The website reCAPTCHA is dedicated to saving time by maximizing human productivity, digital-age-style. CAPTCHAs, short for Completely Automated Public Turing tests to tell Computers and Humans Apart, are used by websites as a defense mechanism against those evil, spam-generating bots. CAPTCHAs are becoming increasingly common on the internet; they offer a sense of security in an age when surfing the web can open people up to all kinds of vulnerability. According to reCAPTCHA’s website, nearly 60 million CAPTCHAs are solved by internet users every day. The philosphy of reCAPTCHA is this: If people are already going to spend time solving CAPTCHAs for everyday web access, why not solve a larger problem in the process?
There are currently several projects in the works aimed at digitizing books. In order to transform these printed books into digital archives, book pages are scanned and then put into text using Optical Character Recognition. reCAPTCHA provides a way to double-check the work of computers by utilizing human web activity. The words from the digitized books that a computer cannot decipher are passed along to a human for verification, thus producing a correct word while simultaneously providing the user secure web access. In essence, reCAPTCHA kills two birds with one stone. Michelle Manafy, editor-in-chief of EContent, says: "While the technologies that make projects like the Internet Archive possible are incredible, they are not perfect. But then again, neither are people. reCAPTCHA allows people and machines to work together to help perfect the digitization process. It also makes me feel like keying in those funny words to authenticate my humanity serves a larger purpose."
To join the movement, webmasters can put reCAPTCHA boxes on their sites. Individuals can also help by using Mailhide, an anti-spam program that allows users to safely post their email addresses on the web by requiring that those wishing to access the full email address solve a reCAPTCHA. reCAPTCHA captures both sides of your nature--making your internet forays more secure while allowing you to contribute to the greater digital good.
Reuters Labs Viewdle Beta
Search engines and technologies continue to emerge at an impressive rate, each improving our ability to find essential information. One area in which tools still struggle to turn up meaningful results is video content search. A recent collaboration between Reuters Labs and search upstart Viewdle, Inc. promises to help shine some light on the challenge of video search, and maybe even result in a solution. This past September, Reuters Labs incorporated the Viewdle beta into its website to search within hundreds of hours of Reuters' video content. Reuters.com will use the Viewdle engine to automatically extract metadata from select Reuters satellite streams and provide real-time indexing and search capabilities.
Viewdle, founded in 2006, is a facial-recognition digital platform aimed at indexing, searching, and monetizing video assets. The nascent company, which has patents in the preparation phase, got a jump-start on the video indexing industry when Reuters Labs decided to launch Search People in Video, powered by Viewdle. When you visit the Reuters site to do a video search, you can type in a name, such as "George Bush," and Viewdle returns clips featuring the president(s) from a collection of Reuters videos. However, these results are not garnered from metadata that has been added to the video footage; rather Viewdle actually looks "inside the clip," frame by frame, to identify appearances of people on screen. As an added enhancement, if a search turns up video with multiple people, you can drag your cursor over different people's faces and the individual's name will pop up, along with a small white box that encompasses his or her face. If you decide to click on someone's face, Viewdle automatically conducts a new video search featuring the person you just selected. While the collaboration will certainly draw attention to Reuters' video content, it also highlights the company's innovative approach--and may well illuminate the challenge of effectively searching video content.
The Royal Society of Chemistry's Project Prospect
The Royal Society of Chemistry's Project Prospect website helps to bring clarification to the technical world of chemistry. The RSC has a deep history dating from the 19th century. Yet as the digital age has evolved, so has the RSC. Its recent enhancement, Project Prospect, is a sign of the times. Project Prospect was officially launched in February 2007, and with it, the RSC became the first primary research publisher to publish semantically enriched articles. The initial goal of the project was to "make the science within journal articles machine-readable, and to add new ways of retrieving and presenting the information within." Before Project Prospect, users searching for chemical compounds typically had to rely on a simple text search, making it difficult to correctly identify and discover the content for which they were searching. In an innovative step, the RSC editors addressed this problem by using text-mining software to annotate compounds, concepts, and data within the article.
The resulting Project Prospect has helped solve problems on both sides of the publishing process. For editors, reviewing and editing chemistry papers can be a daunting task due to the complexity of the content within the papers. The RSC developed a range of editorial tools that allow editors to check the data by cross-referencing it against standard reference databases, thus improving the quality of the paper before it's even published. Once the paper is published, the same information used by the editors in the editing phase can be added later as enhanced metadata. A helpful toolbar enables the reader to navigate through and better understand the content, while the IUPAC Gold Book is a pop-up window feature that allows the reader to click on a specific term within the paper for further clarification. Chemistry papers can often feel intimidating and saturated with information, but the RSC's Project Prospect makes comprehension possible again.
Almost anyone today is beset upon by a deluge of technical jargon, which changes daily as new standards, solutions, and organizations emerge. While an open web search may turn up the meaning of the acronym du jour (amidst a slew of other results), it is nice to have a reliable source of technical vocabulary definitions. What-Is-What is a website that, in its roughly two years of existence, has quickly become a tech bible of sorts. A university student who aims to provide common people with answers to digital queries runs the site, which tries to answer any science questions while emphasizing computer-related issues. What-Is-What defines words, acronyms, ideas, products, and software for all who ask. Michelle Manafy, editor-in-chief of EContent, says, "Confronted with internet alphabet soup and tech jargon, I can hardly keep up with what I'm supposed to know, much less decipher the information I'm supposed to digest on a daily basis. Joining the ranks of other community-built web resources, What-Is-What provides a place to tap into a deep well of others' answered questions as well as providing a forum to ask your own." And that's exactly what the site does: It provides answers to a variety of questions that cross many minds daily. What-Is-What visitors can either search for answers to previously asked questions, or they can ask their own by filling out a simple online form. If you choose to leave your email address, the owner promises to "get back to you personally," but otherwise, questions are generally answered right there on the site's homepage. Either way, What-Is-What.com helps keep people in-the-know and up-to-date on the ever-evolving digital lingo.