Sei sulla pagina 1di 4

Network Applications

Now that you have a working knowledge of what networks are and how you can access them,
the key question is, How do businesses use networks to improve their operations? This section
addresses that question. Stated in general terms, networks support businesses and other
organizations
in all types of functions.
This section will explore numerous network applications, including discovery, communication,
collaboration, e-learning and distance learning, virtual universities, and telecommuting.
These applications, however, are merely a sampling of the many network applications currently
available to users. Even if these applications formed an exhaustive list today, they would not do
so tomorrow when something new will be developed. Further, placing network applications in
categories is diffi cult because there will always be borderline cases. For example, the difference
between chat rooms (in the communications category) and teleconference (in the collaboration
category) is only one of degree.

Discovery
The Internet enables users to access or discover information located in databases all over the
world. By browsing and searching data sources on the Web, users can apply the Internets
discovery capability to areas ranging from education to government services to entertainment
to commerce. Although having access to all this information is a great benefi t, it is critically
important to realize that there is no quality assurance for information on the Web. The Web
is truly democratic in that anyone can post information to it. Therefore, the fundamental rule
about information on the Web is User beware!
In addition, the Webs major strengththe vast stores of information it containsalso presents
a major challenge. The amount of information on the Web can be overwhelming, and it
doubles approximately each year. As a result, navigating through the Web and gaining access
to necessary information are becoming more and more diffi cult. To accomplish these tasks,
people increasingly are using search engines, directories, and portals.
Search Engines and Metasearch Engines. A search engine is a computer program that
searches for specifi c information by key words and then reports the results. A search engine
maintains an index of billions of Web pages. It uses that index to fi nd pages that match a set
of user-specifi ed keywords. Such indexes are created and updated by webcrawlers, which are

6.4
SECTION 6.4 Network Applications

189

computer programs that browse the Web and create a copy of all visited pages. Search engines
then index these pages to provide fast searches.
In mid-2013, four search engines accounted for almost all searches in the United States. They
are, in order: Google (www.google.com), Bing (www.bing.com), Yahoo! (www.yahoo.com), and
Ask (www.ask.com). In addition, there are an incredible number of other search engines that
are quite useful, many of which perform very specifi c searches. (See an article on The Top
100 Alternative Search Engines that appeared on www.readwriteweb.com.) The leading search
engine in China is Baidu, which claimed 78 percent of the Chinese market in April 2013.
For an even more thorough search, you can use a metasearch engine. Metasearch engines
search several engines at once and then integrate the fi ndings to answer users queries.
Examples
are Surf-wax (www.surfwax.com), Metacrawler (www.metacrawler.com), Mamma (www
.mamma.com), KartOO (www.kartoo.com), and Dogpile (www.dogpile.com). Figure 6.10
illustrates the KartOO home page.
One interesting search engine is known as Summly. ITs About Business 6.2 explains how
Summly works.
FIGURE 6.10 The KartOO

home page (www.kartoo.com).


MIS

ITs [about business]


When Nick DAloisio was 12, he created an app called SongStumblr,

which used Bluetooth to let users know what music people nearby
were listening to. Next, he taught himself the basics of artifi cial
intelligence (AI) software and built an app called Facemood, which
updates you on the emotional state of your friends by monitoring
their Facebook status. Over the next 3 years, Nick earned about
$30,000 from profi ts from one-time sales of his apps, netting about
$1.50 for each app download after deducting Apples 30 percent
share of the profi ts. By age 15, he had set up his own company.
(His mother had to sign all of his legal documents because he
was underage.)
In 2011, while studying for a history exam, Nick grew frustrated
with the huge amount of text that showed up on Google search
results. To save himself time, he wrote a program that summarized
content in an easily digestible preview. He used his fl air for
languageshe was studying Latin, Mandarin Chinese, and French
at the timeto create a program, which he named Summly, that
would extract the most relevant sentences out of the long text that
Google searches provide. Microsoft Word had already marketed
a summarizing tool that searched for the most frequent keywords
sequentially. However, Nick was convinced he could do better
with a genetic algorithm that could choose important sentences as
a human does. Therefore, he created a program that detects the
topic of a text and uses that information to determine what metrics
to apply. Then, the program utilizes those metrics to determine, for
example, if numbers are more important than descriptive words,
and to extract critical sentences accordingly.
Nick realized that his program could be used as much more
than just a personal study tool. It could be used as a service for
media companies, fi nancial institutions, law fi rms, or any organization
that faces the challenge of extracting key points from huge
amounts of data. For example, if you visit a Web site that contains
excessive amounts of information, Summly will provide you with
a more user-friendly version of the information condensed into
three to fi ve bullet points. Summly works especially well with news
articles, because it bullet-points the main headlines. Summly also
provides key words for the article.
DAlosio fi rst honed his knowledge of AI and writing computer
code. He then contacted a linguistics researcher at MITs Semantics
Lab and paid him $250 to analyze his algorithm. The researchers
conclusion: Summly constructed summaries in a way that was
40 percent more similar to the process humans use than its competitors.
This fi nding was enough validation for DAloisio to release a free
app (then called Trimit), which ran on an iPhone and received regular

6.2 A New Search Engine


190

CHAPTER 6 Telecommunications and Networking

FIGURE 6.11 Google


Translate.

updates as he refi ned it. He set up a Web site and a demo video of
the app on YouTube. A patent search revealed that, although similar
apps existed in the market, none featured the trained AI component
or the tie-in to the iPhone that Trimit could boast.
Shortly after the apps release, two investment funds contacted
DAloisio via his Web site. One was Horizons Ventures, whose
founder and manager had seen a story about Trimit on TechCrunch
(a technology blog) and was interested in the technology behind the
app. After thoroughly examining DAloisios app, in September 2011
Horizons purchased a 25 percent stake in the product for $300,000.
Over the course of the next 3 months, Summly was downloaded
130,000 times. The downloads were made available free of
charge to build name recognition for the product.
You might ask, if the downloads are free, how does DAloisio
plan to make a profi t? Rather than make money from users, he
licensed Summly to media and fi nancial companies, Web browsers,
and search engines. He will charge these outlets either a regular
subscription fee or a fee for every 1,000 summaries. He has yet to
set the exact pricing for his product.
DAloisio then hired full-time programmers who trained his
genetic algorithm to mimic choices made by humans. This process

was tedious and time-consuming, and it involved reading thousands


and thousands of Web articles (such as Wikipedia, movie
reviews, and many other sources), choosing the three to fi ve most
important sentences, and feeding them into the algorithm.
DAloisio believes his product will be successful because search
interfaces like Google have not changed in many years, despite a
dramatic increase in the amount of content that is available on the
Web. He maintains that he has developed an entirely new type of
search engine that can revolutionize the search-engine fi eld.
In March 2013, Yahoo! purchased Summly for $30 million, and it
hired DAloisio and two other Summly employees. DAloisio pulled
Summlys app off the market, but he claimed that the technology
behind Summly would soon be incorporated into Yahoo! products.
Sources: Compiled from H. Tsukayama, Summly and Its Teen Founder
Snapped Up by Yahoo, The Washington Post, March 26, 2013; B. Stelter,
He Has Millions and a New Job at Yahoo. Soon Hell Be 18. The New
York Times, March 25, 2013; P. Olson, Search Engine Wunderkind,
Forbes, March 12, 2012; J. Wakefi eld, British Designer of Summly App Hits
Jackpot, BBC News, December 28, 2011; M. Kirkpatrick, Summly: New
App Helps You Read All Your Bookmarked Links in Minutes, ReadWriteWeb, December 13, 2011; P. Olson, Teenage Programmer Backed by Hong
Kong Billionaire Li Ka Shing, Forbes, December 13, 2011; C. Bonnington,
Teens IOS App Uses Complex Algorithms to Summarize the Web, Wired,
December 13, 2011; www.summly.com, accessed March 2, 2013.

Questions
1. What are the advantages of the Summly app? Do you see any
disadvantages in using the product?
2. Has DAloisio actually created a new kind of search engine?
Support your answer.

Publication of Material in Foreign Languages. Not only is there a huge amount of


information on the Internet, but it is written in many different languages. How, then, do you
access this information? The answer is that you use an automatic translation of Web pages. Such
translation is available to and from all major languages, and its quality is improving with time.
Some major translation products are Microsofts Bing translator (http://www.microsofttranslator
.com) and Google (www.google.com/language_tools) (see Figure 6.11), as well as products and
services available at Trados (www.trados.com).
Should companies invest their time and resources to make their Web sites accessible in multiple
languages? The answer is, absolutely. In fact, multilingual Web sites are now a competitive
necessity because of the global nature of the business environment. Companies increasingly
are looking outside their home markets to grow revenues and attract new customers. When
companies are disseminating information around the world, getting that information correct is
SECTION 6.4 Network Applications

191

essential. It is not enough for companies to translate Web content. They must also localize that
content and be sensitive to the needs of the people in local markets.
To reach 80 percent of the worlds Internet users, a Web site needs to support a minimum
of ten languages: English, Chinese, Spanish, Japanese, German, Korean, French, Italian,
Russian, and Portuguese. At 20 cents and more per word, translation services are expensive.
Companies supporting ten languages can spend $200,000 annually to localize information
and another $50,000 to maintain the Web sites. Translation budgets for major multinational
companies can run in the millions of dollars. Many large companies use Systran S.A. (www
.systransoft.com) for high-quality machine translation services.
Portals. Most organizations and their managers encounter information overload. Information
is scattered across numerous documents, e-mail messages, and databases at different locations
and systems. Finding relevant and accurate information is often time-consuming and may require
users to access multiple systems.
One solution to this problem is to use portals. A portal is a Web-based, personalized gateway
to information and knowledge that provides relevant information from different IT systems and
the Internet using advanced search and indexing techniques. After reading the next section,
you will be able to distinguish among four types of portals: commercial, affi nity, corporate, and
industrywide. The four types of portals are distinguished by the audiences they serve.
A commercial (public) portal is the most popular type of portal on the Internet. It is intended
for broad and diverse audiences, and it offers routine content, some of it in real time (e.g., a
stock

ticker). Examples are Lycos (www.lycos.com) and Microsoft Network (www.msn.com).


In contrast, an affi nity portal offers a single point of entry to an entire community of affi liated
interests, such as a hobby group or a political party. Your university most likely has an
affi nity portal for its alumni. Figure 6.12 displays the affi nity portal for the University of West
Georgia. Other examples of affi nity portals are www.techweb.com and www.zdnet.com.
As the name suggests, a corporate portal offers a personalized, single point of access through
a Web browser to critical business information located inside and outside an organization.
These portals are also known as enterprise portals, information portals, and enterprise
information
portals. In addition to making it easier to fi nd needed information, corporate portals
offer customers and employees self-service opportunities. ITs About Business 6.3 provides an

Potrebbero piacerti anche