thanks for your interest in using stylo! If you're a beginner, as a general remark I'd recommend taking a look at one of our tutorials,
a slideshow presenting absolute basics or
HOW TO paper - particularly useful when you start changing parameters without using GUI - note examples in the margin.
Regarding your questions:
1) as of 0.6.9 stylo version, UTF-8 is the default expected encoding - if the program finds texts in a different encoding it should read them as well, but the major change is you no longer need to tick UTF-8 checkbox. However, if your files are encoded in a different way, especially if your corpus is in a language using non-Latin scripts or diacritics, R may have a problem with proper delimitation of characters and words. So best - keep your texts in UTF-8. Also, if text files include any emoticons etc. this may cause problems too.
2) as for passing parameters in rolling.delta() and rolling.classify() - of the two only the first one has GUI at the moment (and it is a bit underdeveloped as the function is rarely used since the introduction of the more popular rolling.classify). To pass the parameters include them in the brackets of the function, e.g. rolling.classify(encoding= "UTF-8").
Hope this helps a bit, if the problem persists please write more details about the data you are trying to analyze (number and length of texts /note that default sample size in rolling functions is 5000 words/, language, encoding) and we should be able to help more precisely.
Best wishes,
Joanna Byszuk