Data scientist: How to begin? If i want to be Data Scientist in future,
can somebody advise me sites/books/courses and other good things to learn?
Thanks!
P.S: i know statistical analysis and R/Matlab/Excel/SAS on different levels, but want to rise my skills in it.
 A: I think data analysis is term Tukey coined with his 1960 paper "The future of data analysis".  Tukey's two books on exploratory data analysis is a good place to start. The book with Mosteller is easier to understand than EDA (i.e. the green book in better than the orange one). There have followed books that explain exploratory data analysis better.  See the work of authors like Velleman and Hoaglin.  I think you will find books with titles like "Understanding EDA" and the ABCs of EDA." 
Whether or not the term 'data scientist' includes confirmatory data analysis for you I do not know.  However the term is commonly used to describe researchers who look for patterns in "big" data.  That would fall under the category of data mining.  Data mining is a rapidly developing discipline.  Books by Hand et al and the Tibshirani, Hastie and Friedman book on statistical learning are the best books on that subject (but are highly technical).
Other books regarding statistical pattern recognition include McLachlan's book and the second edition of Duda and Hart's book.  If you search online using an engine like Google you will probably find some downloadable powerpoint presentations on these topics as well as short courses being offered.  The site statistics.com has many short course on a variety of statistical topics including data mining. 
I would also like to suggest looking at Gerry Hahn's book "A Career in Statistics:  Beyond the Numbers."  This book emphasizes what it takes to be a good applied statistician. Gerry and Necip the two authors worked together at GE Research for many years.  Gerry was the manager of the group.  The reader benefits greatly from their judgement which is based on tons of experience working on real world problems.  Although an applied statistician may be a more general term than data analysts/scientist and some may say the two intersect rather than that data scientist is a subcategory of the applied statistician.  But regardless there is a lot of common things that apply to becoming good at either discipline.
