Open Source Could Cut Costs for Federal, State and Local Government

Open Source Could Cut Costs for Federal, State and Local Government (Opinion)
From Government Technology - Solutions for state and local government in the information age
Jun 22, 2009, By Bill Vass

The public sector, especially the federal government, has been slow to embrace what may be the single most important innovation sweeping the technology world -- open source software. It holds the "keys to the kingdom" in making government more efficient in conducting its business, serving citizens securely and better managing tax dollars spent on technology.

Open source technology enables both equal access and the freedom to distribute a technology's source code -- in essence the "language" on which it's built. It includes an accepted set of standards that allow a business or government the freedom to move its critical data and IT services on, off and in-between products from different companies, whenever it chooses.

Think about your car. In the U.S. we've agreed to drive on the right-hand side of the road. We've agreed on traffic rules and signals. The gas pedal goes on the right and the brake goes on the left. These are our "open standards" for driving. Still, we're free to purchase any car from any car company we wish. And if we're not happy with the car we bought, or we don't think the price is fair, we're free to shop around. We don't need a new driver's license or to invest time and money learning to drive a different type of car.

It's the same with the global commercial open source movement. And the first major benefit the government can realize from this freedom and openness is reduced costs. Commercial open source technologies cost less than closed, proprietary technologies. Open source software can typically be downloaded for free. And open source technology and open standards allow multiple vendors to compete fairly for the government's business. Such competition brings more choices, better products and lower prices.

The cost savings are astounding. In the UK, where the government recently announced it will accelerate the use of commercial open source software, it's estimated that the move away from proprietary technologies could save £600 million a year. Here in the U.S., a recent report from MeriTalk estimates federal IT savings could reach "$3.7 billion from open source." These cost savings can't be ignored, especially when taxpayer dollars are footing the bill.

Think, if you will, about government procurement scandals of the past involving unexamined purchases that led to disclosures of $10,000 toilet seats and wastebaskets and other irresponsible spending. Proprietary software is this decade's procurement scandal waiting to happen with government agencies wasting millions on proprietary technologies when equivalent open source technologies are available for just thousands of dollars, or in some cases, free. Creating a formal policy around open source use and procurement brings more transparency and a level playing field to the government technology bidding process.

The second major consideration is security. Ensuring the security and integrity of our government's sensitive data and critical IT systems is crucial. Commercial open source technologies provide more security than their proprietary counterparts because open development means hidden breaches aren't buried in the code. They are ferreted out more quickly and efficiently, and rigorous testing ensures three-letter agency security levels.

Finally, open technologies enable every citizen to better communicate with our government, do business with government agencies, receive government services and access public data. This happens when all government and "citizen-owned" content share common standards and don't require the use of proprietary technologies (word processors and spreadsheets, Web browsers or e-mail clients) to access it.

By no means is open source a new idea. Open source technologies and solutions such as Java, Linux, Solaris, MySQL, Apache and OpenOffice are well established and well proven the world over -- literally billions of people are using them. Some of the world's most successful companies -- like Google and Facebook -- run their businesses on them.

There are a few success stories in the U.S government as well, yet we still lag behind the rest of the world in making use open source. Governments in the UK, Brazil, Denmark, the Netherlands, China and Germany -- just to name a few -- have all engaged in formal, wide-scale deployments of open source technology, with many having articulated specific policy on open source. Without its own open source policy in place, the United States risks widening economic and technological competitive gaps between itself and the rest of the world.

In the end the best performing, most cost-effective technology is always the right technology for the job -- whether it's open source or proprietary. But I do believe our government needs to embrace open source technologies and develop a formal policy for their use. Open source supports many of the new federal administration's stated goals; better security, faster deployment, open government, and more responsible spending.

The time is now for the United States to capitalize on open source technology.

Bill Vass is the president and chief operating officer of Sun Microsystems Federal Inc. and the chief technology officer of Sun's Global Accounts and Industry organization. Vass focuses on providing the U.S. government and elite global accounts with open, secure and efficient solutions to solve unique IT challenges for large-scale organizations.

MJ

喝酒脸红背后的健康隐患

VOA News

A wine-tasting near Beijing, China

Finally, a listener in Taiwan wrote to ask why his face turns red when he drinks alcohol. This effect called facial flushing is a common reaction to alcohol among East Asians. It affects an estimated thirty-six percent of Japanese, Chinese and Koreans.

For many people, even a little alcohol can cause unpleasant effects. Most commonly, their face, neck and sometimes their whole body turns red. People might also feel sick to their stomach and lightheaded. They might experience a burning sensation, increased heart rate, shortness of breath and headaches.

The cause is a genetic difference that some people are born with. It prevents their bodies from processing alcohol the way other people do. But the effects might be more serious than just a red face. Researchers have warned of a link between this condition and an increased risk of cancer of the esophagus from drinking alcohol.

A report about facial flushing appeared recently in PLoS Medicine, a publication of the Public Library of Science. The report says the more alcohol that persons with this deficiency drink, the greater their risk. It estimates that at least five hundred forty million people have the deficiency.

Esophageal cancer is one of the deadliest cancers. It can be treated when found early. But once it grows the chances of survival drop sharply.

Philip Brooks is a researcher at America's National Institute on Alcohol Abuse and Alcoholism. Doctor Brooks says it is important to educate people about the link between the alcohol flushing effect and esophageal cancer. He says doctors should ask East Asian patients about their experiences with facial flushing after drinking alcohol. Those with a history of it should be advised to limit their alcohol use. They should also be warned that cigarette smoking works with the alcohol in a way that further increases the risk of esophageal cancer.

常见中文分词开源项目

中文分词和TF-IDF-月光博客

　　中文分词(Chinese Word Segmentation)指的是将一个汉字序列切分成一个一个单独的词。中文分词是文本挖掘的基础，对于输入的一段中文，成功的进行中文分词，可以达到电脑自动识别语句含义的效果。

　　TF-IDF（term frequency–inverse document frequency）是一种用于信息搜索和信息挖掘的常用加权技术。在搜索、文献分类和其他相关领域有广泛的应用。

　　TF-IDF的主要思想是，如果某个词或短语在一篇文章中出现的频率TF高，并且在其他文章中很少出现，则认为此词或者短语具有很好的类别区分能力，适合用来分类。TF词频(Term Frequency)指的是某一个给定的词语在该文件中出现的次数。IDF反文档频率(Inverse Document Frequency)的主要思想是：如果包含词条的文档越少，IDF越大，则说明词条具有很好的类别区分能力。

　　使用TF*IDF可以计算某个关键字在某篇文章里面的重要性，因而识别这篇文章的主要含义，实现计算机读懂文章的功能。

　　常见中文分词开源项目：

　　SCWS：

　　Hightman开发的一套基于词频词典的机械中文分词引擎，它能将一整段的汉字基本正确的切分成词。采用的是采集的词频词典，并辅以一定的专有名称，人名，地名，数字年代等规则识别来达到基本分词，经小范围测试大概准确率在 90% ~ 95% 之间，已能基本满足一些小型搜索引擎、关键字提取等场合运用。45Kb左右的文本切词时间是0.026秒， 换算完毕大概是 1.5MB文本/秒。支持PHP4he PHP 5。

　　ICTCLAS：　　

　　这可是最早的中文开源分词项目之一，ICTCLAS在国内973专家组组织的评测中活动获得了第一名，在第一届国际中文处理研究机构SigHan组织的评测中都获得了多项第一名。ICTCLAS3.0分词速度单机996KB/s，分词精度98.45%，API不超过200KB，各种词典数据压缩后不到3M.ICTCLAS全部采用C/C++编写，支持Linux、FreeBSD及Windows系列操作系统，支持C/C++、C#、Delphi、Java等主流的开发语言。

　　庖丁解牛分词：

　　Java 提供lucence 接口，仅支持Java语言。

　　CC-CEDICT：

　　一个中文词典开源项目，提供一份以汉语拼音为中文辅助的汉英辞典，截至2009年2月8日，已收录82712个单词。其词典可以用于中文分词使用，而且不存在版权问题。Chrome中文版就是使用的这个词典进行中文分词的。