A Right to Information Assumes It Still Exists

Key Takeaways

• Access to information laws grant a right to request records. They generally say much less about whether those records must be preserved and remain reachable.
• A significant share of web content published a decade ago is no longer accessible through link rot, site closures, and platform shutdowns.
• Most digital preservation is done by nonprofits and volunteers operating without a public mandate or stable funding.
• Accountability depends on being able to check what was said or published at the time. That check only works if the record survives.

Today is the International Day for Universal Access to Information, a UNESCO observance recognizing access to information as a foundation of open government and informed public life. The right it marks is real and worth defending. It also rests on an assumption that nobody guarantees: that the information you have a right to access will still be there when you go looking.

A right to request a document isn’t the same thing as a commitment to keep the document.

Access Rights Assume Preservation That Isn't Promised

Most freedom of information frameworks are built around a request process. A person asks a public body for records, the body responds within some period, and there are exemptions and appeal routes.

That structure works well for records that exist. It does very little about records that don't.

UNESCO's own framework for this observance treats access to information as fundamental to transparency and accountable governance. The organization has a separate instrument, its recommendation on preserving documentary heritage, including in digital form, which addresses the other half. The two are related in practice and usually handled by different institutions with different budgets.

Preservation tends to be the half that loses. It has no requester pushing for it, no deadline attached, and no visible failure when it goes wrong. A missing record doesn't generate a complaint, because nobody knows to complain about something they can't find.

The Record Is Thinner Than People Assume

The internet gave the impression of permanence. Anything published was public, indexed, and findable, and the assumption followed that it would stay that way.

Research from the Internet Archive's Vanishing Culture project, published in April 2026, documents how much of that assumption doesn't hold. Websites close. Links break. Platforms shut down and take their hosted content with them. Organizations reorganize their websites and old pages stop opening.

None of it requires anyone to decide something should disappear. It happens through ordinary institutional churn, and the result is the same either way.

What makes this an accountability problem rather than a nostalgia problem is which records tend to be least stable. Well-funded institutions maintain archives. Major publishers keep back catalogs. The material most likely to vanish is the material nobody was resourced to maintain: local government pages, small publications, community organizations, and the working documentation of bodies that have since been restructured or dissolved.

That's frequently the exact material an accountability question turns on.

Who Is Actually Doing the Preserving

Here's the part I find most striking about how this currently works.

The closest thing the public has to a systematic record of the web is maintained by nonprofits and volunteers. The Internet Archive's Wayback Machine has become the default reference for what a page said before it changed, used by journalists, researchers, courts, and anyone trying to establish what a public body published at a given moment.

It performs that function without a public mandate, without stable statutory funding, and without any requirement that it continue to exist. It's also faced legal challenges that affect what it can preserve and provide access to.

An accountability infrastructure that everybody relies on and nobody is obliged to maintain is a fragile arrangement. It works until it doesn't, and the failure mode isn't an outage anyone would notice. It's a gradual thinning of the record, visible only to the person who goes looking for something specific and finds it gone.

This connects to something we wrote about earlier this year on how digital loss falls unevenly across different communities and kinds of content. The accountability dimension is a different consequence of the same underlying gap. What isn't preserved can't be examined, and what can't be examined can't be challenged.

It also affects the documentation trail in other areas. Platform transparency reports and government request records are only useful as an accountability source if past editions remain available for comparison. A report that exists only in its current version tells you about this quarter, not about a trend.

What Today Is Actually Asking For

Access to information is treated as a right in most legal systems and as a democratic requirement in the framework this day exists to promote. Preservation is what makes that right operational rather than theoretical. Without it, the right amounts to permission to ask about something that may or may not still exist, with no recourse if it doesn't.

That's not an argument for keeping everything forever, which isn't realistic or always desirable. It's an argument that decisions about what survives shouldn't be made by default, through neglect, budget cycles, and the incidental consequences of website migrations.

The International Day for Universal Access to Information exists to affirm that the public should be able to find out what institutions do in their name. Finding out requires a record. Right now, a meaningful portion of that record is held together by organizations doing it voluntarily, and nobody has decided whose job it is if they stop.