Access control

What happens when the rules get more complicated?

Roles make access familiar. Attributes add context. Policies bring rules together. The real difference shows up when a business needs all three and the rules keep changing.

Approaches & tradeoffs

Different strengths. Different costs.

These approaches overlap. A policy can use attributes, and an RBAC wrapper can organize policy sets under role names. The question is where each approach helps and where complexity starts to grow.

RBAC

Roles

Group access around responsibilities, such as Finance Manager.

Finance managers can approve expenses.

Strength
People can understand and administer access through familiar job responsibilities.
Tradeoff
Roles alone don’t describe every situation. Adding separate roles for each limit, region, or exception can create a growing web of roles and assignments.
How Siua builds on it
An RBAC wrapper can associate roles with policy sets. Conditions and shared business rules stay in the policies underneath.
Background reading

ABAC

Attributes

Use details about a person, a resource, and the situation to decide.

The approver and expense belong to the same organization.

Strength
Decisions adapt to the request without requiring a new role for every combination.
Tradeoff
The more details a rule depends on, the more combinations there are to understand and test. Decisions also depend on the quality of the information supplied.
How Siua builds on it
Use typed request data in reusable conditions, and test how those conditions affect the whole decision.
Background reading

PBAC

Policies

Express decision rules as policies that can be managed and tested together.

Allow approval, require verification, and enforce a spending limit.

Strength
Access and business rules have a place to be reviewed, reused, and tested instead of being scattered across handlers.
Tradeoff
Added complexity is the main cost. More policies mean more conditions and interactions to understand, review, and audit. Dedicated policy languages can add another learning burden.
How Siua builds on it
Use familiar code and AI authoring tools, plain-English explanations from static analysis, and organization-wide audits to make policy complexity easier to manage.
Background reading
What about writing the checks in application code?

Application code gives developers expressive logic and familiar tools. But repeating checks across handlers and services makes shared rules harder to find, review, and change consistently.

Siua keeps that expressiveness while giving the decision logic a shared home in reusable policies.

Why we built Siua

An allow shouldn’t erase a requirement.

Maya can approve expenses, but this $2,400 expense exceeds a $2,000 limit. How the rules combine changes what that allow means.

In a first-match rule list

  1. An allow rule matches.
  2. The decision returns immediately.
  3. A later spending check is never reached.

Order determines which rule decides.

Teams must place checks carefully or include requirements inside the rules that allow access.

In a Siua policy set

  1. An allow policy matches.
  2. The enforce policy checks the spending limit.
  3. The requirement fails. The decision is denied.

Another allow cannot bypass the requirement.

Requirements can compose across policy sets without being copied into every allow policy.

Try adding another allow

Reuse the logic

Import shared functions into policies. A change to a common business check has one place to live, using familiar development tools.

Explore shared functions

Keep its meaning

Carry the same execution behavior across applications, including details such as text matching that can otherwise differ between technologies.

Explore consistent execution

Give outcomes more purpose

Observe an evaluation without changing access, or give an exceptional request an approval path.

Explore observations & review

The complexity question

More expressive rules need better tools.

PBAC’s flexibility comes with more logic to understand. Siua’s approach is to make that logic easier to author, explain, and inspect across the organization.

Author with familiar tools

AI assistants and existing editors work with familiar code. Shared functions help teams build on logic they already use.

Explain the policy

Static analysis turns the policy’s conditions and effects into a plain-English explanation for review.

Audit the whole picture

Surface what can happen across the organization, when it can happen, and which policies and conditions shape those possibilities.

Explore how Siua manages complexity

Compare tools

How does Siua compare?

Different policy engines make different choices about authoring and combining rules. Compare those choices with Siua’s approach.

Rego · general-purpose policySiua vs. Open Policy Agent
The difference
OPA uses a language called Rego to write policies. Siua expresses rules as code using familiar programming language syntax.
Tradeoff
Teams adopt Rego and define how their rules produce a decision. General-purpose flexibility also means choosing the decision structure and conventions.
What Siua enables
Use ordinary functions and imports to share business logic across policies, with the IDE and AI tools your team already knows.
When Open Policy Agent fits
OPA is a strong fit for a shared policy platform across infrastructure, configuration, and applications. It also supports running compiled policies in different environments.
Cedar · authorization policySiua vs. Cedar
The difference
Cedar uses its own policy language, with rules that permit or forbid an action. Siua combines allow policies with separately written requirements.
Tradeoff
Authors learn Cedar’s language and express application conditions through its policy and entity model.
What Siua enables
Give a shared requirement its own enforce policy. It can compose with many allow policies, and an extra allow never bypasses it.
When Cedar fits
Cedar fits teams that prefer a language designed specifically for access decisions. Its forbid rules can also express restrictions.
Configurable model · policy rulesSiua vs. Casbin
The difference
Casbin lets teams configure how access rules are combined. Siua provides a built-in way to combine allow, deny, and enforce policies.
Tradeoff
Teams configure and maintain the model, matching rules, and the way matching policies combine. More flexibility means more configuration to understand.
What Siua enables
Combine a shared requirement with allow policies, and associate policy sets with roles through an RBAC wrapper.
When Casbin fits
Casbin fits teams that want to configure their own access model, including role-based and detail-based rules. That flexibility requires configuration and maintenance.
AWS identity and resource policiesSiua vs. AWS IAM
The difference
AWS IAM controls access to AWS services and resources. Siua is designed to express access and business decisions inside your application.
Tradeoff
Teams must understand how AWS identity, resource, and boundary policies interact. That addresses AWS resource access; application business decisions need their own design.
What Siua enables
Bring access and business rules together: decide who can approve an expense and whether it meets the spending rules, in the same policy decision.
When AWS IAM fits
Use IAM to control access to AWS resources; its policies can also place boundaries on permissions.

Explore the policy model