The other day I came across Tradd.us. It is an online text translation service that was developed by a company called Zee, which registered in Brazil. Currently available in private beta, but if you’re interested in using this service, you can register your email address with them, and you shall receive your beta invitation within a single working day.

Tradd.us service is directly applicable to users that seeking for a better Web translation tool, and is converging with its development in mashing up some great platforms such as Google Translator (the core translator), and OpenCalais (semantic Web toolkit). This means that when users type a text on Tradd.us, this mashup will thus translating your text into the language you want while loading any semantics that could be found through the tagging system of such a text.

In their own word, Tradd.us is an innovative and creative approach to text translation. Powered by Google translator and a range of other services we are able to translate, analyze, identify and contextualize important information from your content, in a very quick and organized way.

I like the fact that when using this service, other than its own core translation capabilities, it will generate the underlying meaning of the text as well as point you to the related sites and common synonyms or related terms you have typed into its translate bar. So information is described in a way that not only you learn the text in other language, but you also learn the relationships of groups of texts in the real world.

Tradd.us is created by Irrezeestible Solutions, and its business model is based on Google AdSense at this moment.

I often stumble across and receive beta invites from certain Web startups that seems really awesome for some particular reason and immediately find myself wanting to write a review on it, and Evri is one of them now. Evri, currently available in private beta, considered as a Web startup that perform semantic analysis. It is very much similarly to Twine, that offered semantic technology solutions which I have profiled it in May this year on TagEdge.

As stated on Evri’s site, it is a semantic technology Web startup, with a motto “Search less, understand more.” Thus, for one would like to try it and compare it with some top search engines, such as Google and Yahoo! it will be unfair for Evri. On Evri, you won’t find the search bar, and you can’t perform keyword search on Evri. This contrast to Google or Yahoo!, when you type certain keywords, some sites with high-level of SEO (search engine optimization) skill will rank in the first page. I bet that Evri also not allow someone to craft a really killer little campaign, a sort like Google AdWords campaign, that his/her sites will appear in the sidebar as the sponsored link, and when the potential buyers key in the keyword that match the bidded keyword, they will see these sites and pay-per-click for further action. Understandably, Evri is about content tagging, content discovery and compliance, not about keyword search.

Users at Evri might scream, “What type of value that Evri created for its users?” Temporarily, you’ll only see three type of content that are “Most Popular People,” “Most Popular Places,” and “Most Popular Things” on Evri, all are collected and evaluated by Evri before the data presented to all of its users. Meanwhile, I see Barack Obama top the “Most Popular People” at the time of my writing (See the picture below). On each of these three categories, you’ll find the top ten (10) most popular (located in left-hand side) as well as the top ten (10) most change in popularity recently, either rising or falling on the right. In the middle is the data graph, which is the visual circle connections, click on the connections will display the relevant articles for that people/place/thing or that particular connection. You’ll be surprised by the semantic processing that are done in the Evri’s back-end engine, particularly the ways articles that are processed by Evri, so users can find the underlying information about each connection that might interested them at any one stage.

However, I found the content in existence are a target product for the American, as Evri first launched in the U.S., it is imperative that all the content must suit the taste and characteristics of their first target market.

Twine is one of the startups that is in my waiting list to be reviewed on TagEdge and I recently got a private beta access to it. It is a new Web application that applied the common semantic model, based on the fact that human experts are expensive, and users on Twine can provide valuable search and data information.

To understand what Twine does, and how it differentiate itself from any other web applications on the Web, you need to understand what semantic is. Semantic is refers to the computers’ ways to process the language or code, and understandable, experts usually correlate semantic to information management. But to understand semantic Web, the founder of Twine, has explained it, in a post, entitled, “On the Difference Between “Semantic” and “Semantic Web.” On Twine, what they do is to provide a semantic Wen Web environment whereby an user is given a profile page, and the ability to social bookmark any link or URL address as she like, and Twine will do the rest with the new approach or concept of semantic analysis. Semantic analysis exists in helping you to interpret the information you put on your profile page, categorize, tag, summarize, abstract, and even extract similar link to the data information you like.

Perhaps this semantic technology is highly automated, and apparently not been easily understand by users who never came across this technology. However, the semantic technologies in existence is for the sake of the public. Perhaps you can consider it as a technology to help users create a data information library, discover all the bookmarks that you want to find out, but couldn’t achieve it by one’s own effort. With this technologies, it will grouping all the interest things, stuff in an organized way, and thus, we’ll realize the real relationships of these things with our lives. But expect there is a huge volume of data information out there on the Web, it’s always impossible for us to expect Twine will perform well barely in categorization, classification and tagging automatically on all the data users enter to the site, but more advance this technology evolve, the more it will understand human language, the perception on human towards language or computer code. In short, Twine was trying to make their system “smart” about the environment, but in practice, there’s still a lot of improvement that Twine will need to do, in viewed of many positive feedback, and some are negative out there.

To use Twine, upon you’ve been granted a private access, you’ll be given a profile page. Simple Simply put, it’s like a social network, meet friends with similar interests, access to the data information collected from other Twine users, and join groups, etc. On the profile page, you’ve the options to talk about your latest status, what you’re doing right now, introduce yourself as well as list your own Website on the page. You also provided with a notification page when there’s somebody would like to “connect” with you, and a Twine Digest for you to read as your everyday meal of information.

I like the fact Twine has provided a “Twine This” bookmarklet (drag it to the browser’s toolbar) that provide users to directly add a Web page they like onto their “My Items” of their profile pages. The another way is to email your favorite Web pages through with your own personal email address. All the Web pages you bookmarked are private, only if you purposely create your own public Twine, and add the Web pages you like to the public Twine, the bookmarks will be appeared as public in both “My Items” and public Twine. To illustrate this example, I’ve created a public Twine called “Startup to Watch.” To encourage users to share knowledge or information they came across on the Web, users are welcome to create as many Twines as they like, but manage those strategically. And be cautious you can’t change the URL address of your Twine as it’s impossible for an user to do it at this moment.

The core of semantic technologies is to find and expose content. To achieve this aim, Twine also allowed users to explore “Top 100 Twines”, “New Twines”, and “Top 100 Members.” From the Twines you discovered, you can find the articles, news you’re looking for a long period of time, but couldn’t find it other than Twine provided here. You also will meet some like-minded people in this platform. Meanwhile, there are several public twines that interested me, including the one created by Nova Spivack, Twine’s founder, Ruby on Rails, and Facebook research.

Twine is a project of Radar Networks, a startup based in San Francisco.

The year 2007 is just about to wrap up and there are several predictions on the Web for next year. Whilst several bloggers are arguing on the statement that Web 2.0 will die in year 2008, predicted by an U.K. research firm SciVisum, however in China a nice little Beijing based startup Lonol has quietly launched its homepage with the adoption of Web 3.0 methodologies and technologies.

Lonol is a Web platform that allowed its users to personalize the homepage in addition to aggregate the news content. In a news source (in Mandarin) from Lonol, they have a perspective that Web 1.0 is purely solved the “read” problem when Web surfers surf on the Web. On the other hand, Web 2.0 solved the “write” problem, and with the emergence of Lonol personalized homepage, it will solve the problems of informational needs when users want to get the information on the Web. Since Lonol personalized homepage will initially available in Chinese, and rather than being impediment to non-Chinese users, they’ve planned to add a translation engine that translate all their content to other languages that include: English, French, German, Russian, Spanish, Japanese, Korean, just to name a few.