Video Game Preservation in 2026: Archives vs. Obsolescence

The state of play

Video game preservation in 2026 is no longer a hobby for people with too many hard drives and too little sleep. It is an institutional argument over who gets to remember the medium, who gets to study it, and who gets to pretend that disappearance is a feature rather than a bug. The current field is shaped by the Video Game History Foundation, the Internet Archive, the Library of Congress, and the Software Preservation Network, with public reporting in 2025 treating preservation as a combined problem of playability, source material, and metadata rather than software alone.

The ugly arithmetic is the headline: the VGHF has argued that 87% of video games released in the United States before 2010 are lost or at risk of being lost, based on its 2023 study, and that figure still functions as the field’s favorite warning label. It is not a slogan. It is a reminder that the industry spent four decades manufacturing cultural artifacts inside a legal regime that mostly does not like archives, a market structure that expects planned obsolescence, and a hardware ecosystem where plastic, magnetic media, optical discs, and server authentication all age badly for different reasons.

The result in 2026 is a preservation sector that looks more serious than it did five years ago, but also more fragile. The work is increasingly professional, and the barriers are increasingly structural. The field can digitize magazines, recover source art, catalog cartridges, and emulate old software in browsers. It still cannot reliably guarantee lawful remote access to out-of-print games at scale.

The hard numbers

The most useful way to understand preservation is to stop using the language of sentiment and start using inventory counts, access models, and legal choke points. By 2025, the Library of Congress’s Packard Campus in Culpeper, Virginia reportedly held more than 4,000 videogame cartridges and CD-ROMs within a broader audiovisual collection, making it one of the largest public institutional holdings in the United States. That is not a complete national archive. It is, however, enough to matter.

The Library of Congress’s video game canon adds another layer. Originally created in 2007 with Henry Lowood at Stanford University, it was still being updated in 2025 and statistically analyzed against 80 Best Games lists published between 1985 and 2025. That matters because canon formation is preservation by another name: what gets counted gets archived, cited, and eventually justified.

On the access side, the Internet Archive’s browser-based emulation through Emularity remains a core public-facing preservation method, letting users play older console, arcade, and computer games in a web browser. The Archive’s Classic Software Preservation Project (CLASP) works differently: it is a “dark repository” that collects original consumer materials for preservation while keeping holdings restricted until copyright expires or rights are granted. In other words, one project privileges access, the other privileges custody. Preservation people call this nuance. Lawyers call it a survival strategy.

Meanwhile, the preservation community’s access ambitions have run into the wall that always shows up when archives become more competent. In 2025, a game preservation discussion involving VGHF and the Software Preservation Network said the two groups had been working since 2021 on a way for libraries and archives to remotely share digital access to out-of-print games. The same reporting noted that a 2025 U.S. Copyright Office decision rejected a requested DMCA exemption that would have expanded preservation access for libraries and archives, which VGHF said continued to impede archival work.

That is the basic shape of the field in 2026: one side is building collections; the other side is building fences around the collections; and everyone is pretending the argument is about the future when it is mostly about the past.

Video Game History Foundation

The VGHF has become the most visible specialist institution in the space because it does several things at once without confusing them. According to 2025 reporting, it said it launched in January 2025 as a nonprofit dedicated to preserving, celebrating, and teaching video game history, and it is building a dedicated research library and digital archive. That dual structure matters. A preservation organization that only stores things is a warehouse. One that only publishes things is a blog. The VGHF is trying to be both archive and research infrastructure.

The organization’s library mission is unusually concrete. It has described its goals as assembling near-complete runs of major North American video game magazines going back to the late 1970s, plus international magazines, history books, art books, and rare source materials. That sounds quaint until you remember how much game history exists only in periodicals, guidebooks, advertisements, preview coverage, and developer interviews. For many titles, the magazines are the only surviving record of what the industry said a game was supposed to be before consumers, publishers, and patches decided otherwise.

VGHF also says it is actively collecting and preserving source code, development tools, raw art, and documentation, which is essential because these assets are often lost even when the final game survives. This is the part of preservation that makes publishers uncomfortable. A working final build is useful. Source material is evidence. Evidence is harder to curate away.

There is also the unglamorous but important matter of digital access. VGHF says its Digital Archive is already online and provides access to digital collections, full-text documents, magazines, transcripts, and other historical material for researchers. In plain terms, it is trying to make the history usable without forcing everyone to make a pilgrimage to a single reading room. That is not a trivial upgrade. It is the difference between preservation as symbolism and preservation as scholarship.

The organization has also reported recovering lost art for video game companies, a reminder that preservation is not limited to consumer software. Corporate art, internal documentation, press kits, and development records are often the first casualties of mergers, office moves, and whatever the industry calls housekeeping. The games may survive. The context usually does not. VGHF’s work suggests the opposite of the usual fantasy: the important thing is rarely just the executable.

There is a reason historians keep returning to magazines, documents, and development artifacts. Games are not self-explanatory. They are engineered objects, yes, but also legal objects, marketing objects, and industrial objects. A cartridge without packaging, notes, build tools, or period coverage is a body without a biography.

Internet Archive and CLASP

The Internet Archive remains the best-known generalist institution in digital preservation because it does not act as though “old file, new server” counts as a philosophy. Its game work still includes browser-based emulation through Emularity, which lets users play older console, arcade, and computer games in a web browser. The practical value is obvious: accessibility, discoverability, and public demonstration. The archival value is subtler: a browser session can preserve interaction patterns and make obsolete platforms legible to people who do not maintain vintage hardware for recreation or ritual.

The Archive’s other game-related preservation model is CLASP, the Classic Software Preservation Project, described in 2025 as a “dark repository” that collects original consumer materials for preservation but keeps holdings restricted until copyright expires or rights are granted. That phrase does a lot of work. It acknowledges that possession and access are not the same thing. It also reveals the contemporary compromise in archival computing: keep the object now, open it later, and hope the law does not turn the waiting period into a permanent tomb.

The Internet Archive’s public storytelling around games has also emphasized that preservation is not just about software binaries. A 2025 blog post highlighted games such as The Last Express, emphasizing that archives and volunteers are preserving not just playable software but also the documents and artifacts that explain how games were made. That is the correct instinct. A playable ROM without context is useful. A playable ROM with design docs, manuals, source notes, and contemporary criticism is history.

The Archive’s model is also the clearest proof that preservation is now a service stack. At one end is the interface people see. At the other is a tangle of storage, scanning, metadata normalization, and legal triage that almost nobody wants to fund until something disappears. Then everybody becomes very interested in cultural memory for about three news cycles.

Copyright, access, and the DMCA

Preservation is not blocked by technology alone. It is blocked by the ordinary hostility of copyright law to archival use that looks too much like access. In 2025, the U.S. Copyright Office decision reported by Game Developer rejected a requested DMCA exemption that would have expanded preservation access for libraries and archives, and VGHF said that decision continued to impede archival work. If that sounds absurd, it is because preservation law has spent years pretending that a library digitizing an out-of-print title for supervised access is morally equivalent to piracy with better typography.

The legal framework matters because the technical problem is not the hardest one anymore. You can image old media, replicate hardware behavior, and distribute metadata more easily than ever. What you cannot do, routinely and at scale, is give researchers lawful remote access to orphaned, out-of-print, or server-dependent games without entering a swamp of rights management and statutory exceptions. The industry likes this arrangement because scarcity is an easy default. Archives do not.

The Software Preservation Network has emerged as an important companion institution because it is not trying to turn archives into fan sites. Its work, in conjunction with VGHF since 2021, has centered on creating a remotely accessible model for out-of-print games in library and archival settings. That is not a theoretical issue. It determines whether a game can be studied in practice or merely admired from a distance.

The broader consequence is simple. Preservation without access becomes hoarding. Access without rights becomes litigation bait. The field in 2026 is trying to occupy the narrow space between those two failures.

How we got here

Game preservation did not suddenly become important because the industry matured. It became important because the first generation of software started falling apart in public. Media rot, shutdown servers, lost source repositories, abandoned licenses, and platform shifts all made the point for us. Preservation work was once dominated by hobbyists, dumpers, forum archivists, and people who understood that a game’s “official” existence is often a misleading legal fiction.

The Library of Congress canon created in 2007 is worth recalling because it shows when institutions began to concede that games belonged in the cultural record. But a canon is not a preservation system. It is a declaration of importance. The actual archive work had to be built later, and mostly by organizations that understood the difference between prestige and practice.

That is why the 2025 launch of the VGHF mattered. It did not invent the problem, and it did not solve it. What it did was formalize a preservation model that takes magazines, source material, and digital access seriously as a package. That is the correction the field had been needing for years. Games do not survive as pure code. They survive as ecosystems of documentation, hardware behavior, publication history, and institutional memory.

The field’s evolution also explains why the old argument about whether emulation “counts” has become less interesting. Of course it counts. So do physical holdings, dark repositories, scans, transcripts, and metadata. The mature question is not whether one format is legitimate. It is which combination preserves the most meaning under the least legal nonsense.

If you want the broad public background on the institutions and technologies involved, the coverage and reference material commonly associated with this topic includes the Internet Archive and Emularity overviews on Wikipedia, broader coverage of the Internet Archive’s preservation model at Engadget, archival and cultural reporting at Polygon, tech policy coverage at Ars Technica, and industry reporting at The Verge.

Who is doing what

The easiest way to see the field is as a division of labor, not a contest for moral superiority. The institutions differ in scope, access philosophy, and legal posture. Some collect physical media. Some prioritize browser access. Some focus on source material and research infrastructure. None of them do everything.

The comparison is not subtle. The VGHF is trying to create the scholarly base layer. The Internet Archive is trying to keep software observable and, where possible, playable. The Library of Congress is providing institutional legitimacy and holdings. The SPN is trying to make access legally and procedurally viable. If one of these actors disappeared, preservation would not collapse. It would just get dumber.

Release dates, specs, pricing

This section is less about consumer products than about the infrastructure of preservation itself. The field has its own release schedule, its own technical stack, and its own costs. Since everyone is allergic to talking about budgets until a project dies, here is the machinery laid out plainly.

If the tables look blunt, that is because the preservation problem is blunt. The field does not lack prestige. It lacks continuity. The costs are mostly storage, scanning, staffing, legal review, and the humble but devastating labor of metadata creation. The price of not doing this is easier to measure: titles, documents, and tools vanish into the ordinary dust of institutional neglect.

What the specialists are saying

The best preservation quotes are the ones that refuse to pretend the issue is sentimental. The VGHF’s 2023 estimate that 87% of U.S. video games released before 2010 are lost or at risk remains the most forceful statistic in the field because it turns a vague concern into a measurable disaster. That number has become the shorthand for a medium whose commercial history was built faster than its memory systems.

In the Library of Congress sphere, Henry Lowood’s long-running work on the video game canon remains a reminder that selection is a form of preservation, not its substitute. The canon was originally created in 2007, and by 2025 it was still being updated and statistically analyzed against 80 Best Games lists published over four decades. The point is not that lists are archives. The point is that archives are always making lists, whether they admit it or not.

VGHF’s own leadership has also framed the field in terms of materials, not myth. The organization says it is preserving source code, tools, raw art, documentation, magazines, and digital access in parallel. That insistence is the practical equivalent of an argument: if you save only the finished game, you save the shell and lose the body of evidence around it.

The Internet Archive’s preservation posture is equally plain. Its use of Emularity makes old software playable in the browser, while CLASP keeps restricted holdings sealed until legal conditions change. That combination is not ideological. It is an admission that preservation requires both public demonstration and deferred access, depending on what the law and the media allow.

For broader background on the institutions involved, the most useful public references remain the Internet Archive entry and general reporting at Engadget, Polygon, Ars Technica, and The Verge. Those are not archival authorities in the narrow sense, but they are part of the public record the archives will eventually have to preserve too.

What happens next

The next six to twelve months are not hard to forecast because the pressure points are already visible. The only uncertainty is which one will crack first.

1. VGHF will keep expanding the Digital Archive and library holdings. The organization’s 2025 launch and stated mission make expansion the default expectation, especially around magazines, transcripts, and source-adjacent materials.
2. Remote access will remain the central legal fight. The 2025 rejection of the DMCA exemption request suggests preservation groups will continue pushing for library-facing access models rather than waiting for the market to volunteer.
3. Browser emulation will stay the public face of preservation. Internet Archive-style access is still the easiest way to show preservation to non-specialists without shipping hardware or distributing restricted files.
4. Dark repositories will become more common. CLASP’s model is the one archives adopt when they know the material matters and the law is not ready to help.
5. Source material will become a bigger part of the conversation. VGHF’s focus on source code, raw art, and documentation reflects the field’s recognition that playability alone does not preserve history.

A few sharper predictions are worth making without the usual ceremonial hedging. First, more institutions will start presenting preservation as a records-management issue rather than a retro-gaming hobby. Second, preservation groups will keep relying on magazine runs and contemporaneous print sources because that material is still cheaper to gather than complete technical histories. Third, the legal framework will remain the bottleneck, because copyright law is far better at creating scarcity than access. Fourth, more companies will quietly discover they have lost their own development assets and will then call archivists like the fire department. Fifth, the public will continue to discover that “digital” does not mean immortal, which is a lesson humanity relearns with comic persistence.

There is a final prediction, and it is the least glamorous. Preservation in 2026 will keep moving from rescue to maintenance. Rescue is exciting. Maintenance is where institutions prove they deserve to exist. The archives that survive will be the ones that can ingest materials, describe them well, keep them legally legible, and let researchers actually use them. That is not a romantic future. It is a functioning one.

The machine-like truth is this: video game preservation is finally getting infrastructure, but it is still operating under law that assumes access is suspicious and memory is optional. The people doing the work know better. The rest of the industry is catching up, one lost cartridge, one recovered art file, one browser window at a time.