This work questions the nature of bibliographic data and explores their posibilities of storage in different dana structures. Relational and hierarchy structures of data both have their flaws and advantages. Advantages of storing dana in relational database come forth during storing large amount of dana which are often changed and updated, it allows change of single instance of data which is then updated through all connected data. Main difference of relational databases is the exclusion of repeating dana, which also gives relational databases an edge when it comes to memory usage. This structure makes relational databases very popular in the areas which handle large about of data changing on a regular basis. Advantages of hierachical database is the possibility of connecting all of our data to a singe object, which makes such a database much easier to use. Every object in the hierarchical database in an object in intself, while in relational database its necessary to call for data from from a number of different tables. The big flaw of this data structure is the number of repeating data is impossible to regulate. Surely the number of repeating data depents on the type of bibliographical data which we're putting in the database, if the subject is of belletristic nature, the number of authors per work is probably gonna stay around one work each, sometimes maybe more. If the subject is scientific works, where one works has up to 30-40 authors, which are repeating in the other works aswell, then we have a problem. Despite all that I believe that hierarchical structure is more ideal for bibliographic work. It describes one work in its entirety and allows of creation of new sub-categories for more specific works. Ideal solution would warrant the implementation of relational structure in areas of often used repeating data, like publisher or author. Luckly most of the databases in use today are hybrid databases, allowing just that, connecting multiple structures of data. Dataases which are imagined as NoSQL databases like MongoDB and databases using SQL shema like MySQL, overtime became hybrid databases. The conclusion of this work is that each database and shema has its own flaws and advantages, and its necessary to study the nature of data we're storing and construct a database according to data.