Benjamin Sommer


a tiny efficient and effective bibliography tool

Teaser image of upcoming bibby tool: catalog view


Focus on literature search, and automate daunting repeatable tasks, for advanced typists and small visual displays.

Coming: When it's ready

Expect a first technical preview, maybe in 2019. Only afterwards this release will a list of features along with screenshots and measurements be published, iff released.

Target audience

Individuals and small/medium-sized groups, favoring the typist experience. Linux users with HiDPI visual displays. Others, including Windows and MacOS X may follow.

Who is this tool not for?

Those favoring touch displays, speech recognition, mouse input, and those being afraid to use the terminal. And of course, libraries and other large institutions/organizations.

The Story

During my Master's program of Visual Computing, I wrote bibby 1.x to help me keeping track of a growing private bibliographic catalog. Some features were: hierarchically grouping, searching, indexing along with comments found in PDFs and, notably, easily opening attached documents. With growing projects, I needed a central database to efficiently work with, allowing to easily generate reference sections in the bibtex format for these interleaved projects, across which a singular reference might be used many times. Besides, I need to know that I can rely on the correctness of the imported bibliographic data.

Back at this time, no Linux program shaped up as being satisfiable concerning my demands and what I had in mind. For instance, one tool imported bibliographic data with erroneous UTF-8 back-and-forth conversion while suffering from massive memory leaks (besides of having a cluttered visual interface). Another tool failed to handle well thousand's of such entries with respect to searching. Besides, in a time when notebooks still consist of rather small visual display areas, for instance 30cm to 40cm diagonal, these precious areas may not be wasted by crude visual interface layouts. Redundant information need to be hidden. For instance, when editing or importing bibliographies, all available 2-dimensional euclidean space need to be consumed, not just a forth of it, further forcing to introduce tabs, many buttons or unnecessary scrolling areas, requiring additional user actions such as hand and mouse gestures being avoidable in the first place.

Being in need of a superior tool to satisfy these demands, to let me focus on literature search and research, I sought out to write bibby 1.x. A tool which showed to save precious time during my Master's thesis.

In pursuing optimal visual representation while keeping the program's code base small, being computational and memory efficient, and having few dependencies to enable a long software life cycle, I chose a low-level approach based on OpenGL. Having no complex, constraining middleware, the thereby gained freedom allows to pursue and to experiment with ideas, in automating daunting tasks, in enhancing enjoyment. Literature (re)search is scientifically both interesting and fascinating. And such a tool need to reflect it.

These motivations led me to finally write bibby 2.x.