Xindy is an index processor. Just like MakeIndex, it transforms raw index information into a sorted index, made available as document text with markup that may be processed by TeX to produce typeset book indexes. Unlike MakeIndex, it is multi–lingual and supports UTF–8 encoding, both in the raw index input and in the tagged document output. xindy draws its strengths from five key features. Internationalization is the most important feature point and was originally xindy’s raison d’être: with the standard distribution, xindy knows how to handle 53 languages and dialects correctly out of the box. Markup normalization and encoding support is the ability to handle markup in the index keys in a transparent and consistent way, as well as different encodings. Predefined encodings are not only UTF–8 to support XeTeX, also supported is LICR, the encoding that’s output by standard LaTeX to its raw index files, and TeX/Omega’s low–level output of (Unicode) characters. Modular configuration enables the reusability of index configurations. For standard indexing tasks, LaTeX users do not have to do much except to use available modules. Location references go beyond page numbers. An index entry points to a location in the main text. While most index processors can work only with numbers, xindy features a generalized notion of location references that can be book names, law paragraphs, URLs and other references. Highly configurable markup is another cornerstone. While this is usually not as important for LaTeX users, it comes in handy if one works with other author systems besides TeX. While development of xindy has been dormant for quite some time, the last few months saw a flurry of renewed energy and new work to get xindy in the hand of its potential users. The distribution has been streamlined and is now available in standardized source form, thus paving the road for a future acceptance into TeX–Live.