴民

2014-03-30T13:30:00.000-07:00

如果今天美國的黑人跟金恩博士抱怨：“我們當年流血流汗爭來的民權，都讓這些亞洲人坐享其成。我們都被利用了。” 不知道金恩博士會如何答覆。

如果不是當年這些民權運動人士冒著被警棍毆打、催淚瓦斯洗臉的威脅，爭取民權，今天我們在美國的處境就大不同。我們能夠憑著專業知識找到好的工作，不受歧視。購屋置產，不受排擠。也都是這些前“美國暴民”為我們這些亞洲兄弟姊妹爭取後，才立法保護的。如果沒有財產權，工作權，投票權的保護，上了公車火車，不能隨便挑座位入坐。飛機票也買不了。不管你再優秀，都不會成功。因為“門都沒有”。

有人在第一線冒著人身安全的威脅爭取公民應有的權益，如果不能心存感激，最起碼不要冷言冷語。說他們被利用了？是的，就是我們這些安居樂業的良民在利用他們。

Inheritance of digital assets

2013-12-17T14:12:00.000-08:00

I was wondering about this topic for quite some time. I just stumbled upon this article,

http://www.marketwatch.com/story/who-inherits-your-itunes-library-2012-08-23

This question is not just for estate and inheritance. In one or two generations, ebooks disappear with their owners. There is no longer a physical form to keep for archiving. It is not because of the technology but the fact of non-transferable ownership. Can libraries and museums collect, display, and share digital contents once the writers are no longer alive and the publishers cease to exist? When publishing goes digital, are we handing those issues to commercial companies like Apple and Amazon. If so, these companies are enjoying more dominance and financial success than it is shown on the book.

iPhone nano

2012-12-21T15:35:00.000-08:00

This is my wish for this holiday, a product that I would like to see.

Like an iPhone to an iPod Touch, the size of an iPhone nano should be similar to an iPod nano. It is small and its screen is only good for limited yet well defined functions, such as choosing an album, point-and-shoot camera and address book sync. It will not be a good game device nor a good email reader. It sounds like quite the opposite of all the trendy development nowadays: the bigger, the better. The latest phone is so big that you have to attend to your weight training and fitness program diligently in order to keep your arm strength up with the phone.

But an iPhone nano will be very good for one thing: tethering. It should have the same data connection capability as its cousin. It makes such a device a good companion for iPads and laptops. A user with such a device will not need a separate data contract or go through a complicated data sharing arrangement. When the data connection drains battery too fast, it borrows power from the big device it connects to. In my opinion, a cell phone with a near 5-inch screen serves no purpose well. A cell phone should be compact enough to be dropped in virtually any pocket. And the size of an iPad mini is the minimum to be adequate for a tablet. For people that need a cell phone with a long-lasting battery and a good-sized mobile device, this imaginary iPhone nano is probably a good alternative, if not the best.

Why now? The time is ripe to push further for tablets. The need for data connection is more a given than an after-thought. Wireless carriers may frown at the product as a double-edged sword. Can Android community do this? I have not seen a decent small Android phone yet. All those carrier-branded, dumb-down small Android phones look undesirable. Can Nokia do this? Nokia is very good at building solid yet cute phones. But Nokia may no longer have the influence to persuade carriers to promote such a product. Apple needs an inexpensive phone to combat Android in markets where carrier subsidy is not a preferred business practice. An iPhone nano may fill the bill and sells more iPad.

Apple Map

2012-11-02T14:25:00.000-07:00

It started with a map application, CEO's apology for the bad user experience, and now the departure of a senior executive.

http://www.apple.com/letter-from-tim-cook-on-maps/
http://www.reuters.com/article/2012/10/30/us-apple-changes-idUSBRE89S18H20121030

It is obvious that Apple would want to have a map application of its own. What a smart phone or tablet can provide that a laptop cannot easily match is its mobility and portability. Many applications on iOS give users the ability to make decisions or acquire needed information on the spot in real time. Map data, or say information for location and surroundings, need to be integrated into the platform, iOS, for more interesting applications to happen.

Up to now, Siri cannot have a real conversion with its user. The usage is still limited to question-and-answer in a request-response fashion. When Siri is able to have a conversion, the conversation will have a context. The context may be defined by the state of the conversation on top of other inputs, say, an email, a previous phone call or voice message, or recent locations. Google search and map application may not be able to provide all that Apple needs, not to mention the competition between the two companies.

Recent news reported Apple's executive, the head of iOS, Scott Forstall was let go for the blame of Apple map. I think office dynamics goes beyond one product flop. Apple has produced 'sub-standard' hardware and software before. It is not the first time and it will not be the last time Apple makes a mistake. Google has been working on map applications for years. Even if there is no bug in Apple's map application, it would not be as feature rich as Google's map in this first release. Judging from Mr. Cook's letter, Apple is working on some advanced applications that require integration beyond today's interface. For an iOS developer like myself, it is good to see someone firmly at the helm of the company. It is even more important to see the company make steady progress and eventually grow beyond its legendary founder.

Monster Employment Index Europe

2012-02-05T12:45:00.000-08:00

I have been following the US Monster Employment Index for a few years now. I have not paid much attention to the Monster Employment Index Europe. There was something caught my attention in the latest (January 2012) report.

The summary of the 2012-01 Europe Index reported,

The Monster Employment Index Europe grows 9% on a year-over-year basis.
Germany is the only country to demonstrate strong annual growth at 30%.
The Netherlands, Italy, Belgium and France continue to report negative growth rates.

I use the US Monster Employment Index as a simplistic current-to-leading index for the US economy. I would imagine there are many differences between US and European economies due to the regulations and political reality. The Europe Index reports only on seven countries: Belgium, France, Germany, Italy, Netherlands, Sweden, and United Kingdom. Though not covering all of the European Union countries, it is interesting enough for me to keep an eye on from time to time. I looked back to last year's numbers, this downward trend has started around August/September time. If the Europe Index bears a similar characteristics to its US counterpart, the western Europe except for Germany is close to another recession.

Some media thought that Germany's export has benefited from a weakened EURO. If not, there would have been fewer people that can afford their new BMW and Mercedes. I don't know whether that is true or not, but I wonder Germany can single-handedly pull the whole EU along with it. It is troublesome enough to worry about the bad sovereign debts. And now the economy of Europe is in a shaky state. Maybe that explains why all my mutual fund holdings that have European exposures are in retreat.

Screen size of a smart phone

2012-01-24T00:02:00.000-08:00

The Android phone industry just hands out one oversized Android phone after another.

Let's look at these big phones.

Samsung Galaxy Nexus 4G, 135 x 67.9mm with 1280x800 pixels (aspect ratio 8:5)
Samsung Galaxy SII 125.3 x 66.1mm with 800x480 pixels (aspect ratio 5:3)
HTC EVO 3D 126.1 x 65.4mm with 960x540 pixels (aspect ratio 16:9)
And Samsung Galaxy Note trumps all of them at 146.9 x 9.7 mm with 1280x800 pixels.
(For reference, iPhone 4S 115.2 x 58.6 mm with 960x640 pixels)

I don't understand why all the obsessions with a gigantic screen on a phone. Without a big hand, one can hardly wrap his/her thumb around to touch the right spot on the screen with one hand. A purse or bag is required to carried it around. Putting it in a pocket is possible but it takes away too much precious real estate of a pocket. And a big screen burns battery much too fast.

Another problem is the aspect ratio of the screen. It already presents a design challenge for me when designing a universal game for both iPhone (3:2) and iPad (4:3). But I can make an excuse between a phone and a tablet. Stretching images into a different aspect ratio does not make it picture perfect. The other choice is to fill un-used space with background color or image. There are many blogs and articles discussing techniques dealing with this issue. But none has talked about how to test and see the user interface on different screens without buying all of them.

Bigger screens will not make those products more appealing to consumers, when human factors are taken out of product design. I also wonder whether Samsung and HTC care to tend Android application and game developers. The two leading Android brands behave like pure hardware manufacturers that don't seem to appreciate the importance of consistency to the ecosystem which adds value to Android phones. At the same time, I heard it is the carriers who drive product specifications. Well, carriers have never had a love for third party developers.

Kindle Fire

2011-11-23T11:46:00.001-08:00

Kindle Fire is not the champion for Android tablets, but it is the right product for Amazon.

I watched review and comparison videos on YouTube as I have not bought a Kindle Fire yet. From those videos I had the impression that Kindle Fire is not as fast as Apple's iPad 2. The page loading and refresh speed is slower than that on iPad 2. The touch response does not seems to be as smooth. By most measures for a web browsing device and a game console, I doubt it can compete with iPad 2. Nevertheless, I still think it is a good and smart product.

I have complained about last generation's Kindle reader earlier this year. This new generation of Kindle answered my complaints. The touch screen takes away the clumsiness for users to interact with contents. Compared with the previous generation, Kindle Fire provides all the needed improvements in terms of color display, faster page flipping and rendering. It also provides a comparable video experience as iPad 2. Given its price, it is an adequate tablet but an outstanding content viewer. It is exactly right up Amazon's alley for all the digital contents it wants to sell.

Kindle Fire is a smart product with a right compromise and feature focus. I may buy one for my daughter to read on. But I am sure she will ask for my iPad when it comes to Angry Birds and Fruit Ninja. Young users do not settle for their game experiences.

No more mobile Flash

2011-11-14T22:50:00.000-08:00

Adobe gave up on mobile Flash.
http://www.reuters.com/article/2011/11/09/us-adobe-apple-idUSTRE7A84NO20111109

Now it is even more obvious that HTML5 is the technology of choice. HTML5 originated from the grudge of companies that wanted to break free from a dysfunctional consortium, W3C. It is not every day that an open standard comes out winning. Nokia/Symbian used to be the major licensee of mobile Flash. But Symbian is no longer a viable platform. Windows Mobile is going with Silverlight. Google has been a major champion for HTML5. It is up to Adobe to work on various hardware compositions with Android licensees. I don't know whether Adobe had planned for all the factors or whether Adobe has done its best to meet the challenges head on. The consequence of the announcement is that Adobe has not just given up on mobile Flash, it also sent a disturbing signal to its developer community.

Developers invest in technologies that they can leverage upon. When Adobe stopped the development of mobile Flash, it means all the technical know-how and the digital assets invested in Flash are not going to be applicable on mobile devices. Mobile devices are the new frontier. If developers have no Flash but HTML5 on mobile devices, the future investment of their money, time, and brain power will go to HTML5. The death of mobile Flash is not just the end of Flash on mobile devices. It will soon drag down the desktop Flash with it.

HP's next tablet is Windows 8

2011-11-12T16:08:00.000-08:00

HP's new CEO, Mrs. Whitman, said the next tablet product from HP will use Windows 8.

http://moconews.net/article/419-hps-whitman-were-going-to-make-windows-8-tablets/

Two features caught my attention when I took a look at those Windows 8 introduction videos for the Metro Interface, http://www.youtube.com/watch?v=ihcIlg37QKU and http://www.youtube.com/watch?v=BHrcz7zcm_8. One is the touch area outside the visible part of screen and the other is the support of the original windows interface. A touch area outside the screen requires the same high density touch sensory to be integrated beyond the viewing portion of display. The desktop Windows user interface requires more memory. Both add to the BOM cost.

Based on the iPad 2 3G tear down, the BOM cost is around $326. This number came with the fact of Apple's own A5 processor and the low memory usage enjoyed by iOS. A WiFi-only iPad 2 would cost even less. Enough hand-held Windows devices had suffered under-provisioned hardware and failed so far. HP's decision of adapting Windows 8 will put itself between the choice of higher cost per unit or unsatisfactory user experience. Mrs. Whitman has one shot on this one. A decision of using Windows 8 may seem to be a no-brainer. It may turn out to be an ill-thought decision because it seemed to be such a no-brainer.

Will HTC have its own OS

2011-09-13T23:13:00.000-07:00

Rumor has it that HTC is contemplating on buying an OS for itself.
http://www.cellular-news.com/story/50859.php

"We have given it thought and we have discussed it internally, but we will not do it on impulse." HTC Chairwoman, Cher Wang said in an interview. She went further on, "Our strength lies in understanding an OS, but it does not mean that we have to produce an OS."

It takes approximately 40 to 50 million dollars annually to maintain a team of 200 for mobile OS R&D. It is quite affordable to today's HTC. But what will be the strategic position for this OS? If the new OS is to replace Windows Mobile OS, an annual operating cost of $40 million can justify some $70 to $100 million royalty paid yearly for Windows Mobile licenses. But Windows Mobile OS is not the one making HTC feel naked. It is Google's Android, which does not have a royalty based license.

Maybe the whole thing is just some psychological response to Google's buying Motorola Mobile. Or maybe it is because HP's webOS is available at this inopportune time. HTC did not express this idea in such a way as "We'd like to build our own team and our own OS". HTC actually had an small Linux team before. Without a strategy in mind and a concise goal to develop OS in a culture dominated by hardware business, acquisition will not add value. The interesting part is that, though everyone in the mobile industry thinks OS makes a difference, not many companies got it right or even bother to approach it. If Google is to build its own hardware business, what can HTC do? Without an alternative, what HTC should keep an eye on is companies like ZTE, not Google's Moto.

Steve Jobs resigned

2011-08-25T13:54:00.000-07:00

Steve Jobs announced his resignation as CEO of Apple.
http://www.apple.com/pr/library/2011/08/24Steve-Jobs-Resigns-as-CEO-of-Apple.html
http://www.apple.com/pr/library/2011/08/24Letter-from-Steve-Jobs.html

Jobs turned Apple around from the brink of bankruptcy. It was no small feast to say the least. What he and Apple impressed me most was the way Apple jumped in and woke up a very competitive and already booming cell phone industry by introducing iPhone. In 2007, more than 1.1 billion units of cell phones were shipped, while there were only 271 million units of PC shipped in 2007. Among those, more than 110 million units were smart phones. The number one cell phone vendor in 2007, Nokia, owned at least 36% of the overall cell phone market and over 50% of the smart phone market. Nokia had a record revenue in 2007 at €51 billion, which was approximately $80 billion dollars on an average exchange rate of 1 EUR to 1.5 dollar. While the whole electronic-hardware industry looked at the cell phone industry as the main source of demand, Apple walked in with iPhone and told everyone the king has no clothes on. Given an average lead time of two years for a brand new cell phone, the iPhone project might have started in early 2005. In 2005, North America was only a barren field for smart phones.

Apple's stock fell only slightly to $373.72, a drop of $2.46 or 0.65% today. It seems that the market believes Apple now stands firm on the ground, not on Jobs' shoulder. No matter it is an impression or a reality, Apple will eventually out-grow Jobs. Some people in the Valley refer Jobs as "micro-management". That is an expression of their own preferences. After all, Jobs delivered products, revenue and profit to underscore his leadership style. Not many CEOs on earth can say the same, no matter what their preference in leadership styles is.

Seoul Sausage Co.

2011-08-23T12:20:00.000-07:00

I just heard about this company from a friend who went to the 3rd Annual San Francisco Food Festival.

I was told its sausages were sensational. What is more amazing is that this company has no store front, not in the on-line or mail order business. It only does catering and participating in various festivity events. On the web site, the company states,

In less than one year this little dream has turned into the longest lines at street fairs, requests for private orders for wineries/chefs, catered events from movie studios & Fortune 500 companies, and requests to attend various festivals around the country. All currently with no food truck and no retail space, but simply word of mouth...

Of course, there are Facebook, Twitter and Yelp in the mix for such a viral marketing effect. This is a great example of finding a blue ocean in a red sea. When I searched it online, the first few hits were the company's web site, Yelp review, Twitter link and Facebook page. On Yelp review, it has only Los Angeles, CA as a nominal address, but the area code of the telephone number is somewhere near San Jose, California. The telephone number in the company web page shows another area code from San Francisco. You can imagine how dynamic its operation is and a form of telecommuting the company employs.

This company addresses the real value its customers appreciate and cleverly bypasses the overhead that traditionally burdened this industry. At the same time, it leverages all the free marketing resources made available to it. This is definitely a new type of company that deserves a little more attention.

HP/Palm webOS deserves a true steward

2011-08-22T14:49:00.000-07:00

Now HP's executive is saying "We stand by webOS", after offically abandoning it.
http://www.bloomberg.com/news/2011-08-19/hewlett-packard-webos-software-bought-with-palm-not-dead-dewitt-says.html

I received the WebOS developer blog in my email with the title "The next chapter for webOS", signed by Richard Kerris VP webOS Developer Relations. In it, it claims the development and innovation will continue, although the planned developer events around the world may have changes. And one of HP's Senior Vice President Stephen DeWitt said in an interview, "The webOS is not dead... We’re going to continue to evolve it, update and support it. We stand by it." If you look at the price of webOS tablet went down in the past weekend, those words carry neither weight nor meaning.

In my previous blog, I believed HP should go for products, not licensing business. Even now there will be no webOS based product from HP, licensing it still makes no sense. The licensing fee will not recoup the R&D cost, as shown by Microsoft Windows Mobile revenue. A cost center will not result in a vibrant environment for innovation, which will kill webOS sooner than later.

HP should find a suitor for webOS and sell it, like Ebay eventually did for Skype. WebOS is modern and has a lot of value in it. It is not a burning platform like Symbian and Nokia Series-60. Find a worthy home for the team before they jump ship to Apple and Google and leave the webOS an empty shell. After all, Palm's office is only 5 to 10 minutes drive away from Google and Apple's headquarters.

HP gave up on webOS

2011-08-19T12:33:00.000-07:00

HP decided to stop producing tablets and smartphones based on webOS.
http://www.cellular-news.com/story/50543.php

It is really a sad yet harsh footnote to the cell phone industry. I wonder what webOS folks would think if they look at the history of Google purchasing Android, Inc. and the rise of Android phones. A little more than four years (August 2005 to April 2010) separated the fate of the two companies. Should I say the vision and execution of a company made such differences? Excellent technologies can go only so far.

Now HP is going back to a Business-to-Business (B2B) company. Look back not too far into HP's history. Carly Fiorina took HP past the point of no return to a PC OEM company. The low-margin yet big-volume business impacted the company culture so much that the next CEO wanted a strategy for consumer products. Or is it the big share of PC revenue demanded a consumer product oriented CEO? So the next CEO, Mark Hurd, bought Palm and webOS. When Mr. Hurd left HP, the CEO followed, Léo Apotheker, decided otherwise. People can say anything about his decision on discontinuing webOS. But no one can say he made a wrong decision in buying Palm/webOS, because it was not his decision. That fact makes him safe to call the shot in front of the board.

I registered as a webOS developer almost as soon as it is open to the public. My attention to the platform however diminished as it has never gathered enough momentum. I hope this blog will not become an obituary for webOS. But hope against hope, who is to challenge Apple's supremacy of vertical integration?

Google Buys Motorola Mobility for its patent portfolio

2011-08-17T11:49:00.000-07:00

Google is going to buy Motorola Mobility for $12.5 billion.
http://www.cellular-news.com/story/50468.php

The patent war in the cell phone industry has not ended but Motorola Mobility has benefited from it already. Google is an Internet service company, a software company, but not a manufacturer, not an expert in logistics. This is purely a acquisition for Moto's patent portfolio. I do not see Google running Motorola Mobility and producing Google phones vigorously.

Some analysts thought Microsoft might be a winner in this deal, because of doubts and distrusts towards Google among the Android licensees.
http://www.totaltele.com/view.aspx?ID=467053
http://allthingsd.com/20110815/u-s-carriers-silent-on-motoroogle-but-france-telecom-gives-it-a-thumbs-up/
It is just a wishful thinking, in my opinion. Microsoft was in the market licensing its Windows Mobile OS when Nokia had 60% of smartphone market share in 2006-2007. The fact of dominant Nokia did not help Mircosoft to a strong second place. Instead, RIM's Blackberry filled the gap and rose to the second place in 2008 and 2009. When Apple introduced iPhone in 2007 and made its stand in 2008, Android was still in its infancy. Most vendors went to Google and invested their own resources to develop Android phones instead of diving deeper into Windows Mobile. Why is that? I have my thoughts on that, though without direct proof. It suffices to say Windows Mobile is not attractive enough to OEM vendors given all the incentives and the business environment.

Maybe Google will spin off Motorola Mobility and share the patent portfolio with it once the acquisition is done. After all, it is all about patents and it is better for Google and Motorola Mobility to run separately.

Patent suit Nokia vs Apple

2011-06-19T13:28:00.000-07:00

Nokia and Apple settled their law suits over patent infringement out of court.
http://www.cellular-news.com/story/49560.php
http://www.zdnet.com/blog/btl/nokia-likely-netted-600-million-plus-in-apple-patent-settlement/50590

It is said the settlement included a one-time payment between $550 million to $600 million. Apple and Nokia agree on some cross-licensing terms, with a patent license fee up to $11.5 per iPhone sold paid to Nokia. It looks like a victory to Nokia from the result. It is also in a way good for Apple to conclude this fight with Nokia. Based on the latest financial reports from both companies, as of Q2, 2011, Apple has $29.2 billion in cash or cash equivalents, while Nokia has about $16.5 billion. The amount of $600 million is approximately 2 percent of what Apple can pull out of its pocket. With the quarterly iPhone shipment reaches 18 million units, the royalty payment to Nokia is close to $210 million. With Q2 2011 EBITDA income at $7.9 billion, this is affordable to Apple. It is though a boost to Nokia's quarterly EBITDA income of $778 million.

At this point, Nokia has lost one quarter of its global cell phone market share (from 36% down to 27%) and over 50% of its smart phone market share (from around 55-60% to between 20-25%). Nokia lost not just market share, revenue, but also the waning cost advantage associated with scale. Did its patent portfolio protect the company and its shareholders from those aggressive competitors? This patent portfolio did not even help Nokia executives to keep their jobs.

There are two other threads of events worth watching for. One is the patent fight between Chines vendors Huawei and ZTE. Will this domestic dispute result in any legal precedent or will it settle out of court? http://www.cellular-news.com/story/49538.php

The other one is the patent auction by Nortel and Google's intention in Nortel's patents.
http://www.nytimes.com/2011/04/05/technology/05google.html
http://www.reuters.com/article/2011/06/14/us-nortel-idUSTRE75C5WT20110614

A complacent company sued an innovative company and get paid. Do patent laws really encourage innovations?

HP webOS for OEM to license

2011-06-02T22:05:00.000-07:00

HP is entertaining the idea of licensing its webOS to other vendors. Or is this just an indication that HP has not sorted out its strategy for webOS?

http://www.reuters.com/article/2011/06/01/hp-ceo-idUSN0116927120110601

It can be a good news for the industry. One more contender means more choices. But this is just on the surface. Windows Mobile has shown that OS licensing is not a lucrative business in the mobile phone industry. If webOS can collect US$10 a piece, a volume at 3.5 million units a quarter, which is roughly Windows Mobile's 2011-Q1 shipment, makes US$140 million a year. It does not seem to be a reasonable payback for the US$1.2 billion paid for acquiring Palm. It may not even be enough for the operation cost for the business unit. Google Android makes money from advertisement revenue, not licensing fees. What is the leverage that HP can get by licensing webOS out? In the meantime, HP has to make it cost-effective for device vendors to invest in webOS. How many vendors have the extra budget and human resource to take on another OS, chipset, and board support package integration? HP did not talk or hint on intellectual property indemnification. With Microsoft onto everyone who is licensing Andorid, that is also an issue HP has to address when going to partners. http://www.zdnet.com/blog/hardware/microsofts-next-cash-cow-android/12998

The talk from HP's CEO seems to be a spillover of internal disagreement on future directions. It has been a year since HP acquired Palm. There can be some pressure built up over what webOS can do for HP. But it is clear to me, licensing webOS out will not help HP, only to create distractions.