What to make of OMB's software licensing initiative
- By Michael Garland
- Dec 23, 2015
On a sleepy Monday, just as many industry and government workers took a break for the holidays, the Office of Management and Budget's Anne Rung and Tony Scott released the most dramatic and potentially tax-saving "draft" category management plan for commercial software in the history of federal IT acquisition.
On the heels of their desktop and laptop memorandum of October, Rung and Scott have attempted to square the corner on the government's unnecessary and wasteful duplication of expenditures on commercial IT products. Using the Federal IT Acquisition Reform Act and OMB's own category management initiatives as the springboard, the agency is aggressively seeking to leverage the massive buying power of the U.S. government in an attempt to force software companies to treat the government more like a single user.
Here, in a nutshell, is the long-standing problem: Commercial software companies tend to license their products to the smallest economic unit inside a federal agency. A copy of a popular commercial product might be licensed for use to a specific user, for a specific sub-agency or (when lucky) at the entire agency enterprise level.
This licensing approach serves the software industry well because it tends to preclude a more informed view by government of what it is collectively spending for any particular product. In practice, it allows software vendors to carve up the universe of potential government consumers to craft the most advantageous deal possible under the circumstances.
A particularly savvy agency has a chance to get a much better deal than an agency with less experience negotiating a software license for the exact same product, but that hard-won bargain is not then available (or even known) to others across government.
This advantage for industry hasn't necessarily been malicious, and it has been totally supported by the way Congress allocates budgets on a program or agency basis, which has historically made it almost impossible for the government to aggregate its commercial software requirements for a single purchase.
The core of Rung and Scott's approach to solving this problem is centralized information sharing. Each Chief Financial Officers Act agency will have to appoint a software czar who will take ownership of the agency's software portfolio in a way that forces a comprehensive understanding of what commercial software assets that agency owns and what it spends on them. This information, along with specific detail on negotiated prices, will be published internally for all government agencies to see.
Additionally, the newly formed Enterprise Software Category Team -- co-managed by the General Services Administration, the Defense Department and OMB -- will endeavor to establish governmentwide licensing agreements that will become mandatory for use, unless an agency can document a cost savings by choosing an alternative method.
The engine that will drive the ability to manage software assets is the requirement that agencies finally deploy Continuous Diagnostics and Mitigation tools. Collectively referred to as CDM, these software agents deploy on networks and aggregate information about what software resides on which device. When deployed correctly, the tools can provide important detail down to the version and patch level of software.
As the Government Accountability Office rightly pointed out in 2014, most government agencies had little ability to list their software inventories and as a result were vulnerable to cyberattack because some software was hopelessly out of date. And from a buying perspective, an agency cannot possibly aggregate its spending for a better deal if it has no idea what it currently owns. Government maximization of CDM tools will take a while, but it is an incredibly important first step.
The foundational steps outlined in Scott and Rung's memo will likely save taxpayer money immediately, but the long-term ramifications are even more dramatic. When FITARA was contemplated, the House committee that drafted the original version was taken with a report from the United Kingdom, which proclaimed that the U.K. government was attempting to force commercial software providers to license software directly to the Crown.
This meant that, where such an agreement could be struck, commercial software licenses were written for use for the government as a whole. In this context, software purchased for use at the Ministry of Defence but no longer needed there could be shifted to the National Health Service.
The advantage of this approach is startlingly obvious and is one that industry already enjoys. In point of fact, when a large company like Walmart licenses commercial software for its accounting department, it is most certainly free to transfer it to its marketing department at a later time, as long as it does not exceed the licensing quantity metric -- typically the number of users.
When CDM tools can finally provide the government with an accurate assessment of its software assets, the next logical step will be to demand that software companies allow assets to move freely anywhere in the government.
So kudos to Rung, Scott and the category management team at OMB. Their latest initiative to gain control of the software estate puts real meat on the bone and is a profoundly useful step in the right direction.
Michael Garland is the founder of Garland, LLC, a consulting firm that advises clients on issues related to federal procurement law and the business of IT.