I recently testified before a joint hearing of the House Oversight Committee Subcommittees on IT and Government Operations alongside government and industry representatives on the roles of the General Services Administration’s (GSA) 18F and the U.S. Digital Service (USDS) in providing information technology (IT) services to federal agencies. The discussion was robust, with Committee members and panelists agreeing that the programs’ overall objectives of establishing a user-centric focus in IT, changing federal IT programs from the waterfall development to agile development, and injecting an outsider perspective to solve government missions were all laudable outcomes to pursue.
Some of the other takeaways I gathered from this opportunity include:
- While 18F’s and USDS’s objectives have broad support, both of these entities’ activities are largely unknown to their stakeholder audiences. In my testimony, I noted that it was not clear what these programs are, what they are not, and how they can be expected to operate. Everyone, including the heads of these entities, agreed that more could be done to make their efforts transparent.
- 18F only works on small-scale projects, however, their reach and influence can go beyond their deliverables. Commissioner Phaedra Chrousos, the new commissioner of the Technology Transformation Service at GSA overseeing 18F’s activities, noted that 18F has only worked on small projects that can be delivered quickly. They are also currently not self-sustaining, but their business model dictates that they should be.
- USDS does consult with agencies on larger project challenges, but it lacks needed focus. The Committee was clear that it preferred the service use the Government and Accountability Office (GAO) High-Risk List to identify projects and determine the distribution of resources for the programs most in need of transformation.
While it was good to get these and other elements of the discussion out in the open, including the findings and recommendations of the GAO report released in conjunction with the hearing, several issues and questions remain unresolved and I hope that the hearing is just the beginning of a continued dialogue to address them.
These unresolved questions and issues include:
- How do we get traditional IT vendors involved in innovative approaches to solving the government’s IT challenges? We have to find ways to permit out-of-the-box thinking in response to more of the solicitations the government releases.
- These programs don’t scale and at least one of them has no plans to do so. The hearing discussion did not identify an effective way to inject new technologies, agile development and agile acquisitions into the quest to modernize large-scale government IT programs, so do we need to task one or both of these efforts with addressing these questions or finding another way?
- 18F’s and USDS’s heavy reliance on open source and custom coding raises serious concerns. While open source is an entirely appropriate option for agencies to consider when looking to acquire software solutions for mission needs, we need to ensure that these are not the only options available as different missions have different demands. Making open source software the only preferred solution doesn’t adhere to the tenets of tech neutrality. It also limits options and competition, which means we may not be achieving best value for the taxpayer.
- Questions remain about what the Technology Transformation Service at GSA will be doing. By their own admission, 18F is a relatively small program focused on small tech projects, so why does it need such a big home? Creating an entirely new service at GSA needs more socialization so everyone – GSA employees, the vendor community and the broad government customer base – fully understands the justification for its’ creation and how it fits into the existing mission areas.
- These entities cannot continue to operate as both buyer and seller. There is an inherent conflict of interest when the entity that consults on how to solve a problem then also offers to sell you the solution. The government outlawed such arrangements for contractors, so they should not be permitted to create such a dynamic internally.
- Congress needs to establish hiring authorities. Using presidential appointments, which expire at the end of the president’s term, doesn’t create a sustainable means of identifying and onboarding the tech talent these activities seek. In order to attract top talent, such authorities should balance rapid onboarding, non-career stints of service, and preservation of interests in the private sector. That includes ensuring lawmakers protect the interests of the agencies and the taxpayer for things like personal or organizational conflicts of interest.
- The conversation needs to continue about 18F and USDS after this presidential administration. The hearing and report raised more questions and unknowns that still need to be addressed and determined. It is not clear how these entities are different in function, and even after the hearing, it has not been adequately explained why we need both. There are also unanswered questions about how these activities engage with the IT vendor base in the public sector, and how are they selecting the products and services they are recommending to their agency clients.
There are completely rational and acceptable answers to all of these questions. Without knowing the answers, however, we risk creating a bifurcated acquisition system that could add unneeded complexity to a dysfunctional acquisition system, undercut competition, and not deliver the best value and innovation to the taxpayer. These activities hold a lot of promise, but will they be able to address these and the other concerns in the short time that this administration has remaining?