I wanted to explore the linguistic limitations in using Textual Analysis. I hunted for digitized copies of certain Tamil texts. There are lots of archives but no tool sophisticated enough to conduct a Text Analysis. So I began by playing around with Google's N-gram viewer. I wanted to see the usage of words 'Tamil', 'Telugu', 'Malayalam' and 'Hindi. I tried it in two different corpus of texts - English and German. There are interesting peaks in the graph at totally different time periods! When I picked specific years to see where and in what context the words were used, language was a barrier. 

English corpus 

German Corpus

Then I picked a popular text in English - 'Alice in Wonderland' by Lewis Caroll and ran it under various Text Analysis tools. Here they are:

In Wordle

But depending on what text I used (there are lots of digitized 'Alice in Wonderland' versions out there!), the image could also look like this

A particularly interesting tool was on Many Eyes. This is one of the visualisations for the data set Alice in Wonderland. There are lots of other possibilities for visualising the same set of date. But I wonder if it is possible in some way to map Alice's journey and create an atlas to trace her physical journey through Wonderland? Now such a tool will be interesting! Imagine what it can do to 'Gulliver's Travels','Ulysses' or any text that has a journey, for that matter! If that map can have a 'sentiment reader' component like booklamps, then even internal journeys of characters can be mapped. That will be truly exciting!