Tim Bray, the co-inventor of XML shares in detail his understanding of and hopes for search engine technologies on his On Search, the Series blog series of essays. Particularly interesting in these articles posted between June and December 2003 is the focus on the user experience in searching and the expressed hope for future developments which would make search engines and search features usable.
One of the articles, On Search: I18n, focuses on issues of internationalization. When a web site uses the usual 26 characters of the modern English alphabet, search engine optimization is, one could claim, trivial. The challenge is when a web site uses characters which belong to another of the European languages (or why not a totally different language group). The easiest solution, of course, would be to approximate the non-English version, i.e. Ä, to the closest English version of a character — A. Sure enough, most people would search for information using the standard 26 Latin / English characters. Consequently, a web site should be able to rank highly against keywords written in those standard character sets.
Things get less trivial and more complicated when more and more non-English speakers join the web. Many of them would be used to typing non-standard characters that happen to be standard in their own language. Both search engines and search engine marketers will have to prepare for that day, as it is nigh.