Monday, December 4, 2023

Open Source Now is Ubiquitous in Enterprise Computing

Though there was a time, decades ago, when open source might not have been widely trusted and used in enterprise settings, that time has gone, as virtually all enterprises use open source code. 


Black Duck Software in 2020, found that open source software is used in 98 percent of enterprises. The study also found that open source software accounts for an average of 60 percent of the code base in enterprise applications.


The Linux Foundation in 2021 argued that open source software generates $6 trillion in economic value each year, and supported 16 million jobs worldwide.


Most studies do suggest substantial use of open source by enterprises. 


Study

Usage

Date of Publication

Publisher

Black Duck Software

70%

2022

Black Duck Software

North Bridge

66%

2022

North Bridge

Linux Foundation

80%

2022

Linux Foundation

Opensource.com

50%

2022

Opensource.com

OSS Impact Report 2023

65%

2023

The Linux Foundation

Open Source Software: Economic Impact and Benefits

60%

2022

BSA

The Value of Open Source: A Study by the Open Source Initiative

55%

2021

OSI

Open Source Software: The Real Revolution

50%

2020

O'Reilly Media


There might always be some tension between a community-driven open source software initiative and vendor-led or vendor-organized open source projects, but there arguably also are advantages, including less risk of sudden abandonment, greater resource commitments and more direction.  


On the other hand, concerns might be raised about vendor influence or control, sustainability if the vendor loses interest (though that can be an issue for any open source initiative), community engagement or transparency. 


Consider Android (operating system), Kubernetes (container orchestration platform), Azure DevOps (DevOps services), TensorFlow (machine learning platform), .NET (cross-platform development framework), GitHub (hosting service), Cloudant (NoSQL database, OpenShift (container platform), Hyperledger Fabric (open-source blockchain), Red Hat Enterprise Linux, OpenShift (container platform) or JBoss EAP (application server), Prometheus (monitoring and alerting toolkit), OpenTelemetry (instrumentation framework), Jaeger (distributed tracing system), Apache Spark (distributed data processing framework), PyTorch (machine learning library), React Native (JavaScript library for building user interfaces for mobile appsGrafana (data visualization platform) or Apache Kafka (distributed streaming platform).


No comments:

Have LLMs Hit an Improvement Wall, or Not?

Some might argue it is way too early to worry about a slowdown in large language model performance improvement rates . But some already voic...