You can certainly create custom dictionaries and flashcards with both; in the custom dictionary editing screen, simply choose the appropriate recognition mode in the handwriting recognizer (or pick the appropriate character out of the radical table) and the software will quite happily accommodate both simplified and traditional characters in the same entry. When importing cards, if you use UTF-8 format that will allow you to use simplified and traditional characters in the same text file (and our software should import them correctly).
As far as switching between traditional and simplified characters in an entry, that's a little trickier - there is a way to do it (one we use in our own data files), but it might be difficult to do with an average text editor. We're considering whether or not to add an easier option for this (some sort of a tag, like <TC>(chars)</TC>) in the finished version - we probably will if we have time. Basically the dictionary will automatically convert characters in entries from simplified to traditional if their simplified form never appears in traditional text, but you need to code in the traditional characters manually if the simplified ones sometimes do appear (e.g. gan1, ji1) or if there are multiple traditional characters which correspond to a single simplified one and you're using one of the less common variants (e.g. li4, mian4).
This doesn't currently happen in the Oxford data files, BTW, because we haven't gotten around to coding them up with the correct traditional variants yet; we're already working on some other major changes to them (chief among them the addition of several hundred new entries from the 3rd Edition Oxford dictionary) and once we've done that we'll code in the correct traditional variants as well, but for now we think it's better to have characters remain simplified rather than be incorrectly converted to traditional.
