Connection to Web search engine   

Hi Valentin. Unfortunately proxy connections are not currently supported when creating a document collection from the web. Thanks for notifying us of this limitation - we will investigate this for future releases.

In answer to the last part of your question, we connect to Google search for this feature http://www.google.com/search.

Kevin

Hi Valentin. At present LanguageWare annotators only support UTF input. This is standard for Java applications. However, it is not hard to convert from other encodings to UTF format. Java has support for this when reading from files - I believe that if you specify the charset of the file, it will automatically convert it to UTF format. There are also tools such as ICU (International Components for Unicode) that have good support for doing this conversion in a program.

For example, the Document Collection Creator in the workbench checks the character encoding set in HTML meta tag. Then it tries to convert from specified (in HTML meta) encoding to UTF-8. For Russian web pages, you would probably want to do something similar before passing the text to the LanguageWare annotators.

Hope this helps you.

Kevin