Weak-Linear Types

Hector Gramaglia

2024-02-19

Abstract:Computational interpretations of linear logic allow static control of memory resources: the data produced by the program are endowed through its type with attributes that determine its life cycle, and guarantee safe deallocation. The use of linear types encounters limitations in practice, since linear data, in the traditional sense, do not so often appear in actual programs. Several alternatives have been proposed in the attempt to relax the condition of linearity, adding coercions to the language to allow linear objects to be temporarily aliased. In this work we propose a new alternative, whose virtue is to preserve the simplicity and elegance of the original system.

Programming Languages

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is the limitations of linear type systems in practical programming. Specifically, linear data in the traditional sense does not often appear in actual programs, which restricts the wide application of linear types. To solve this problem, the paper proposes a new method to relax the linear conditions, allowing read - only access to linear data of basic types while retaining the simplicity and elegance of the original system.

### Detailed Explanation

1. **Background and Motivation**:
- The computational interpretation of linear logic allows for static control of memory resources: data generated by a program has properties through its type, which determine their life cycle and ensure safe release.
- However, linear data is not common in actual programs, so a method needs to be found to relax the linear conditions to make it more suitable for practical programming requirements.

2. **Limitations of Existing Solutions**:
- Previous research has attempted to allow temporary aliasing of linear objects by introducing coercions, but this increases the complexity of the system.
- The paper points out that although these methods are effective, they fail to fully retain the simplicity and elegance of the original linear system.

3. **Solution Proposed in the Paper**:
- Introduce a new qualifier `hi` (hidden qualifier) to facilitate read - only access to basic linear data.
- Modify context splitting to manage sub - structure properties without adding new modalities.
- In this way, flexible access to linear data can be achieved without significantly increasing complexity.

4. **Main Contributions**:
- Propose a new linear application language that supports a weakened form of linear data, allowing read - only access.
- Retain the simplicity and elegance of the classical linear type system proposed by Walker [15].
- Through the introduction of the hidden qualifier `hi` and pseudo - split of contexts, achieve flexible management of linear data.

### Formula Representation

- Context Split is defined as an \((n + 1)\)-ary relation:
\[
\Pi_1\circ\ldots\circ\Pi_n=\Pi
\]
For simplicity, it is defined for the case of \(n = 2\).

- Pseudo - Split is defined as:
\[
\Pi_1\sqcup\ldots\sqcup\Pi_n=\Pi
\]
When a linear base type appears, for example \(x:\text{li }B\), then:
\[
(\Pi_1,x:\text{hi }B)\sqcup\ldots\sqcup(\Pi_j,x:\text{li }B)\sqcup\Pi_{j + 1}\sqcup\ldots=\Pi,x:\text{li }B
\]

### Conclusion

By introducing the hidden qualifier and pseudo - split of contexts, the paper provides a new method to relax the linear conditions, making the linear type system more suitable for practical programming requirements while maintaining the simplicity and elegance of the original system. This provides a theoretical framework for further research on the relationship between sub - structurality and in - place updates.