At first sight Unidatab seems difficult to use. On one hand, this is due to the fact that like other relational devices, Unidatab requires some acquaintance with command line tools and SQL, and is meant to be used by programs as well as people. On the other hand one has to remember that Unidatab is still purely experimental, the code contains unfinished functions, uneasy peculiarities and requires many obvious improvements. On top of that Unidatab is too slow in some actions like search function. This tardiness will not be be felt as prohibitive for a single user, but it may impede proper working in the multi-user mode Unidatab has been designed to handle.
However, once mastered the basics, both the web and the graphical interfaces allow to take more easily advantage of the common features of this software (some tasks will always require the command line tool unid-adm). In spite of all its weaknesses Unidatab will reveal handy to everyone who wanted to experiment XML like data trees in SQL context.
Instead of writing a long manual, I’ll blog tips and class them under tutorial categories. Feel free to post questions and/as comments.