If you are reading this as a printed thesis, it is likely a shortened version of the full document it is part of. To conform to the expectations of a PhD thesis many parts are removed that are irrelevant for the basic presentation of the work done during the thesis.

However, a fellow researcher who wishes to understand all the details, in particular in terms of reproducibility of the results, the full document should be read instead. If possible, a PhD thesis should be a tome of knowledge about the topic that allows the interested reader to absorb as much of the authors knowledge as possible to help with the continuation of the research. Crucially, this includes details about how results were obtained and access to results in numerical formats, which is both commonly lacking in regular PhD theses.

For this reason the idea of an extended thesis was born, found at ¹

https://phd.vindaar.de.

There you will not only find a PDF of the thesis you are likely reading right now, but also:

• a PDF version of the extended document,
• an HTML version of the extended document,
• the raw Org document of the thesis ²,
• a link to the repository of the thesis,
• a link to a repository of the entire raw datasets used in the thesis (plus reconstructed data) following the FAIR (Wilkinson et al. 2016) guiding principles,
• a large amount of additional documents of notes taken during research and development.

The extended document contains many additional subsections tagged 'extended', similar to the 'Intro' tag (see the top right of the page) of this chapter. These either contain personal thoughts about why certain decisions were taken or additional research done that either lead nowhere or simply is not very relevant for the thesis. Maybe most importantly, you will find a subsection for every single plot and table included in this thesis (that was created by me) with the commands or code snippets to produce them! In addition, the linked git repository of the thesis not only contains all the figures included in the thesis as vector graphics, but also a CSV file for every plot. ³ Use them freely under the only condition to reference this thesis when you do so.

In essence the idea is to provide a fully reproducible thesis.

Furthermore, every thesis, especially those relying on large pieces of software, contains mistakes, bugs, wrong assumptions and more. Most of these are not known to the researcher, in case of wrong assumptions possibly due to lack of knowledge in a specific topic. Other times shortcomings are known, but left out for convenience. In this thesis I try to be transparent about known shortcomings. There are almost certainly bugs in the referenced code that I'm not aware of, bad assumptions about certain things, etc. Where I am aware of sketchy choices and mistakes, I will highlight them honestly.

During the work of my PhD I have written and contributed to a large number of free software / open source projects. The TimepixAnalysis ⁴ project – the code base at the heart of all data reconstruction and analysis for this thesis – is just one more such project. What this means is that I have no intention of abandoning it in the future, even if I leave academia. As long as there is demand for it (for example for a GridPix3 detector used at BabyIAXO), I'll be happy to maintain and extend it as needed. Future reader, don't let "last updated N years ago" scare you! Just open an issue on Github ⁵.

Note that I'm always available for questions about my work or this thesis either via Matrix (@vindaar:matrix.org) or on other channels (Twitter / Discord / Github @vindaar) or plainly by email phd@vindaar.de. Please do not hesitate to contact me, even if it is several years since this thesis was initially released.