It’s been much more than a century in internet time since the founder of the World Wide Web stood before the first International Open Government Data Conference (IODC) and laid out a vision for countries to unlock their files and share them with the world. In the six years since Tim Berners-Lee called on governments to put data on the web, dozens of nations have done so, convening at the second International Open Government Data conference in D.C. to check in on our collective progress and challenges, and again at the third conference in Ottawa in 2015. I’ve attended them all, in one capacity or another. Here are my takeaways from the 2016 iteration in Madrid, Spain, using Wired’s “Wired, Tired, Expired” frame.

What’s wired

1. Recognition of the impact of massive leaks on the world in a plenary panel that addressed Wikileaks, the Panama Papers and the role of whistleblowers that grappled with the impact of releases on this scale, their impact on policy and politics and their role in holding power to account. This touches upon what we’ve said about how weaponized transparency can affect specific countries, with respect to their policies and programs around open data and freedom of information, and the broader scope of global opinion about transparency, technology and journalism. There’s a reason ethics and responsibility loomed large in the discussion in Madrid: both matter.

2. Explicit discussion of how privacy and transparency balance, including the introduction of draft transparency and privacy principles from Access Info that usefully articulate what is public and what is private, and how governments should approach these questions. Sunlight has been thinking about how to protect privacy while releasing open data, applying a public interest balancing test to transparency, and applying these principles to voting data. Our work with Seattle on its open data policy specifically kept this balance in mind.

3. Focus on inclusion and exclusion, from who is represented in open data to who is represented on stage to who is consulted about policies or disclosures. As a technology-dominated space, open data conferences have been typified by an obvious lack of diversity on the stage. That changed a bit in 2016, although much work remains to be done: According to the conference data, attendance was roughly equal between the sexes but many “manels” remained.

4. Grappling with the impact of politics and transitions on open data, from Canada’s successful shift in governments to the upcoming election in the United States to dozens of countries and municipalities around the word. Advocates and officials alike have to think through how and where open data is institutionalized and connected to rights to access information, including the engagement of political allies across the ideological spectrum, or risk losing hard-fought gains. Public optimism about open government and open data has to be weighted with a lack of understanding or knowledge about how data underpins apps and services, and the historic lows in trust in institutions, from government to media.

5. Madrid. The Spanish people were extraordinary hosts to the conference, beyond the terrific refreshments to the city itself. The IODC itself was bigger and more professionally produced than ever, illuminated in red lights, animated by a thumping soundtrack and livestreamed online. The gigantic venue was a startling contrast to the first conference in a theater in the Interior Department in D.C., where I struggled to maintain a liveblog and audio stream over flaky wifi.

What’s tired

1. Dropping open government from the open data conference. In 2010, the first International Open Government Data Conference featured high aspirations for the release of data from the U.S. government, validated by the web’s creator and committed to by a young Obama administration that was borne into office by a swell of internet-infused optimism. Some of the aspirational hopes expressed by the participants have come to pass, including Berners-Lee, but the extension of open data to a broader frame of nonprofits, corporations and academia has diluted the event’s initial focus on accountability and transparency and introduced other competing goals without bringing in the startups, campaigns and technology companies that are presenting state of the art data science in other contexts.

2. Openwashing and openwishing. In 2016, what we wish open data will be in the world should be tempered with the experiences of the past decade, from successes to failures to the muddles in between. Similarly, politicians shouldn’t be allowed to claim credit for thousands of datasets on thousands of platforms that comprise links and PDFs with poor quality data, nor substitute publication of open data that’s useful for economic outcomes for those that are relevant to accountability. When a host country has yet to meet many of its own commitments to freedom of information, open data and open government, openwashing represents a risk to legitimacy that shouldn’t be brushed away. When Russia’s open data working group has a place in the discussion, it’s worth remembering that the country is a case study for how open data can be focused on services without press freedom, revealing corruption or holding power to account.

To her credit, Suneeta Kaimal, the chief operating officer of the National Resource Governance Institute, highlighted the weakness of many open data commitments in Open Government Partnership and led a discussion about what can be done to improve not just the scope and ambition of the goals but ensuring that countries achieved direct impact upon the lives and knowledge of their people. Similarly, Anne Jellema, the director of the Web Foundation, warned the conference of the risks that a closing world poses to the open agenda. If open data ends up coexisting with authoritarianism, we will all lose more than ground an argument.

It’s critical that advocates invest in achieving legislative successes like the DATA Act that combine transparency and accountability with structural reforms regarding how a nation tracks and discloses its spending to the public through an open, iterative process. I saw many presentations that demonstrated many attendees of IODC understand this. It’s up to all of us to hold governments accountable for following through on their rhetoric and publishing what they spend and disclosing what they do in our name.

What’s expired

1. Novelty. The first decade of open data as an idea is almost over, along with the freshness of the notion that governments should use technology more effectively to open themselves up. On one hand, that might account for the relative lack of announcements of new platforms or dataset releases or the launch of new companies or legislation. On the other, the increasing maturity of the expanding open data community around the world puts a higher premium on researchers, leaders and officials sharing their disappointments and failures with their peers, enabling new participants to learn from mistakes and leapfrog ahead to make new ones. Now that open data isn’t new, researchers and governments can and should dig into what’s working — and what’s not. After a decade of working on open data as an idea — and many more before that working to access and share documents through freedom of information laws — the deadline for more evidence is getting close. Politicians will always question transparency, which puts a premium on demonstrating why it matters in terms that the public understands and can apply in their lives.

2. The relative absence of politicians and journalists on the stage and the conference in general. As the master of ceremonies at last year’s International Open Data Conference in Ottawa and moderator of several discussions, I couldn’t help but notice that a single panel on data journalism in Madrid was the closest that a key infomediary in the space came to the getting a stage for the experience of journalists trying to get data out of government.

Friday’s discussion about leaks and journalism featured two journalists was moderated by a foundation staffer. While the questions that panelists fielded in many contexts were relevant, there wasn’t enough followup discussion that pushed for more evidence of impact or directly engaged with some of the unresolved challenges around why the world hasn’t changed more as a result of open data disclosures, or how it happened if it did. If publishing and analyzing open data fulfills its promise for holding power to account that advocates around the world want to see, journalists will be a critical part of that story – and politicians and corrupt will be threatened by that outcome. That means that engaging with the role of politics and media can’t be an afterthought, nor explicitly talking about power and influence.

3. Hundreds of mentions of open data “portals” where the broader public is expected to visit, download open data and put it to use, or download government apps. Platforms publish data for re-use, where consumer platforms, applications and services. Portals host data for use on the site.

Publishers can and are engaging the public in helping to make sense of PDFs and convert them to open data, but governments need to move away from the “if we build it, they will come” frame for publishing open data and move aggressively to respond to demand, including analyzing existing freedom of information requests to prioritize proactive disclosures, and releasing data in cooperation with civil society partners that enable specific missions, from education to environment to campaign finance to budgeting to public health. In a moment when billions of people are accessing the Internet using smartphones and technology companies are integrating and using data through conversational appliances like Google Home and Alexa, it’s critical that governments double down on creating open data at the source, cataloging and archiving it in modern databases, analyzing it to improve operations and regulatory actions, and disclosing it in bulk for reuse.