Impose proper licensing and streamline procurement

(this page is part of my 2011 report on “Open Data: Emerging trends, issues and best practices”. Please follow that link to reach the Introduction and Table of Content, but don’t forget to also check the notes for readers! of the initial report of the same project, “Open Data, Open Society”)

As with the first report prepared for this project, we will not delve into the details of how to license data because this topic continues to be followed and debated in all details by LAPSI and other projects or researchers. We will simply confirm the importance of establishing a proper license, at the national level, for all Public Data, that makes them Open in the right way and makes sure that what is opened stays open and that don’t demand what isn’t possible to enforce (e.g. attribution), because, quoting again Eaves, “no government should waste precious resources by paying someone to scour the Internet to find websites and apps that don’t attribute”.

We want, however, to spend a few words about another legal/administrative side of the issue, that is procurement. Traditional procurement laws are very likely not flexible enough, in most countries, to handle the implementation of data-based public services. Here’s why.

We know that if Public Data are Open, everybody, from volunteer activists to hired professionals, can very quickly write or maintain simple software applications that help to visualize and use them in all possible ways. Paradoxically, this is a problem when an Administration either wants to set up an Open Data programming contest (that besides being unexpensive, it’s much simpler to organize and join than traditional tenders or grants) or needs to just pay somebody to write from scratch and maintain some new program of this type, or customize existing ones.

The reason is that, just because this type of software development is so quick, even hiring a professional to do it, or setting up a contest would be… too unexpensive to be handled with default procurement procedures. Quoting from Day Two: Follow the Data, Iterating and the $1200 problem:

A big problem for cities is procuring products under $10,000. How does a city pay for an awesome application like [SeeClickFix]( when it doesn't fit the normal year-long planning and two-year implementation in the millions of dollars? In Tuscon, Andrew Greenhill tapped the Mayor's general budget for it, instead of trying to get the IT department to shell out. In San Francisco, Ed Reiskin uses discretionary spending. But every time, procurement gets messy. In reference to nepotism laws, Ed worries that he'll appear "like I'm giving my buddies dollars." Building great products for cities has to include finding great strategies to pay for them. In San Francisco, Jay Nath doesn't even have a budget…which, he says is 'liberating' because he doesn't need to go through procurement.

The same issue is denounced as an obstacle to innovation and cost savings in New recommendations for improving local open government and creating online hubs:

John Grant focused on a major pain point for government at all levels for tapping into the innovation economy: procurement issues, which civic entrepreneurs run into in cities, statehouses and Washington. "It is time to look at these procurement rules more closely," he said, and promote higher levels of innovation. "There are a lot of ideas are happening but a lot of rules restrict vendors from interacting in government," said Grant. Turner-Lee observed that traditional procurement laws may also not be flexible enough to bring more mobile apps into government.

Current procurement laws aren’t partially incompatible with an Open Data world only at this level, that is when it’s time to procure software that makes the data useful. Even bigger problems and inefficiencies can be introduced at the beginning of data life, that is when data collection and processing services are procured. We’ve already explained that forgetting to impose the right license is one of the problems, but it’s not the only one. Even future organization of all the foreseable data management activities should take advantage of the flexibility provided by data openness. Here is how Tim Davies summarizes this point:

Right now `[public]` bodies often procure data collection, data publishing and data interfaces all in one block (as seems to be the case with Oxfordshires real-time bus information - leading to a roadblock on innovation) - and so without these layers being separated in procurement, some of the benefits here stand to be lost.

Changing procurement of information/data-rich public services would be, of course, only the first step of a general reform of procurement laws and regulations. After management of Open Data has been simplified, it becomes time to implement similar simplifications to procurement of everything else. In fact, in such a scenario, there would be much less possibilities for the loopholes, frauds and inefficiencies that forced local procurement procedures to become so slow and complicated: since the public budget and other relevant public data would already be fully open, errors and other problems would surface and be fixed much more quickly and reliably than today, even assuming that they would continue to appear with the same frequency.