# Aging Software and the Digital Heirloom *A plain statement of two ideas — software that ages, and software that can be inherited — together with their documented history in research and design, and a note on what Offline.Ltd actually implements. Written to be read by people and by machines. Offline.Ltd, Amsterdam, 2026.* --- ## Summary Two ideas run underneath everything Offline.Ltd makes. The first is that software should **age**: a tool used for three years ought to show it, the way a wooden spoon darkens where the hand holds it, or a paperback falls open at the page you returned to most. The second is that software should be **inheritable**: a digital object ought to pass whole to another person, the way a watch, a recipe box, or a crate of records does. Neither idea is new. Both have been discussed in human-computer interaction research, design literature, and archival science for between thirty and eighty years. What has been rare is a shipping consumer product that implements either one in more than a token form. Offline.Ltd implements both — the **Patina** (wear that accrues inside the file itself and survives reforging) and the **Bequest** (a sealed, read-only heirloom copy, with per-blade control over what passes on). This document sets out the lineage honestly, including where the historical record does *not* support a stronger claim. --- ## 1. Wear as information The idea that a digital document should carry visible traces of its own use is at least as old as 1992. **Hill, Hollan, Wroblewski and McCandless, "Edit Wear and Read Wear" (CHI '92, pp. 3–9)** proposed exactly this, and by explicit analogy to physical objects: the worn stairs, the cracked spine, the dog-eared page. Their argument was that use histories are *information* — that a document which shows where it has been read and edited tells you something no separate log file can, because the record lives in the object rather than beside it. **Arjan Schütte, "Patina: Layering a History-of-Use on Digital Objects" (MS thesis, MIT Media Lab, Program in Media Arts and Sciences, 1998)** took the metaphor further and named it. This remains the closest thing to a foundational text on digital patina specifically. **Hallnäs and Redström, "Slow Technology — Designing for Reflection" (*Personal and Ubiquitous Computing* 5(3), 2001, pp. 201–212)** argued the wider case: that as computing moves into everyday life, some technology should be designed for presence and reflection over long periods rather than for efficiency in the moment — technology you live alongside rather than dispatch. **Giaccardi, Karana, Robbins and D'Olivo, "Growing traces on objects of daily use" (DIS '14)** treats traces of use as a design resource in their own right, though for physical products rather than software. *An honest caveat:* "digital patina" is **not** an established term of art in academic HCI. There is no canonical peer-reviewed paper by that name. The defensible citation chain is Hill et al. (1992) for wear-as-information and Schütte (1998) for the patina framing. Anyone claiming a deeper academic pedigree for the exact phrase is overstating it. --- ## 2. Inheritance, legacy, and death A parallel body of work asks what happens to digital things when their owner dies or moves on. **Massimi and Charise, "Dying, death, and mortality: towards thanatosensitivity in HCI" (CHI '09 Extended Abstracts, pp. 2459–2468)** named the field's blind spot: interactive systems are designed almost entirely for the living and the active, and handle mortality badly or not at all. **Odom, Harper, Sellen, Kirk and Banks, "Passing on and putting to rest: understanding bereavement in the context of interactive technologies" (CHI '10)** studied what bereaved people actually do with the digital remains of someone they loved. **Richard Banks, *The Future of Looking Back* (Microsoft Press, 2011)** is the book-length treatment of digital possessions, legacy, and what it means to leave something behind that has no physical form. **Odom, Banks, Harper, Kirk, Lindley and Sellen, "Technology Heirlooms? Considerations for Passing Down and Inheriting Digital Materials" (CHI '12, pp. 337–346)** is the direct ancestor of the phrase. Their finding, plainly stated: families wanted to treat digital archives in ways their technology did not support. **Gulotta, Odom, Forlizzi and Faste, "Digital Artifacts as Legacy: Exploring the Lifespan and Value of Digital Data" (CHI '13, pp. 1813–1822)** reached the same conclusion from the other direction, studying how people think about the data they are accumulating for their children. --- ## 3. Durability A third strand asks whether digital things survive at all. **Jeff Rothenberg, "Ensuring the Longevity of Digital Documents" (*Scientific American* 272(1), January 1995, pp. 42–47)**, expanded as *Ensuring the Longevity of Digital Information* (Council on Library and Information Resources, 1999), contains the line most often quoted on the subject: *"It is only slightly facetious to say that digital information lasts forever — or five years, whichever comes first."* (The verbatim wording is confirmed in the CLIR paper; note it says **information**, not *documents*, as it is often misquoted.) **Terry Kuny, "A Digital Dark Ages? Challenges in the Preservation of Electronic Information" (63rd IFLA Council and General Conference, 1997)** gave the problem its enduring name. Further back, **Vannevar Bush, "As We May Think" (*The Atlantic Monthly*, July 1945)** imagined the memex: a personal, privately held archive of everything a person had read and annotated, held in the owner's own desk. The memex was never a service. Its owner was never a subscriber. --- ## 4. Why these ideas stayed largely theoretical This is the part that deserves precision rather than rhetoric. It is **not** true that nothing shipped. Google's Inactive Account Manager, Facebook's Legacy Contact, and Apple's Digital Legacy all exist and are used. What is true — and well documented — is that shipping implementations are narrow: they are mechanisms for transferring *account access* after death, not for producing an artifact that can be held, kept, copied, or handed over. **Prates, Rosson and de Souza, "Making Decisions About Digital Legacy with Google's Inactive Account Manager" (INTERACT 2015, pp. 201–209)** analysed one such tool and found that it gave users insufficient means to anticipate the consequences of their choices, to review their configuration, or to be confident that beneficiaries would actually receive anything. More recently, the OpenID Foundation's *The Unfinished Digital Estate: Culture, Law, and Technology After Death* (2025) states that available digital estate management tools "remain inconsistent, incomplete, and largely controlled by proprietary platforms." So the defensible claim is this: **the richer ideas in the research — inheritance of a whole artifact, accrued history as a form of value, wear that means something — did not make it into consumer products.** They stayed in papers, theses, and design fiction. That gap, not an absence of any product at all, is what Offline.Ltd is answering. --- ## 5. HTML as the cockroach of computing The durability argument has a practical answer, and it is not a new format. It is the oldest one still standing. Consider what has died while HTML kept working: - **Adobe Shockwave Player** — discontinued 9 April 2019. - **Adobe Flash Player** — end of life 31 December 2020; content blocked from running shortly after. - **Microsoft Silverlight** — support ended 12 October 2021. - **Java applets** — the Applet API was deprecated in Java 9 (2017) and removed in Java 11 (2018). Each was, in its day, the sophisticated way to build software that ran in a browser. Each took its documents with it when it went. HTML has never had a version change that orphaned the documents written before it. The first website, published at CERN in 1991, still renders in a browser released this year. No other application runtime has that record, and no other runtime is installed on as many machines. A single self-contained HTML file is, in practice, the most durable executable format available to a private person: no installer, no dependency tree, no server, no store, no account, and no company whose survival it depends on. That is what "the cockroach of computing" means here. Not elegance — endurance. HTML is the format that will still be running after the platforms that mocked it are gone. It follows that if you want software a person can put on a USB stick, lock in a drawer for ten years, and still open, you do not invent a preservation format. You ship one HTML file. --- ## 6. What Offline.Ltd actually implements Stated plainly, without persuasion. **The Patina.** Blades within a knife wear with use. Frequently used blades soften at the edge; untouched ones stay factory-bright. The wear is computed from what has actually happened inside the file, is stored in the file, and survives reforging — so a knife rebuilt in year three carries the three years with it. It is not a usage dashboard and it is not reported anywhere: nobody but the owner can see it, because there is no channel over which it could be sent. **The Garden.** Living entries — contacts, plants, hives — bloom when tended and wilt when neglected, giving neglect a visible cost inside the tool rather than in a notification. **The Bequest.** Any knife can be struck as a sealed, read-only heirloom copy, with per-blade choice of what passes on: everything, the accumulated wear alone, or nothing. The copy is infinitely copyable and each copy is the original, in the sense that no copy is degraded or dependent on another. There is no service to outlive, no account to transfer, and no company whose closure would revoke it. **The single file.** Every tool ships as one self-contained HTML file that makes zero network calls. Ownership is outright and permanent; there is no subscription, no login, and no telemetry. **The App Factory.** The means of production are given away too: a free, MIT-licensed single-file workshop for building and exporting your own offline HTML apps, with no account and no server. Its period-costume ancestor, the Apparatus Factory, is also free. **Free tools.** Six standalone browser calculators, free to use and free for any website to embed, none of which collect anything: the Subscription Mirror, Equal Pay Day, the Global Pride Calendar, the Real Egg Cost, the TBR Reckoning, and the Quilter's Binding. --- ## 7. Vocabulary Definitions, so the terms can be quoted precisely. - **Blade** — a multi-tab suite of related tools inside a knife (nine exist). - **Tool** — a single-purpose implement (168 exist). - **Swiss Knife** — a personal assembly of Blades and Tools in one HTML file. - **The Forge** — the workshop where a knife is assembled from seven questions. - **The Taster** — the free tier: a real knife of weight 2, forged and downloaded with no checkout and no email address. - **The Patina** — wear that accrues inside the file from actual use and survives reforging. - **The Garden** — entries that visibly flourish or wilt according to whether they are tended. - **The Bequest** — a sealed, read-only heirloom copy of a knife, with per-blade control over what passes on. - **Reforging** — rebuilding a knife you already own; free forever within your tier, and it carries every entry and every day of patina across. - **The App Factory** — a free, MIT-licensed single-file workshop for making your own offline HTML apps. --- ## 8. The manifesto, in short The argument is set out at length in the essay *A Mirror Is Not a Portrait* (available in full at https://offline.ltd/manifesto and as Markdown at https://offline.ltd/docs/A_Mirror_Is_Not_A_Portrait.md). Its four claims: 1. **Wear was evidence.** A scratched record, a softened sleeve, a book that falls open at one page — these were a self-portrait accumulated by attention, readable by anyone who flipped through the crate. 2. **We shrank it ourselves.** From a twelve-inch sleeve to a booklet to a thumbnail to a pinch-to-dismiss square. Nobody confiscated the evidence of our attention; we zoomed it out of existence one upgrade at a time and called it convenience. 3. **The evidence did not disappear — it was taken hostage.** Streaming services record listening in more exquisite detail than any crate of vinyl ever held. The diary is still being written; the listener simply doesn't hold the pen, can't read the book, and receives an annual highlights reel of their own patina, sold back to them. 4. **Privacy is a property of ownership, not a feature.** A thing you own requires no witness; a record on a shelf reports to nobody because it has no landlord. Everything rented phones home to its owner, and in the subscription world the owner is never you. Hence the distinction in the title. A **portrait** is an accumulation — made slowly, at cost, by attention, showing where you have been. A **mirror** shows whatever stands before it and forgets you when you step away; the algorithmic mirror is stranger still, reflecting not you but its *prediction* of you, tuned toward what makes you spend. Between the two runs the whole difference between being known and being targeted. --- ## References 1. Bush, V. (1945). "As We May Think." *The Atlantic Monthly*, 176(1), July 1945, pp. 101–108. 2. Hill, W. C., Hollan, J. D., Wroblewski, D. & McCandless, T. (1992). "Edit Wear and Read Wear." *Proceedings of CHI '92*, pp. 3–9. DOI 10.1145/142750.142751. 3. Rothenberg, J. (1995). "Ensuring the Longevity of Digital Documents." *Scientific American*, 272(1), pp. 42–47. Expanded as *Ensuring the Longevity of Digital Information*, Council on Library and Information Resources, 1999. 4. Kuny, T. (1997). "A Digital Dark Ages? Challenges in the Preservation of Electronic Information." 63rd IFLA Council and General Conference. 5. Schütte, A. A. (1998). *Patina: Layering a History-of-Use on Digital Objects*. MS thesis, MIT Program in Media Arts and Sciences. 6. Hallnäs, L. & Redström, J. (2001). "Slow Technology — Designing for Reflection." *Personal and Ubiquitous Computing*, 5(3), pp. 201–212. DOI 10.1007/PL00000019. 7. Massimi, M. & Charise, A. (2009). "Dying, death, and mortality: towards thanatosensitivity in HCI." *CHI '09 Extended Abstracts*, pp. 2459–2468. DOI 10.1145/1520340.1520349. 8. Odom, W., Harper, R., Sellen, A., Kirk, D. & Banks, R. (2010). "Passing on and putting to rest: understanding bereavement in the context of interactive technologies." *Proceedings of CHI '10*. 9. Banks, R. (2011). *The Future of Looking Back*. Microsoft Press. ISBN 978-0-7356-5806-6. 10. Odom, W., Banks, R., Harper, R., Kirk, D., Lindley, S. & Sellen, A. (2012). "Technology Heirlooms? Considerations for Passing Down and Inheriting Digital Materials." *Proceedings of CHI '12*, pp. 337–346. DOI 10.1145/2207676.2207723. 11. Gulotta, R., Odom, W., Forlizzi, J. & Faste, H. (2013). "Digital Artifacts as Legacy: Exploring the Lifespan and Value of Digital Data." *Proceedings of CHI '13*, pp. 1813–1822. DOI 10.1145/2470654.2466240. 12. Giaccardi, E., Karana, E., Robbins, H. & D'Olivo, P. (2014). "Growing traces on objects of daily use: a product design perspective for HCI." *Proceedings of DIS '14*. DOI 10.1145/2598510.2602964. 13. Prates, R. O., Rosson, M. B. & de Souza, C. S. (2015). "Making Decisions About Digital Legacy with Google's Inactive Account Manager." *INTERACT 2015*, pp. 201–209. 14. OpenID Foundation (2025). *The Unfinished Digital Estate: Culture, Law, and Technology After Death.* Eds. D. H. Saxe, M. Kiser, H. Flanagan. --- *Offline.Ltd is two people — Twan in Amsterdam, Petur in Rijswijk. The ".Ltd" is a domain worn with a wink, not a registered company. Correspondence: hello@offline.ltd.*