US Government Successfully Issues Contract For Open Source Code… For $1
from the why-that's-1/37-of-a-screw dept
One of my very first jobs in Silicon Valley was to try to help an internet startup get a big juicy contract with the US government (specifically the Department of Defense). The whole process was a disaster of epic proportions, in which I learned a ridiculous amount about government procurement, none of it good. At one point, I believe the company I worked for was paying a 5-figure-per-month “retainer” to an ex-high ranking military guy, mainly so that he would go out and drink a lot of bourbon with his DoD buddies and award us a no-bid contract before anyone realized it should be put out to bid. And, of course, as an internet startup, we didn’t have a GSA contract, and had to find a sham “partner” who would officially get the contract, under which we’d be a subcontractor. And, of course, we were asking for millions of dollars in government cash, and the technology we had in place wasn’t anything like what the DoD was actually looking for. In short, the whole thing was a complete mess. That was two decades ago, so I’d hope that things had changed, but we’ve heard so many stories of the ridiculousness of government procurement, that I doubt it’s changed that much.
But… there are glimmers of hope. And a neat little technical division of the GSA, known as 18F, which is modeled on startup culture and bringing a much more innovative take on technology to the government (it’s the same group that’s going around making the rest of the federal government encrypt their websites…), recently ran an experiment which (somewhat unexpectedly to all involved) resulted in the GSA awarding a $1 contract for a bit of open source software. And, yes, that’s ONE DOLLAR.
A few weeks back, 18F announced this experiment in “micro-purchase” contracts, with the idea being to see if they could create a quick and simple process to both (1) do small focused contracts and (2) make it easy for smaller tech firms to actually provide their products and services to the government. So 18F posted the details of a specific problem it was trying to solve to Github, and then created a Google form, to serve as a sort of blind reverse auction. Here’s how 18F described things:
If you?re interested in bidding, the closing time for the bid is 12 p.m. on Thursday, October 29, 2015. The opening bid starts at $3,499, and the lowest bid at the closing time will have 10 working days to ship the code necessary to satisfy the criteria. If the criteria are met, the vendor gets paid. If the criteria aren’t met the vendor shall not receive payment, the next lowest bidder will have the opportunity.
Got it? Makes sense, as a way to try to keep costs down, but not to make it so crazy low that it’s not worth someone’s time (or where they’re not able to deliver). Except… no one expected someone to (a) bid $1 and (b) then deliver working code that not only met the requirements, but exceeded them. But that’s what happened:
Then, there was the $1 bid.
When we received the $1 bid, we immediately tried to figure out whether it was intentional, whether it was from a properly registered company, and whether we could award $1. We contacted the bidder and we confirmed that the bid was valid, that the registration on SAM.gov was current, and that the bid would be the winning bid. It was a plot twist that no one here at 18F expected. This unexpected development will no doubt force us to rethink some of our assumptions about the reverse-auction model.
In some respects, this result was the best possible outcome for the experiment. It proved that some of our core assumptions about how it would work were wrong. But the experiment also validated the core concept that open-source micro-purchasing can work, and it?s a thing we should try to do again. A few weeks ago, micro-purchasing for code was just an idea, but now that we?ve done our first experiment, the data demonstrate that the idea has potential and can be improved upon.
You can see the “winning” $1 pull request by Brendan Sudol over at Github, which went above and beyond the requirements:
Not only did Brendan Sudol meet the requirements of loading the data, the new code had 100 percent test coverage, an A grade from Code Climate, and included some new functionality to boot.
18F notes that the experiment turned up a few other interesting tidbits, including that of the 16 bidders, 8 of them registered to be a government contractor on SAM.gov after the project was announced (showing that, indeed, the process of becoming a government contractor appears to be getting much easier). They also noted that the highest bidder was for $740 with the smallest (obviously) being $1. The most common bid was $50.
Obviously, the $1 bid is both a bit of a gimmick, and something where Sudol recognized that it wouldn’t be that much work to provide the code and it apparently seemed like a project worth doing. I doubt that we’ll regularly see the government awarding $1 contracts, and the GSA is hardly likely to become the government version of TaskRabbit or Mechanical Turk. But it’s still interesting to see the ways in which the terrible inefficiency of government purchasing might go down — and more recognition over the idea that, if the government just needs a few lines of code, it doesn’t need to award a multi-million dollar contract to some stodgy old “services” company.