SAS is promoting the idea that our customers can use programming languages which they're already familiar with (currently Java, Python, Lua, and R) to gain access to the analytic power of SAS.
When a software is described as open source, that means that the programming code is visible for you to see. Open-source software is a critical part of today's modern IT infrastructure. To this end, SAS is leveraging other open-source, third-party software to deliver features and functionality using standardized technologies already familiar to programmers, sysadmins, and analysts.
The promise of open-source software is that if you're a programmer, then you can read the code, follow it, understand it, and verify it's written to behave as promised. You can even identify problems and deficiencies. And then you even have the ability to change the code yourself and, if you want, submit those changes back to the managing entity for consideration to include in the next release.
Most open-source software is freely available. Since the source code is always viewable, then it's unusual to try and enforce a software usage license of the kind you typically see with closed-source software (like SAS or Microsoft Office). Instead, companies which make money from open-source software usually do so by having their customers pay "only" for support and training services. For example, CentOS is a freely-available, open-source Linux operating system - but there's no official support organization you can call with problems. If you have a problem with CentOS, then you'll likely use Google to search for a solution, or post a query on a web site like StackOverflow.com or SuperUser.com, or else if you like to get your hands dirty, then learn how to program operating system frameworks and make your own fixes. Alternatively, the Red Hat company offers Red Hat Enterprise Linux. Besides the name and a few details, RHEL is essentially the same thing as CentOS and its source code is also freely viewable. However, to get RHEL as well as timely software updates, you must enter into a paid agreement for Red Hat's support services. With those services, if you have a problem with RHEL, you can pickup the phone and call a real person, negotiate service-level agreements, contract for consulting services, etc.
From a SAS perspective, our customers can run their SAS software in environments with access to real support teams similar in concept to our own SAS Technical Support. So while SAS software can technically function on a machine using CentOS as the operating system, customers can use RHEL - or really, any technically compatible OS - for which the customer has a direct line to call for OS support issues (whether that OS is open or closed source doesn't matter).
This freely available open-source software model with paid support services is utilized by lots of software vendors, not just for operating systems but also for databases, application engines, programming languages, and much more. And we use several of those technologies as part of our SAS solutions offerings.
For SAS 9.4:
For SAS Viya:
As you can see from this list, the SAS 9.4 and SAS Viya rely heavily on open-source software to deliver powerful and flexible analytics capabilities. Take away the application-level open-source technologies and we still support running on open-source Linux operating systems. And even if we drop the operating systems, there are still open-source technologies employed which I haven't covered here, including TLS (formerly SSL), Kerberos, LDAP, and Hadoop.
With SAS Viya, we also provide openness to our broader analytics platform in the form of RESTful HTTP APIs and third-party programming languages so that the power of SAS can be used by an ever-growing community.
Rob Collum is a Principal Technical Architect with the Global Architecture & Technology Enablement team in SAS Consulting. When not looking at open and freely available source code, he also enjoys sampling coffee and m&m’s from SAS campuses around the world.