Announced today, the Internet Archive has been designated as a federal depository library by Senator Alex Padilla. The designation was made via letter to Scott Matheson, Superintendent of Documents at the U.S. Government Publishing Office.
“The Archive’s digital-first approach makes it the perfect fit for a modern federal depository library, expanding access to federal government publications amid an increasingly digital landscape,” Padilla said in a statement to KQED. “The Internet Archive has broken down countless barriers to accessing information, and it is my honor to provide this designation to help further their mission of providing ‘Universal Access to All Knowledge.’”
Internet Archive’s founder and digital librarian Brewster Kahle remarked on the designation:
“ I think there is a great deal of excitement to have an organization such as the Internet Archive, which has physical collections of materials, but is really known mostly for being accessible as part of the internet,” Kahle said. “And helping integrate these materials into things like Wikipedia, so that the whole internet ecosystem gets stronger as digital learners get closer access into the government materials.”
The grant will support development of the “Today’s News for Tomorrow” national program by Internet Archive, IRE, and The Poynter Institute to provide infrastructure, preservation services, training, and community building that enable local newsrooms and journalists to ensure the archiving and perpetual access of their their publications, digital assets, and other materials. As the first draft of history, local news published today is a critical resource documenting the lives and stories of the American people as well as an essential record for use by students, historians, and researchers. The “Today’s News for Tomorrow” program will address the financial and operational challenges that many local news organizations face in managing and preserving their digital materials both for their immediate internal needs and the future information needs of their communities.
The Press Forward funding will allow the program partners to provide infrastructure and training to over 300 newsrooms and journalists across the country, with a focus on vital local online news that is particularly at risk. Internet Archive’s web archive has long been an essential resource for journalists in their reporting. Pairing Internet Archive’s preservation infrastructure and services with IRE’s and The Poynter Institute’s experience in training and community support for journalists will further Press Forward’s goal to strengthen communities by revitalizing local news. The “Today’s News for Tomorrow” program also builds on Internet Archive’s successful “Community Webs” national program which has received nearly $3M in funding to provide preservation services and cohort-based training to over 275 libraries, museums, and municipalities from 46 states and 7 Canadian provinces in support of their work documenting the history of their communities.
We thank Press Forward and The Miami Foundation for their support of “Today’s News for Tomorrow.” We are excited to work closely with IRE and The Poynter Institute supporting newsrooms and journalists and are honored to be part of the group of organizations receiving funding as part of Press Forward’s Open Call on Infrastructure. The full list of recipients is available online at pressforward.news/infrastructure25.
A group of librarians and cultural heritage workers from across the country recently convened at two events hosted by Internet Archive’s Community Webs program. Made possible in part with support from the Mellon Foundation, the meetings allowed librarians from across the country to discuss shared challenges and opportunities around documenting, preserving, and sharing the unique culture and digital heritage of their communities.
Community Webs members in Philadelphia for the 2025 Community Webs National Symposium
Launched in 2017, Internet Archive’s Community Webs program provides public libraries and similar organizations with the tools and support they need to document local communities. Members of the program receive access to Internet Archive’s Archive-It web archiving service and Vault digital preservation service, have coordinated on funded digitization projects to bring local history collections online, and receive training, technical support, and opportunities for collaboration and professional development. There are now over 260 members of the program from across 46 states, 7 Canadian provinces, and a growing number from outside of North America.
Attendees at a workshop led by Queens Memory Project founder Natalie Milbrodt
The first of these events was held on May 9 at Internet Archive Headquarters in San Francisco and brought together a small group of public librarians interested in launching new community-focused local preservation initiatives. As local information hubs and community connectors, public libraries play a critical role in the preservation and access of local history. Over the course of the day, attendees engaged in exercises and discussions that helped them develop plans to support this work in their communities.
Community Webs members view highlights from the Parkway Central Library Special Collections
The 2025 Community Webs National Symposium was held on June 25 and 26 in Philadelphia ahead of the American Library Association annual conference. This two-day event brought together 40 Community Webs members representing a range of cultural heritage institutions. Attendees participated in workshops on community archiving and digital preservation led by Queens Memory Project founder Natalie Milbrodt and Digital POWRR instructor Danielle Taylor, listened to presentations from Community Webs members on local projects they are leading in their communities, and toured the Parkway Central Library Special Collections.
A main goal of the Community Webs program is to create opportunities for multi-institutional collaboration across organizations devoted to preserving local history. In-person events like these provide a forum where members can build relationships, exchange ideas, and develop skills. By supporting the work of these cultural heritage practitioners to preserve local knowledge, Internet Archive is able to move closer to achieving its mission of “Universal Access to All Knowledge.”
This October, the Internet Archive’s Wayback Machine is projected to hit a once-in-a-generation milestone: 1 trillion web pages archived. That’s one trillion memories, moments, and movements—preserved for the public, forever.
We’ll be commemorating this historic achievement on October 22, 2025, with a global event: a party at our San Francisco headquarters and a livestream for friends and supporters around the world. More than a celebration, it’s a tribute to what we’ve built together: a free and open digital library of the web.
Join us in marking this incredible milestone. Together, we’ve built the largest archive of web history ever assembled. Let’s celebrate this achievement—in San Francisco and around the world—on October 22.
Here’s how you can take part:
1. RSVP Sign up now to be the first to know when registration opens for our in-person event and livestream. RSVP now
2. Support the Internet Archive Help us continue preserving the web for generations to come. Donate today!
3. Share Your Story What does the web mean to you? How has the Wayback Machine helped you remember, research, or recover something important? Submit your story
Let’s work together toward October 22—a day to look back, share stories, and celebrate the web we’ve built and preserved together.
A recent legal decision has reaffirmed the power of fair use in the digital age, and it’s a big win for libraries and the future of public access to knowledge.
On June 24, 2025, Judge William Alsup of the United States District Court for the Northern District of California ruled in favor of Anthropic, finding that the company’s use of purchased copyrighted books to train its AI model qualified as fair use. While the case centered on emerging AI technologies, the implications of the ruling reach much further—especially for institutions like libraries that depend on fair use to preserve and provide access to information.
What the Decision Says
In the case, publishers claimed that Anthropic infringed copyright by including copyrighted books in its AI training dataset. Some of those books were acquired in physical form and then digitized by Anthropic to make them usable for machine learning.
The court sided with Anthropic on this point, holding that the company’s “format-change from print library copies to digital library copies was transformative under fair use factor one” and therefore constituted fair use. It also ruled that using those digitized copies to train an AI model was a transformative use, again qualifying as fair use under U.S. law.
This part of the ruling strongly echoes previous landmark decisions, especially Authors Guild v. Google, which upheld the legality of digitizing books for search and analysis. The court explicitly cited the Google Books case as supporting precedent.
While we believe the ruling is headed in the right direction—recognizing both format shifting and transformative use—the court factored in destruction of the original physical books as part of the digitization process, a limitation we believe could be harmful if broadly applied to libraries and archives.
What It Means for Libraries
Libraries rely on fair use every day. Whether it’s digitizing books, archiving websites, or preserving at-risk digital content, fair use enables libraries to fulfill our public service missions in the digital age: making knowledge available, searchable, and accessible for current and future generations.
This decision reinforces the idea that copying for non-commercial, transformative purposes—like making a book searchable, training an AI, or preserving web pages—can be lawful under fair use. That legal protection is essential to modern librarianship.
In fact, the court’s analysis strengthens the legal groundwork that libraries have relied on for years. As with the Google Books decision, it affirms that digitization for research, discovery, and technological advancement can align with copyright law, not violate it.
Looking Ahead
This ruling is an important step forward for libraries. It reaffirms that fair use continues to adapt alongside new technologies, and that the law can recognize public interest in access, preservation, and innovation.
As we navigate a rapidly changing technological landscape, it’s more important than ever to defend fair use and support the institutions that bring knowledge to the public. Libraries are essential infrastructure for an informed society, and legal precedents like this help ensure they can continue their vital work in the digital age.
Louis Brizuela says managing the microfiche digitization center for Democracy’s Library gives him a sense of pride. “I feel like I’m making a difference,” said the 28-year-old who lives in the Bay Area. “We’re scanning and preserving all this really cool content.”
Brizuela and his six-person team are currently digitizing U.S. Supreme Court case documents and government records from Canada dating back to the 1930s. The documents are stored on microfiche cards, a flat, film-based format commonly used from the mid-20th century for preserving and accessing paper records, which requires a specialized reader for viewing—making the information contained on the cards difficult to access. “It’s useful for law students or anybody – and it’s free to use without borders,” he said. “Also, it’s valuable for the sake of archiving so information doesn’t get lost.” Next, Brizuela said he’s looking forward to receiving a donated collection of microfiche with images of Sanskrit Buddhist tablets.
Anyone can watch the crew in action on a livestream of the microfiche scanning operation (https://www.youtube.com/live/aPg2V5RVh7U). Activity occurs Monday–Friday, 7:30am-3:30pm and 4:00pm-midnight U.S. Pacific Time (GMT+8)—except U.S. holidays. Mellow lo-fi music plays in the background during working hours and continues with various video and still images from the Internet Archive’s collections rotating on the feed when the digitization center is closed.
During the livestream, one camera is focused on an operator feeding microfiche cards beneath a high-resolution camera; another other provides a close-up view of the material. Each page is processed, made fully text-searchable, and added to the Internet Archive’s public collections. Researchers and readers can easily access and download the documents freely through Democracy’s Library.
Brizuela said the staff has embraced the public window on their work. He joined the Internet Archive in February and hired people who were willing to be on camera and understood the potential benefit of the exposure. “It’s not like ‘Oh, Big Brother is watching’,” he said, noting the employees have fun with the situation. “We’re not robots. We do show our characters. We’re human.”
The team is leaning in, Brizuela said, suggesting they dress up in costumes for Halloween and maybe wearing elf hats at Christmas to add a festive touch to the project. They also answer questions in a live chat with viewers.
Brizuela comes to this position from a varied career working in the military, medical fields, retail and web development. He’s long had an interest in photography, particularly shooting and developing his own 35mm film. So, Brizuela said, it was not hard to pick up how to operate the custom-built scanner and oversee the digitization process.
Louis Brizuela stands in front of a custom-built microfiche scanning workstation.
Every morning, the team huddles up in the small digitization center to talk about the previous day’s completed pages and map out the upcoming work. Brizuela watches over and QA’s the scanning done by the team. Depending on the type of collection, each scanner can scanhundreds of cards a day.
Brizuela describes the vibe in the microfiche digitization center as pretty relaxing, with staff members chatting and interacting while they work. Often, they have headphones to listen to an audiobook or podcast. “If they are listening to music, sometimes they bust a dance move, or bob their head to get in the groove. People enjoy seeing that,” Brizuela said.
Here’s a resource sharing tip for our community of librarians:
RapidILL members have an option to include the Internet Archive as a potential supplier for their borrowing requests. If you are interested in providing your users with access to the Internet Archive’s collections through your RapidILL workflows, please complete this form.
As an interviewer, he knew how to keep things light, conversational. He got the information he needed, and wrote articles based on what answers the subject provided, but he did it in a way that never felt like he was prodding, or intending to catch someone out.
He held a number of positions in journalism but one of the most memorable was as a Hollywood correspondent for the NY Post, where he would write up interviews with on-the-rise celebrities or long-established actors and directors about their current project and what they’d learned. If you’ve ever read a typical Sunday newspaper magazine with a couple pages of interview with a contemporary star of stage or screen, you’ve settled in with Bob’s bread and butter for decades.
Bob would share his interview tapes with his family, scrawled with all sorts of markings and ranging with dates from the 1960s to the 1980s. Ultimately, they came to the Internet Archive as a physical donation with the intention of being digitized and put up for all to enjoy.
A selection of Lardine cassettes from the original physical donation
For a number of years, after being donated, classified, and assigned an inventory number, the tapes were stored waiting to join a digitization queue. In 2025, the box was opened to be digitized using a tape setup and converted to .WAV sound files.
Tape Digitizing Setup – TASCAM 122mkIII deck to MOTU M4 USB Interface to Audacity
The box of audio cassettes, excepting a few in need of repair, are now digitized into the Interview Tapes by Bob Lardine collection at the Archive. 57 separate recorded interviews with celebrities, and two compilations of tapes, discussed further below.
Most people will be naturally drawn to the celebrity interviews. With names like George Peppard, Sharon Gless, Ricardo Montalbanand more, they represent a killer lineup of recognizable names, especially if you experienced television in the 1970s and 1980s. Many of these tapes were recorded during the height of their careers (Peppard in the middle of A-Team, Gless while appearing in House Calls, and Kate Jackson just starting out on Charlie’s Angels) and they are more than happy to talk through their biographies and thoughts while in the salad days of nationwide celebrity status.
Which is fine, but you should know – the tape quality is spectacularly terrible.
Recorded, as they were, on the tables of restaurants, in dressing rooms or sitting on set between scenes, the goal of these recordings was clearly for Bob to use as backups to notes he was taking on paper. In the modern era of podcast microphones and post-processing software able to be recorded next to moving vehicles with no problem, the tape recorder in use was likely to be a simple affair, and one left in the same place even as people shifted around or looked in the wrong direction while talking.
But as muddy as the interview tapes can be, they still do the job. In her 1978 interview, Olivia Newton-John talks about her accent and sketches out her plans for her future career, and the listener can follow with little trouble. Erik Estrada talks about his health regimen and his plans to support his extended family, recorded in what sounds like a small room. And Robert Urich talks about feeling betrayed by various press interviews, showing how trustable Bob Lardine is in conducting his.
Ultimately, the tapes are legible. And, once your ear adjusts to the situation, wonderfully personal. These are workers, craftspeople, artists, taking time out from their day to share their current worries, considerations and plans. They speak, not so much as a performer providing entertainment at a microphone for a “personal moment” during a concert or appearance, but people with a job sharing how they got there, and where they are going.
No interview shows this better than the 1975 interview tape with Henry Winkler.
With Happy Days now in its third season, Winkler has been given co-starring status in the series with Ron Howard. Fonz-mania, years away from famously “jumping the shark”, has him in stadiums with 25,000 people cheering for him. Under any measurement, he is experiencing super-stardom, with the sky the limit.
But in this tape, Winkler is the picture of humility. He talks about how nobody keeps the throne for long, how it can all disappear overnight, and what steps he takes to mentally prepare for that change. He fears typecasting (which turned out to be a legitimate concern in the 1980s) and opens his sketched-out plans for what to do about that. Through it all, he’s an artist who cares about his art, and is doing his best to keep a level head through a gauntlet of hyperbolic fame.
It’s worth nothing that our obsession with celebrity means that many of the basic facts about these interviewees is known – where they were in May of 1975, or what the actual name of a production they were working on became. We have a literal deluge of knowledge about their marriages, divorces, places of residence. From these known facts, we can surmise a lot about what these tapes are talking about. If only this were the case with so many other cultures, now-lost places or people.
This collection would already be hours of insight and materials, but there’s just a little bit more.
Alongside these celebrity interviews, Bob also had tapes from the 1960s for a radio program called The Jewish Hour. Broadcast out of Phoenix, Arizona, and syndicated elsewhere, this radio show contains a variety of interviews, appearances and performances aimed from a Jewish perspective. There appears to be very little information about this show online – and while there might be a library or archive that has records of this show, there is nothing currently obvious to find. Until now: Lardine’s tapes have recordings, as well as related taped-off-radio recordings of interviews and shows covering historical people and events of the time. Without these tapes, there seems to be very scant recorded evidence of them available.
We’re always happy to take donations of audio cassettes like this, and look forward to continuing the process of bringing them online. Who knows what other lost treasures lurk in the world?
A very large thank you to Bob Lardine’s family for their donation of these tapes, as well as friends of the Internet Archive who helped fund purchase of the tape decks used for playback and digitization.
As many iconic works have entered into the public domain since 2019, there has been a surge of horror film adaptations. These horror adaptations have received strong critiques for their deviation from or failure to say something unique about their source material. Ultimately, this criticism has spilled over into skepticism about the public domain itself, framing it as a creative dead-end. This critique, however, overlooks the underlying benefit of the public domain: the ability for anyone, not just corporations, to create their own version/adaptation of the same work. Despite consistent criticism surrounding public domain horror adaptations, a further study of these works reveals underlying contemporary industry conditions that lead to their creation, and demonstrates the enduring importance of the public domain in enabling creative freedom beyond pure corporate control.
These adaptations exist within the current characteristics of contemporary filmmaking; a type of filmmaking largely driven by financial risk-aversion that relies on Intellectual Property (IP) adaptations rather than original stories to guarantee audience attendance and big money earnings. Look no further than April 2025’s A Minecraft Movie that relied on the Minecraft IP to pull in over $150 million in a single weekend in the United States and Canada, as well as over $900 million worldwide across its theatrical run. As studios continue to embrace IP and risk-aversion as rules of the game, creators must either find ways to craft original stories within these confines, or find another way to keep the cost down, such as working in a historically proven low-cost genre: Horror.
Horror films are a popular selection for filmmakers as they can be made more economically compared to other genres by utilizing fewer elements such as limited locations, small casts, and visual ambiguity to enact the horror/unease. There is a long lineage of economical horror films that set off careers including John Carpenter’s Halloween, Sam Raimi’s The Evil Dead, and Mike Flanagan’s Absentia. Each film was made for less than $500,000, unadjusted for inflation, and launched careers of well known and successful filmmakers. While each film is hugely varied and different from one another, they are all connected by one common element: being original stories. But when IP is heavily guarded and protected by risk-averse studios, it makes sense to turn to the public domain for creative freedom as an independent filmmaker working with budgetary constrictions.
Though shaped by the same constraints and standards, the resulting films vary wildly. Some horror films adapted from public domain works lean heavily on shock value while others take a more reflexive approach, using the tools of horror to comment on copyright itself. In 2022, Winnie-the-Pooh: Blood and Honey utilized the shock elements to draw an audience, while 2025’s Screamboat embedded a metatextual critique of copyright lengths.
An iconic illustration by E. H Shepard from the first Pooh book. This iconography has helped to make the Pooh stories recognizable worldwide.
In 2022, shortly after 1926’s Winnie-the-Poohentered the public domain, filmmaker Rhys Frake-Waterfield, whose earlier indie films received little attention, announced Winnie-the-Pooh: Blood and Honey. By adapting the Pooh stories, Frake-Waterfield utilized their near-hundred year history and public recognition to garner immediate attention to his new production, highlighting the benefits that IP provides even in an area of filmmaking that is historically amenable to emerging filmmakers.
Since its release, critics and viewers alike have highlighted the deviation from the childhood source material through a gory slasher adaptation. These critiques are reasonable and definitely feel notable to viewers as a dead-eyed Pooh Bear and tusk-bearing Piglet eat Eeyore. However jarring the contrast of the adaptation to the source material may be, it does not undercut the value of the public domain. In creating this adaptation, it acts as a celebration of the public domain as a vehicle for filmmakers and other creatives to remix old works for their own creative and commercial benefit and not just the benefit of select corporate IP holders.
The Blood and Honey film is an adaptation that does not utilize the Pooh stories for much more than audience familiarity. It utilizes the public domain works primarily as a shock factor to attract audience attention. Generally it grafts the iconography of these stories onto an indie horror film that would remain fundamentally unchanged if all of the Pooh elements were stripped away. Beneath the surface of this iconography is a standard slasher film playing in the mold of what has come before.
The original Pooh stories, like The House at Pooh Corner (1928), can be fully read online.
This adaptation does not diminish the original stories that still exist and are available to everyone. Nor does it create a new monopoly on the stories, as these Pooh stories remain in the public domain. Instead, it highlights the underlying conditions of filmmaking that surround the film during the time in which it was made. By entering the public domain in the 2020s, newly public domain works give rise to modern adaptations that reflect the popular trends of the moment. They fit within the confines of the corporate, risk-averse IP conditions that drive filmmaking. And yes, many are becoming franchises, itself a reflection of the current moment. Frake-Waterfield has expanded upon his original Blood and Honey film with a direct sequel as well as the greater Twisted Childhood Universe, pulling from other public domain works such as the original Bambi and Peter Pan.
Similarly to Blood and Honey, the recent Screamboat adaptation of Steamboat Willie by Steven LaMorte is also a grafting of a public domain work onto a more standard narrative. In a 2025 interview with Paul Marsh, LaMorte reveals that he had been working on a Staten Island Ferry horror film since the early 2010s. However, following Steamboat Willie’s passage into the public domain in 2024, LaMorte reworked the film into an adaptation. In contrast to Blood and Honey, Screamboat functions as a metatextual film commenting not only on the original work, but also the nature of the public domain. It is not solely a horror film based on a public domain work, but a horror film about corporate copyright terms and how these long terms may alienate creators from their original works. This perspective becomes especially vivid in the film’s midsection, which recounts the story of Willie’s separation from Walt Disney in a visually striking animated flashback.
Original animation from Steamboat Willie (1928) that inspired Screamboat (2025).
Utilizing animation reminiscent of the original Steamboat Willie cartoon, the film recounts an old man’s tale of how Willie was separated from his creator, an animated depiction of Walt Disney. Much like in real life, the film too omits inclusion of Ub Iwerks as a creator of Mickey Mouse, reinforcing how authorship itself can be obscured by copyright mythologies. In the course of the tale, Walt falls overboard leaving Willie behind locked away in the ferry’s underbelly. Upon Willie’s release, after ninety-five years, he goes off on a rampage killing and terrorizing anyone that he comes across. Willie’s violence is framed not just as horror, but as retribution—an eruption of neglected cultural memory finally freed from captivity.
The middle animation segment of Screamboat utilizes the public domain nature of Steamboat Willie by formally adapting something that was previously restricted by copyright. This unique passage during the film’s middle point sticks in the viewer’s mind, elevating the work a step beyond pure shock value. It instead evokes an iconic character to examine the legacy of copyright control. Through Willie’s violent acts, the film suggests that long copyright terms can turn cultural icons into imprisoned relics. Screamboat critiques the copyright maximalism that the Disney company helped enshrine, using one of Disney’s earliest icons. Together, Blood and Honey and Screamboat reflect two poles of public domain horror—one exploitative, the other expressive. But both are artifacts of a specific cultural and creative moment.
Pulling from a long public domain tale, West Side Story adapts Shakespeare’s Romeo and Juliet to 1950s New York City.
Placed in the context of this broader moment in filmmaking, public domain horror is not an aberration but a logical outcome. Despite this, the context that surrounds the films now will not always be in living memory. In many years, when reflecting on this particular filmmaking environment, these horror adaptations might be seenas an odd and quirky moment of filmmaking. In actuality, these films are emblematic of the cultural moment in which they were produced, highlighting an evolving landscape of intellectual property and creative voice. Ultimately, these films probably won’t reach the same cultural impact as adaptations of other public domain works like 1961’s West Side Story (Romeo and Juliet) or 1959’s Ben-Hur (Ben-Hur). Still, they will remain important and interesting cultural artifacts that inform future generations as snapshots and reflections of the conditions in which they were made. Looking back at the past through creative works informs us of the societal and creative mores of that moment, and helps to anchor us in a contextual reference point to our own moment. Maybe these films will be celebrated themselves when they inevitably enter the public domain… in nearly 100 years.
Screams in the Vault: Public Domain Horror in the Age of IP by Sterling Dudley is marked with CC0 1.0
We are excited to announce a new version of GifCities, Internet Archive’s GeoCities Animated GIF Search Engine!
GifCities was a special project of Internet Archive originally done as part of our 20th Anniversary in 2016 to highlight and celebrate fun aspects of the amazing history of the web as represented in the Wayback Machine. Since then, GifCities GIFs have been used in innumerable web projects, artistic works, and in the media and press, including this internet-melting combination of GifCities GIFs and the British Royal Wedding in this New York Times article and the avant-GIF “GifCollider” exhibit at Berkeley Art Museum & Pacific Film Archive.
The new version of GifCities includes a number of new improvements. We are especially excited at the drastic improvement in “GifSearchies” by implementing semantic search for GifCities, instead of the hacky old “file name” text search of the original version. Note that via the “Special Search” tab, you can still search only by the old text-based index, for, uh, search index nostalgia purposes. But wait, you say, I need a specific-sized animated GIF for my project! No problemo, you can also now search for GIFs by size too. State of the world got you a bit anxious, dear reader? Calm thyself with 51 results pages of 150×20 pixel blinkies.
We also updated the results interface to add pagination for results instead of the original infinite scroll — that would sometimes paralyze your browser after a non-infinite amount of scrolling. And as with GifCities v1, each GIF links to the archived GeoCities page on which it originally appeared, many of these wonders unto themselves.
Finally, in the spirit of “sharing is caring” and “every GIF is truly a GIFT,” we have added the ability to make GifGrams that you can share with your special someones. Nothing says “I’m thinking of you” more than a custom webcard of GifCities GIFs and inspirational text, like “Hang in There” or “The Web is Yours for the Making.”
GifCities’s new semantic search index used a model called CLIP-ViT L/14 to analyze each frame of the GIFs and searching applies a “nearest neighbors algorithm” to find GIFs that match a vectorized query, allowing you to enter nuanced searches like “blue sparkling border‘ or (everybody’s favorite performing rodent) ‘”dancing hamster” and get matching results. Need more wolf snowglobe GIFs in your life? Yes, you do. Here ya go.
Thanks to all the GifCities enthusiasts out there that send us many messages each week on how they are using and enjoying the site. More details on GeoCities and GifCities are in the About page and a special thanks to enthusiast Ben Friesen who helped prototype using CLIP on GifCities. Now go browse some GeoCities GIFs and send some GifGrams!
