TopicWeb front-end testing. Large scale analysis and techniques improvement
SupervisorsArnaud Blouin, Olivier Barais, Djamel-Eddine Khelladi
Duration3 years
Start Date01/12/2024
End (estimate)30/11/2027
SiteInstitute for Research in Computer Science and Random Systems IRISA

Project structure

00 Antilibrary

An antilibrary is a collection of books that are owned but have not yet been read. The term was coined by Nassim Nicholas Taleb in The Black Swan: The Impact of the Highly Improbable. Hence, the antilibrary includes all the notes that are not yet written

01 Notes

All notes taken so far. Each note has a unique ID and one or more tags that classify its content, making it easier to find notes related to the same topic

02 Literature

Books and articles are summarized here. The main concepts and ideas extracted from articles or books are linked to notes included in the previous section

Books: each book is organized in a separated folder inside /books, which contain:

  • root note (the book index): indicates all the chapters that have been studied
  • canvas: a graphical representation that summarizes the main ideas of the book
  • chapters: contains a note for each chapter

Papers: contains all the papers, grouped in thematic sub-folders

03 References

This folder includes all references used by articles or notes in general. References are in BibTeX format and (may) include a link to the original source, if any

04 Canvases

This folder includes all canvases used to summarize or map concepts

05 Dataset

This folder includes datasets or links to dataset

06 Courses, Webinars

This folder includes courses and webinars notes

96 Meta

This folder includes dataviewjs scripts used to analyze the notes and to get useful statistics

97 Assets

Images, GIFs and video used

98 Templates

Templates are boilerplate notes to standardize the note taking process

99 Output

This folder contains the scripts output, for example a .bib file which includes all the references used for a specific project.

Scientific paper notes workflow

A new paper is taken into consideration

  • The paper reference is inserted into 03 References, with its BibTeX ID
  • A new note is inserted in 02 Literature/papers. If a relevant sub-folder is not found, a new folder must be created
  • The paper’s notes are build on top of the Temp_Literature template
  • The paper’s main concepts and the author’s approach to the problem belongs to the paper’s note scope, whereas the details of the topic addressed by the study belong to 01 Notes. The latter includes low-level note, while 02 Literature/papers just highlights the unique approach or the solution proposed by the authors. In this way low-level concepts grow in details and nuances as more papers are added
  • Low-level notes always include references to documents that discuss that concept. This references are both in the paragraph, justifying important statements, and in the footer, referring to the context in which that concept was addressed. I.e., cookie banners and XSS (cross site scripting) (Klein, Musch, et al., 2022).
  • Each document and note must have its own tags, which facilitate the process of information retrieval
  • Custom property might be added to group specific papers. For example, papers that are actually part of a SLR (systematic literature review) present the property SRL, whose value is the URL of the OverLeaf project. This ease the retrieval of all the scientific studies used in a specific project. Obviously, a paper may be part of more projects: in this case, a new custom property is added
  • Custom Bookmarks are added when new custom properties are needed

QuickAdd Automations are exploited to simplify the workflow

  • Ctrl + P / QuickAdd:Run QuickAdd / Add new paper and reference: after inserting paper’s folder BiBTeX code and optional URL, it generates the new notes linked together with the correct reference
  • Ctrl + P / QuickAdd:Run QuickAdd / Generate .bib file: generate a .bib file which contains all the BibTeX references that are used in the notes that display the requested custom property

Contacts

GitHub | LinkedIn | Portfolio