Our mission is to build free software tools to facilitate the study and analysis of Free and Open Source Software. Today we are best known for finding software licenses in text.
Installing the FOSSology software:
creates an empty software filesystem repository
creates a database (PostgreSQL) for metadata storage and retrieval
provides web and command line interfaces to populate the software repository
provides web interface for viewing reports
provides a batch subsystem for running lengthy analyses and reports
provides engines (run from the web or cli) for:
License analysis (analyzes EVERY file for license information)
-
metadata extraction from libextractor (jpg headers, pdf, doc, …)
file type
executing ad hoc sql
executing ad hoc scripts
A short paper on FOSSology using v 1.2 examples
Much faster and more accurate license detection. We call this the Nomos License scanner because it is based on a license scanner used internally by HP for several years. It's fast, it's accurate, but currently is unable to highlight the found license when you view the file and it requires one to update and recompile source code to add licenses. We are evaluating how to put these features in (back in) to a future release of FOSSology.
The previous license scanner (bSAM) has been moved to “Obsolete” in the top menu. We encourage people not to submit anything to the bSAM agent because we plan on removing it in a future version (probably v 1.3). Obsoleting bSAM also means that the license groups feature will go away (replaced by buckets, see below).
Copyright/
URL/email scanner. This is a new agent that beta users have liked, but has one flaw, it gets many false positives. This is unfortunate but beta users have still found this feature very usable and useful. The next version will do a much better job filtering out the false positives.
Ability to customize reporting categories. For example, “copyleft licenses”, “ship hold licenses”, … these are defined by the administrator of the system you are using. These categories are called buckets and a set of buckets used together is called a bucket pool. When 1.2 is installed, a very simple demo bucket pool will be created. To use it, you must go into Admin > Users > Account Settings and set your Default Bucket Pool. Do this before submitting a job to the bucket agent.
Much faster report (web page) generation
Cataloging both RPM and Debian package data
New in version 1.0.0:
Massive improvements to web-based user interface responsiveness, 50x or better speed-up, with retrieval and display of license reports in just seconds, for even the biggest reports.
Huge improvements in license analysis speed, as much as 10 times faster, with analysis of full linux distribution DVD images in 24-48 hours (depending on system performance, CPU, memory, etc)
License detection accuracy improvements - the library of licenses and terms that we search for and report has been carefully tuned based on lots and lots of real-world testing
Completely redesigned build and release system, makes building and installing FOSSology from source code much easier, and also enables creation of FOSSology packages for common Linux distributions
Software packaging of FOSSology for the Debian Linux distribution - For the first time, the FOSSology project has a complete set of packages to enable simple apt-based installation of FOSSology on most DEB-based Linux systems.
(beta) Software packaging of FOSSology for the Fedora distribution, using native RPM packages. These RPM packages should provide streamlined installation on most RPM-based systems (Redhat, SuSE, etc)
Automated test suite to provide web-UI testing and validation by the FOSSology project and any other downstream users
Self-test agent that allows FOSSology to inspect itself and report on common configuration problems that could impact FOSSology operation
View the complete release notes for FOSSology 1.0.0 for more details:
release notes
Celebrating 1 year of FOSSology!
Our mission is to build free software tools to facilitate the study and analysis of Free and Open Source Software. Today we are best known for finding software licenses in text.
