Computational science: shifting the focus from tools to models.

Area of Science:

• Scientific research methodology
• Computational science

Background:

• Computational techniques have significantly advanced scientific research.
• Researchers increasingly use computation for big data analysis and complex model predictions.

Purpose of the Study:

• To highlight the limitations of traditional scientific articles in publishing computational research.
• To argue against the overemphasis on software tools at the expense of models and data.
• To propose a solution for integrating computational models and data into the scientific record.

Main Methods:

• Analysis of current scientific publishing practices.
• Discussion of the long-term implications of data and model accessibility.
• Proposal of a new framework for scientific publication.

Main Results:

• Traditional articles cannot accommodate large datasets or complex computational models.
• Crucial research components (big data, complex models) are excluded from the scientific record.
• Scientific models and data are often locked within proprietary software formats.

Conclusions:

• The current emphasis on software over models and data is detrimental to scientific progress.
• Reversing this trend by prioritizing data and model publication is essential for long-term scientific advancement.
• Integrating computational models and data into the scientific record will enhance reproducibility and collaboration.