GONE IN A FLASH

How Adobe Flash, once the face of the web, fell to the brink of obscurity—and why it’s worth saving

Adobe Flash was the future, until it wasn’t. It emerged in the late 1990s, at a time when the internet was at its ugliest, when the most exciting thing one might find on a webpage was a low-resolution GIF, or blinking text. Flash brought web-based video, animation, and interactivity into ubiquity; it allowed designers and developers alike to make a new kind of rich content that would work on any computer or browser.

If bulletin-board systems marked the modern internet’s infancy, and AOL chat rooms its childhood, then Flash was the web’s adolescence—its weird, rioting teenage years. Flash wasn’t just a tool or a technology; it was its own genre. Quirky Flash-based cartoons and video games proliferated in the oughts, identifiable by their vector art style and striking departure in tone from traditional media. The world had never seen a cartoon quite like Homestar Runner or a video game like QWOP before Flash made such things possible.

The Flash Player plugin, a proprietary piece of software required to load Flash content in a web browser, was originally a big part of what made the technology so revolutionary. It guaranteed the content would look and behave the same way for anyone who loaded it, regardless of what type of browser or computer they were using. But once technologies capable of running natively in web browsers gained steam, that same plugin requirement became a liability.

Inching toward irrelevance for nearly a decade now, Adobe Flash is hardly the only technology being trampled by the ceaseless march of technological progress (remember RealAudio?). In 2014, two researchers at the Austrian Institute of Technology looked at some popular file formats to assess which are at the highest risk of going obsolete:

flash-archiving

The file formats found to be at the highest risk were MAC (macro files), SXW (OpenOffice documents), and DXF (AutoCad files)—all outdated formats with niche user bases. Flash wasn’t included in the study, but also would have fallen into the high-risk category: It’s a complex and largely closed format that has gone through many versions, and has just one vendor supporting it.

That doesn’t just mean Flash is at risk of becoming obsolete; it also makes Flash content difficult to preserve for future generations. Internet archivists are already dealing with that difficulty, and recent developments in browser support for Flash are making their task all the more urgent.

The long good-bye

The earliest signal of Flash’s fall came in 2007, when Apple decided not to support it in the newly introduced iPhone. At the time, the fifth version of HTML was about to emerge, and promised to replace some of the functionality Flash provided. With the nascent mobile web in mind, developers across the world began moving away from Flash and toward HTML5.

In 2010, Steve Jobs posted a note to the Apple website to clarify his thoughts on Flash, and further cement HTML5 as the new standard. The problem with Flash, he wrote, was that it was insecure and resource-intensive, and its plugin was overly proprietary. Jobs made the case for HTML5 and JavaScript, pointing out that they were based on open standards that web browsers could build on.

“Though the operating system for the iPhone, iPod and iPad is proprietary, we strongly believe that all standards pertaining to the web should be open,” he wrote, knowingly throwing stones from inside Apple’s very proprietary glass house. “Rather than use Flash, Apple has adopted HTML5, CSS and JavaScript—all open standards.”

Apple’s hardline stance was ominous, but Flash was still available on Android devices, and could be used to create native apps for iOS. More important, it remained an integral part of the web. Facebook’s wildly popular suite of games, including FarmVille and Words with Friends, was built on Flash, as was YouTube’s video player. A few months after Jobs’ note, a post on YouTube’s developer blog said that although the company had made most of its videos available on HTML5, the technology wasn’t quite ready for primetime.

“While HTML5’s video support enables us to bring most of the content and features of YouTube to computers and other devices that don’t support Flash Player, it does not yet meet all of our needs,” the post said. “Today, Adobe Flash provides the best platform for YouTube’s video distribution requirements, which is why our primary video player is built with it.”

The idea that HTML5 could fully replace Flash was based on widespread misconceptions about the former’s capabilities. Flash included advanced video-serving features, like DRM and full-screen support, and provided designers and developers with a sophisticated app for creating interactive animations. Coding was optional. HTML5, on the other hand, was just a markup language: structured code that tells web browsers where elements should be rendered, and how they should be formatted. What made it distinct from previous versions of HTML was that it added some new features, like the ability to directly embed audio and video files that would be playable in web browsers.

HTML5 also added support for vector graphics, but they could not be animated or made interactive without the use of JavaScript and CSS, both of which require coding expertise. Jobs took care to name those technologies in his note, but the nuance was somewhat lost in public discourse. That misunderstanding, it turned out, was bad for Flash. HTML5 became a catch-all term for the New Web: a pure, plugin-free internet experience that worked as well on the phone as it did on the desktop.

By 2011, that idea had gained enough momentum that even Adobe acknowledged the changing tide. It ceased production of the Flash Player for Android and released a product called Edge Animate—a new way to create HTML5 content. “HTML5 is now universally supported on major mobile devices, in some cases exclusively,” the company said in a press release. “This makes HTML5 the best solution for creating and deploying content in the browser across mobile platforms.”

The brink of obscurity

Over the next several years, HTML5 and related technologies continued to overtake the web. At the beginning of 2015, YouTube announced that its HTML5 video player was now the default. Just a few months later, it was revealed that hackers had been using an exploit in the Flash Player to inject surveillance software onto users’ computers. Facebook’s chief security officer called on Adobe to set an end-of-life date for the plugin.

Adobe discontinued Edge Animate at the end of 2015, then renamed the Flash Professional app to Animate CC. Upon announcing the rebranding, Adobe said Animate CC would be its “premier web animation tool for developing HTML5 content while continuing to support the creation of Flash content.” The company didn’t set an end-of-life date for the Flash Player plugin—it still hasn’t—but said in its announcement that developers should abandon the platform.

“Looking ahead, we encourage content creators to build with new web standards,” the company said. In a separate blog post in November 2015, an Adobe product manager noted that “over a third of all content created in Flash Professional uses HTML5.”

Facebook, meanwhile, became Flash’s last bastion. Some 650 million users were still playing Flash-based games on the social network’s website, but Facebook had for years been looking for a way to make its games work on its mobile apps. It had attempted to launch an HTML5- and JavaScript-based games platform in 2011, but quickly scrapped the project.

Then, just last month, Facebook took another shot. It launched Instant Games, “a new HTML5 cross-platform gaming experience, on Messenger and Facebook News Feed for both mobile and web.” The social network hasn’t revealed plans to shut down its library of browser-based Flash games, but the plugin’s grasp on that niche section of the web is clearly diminishing.

The preservationists

Facebook isn’t the only tech giant with a stake in Flash’s future. Earlier this month, Google announced changes to its Chrome browser that will block Flash by default. Users will still be able to access Flash content by clicking an opt-in button, but the move suggests a ticking clock on Chrome—the most popular web browser in the world—dropping support for Flash altogether. When that day comes, trying to play Flash games or watch Flash cartoons will be much like trying to play a cassette without a tape deck.

So how do we preserve Flash content if there’s no way to load it?

Creators of Flash content can update their work to more modern formats—cartoons and animations can be converted to video, and the vector graphics behind them can be moved to programs like Adobe Illustrator. Video games are more complicated, but can be saved as executables that will run on Windows and OSX.

But websites made entirely in Flash, with their glamorous button animations and fuzzy text rendering, are another beast entirely. Unlike cartoons and games, websites tend not to be thought of as discrete artifacts—when they get updated, previous designs are for the most part overwritten. For those websites, and for cartoons and games that no one bothers to convert to new formats, the future of Flash will depend on the efforts of internet archivists.

This effort is already underway. The Internet Archive and the Archive Team are currently saving Flash files. The website oldweb.today, which allows visitors to access archives of the internet past, provides emulations of vintage browsers, which will be necessary for viewing Flash content should modern browsers stop supporting the plugin entirely. Oldweb.today creator Ilya Kreymer says that even running Flash in an emulated browser presents a number of technical hurdles, and his description of those logistics is headache-inducing:

The first [challenge] is finding the right version of Flash for the right browser. Chrome on Linux bundled Flash up to Chrome 53, but changed the distribution for 54, so we were unable to get Flash working with a newer version of Chrome. Chrome 53 may well be the last version of Chrome that we offer with Flash. For Firefox, we are using the latest Adobe Flash plugin, which is still widely available as part of Ubuntu and other common Linux distributions.

Archiving Flash projects en masse is also far from straightforward, Kreymer says. Often, there are a collection of files that make up a Flash presentation, all of which need to be chained together in a particular way for it to work. (Think of it as a mixtape that spans multiple cassettes.) That makes it difficult to write a program that will find and archive the content automatically.

“All media deteriorate”

Archivists and programmers are working on ways around these issues—see: Kreymer’s Webrecorder project—but the preservation of Flash will always rely on many moving parts. The work being done by volunteers and nonprofit organizations is encouraging, but imagine the longevity of something like your family photos being dependent on such a tentative infrastructure. Wasn’t digital information supposed to be more reliable than this?

Not exactly, says Trevor Owens, a digital librarian and archivist at the Institute of Museum and Library Services.

“All media deteriorate, just at differing rates,” Owens says. “Most information eventually disappears. Historically it has only been a small amount of objects and records that have persisted long-term.”

Indeed, the historical artifacts that we as a species have managed to hold onto likely make up only a fraction of what we’ve lost. The near-mythical loss of innumerable ancient books at Egypt’s Library of Alexandria comes to mind, as does the loss of every movie Fox Pictures produced before 1935 (the movies were kept on flammable nitrate film). Digital media may be safer from the elements than their analogue predecessors, but still face a different sort of danger.

All digital content we produce today requires an interface, a program that can interpret and present that content to human eyes and ears. Our digital photos are not self-contained objects that we can look at without the aid of such a program. Neither are our audio files, videos, interactive web pages, or even our text. Every byte of it is dependent upon an interface, and interfaces become obsolete all the time.

Each file format in the obsolescence chart at the top of this page is subject to that risk, no matter how ubiquitous it is today—even GIFs, the format with the lowest risk score. GIFs filled the internet before the emergence of Flash, and still do today, and probably will long after Flash is gone. But the web browsers and other apps that can open GIFs will keep changing; priorities will keep shifting, and one day the future of GIFs will be in jeopardy.

Also sitting precariously in the middle of that risk spectrum? Another format with multiple versions and enough complexity to warrant fear over long-term sustainability, which just a few years ago was considered a suitable replacement for Flash: HTML.

home our picks popular latest obsessions search