F
← Back to Journal

Futura's Solution

The missing memory album for newlyweds

•
Stefano
weddingmemoryalbumprivacyownershipbeautyforever

Introduction

Futura aims to be a digital multimedia memory album for newlyweds. Spouses can collect the memories of their wedding, share them with their loved ones, and, thanks to Web3 technologies, preserve them in a space that belongs to them. Forever.

In this essay we outline the problems that Futura solves and describe the approaches we're taking to address them. We are still in early stage development, so some work still needs to be done.

The Problem

Couples don't own their wedding galleries the same way they own their wedding album. Most online gallery hosting systems, which are also used in the wedding photography space, address the needs of photographers more than those of newlyweds. This is already visible in how the most prominent services present themselves: Pixieset positions itself as a "Client Photo Gallery for Modern Photographers," Pic-Time as "Online Photo Galleries for Professional Photographers," and so the others.1 The primary customer being addressed is not the client of the photographer, in our case the couple, but the professionals themselves.

This approach makes sense from a business perspective. Photographers shoot weddings and other events every day, while people usually marry once, or very few times in a lifetime. Gallery platforms therefore optimize for recurring professional users rather than for one-off clients. But the consequence of this market logic is that wedding galleries are usually created and managed under the photographer's account. Couples are granted access to them, but only for a limited period — which may be also long. They are dependent on the photographer's continued subscription and goodwill with no possibility to redeem the gallery, take ownership of it, and maintain it as a persistent digital space tied to their wedding.

Galleries are not the only way couples get their wedding photos. They also typically receive a physical album with a selection of the best images and a complete digital copy — either on a physical drive or via cloud download. But they are then left alone with the task of preserving these digital files properly over time. That this is an issue is evident from the large number of guides published by wedding photographers, explaining to clients how to safely store their wedding photos.2 Redundancy, of course, is always presented as the key principle in their suggestions: multiple copies, multiple locations, and at least one copy in the cloud.

But what's revealing is which cloud solutions are recommended. Almost without exception, they suggest mainstream consumer storage platforms such as Google Drive, Dropbox, iCloud, or Amazon Photos. This exposes a clear disconnect. On the one hand, photographers rely on specialized gallery platforms to present and deliver wedding images. On the other hand, couples are expected to preserve those same images long-term using generic cloud storage systems that are not designed around the specificity, structure, or symbolic weight of a wedding gallery. There is no dedicated digital album system that serves as the counterpart to the physical wedding album.

How Futura Solves It

Futura is that dedicated digital album. It offers newlyweds true ownership of a beautiful, private digital space designed specifically for their wedding memories — something far more meaningful than a folder on Google Drive — and a way to share them with the people they love and preserve it forever.

Beauty, ownership, foreverness, and safe shareability are the four core features on which Futura is built.

Beauty. Before anything else, Futura must be beautiful. Wedding memories demand a space that honors their significance—not just functional storage, but a meaningful, elegant experience. This is why we have James Dominique Barranger3 and Massimiliano Muner4 serving as art directors, ensuring that every aspect of Futura reflects the importance of what it preserves.

Ownership. Futura gives to the newlyweds true ownership over their gallery, not only for the simple reason that they hold the gallery account, as they could with any other cloud gallery systems, but thanks to web3 technologies5, they can have control over the digital space that hosts their gallery, even beyond the existence of Futura itself. Within the Web3 ecosystem ICP, the Internet Computer, stands out for its capability to host a full-stack application on-chain. And therefore we picked up ICP for our Web3 backend.

The backend of an app is not a monolith, but a complex system. There is normally a computational layer that performs operations, a Database that holds the data, and for heavyweight assets, like photos or videos, you use a so called Blob storage. The Blob storage is the place where the memories of the couple's wedding album are physically stored. We will discuss our choices regarding the storage layer in the next paragraph, dedicated to the challenge of storing memories forever.

Foreverness. Futura's promise to store your wedding gallery forever is a bold one, and maybe a naive one.6 Nothing seems to be forever. We mean 'forever' as a really long time, at least lifelong, which is bold as well. 7 How to tackle the challenge?

The first lesson when you enter the space of long-term storage strategies is that redundancy is the key. The famous 3-2-1 backup rule says you should have at least 3 copies of the data you want to store, in two different media, one of them offsite, i.e. somewhere else than where you are keeping the other two. This means that whatever storage solution we choose, if Futura aims to be the service that gives you a reasonable peace of mind about your data being there for future generations and not only the 1, the off-site copy, in the 3-2-1 rule, the chosen system can't be the only one. This is the first constraint: multiple storage solutions are necessary.

Beyond that, there are two more constraints. Second, an in-house solution would be foolish. Even if we implemented full 3-2-1 redundancy within our own infrastructure, it would all still depend on Futura's survival as an organization. We are a young startup, a sprout—far too fragile to promise 'forever' based on our own resources alone. Maybe one day Futura will be a trusted institution like the Internet Archive, but we are not there yet. We need to rely on third party services whose stability is stronger than Futura's own.8 Third, there's a UX constraint: we want to offer forever storage for an upfront payment, not as a subscription. "Your data will be stored forever as long as you subscribe" doesn't make much sense.

Most of the cloud storage services which offers an API don't offer lifetime storage.9 A possible strategy to tackle the problem would be an endowment system10, yet in the blockchain horizon we adopt, it would shift the responsibility for foreverness from a computational mechanism to financial management, and the financial model would become the weak chain element.

At the moment of writing we are still in an early development stage and we are tinkering with different technologies.11 We will shortly present the solutions we have implemented so far and the ones we plan to implement: ICP, S3 (AWS), Arweave, and a self-hosted solution, giving users the possibility to choose which one they want to use.

Since Futura was born in the ICP ecosystem and we chose ICP as the Web3 backend, it's natural to think of ICP as one possible Blob Storage solution. A smart contract (canister) has 500 GiB of stable memory, which in ICP terms is the non-volatile memory. Storing a selection of 50 photos in JPEG format in full resolution for 100 years would cost around $500.12 It's worth noting that DFINITY is preparing to roll out an updated blob storage service as part of the Caffeine Phase III upgrade (targeting early October 2025), which will dramatically decrease the price of storage, bringing 1 GiB for 100 years to approximately $25.13 As of January 2026, the feature has not been rolled out yet.

Arweave represents a fundamentally different approach to permanent storage. Unlike ICP, where data persistence depends on continued operation and payment, Arweave is designed specifically for permanent, immutable storage through what they call the "permaweb." The system uses an endowment model: users pay once upfront, and that payment is invested in a fund that generates returns over time to compensate miners for storing the data indefinitely. The economic incentive structure is designed to make data storage profitable for miners for at least 200 years. This makes Arweave particularly appealing for our use case, as it directly addresses the "foreverness" promise through its core architecture rather than as an add-on feature. However, Arweave storage is immutable—once data is uploaded, it cannot be modified or deleted. This creates both a strength (true permanence) and a constraint (no flexibility for updates or corrections).

Regarding AWS, to paraphrase Gertrude Stein's "a rose is a rose is a rose," AWS is AWS is AWS. It represents both the state of the art in cloud storage solutions and the quintessence of centralized architecture. As the dominant player in the market, direct or indirect engagement is unavoidable. We therefore integrate it as part of our offering and maintain it as a practical laboratory. That said, couples will always have the possibility to avoid AWS completely and choose other storage solutions. We will discuss the tension between Web2 and Web3 approaches in the paragraph about flexibility.

As for the self-hosted solution, we tagged it as foolish, and yet we are going to implement it. This is just part of our discovery path and developer fun.

Privacy. Data will be encrypted based on the couple's wishes. ICP, through VetKeys, offers a wonderful technology that makes encrypted shared content shareable in a safe way. The other way we address privacy is by making our code open source, so that it is public what we are doing with your data.

Flexibility, Modularity, Freedom. Let's be honest. Most people don't care about data ownership, don't care about privacy, don't care that software is open source, since they don't even know what open source means. When you say 'wallet' they think about the leather one, and they are happy when the website they need to register into offers Google as sign-up option. We built Futura to please the web3 maximalist and the normal folks as well. We therefore adopted a modular approach for the architecture of the app, so that people have the freedom to choose.

Modularity means that we have a double backend—a Web2 backend and a Web3 backend—and that you can choose among different data storage solutions and combine them. We will also have a double frontend solution for the most picky ones. We will describe the architecture of Futura in more detail in another dedicated article soon, so stay tuned.14

Footnotes

Footnotes

  1. Besides the already mentioned Pixieset and Pic-Time, other gallery platforms positioning themselves for photographers include: ShootProof ("Online Photo Galleries for Photographers," "Created By Photographers, For Photographers"), CloudSpot ("Online Galleries for Photographers"), Zenfolio ("Website & Gallery Solutions for Photographers"), Picdrop ("Best Photo Sharing Platform for Photographers"). ↩

  2. Some examples we found online include: (1) Heather Sham Photography, "Backup Wedding Photos"; (2) Jose Melgarejo, "How to Store Your Wedding Photos"; (3) Helena & Laurent, "The Ultimate Guide to Backing Up & Storing Wedding Photos"; (4) Caitlin Elizabeth, "How to Safely Store and Backup Your Wedding Photos"; (5) Heidi Talic Photography, "How to Backup and Safely Store Your Photos"; (6) Love Life Images, "Your Guide to Digitally Storing Your Wedding Photos"; (7) Andrew Franciosa, "How to Back Up Your Wedding Photos"; (8) S. Arnold Photo, "How to Store and Organize Wedding Photos"; (9) Victoria J Photography, "Storing Your Digital Wedding Photographs"; (10) Saywell HQ, "How I Keep Your Wedding Photos Safe"; (11) Emily Nicole Photography, "How to Safely Store Your Wedding Photos"; (12) Kelly McPhail, "How to Safely Store Your Digital Photos"; (13) DK Photo, "7 Tips: Storing Digital Wedding Photos Forever". ↩

  3. James Dominique Barranger is a Lettering Artist, Graphic Designer, and Art Director with significant experience in international marketing and advertising. He co-founded La Calzoleria, a cultural space and bar in Rome. His main passion is creating lettering designs and illustrations by hand. See behance.net/jdbarranger. ↩

  4. Massimiliano Muner is an Italian visual artist and photographer from Trieste, known for his experimental use of Polaroid and analog photography. He is the owner of the photography agency Spazio White, which explores all facets of commercial photography including weddings, and co-founder of the association Silver AgE and creator of the Fotografia Zeropixel festival. See massimilianomuner.com. ↩

  5. The passage from web1 to web2 to web3 can be explained as the evolution of the fruition of the internet by normal users: a read, write, own evolution. Web1 was the initial phase, where normal users were able mainly to read content on the websites. Web2 was the phase where people were able also to produce content. This is the season of the blogs. The blog platforms themselves, like YouTube or Instagram today, were owned by a company, but the users were able to easily publish new content. With the rise of blockchain technologies the normal user is able to own portions of the web. This is made possible through the creation of decentralized or distributed systems, whose stability is ensured by complex cryptographic mechanisms, which are basically not owned by anyone. See https://en.wikipedia.org/wiki/Web3 ↩

  6. We are used to associating blockchains with the idea of immutability. Content saved "on-chain" is commonly understood as being stored forever, or at least for as long as the blockchain itself exists. Our team is relatively new to Web3 technologies. Beyond the general exposure that most people have to blockchain as a cultural phenomenon—largely shaped by Bitcoin and cryptocurrencies—we had limited technical familiarity until May 2023, when we participated in our first Web3 hackathon. Personally, my prior experience with blockchains was limited to following a programming tutorial years earlier that implemented a simple blockchain in Python, which contributed to imprint in my mind the idea of on-chain content being immutable. At its core, a Bitcoin-style blockchain is a distributed ledger maintained by multiple nodes, where cryptographic mechanisms and economic incentives ensure that ledger entries cannot be forged or retroactively altered. The central idea is that stability and immutability are achieved without reliance on a centralized authority. In this context, "on-chain" naturally came to mean data that is legitimately recorded and will persist for as long as the blockchain itself continues to exist. This interpretation is broadly valid for Bitcoin, where the blockchain essentially is the ledger, and where the primary function of the system is to record and preserve transactional history. Blockchains, however, have undergone a profound evolution. With Ethereum, the blockchain is no longer only a ledger for financial transactions but a permissionless distributed system capable of executing arbitrary programs. The metaphor shifts from a single-purpose financial ledger to something closer to a "world computer." While an immutable ledger and native currency remain central components, the focus expands toward programmability, decentralized applications, and governance mechanisms rather than the immutability of all data associated with the system. This evolution also complicates the meaning of "on-chain." On Ethereum, immutability strictly applies to transactions, smart contract code, and contract state—that is, to data that is admitted into consensus. Large data such as photos are almost never stored directly on Ethereum, because doing so would be prohibitively expensive. Instead, smart contracts typically store hashes or references to data that lives off-chain, allowing verification of integrity without storing the data itself. In these cases, what is immutable is the commitment to the data, not the data itself. If the external storage disappears, the on-chain reference remains, but the content may no longer be accessible. The situation differs further in systems such as the Internet Computer, where application data can be stored within the system itself but remains mutable at the application level. In this context, "on-chain" resembles hosting data within a distributed computational platform rather than committing it to an immutable ledger. As a result, being "on-chain" does not universally imply permanence. The widespread association between blockchains and immutable content reflects the historical model established by Bitcoin, but it does not uniformly apply across newer blockchain architectures. Summarizing: On Bitcoin, all on-chain data is immutable and persists for as long as the blockchain exists, but the system is not designed to store large data such as photos, which would be impractical or economically prohibitive. On Ethereum, it is technically possible to store a photo directly on-chain, and such data would be immutable and persist for the lifetime of the blockchain; however, storing even a few megabytes would cost on the order of hundreds of thousands to millions of euros, making it unrealistic in practice. On the Internet Computer, photos can be stored directly at reasonable and potentially cloud-competitive costs, but this data remains mutable and its persistence depends on continued operation and payment rather than on strict ledger immutability. It is also important to note that Bitcoin, Ethereum, and the Internet Computer are only a few prominent examples within a much broader landscape. Today, there are hundreds of different blockchains, each making different trade-offs around decentralization, cost, performance, and data persistence. In particular, around Ethereum a rich ecosystem of so-called layer-2 networks and sidechains has emerged. These secondary chains are designed to reduce costs and increase scalability by moving computation and data away from Ethereum’s main chain, while still relying on it—directly or indirectly—for security and settlement. As a result, what “on-chain” means, how long data persists, and what guarantees are provided can vary significantly depending on the specific blockchain or layer being used. ↩

  7. For a good discussion of the problem of long-term digital storage, see Maxwell Neely-Cohen, "Century-Scale Storage," https://lil.law.harvard.edu/century-scale-storage/ ↩

  8. But even if we would not be concerned with true data ownership and we would consider as main storing system the mainstream Web2 solutions, and we would consider the most prominent among them, AWS, it is noteworthy what was Jeff Bezos self-assessment about the future of Amazon during an internal meeting in 2018 where he responded to a question about Sears's bankruptcy by telling employees that "Amazon is not too big to fail… In fact, I predict one day Amazon will fail. Amazon will go bankrupt. If you look at large companies, their lifespans tend to be 30-plus years, not a hundred-plus years." This episode is reported in Maxwell Neely-Cohen, "Century-Scale Storage," https://lil.law.harvard.edu/century-scale-storage/ ↩

  9. Several consumer services advertise "lifetime" cloud storage, though this typically means for as long as the company operates, not a legal guarantee. Main providers include pCloud (500 GB to 10 TB lifetime plans), Icedrive (starting from 2 TB), Internxt (privacy-focused, approx. 3 TB), Filen (zero-knowledge encrypted), and Koofr. FOREVER uniquely guarantees storage for your lifetime plus 100 years. Major players like Google Drive, Dropbox, and iCloud do not offer lifetime plans. Important caveat: "lifetime" equals service lifetime—bankruptcy, acquisition, or business model change can end the deal. ↩

  10. An endowment is a financial mechanism where funds are invested to generate returns, with only the investment income (not the principal) being used to support ongoing operations. This allows for long-term sustainability through a one-time upfront payment. American universities commonly use endowment systems to fund their operations indefinitely. A similar strategy is used by permanent.org, which, like Futura, operates in the long-term digital preservation space. See https://www.permanent.org/blog/how-our-endowment-works/ ↩

  11. From a genetic perspective, after the first experiments on ICP, we built a proof of concept of Futura as a Next.js app for ease of development. We started with Vercel Blob Storage, since you get it for free from Vercel. We then switched to S3 by AWS as the most mainstream storage solution, to get acquainted with the system - keep your friends closer and your enemies closer - but also because all the links to the assets generated by Vercel Blob are inherently public. Then during ETH Cannes 2025 we played around with SUI/Walrus, and during the WCHL we finally implemented ICP as well. Arweave has been on our radar since a long time, since they market exactly what we are looking for. ↩

  12. In contemporary wedding photography, the photographer typically delivers a curated gallery of approximately 500 edited images, although this number varies with the scale of the event and the duration of coverage; final collections may range from about 300 to as many as 1000 photographs. For high-resolution exports, individual files are most often provided in JPEG format, whose upper size limit is roughly 20 MB per image. The physical wedding album, by contrast, comprises a much narrower selection of around 50 photographs. Consequently, the complete digital delivery generally amounts to about 10 GiB, while a brief online selection shared for convenience is close to 1 GiB. Regarding ICP storage costs: storing data costs the same whether it's in heap or stable memory. On a 13-node subnet, 1 GiB for 1 year costs approximately 4 trillion cycles (approx. $5.35). Therefore, storing 10 GiB for 1 year costs approximately 40 trillion cycles (approx. $53.50). See https://docs.internetcomputer.org/building-apps/essentials/gas-cost#storage ↩

  13. DFINITY is preparing to roll out the updated blob storage service as part of the Caffeine Phase III upgrade, targeting early October 2024. Blob storage will let apps store large files like photos, videos, and documents directly on-chain at a fraction of today's cost. Dominic Williams announced this rollout and shared a target of about $0.025 per GB per year, which is significantly cheaper than AWS S3 Standard ($0.276), Wasabi ($0.084), Backblaze B2 ($0.072), and Storj ($0.048). https://x.com/dominic_w/status/1955447139347337491 ↩

  14. The basic idea that we implemented for the backend: for the Web2 architecture, we use a 'normal' centralized database, where Futura keeps the couples' wedding albums. Imagine a library holding all wedding albums, where couples can access only their album. For the Web3 architecture, we have an architecture in which all data regarding a couple live inside a capsule. All the capsules live in a central archive, in a space owned by Futura, but as capsules they can depart toward their own canister and live in a space owned by the couple alone. ↩