Relational Diagrams are a complete and sound diagrammatic representation of safe relational calculus. They are (𝑖) proven unambiguous, (𝑖𝑖) proven relationally complete, and (𝑖𝑖𝑖) are able to represent all relational query patterns for unions of non-disjunctive queries. Our anonymously preregistered user study shows that Relational Diagrams allows users to recognize patterns meaningfully faster and with higher accuracy across different schemas than SQL. We envision a future in which a user dictates queries while interacting with a relational database, and the system visualizes the queries back, enabling the user to verify their correct interpretation

Intersection.png

Reference

On the Reasonable Effectiveness of Relational Diagrams: Explaining Relational Query Patterns and the Pattern Expressiveness of Relational Languages
Wolfgang Gatterbauer, Cody Dunne
SIGMOD 2024 (to appear)
preprint | arXiv:2401.04758 (long version) | OSF supplements | OSF user study preregistration | Executed user analysis code
Proposes a semantic definition of relational query patterns, which allows us to analyze the relative pattern expressiveness of relational query languages. Also proposes "relational diagrams", a natural diagrammatic representation of tuple relational calculus. 
@article{SIGMOD2024:GD,
  author = {Wolfgang Gatterbauer and Cody Dunne},
  title = {On the Reasonable Effectiveness of Relational Diagrams:
          Explaining Relational Query Patterns and the Pattern Expressiveness of Relational Languages},
  journal = {Proc. {ACM} Manag. Data},
  volume = {2},
  number = {1}
  pages = {61:1--61:27},
  year = {2024},
  doi = {10.1145/3639316}
}

Funding

This work has been supported in part by the National Science Foundation (NSF) under award numbers IIS-1762268 and IIS-1956096, and conducted in part while Wolfgang Gatterbauer was visiting the Simons Institute for the Theory of Computing. Any opinions, findings, and conclusions or recommendations expressed in this project are those of the author(s) and do not necessarily reflect the views of the Funding Agencies.

National Science Foundation Simons Institute for the Theory of Computing

Related papers

A Tutorial on Visual Representations of Relational Queries
Wolfgang Gatterbauer
PVLDB, 16(12), pp. 3890-3893, 2023
ACM | PVLDB | arXiv:2308.10319 | Tutorial slides | gs | bib
Tutorial page: Visual query representations
Surveys the key visual metaphors developed for visual representations of relational expressions.
Principles of Query Visualization
Wolfgang Gatterbauer, Cody Dunne, H.V. Jagadish, Mirek Riedewald
IEEE Data Engineering Bulletin, Special Issue on Widening the Impact of Data Engineering through Innovations in Education, Interfaces, and Features, 45(3):47-67, 2022
IEEE | arXiv:2208.01613 | gs | bib
Discusses the principles of relational query visualization and its potential for simplifying user interactions with relational data
Stratisfimal Layout: A modular optimization model for laying out layered node-link network visualizations
Sara Di Bartolomeo, Mirek Riedewald, Wolfgang Gatterbauer, and Cody Dunne
IEEE VIS, Vol 28, Issue 1, pp. 324 - 334, 2021
IEEE | OSF preprint (long version) | OSF supplements | gs | bib
Project page: Stratisfimal Layout
Proposes a comprehensive approach to make complicated networks (such as those with layers and grouped nodes) easier to read. The idea is to formulate various readability criteria (notably crossing minimization and bendiness reduction) as a modular and customizable constraint optimization problem.
QueryVis: Logic-based Diagrams help Users Understand Complicated SQL Queries Faster
Aristotelis Leventidis, Jiahui Zhang, Cody Dunne, Wolfgang Gatterbauer, HV Jagadish, Mirek Riedewald
SIGMOD, pp. 2303–2318, 2020 (SIGMOD reproducibility award, announcement)
ACM | preprint | arXiv:2004.11375 (long version) | OSF preprint (long version) | video (12min) | OSF supplements | reproducibility instructions | reproducibility report | gs | bib
Project page: QueryVis
Shows that logical diagrams automatically created from SQL queries help users understand the queries faster and with fewer errors than SQL itself. Our ultimate goal is to allow users of SQL to reason about queries in terms of "diagrammatic SQL patterns" based on first-order logic.
QueryViz: Helping users understand SQL queries and their patterns
Jonathan Danaparamita, Wolfgang Gatterbauer
EDBT 2011, pp. 558-561 (System demonstration)
ACM | preprint | gs | bib
Project page: QueryVis
Proposes QueryViz, a light-weight add-on to existing databases. Copy your SQL query into the interactive interface at http://queryviz.com and look at the visualization of the query's relational pattern.
Databases will visualize queries too
Wolfgang Gatterbauer
PVLDB, 4(12):1498-1501, 2011 (VLDB challenges and visions track)
ACM | PVLDB | preprint | narrated slides (16MB) | slides | video (19min) | gs | bib
Project page: QueryVis
Describes a new human-query interaction in which users re-use existing queries as templates to compose their own queries. This interaction is made possible with new automatic query visualization tools (such as QueryViz, or now QueryVis) which help users understand SQL patterns quickly.

Related web pages

Tutorial on Visual Query Representations (VLDB'23)
Surveys the key visual metaphors developed for visual representations of relational expressions.
Stratisfimal Layout (VIS'21)
A modular optimization model for layered node-link network visualizations, as needed for query visualizations
QueryVis, formerly known as QueryViz (SIGMOD'20, EDBT'11 demo, VLDB'11 vision)
Precursor to Relational Diagrams