My parents have a box of photographs. It is a bad archive by every technical measure — unsorted, undated, fading, vulnerable to a single house fire. It has also survived four moves, two countries and about fifty years, and I can open it right now without anyone's permission.
I have roughly a decade of my own life on platforms, and I could lose the lot on a Tuesday because of a policy update, an automated ban, a country-level block, or a company deciding a product line no longer fits the strategy.
That gap is the thing I have not stopped thinking about since I started working on tools that pull files out of feeds.
From files to feeds
There was a period, roughly the two decades either side of the millennium, when consumer computing was built on files. A photo was a JPEG in a folder. A song was an MP3 you could copy to a second drive. If the software you used to open it went away, the file stayed — possibly annoying, but yours.
Streaming ended that for music and video, and short-form apps finished the job for the personal stuff. What you have now is not a file. It is an entry in someone's database, rendered to you through an app, retrievable only as long as three separate things stay true: the company exists, your account is in good standing, and the content has not been removed by a rule that did not exist when you posted it.
None of those three is under your control. All of them fail regularly.
Ephemerality stopped being a bug
The part I find genuinely remarkable is that disappearing was deliberately designed in, then marketed as a feature, and we all agreed.
Stories vanish after a day. Some formats are built to be unsaveable by design. The interface pushes you toward posting more rather than keeping anything, because a full archive is not what generates engagement — a fresh post is. There is no export button anywhere near the create button, and there never will be, because those two actions serve opposite business goals.
An app optimised for you to produce constantly has no reason to be good at helping you keep what you produced. The incentives simply do not point that way, and no amount of complaining will make them.
The failure modes are not rare
Everyone treats archive loss as a freak event. From where I sit it looks routine:
- Bans and false positives. Automated moderation at scale means a percentage of accounts go down wrongly. If you are in that percentage, your appeal goes to a queue, and your library is behind a login you no longer have.
- Regional blocks. Whole countries have lost access to entire platforms with days of notice. The content did not move; the border did.
- Rights takedowns. A track in the background gets a claim years later and your video is muted or gone, retroactively.
- Products being switched off. Every large technology company has a graveyard of discontinued services. Some gave a year's warning and an export tool. Some did not.
- The ordinary ones. A phone lost, a password unrecovered, a relationship ending and an account being deleted in a bad week.
Nobody archives until they have lost something. That is the pattern, and it never changes. The people who write to a service like this asking how to get a video back are almost always already past the point where it was possible.
My guess about where this goes
I expect saving to get harder, not easier, and I want to be clear that this is a prediction rather than a report.
The pressure comes from three directions at once. Rights holders — music especially — keep pushing for tighter control over how audio leaves a platform. Platforms themselves have an obvious interest in content not travelling to competitors. And the technical means to enforce both keep getting cheaper: stricter APIs, segmented and encrypted delivery, tighter binding between a stream and a session, client attestation that treats anything but the official app as an attack.
None of that is aimed at you personally. It is aimed at scrapers, at rival products, at automated re-uploaders. You are collateral, in the same way you are collateral when a game you paid for stops working because its authentication server was retired.
My rough expectation for the next few years: the official export paths get slightly better because regulators in some regions now require data portability, while the unofficial paths get considerably harder. Which means the gap between "the platform's idea of your archive" and "an actual archive you can open" widens.
What I would actually do about it
Not a manifesto, just the habits I think hold up.
Keep the original. The version you upload is the best version that will ever exist. Everything after that is a compressed, stamped, re-encoded derivative. Get the master off the phone and onto something boring — an external drive, a NAS, a cloud folder you pay for — before it is the only copy.
Do not treat a platform as storage. It is a distribution channel. The distinction matters the day the channel closes.
Export on a schedule, not on a scare. Most large platforms offer a full data export. It is slow, the format is awkward, and almost nobody does it twice. Once a year is enough to change the outcome entirely.
Save the things you would grieve. Not everything — most of it is genuinely disposable. But the video of your kid, the last clip of someone who is gone, the tutorial from a channel that has since deleted everything: those cost you a minute now and are unrecoverable later.
Accept that keeping is now an active habit. This is the real shift. Keeping used to be the default state of a file; you had to try to lose it. Now losing is the default and keeping is the thing that takes effort.
The uncomfortable part
I am aware that I am arguing for personal archiving while a good deal of the industry argues that content should stay where it was published. I do not think those positions are as far apart as they are made to sound. Nothing here is an argument for taking other people's work — that remains theirs whatever container it sits in.
It is an argument that the default should not be total dependence on a company's continued existence, continued goodwill, and continued interest in a product line, for access to your own life.
My parents' shoebox is a terrible archive. It has one advantage that every service I have ever used lacks: nobody can revoke it.

Comments
Post a Comment