2025 Language Analysis

The Language Speed Hierarchy 2025

PowerShell merges 6x faster than C in 2025. 25 languages ranked by cycle time.

6x

Speed Gap

PowerShell (4h) vs C (24h)

Based on 802,979 merged PRs | GitHub Archive / BigQuery | October 2025

The Speed Rankings

PowerShell4h🏆 Fastest6xC24hSlowestMedian merge time

All 25 programming languages ranked by median PR merge time. The fastest languages merge in hours; the slowest take nearly a full day.

Median Merge Time by Language (Hours) - Top 15

0h5h10h15h20hPowerShellDockerfileShellNixTypeScriptJavaScriptRubyHCLYAMLC#HTMLPythonVueScalaCSS
RankLanguageMedian (h)Avg (h)PRs AnalyzedAvg PR Size
#1PowerShell4h33.2h1,292234 lines
#2Dockerfile11h51.6h1,174110 lines
#3Shell14h64.4h5,230148 lines
#4Nix14h61.1h1,36677 lines
#5TypeScript15h59.5h43,177503 lines
#6JavaScript15h61.6h16,904461 lines
#7Ruby15h66.8h3,534174 lines
#8HCL15h56.5h1,492133 lines
#9YAML15h74.7h1,41629 lines
#10C#16h62.2h6,762463 lines

"PowerShell PRs merge 6x faster than C PRs: 4 hours vs 24 hours median."

The Slow Movers

At the other end of the spectrum, C takes 24 hours median to merge. Swift (23h), C++ (21h), and Java/PHP (20h) also sit in the slow lane.

Go19h

13,726 PRs analyzed

Rust19h

8,374 PRs analyzed

Kotlin19h

4,478 PRs analyzed

Jupyter Notebook19h

1,759 PRs analyzed

Dart19h

1,105 PRs analyzed

Java20h

14,940 PRs analyzed

These languages aren't slow because they're bad—they're slow because they're used in domains that demand careful review. Systems programming, safety-critical applications, and legacy codebases require more scrutiny.

Why the Difference?

The 6x speed gap between PowerShell and C isn't about language syntax. It's about ecosystems, use cases, and review culture.

Community Size & Activity

TypeScript leads with 43K+ PRs. Larger communities mean more active reviewers and faster turnaround. Smaller communities (Swift: 1.1K) have fewer available reviewers.

PR Complexity

Dart PRs average 844 lines; Shell PRs average 148 lines. Infrastructure code (PowerShell, Dockerfile, YAML) tends to be smaller and more straightforward.

Domain Requirements

C/C++ dominates embedded systems and kernel development where bugs can be catastrophic. More thorough review is a feature, not a bug.

Automation Culture

DevOps languages (PowerShell, HCL, Dockerfile) have strong CI/CD integration. Automated tests provide confidence, speeding up human review.

"TypeScript and JavaScript tie at 15 hours median merge time—the web stack moves fast."

Cycle Time vs Community Size

Larger communities don't automatically mean faster reviews. TypeScript has 43K PRs and 15h median; C has 2.8K PRs and 24h median. The relationship is more nuanced.

PR Count vs Median Merge Time (bubble size = avg PR size)

06121824Median Merge Time (hours)015K30K45K60KPRs Analyzed

Fast Tier (4-15h)

9 languages

PowerShell, Dockerfile, Shell, Nix, TypeScript...

Medium Tier (16-19h)

11 languages

C#, HTML, Python, Vue, Scala, CSS, Go, Rust, Kotlin, Jupyter Notebook, Dart

Slow Tier (20-24h)

5 languages

Java, PHP, C++, Swift, C

Year-over-Year Changes

Comparing to our 2024 study (3.4M PRs), the hierarchy is largely stable. PowerShell remains the fastest, C remains the slowest. But the gap has widened from 4x to 6x.

Speed Gap Widened

From 4x (2024) to 6x (2025). Fast languages got faster; slow languages stayed slow.

PowerShell Got Faster

PowerShell dropped from 6h to 4h median—a 33% improvement in cycle time.

Caveats

Correlation, not causation

Languages don't determine review speed—teams, domains, and cultures do. C is slow because it's used in cautious environments, not because of syntax.

Public repos only

GitHub Archive captures public repositories. Private enterprise repos may show different patterns—especially for languages like C# that dominate in enterprise.

Sample size varies

TypeScript has 43K PRs; Swift has 1.1K. Languages with smaller samples may be less representative of true median values.

🔥 Our Take

Language matters less than team practices.

A 6x speed difference sounds dramatic, but it reflects ecosystem cultures more than technical constraints. Your TypeScript team can be slower than a well-organized C team. The real question isn't "what language should we use?" but "what review culture are we building?"

The takeaway: Don't blame your language. Benchmark against peers in your domain, set appropriate SLAs, and focus on the practices that actually move the needle—PR size, reviewer availability, and automated testing.

Related Research

Methodology

This analysis is based on 802,979 merged pull requests from GitHub Archive / BigQuery during October 2025. Language is determined by GitHub's primary language detection for each repository. Merge time is calculated from PR creation to merge event. Languages with fewer than 1,000 PRs were excluded for statistical reliability. For full methodology, see the complete study.

See your team's language breakdown

CodePulse shows you merge times by language, repository, and developer.