Saving the Description and Everything Around It

What is in the file and what is not

This surprises people archiving a channel or saving research material. The video is the smallest part of what a page contains, and it is the only part a downloader retrieves.

Descriptions frequently hold the most durable value: sources, links to papers, timestamps, credits, corrections. A tutorial without its linked resources is often half a tutorial.

The simple method that works everywhere

1. Expand the description. Click ‘more’ — the collapsed view hides most of it, and copying without expanding gets you three lines.
2. Select and copy the whole thing. Including the timestamps, which are plain text.
3. Paste it into a text file next to the video, named the same way.
4. Add the video's own address at the top, plus the date you saved it.

Unglamorous and completely reliable. A plain text file beside the video will still open in thirty years, needs no software, and survives every format change that might affect the video itself.

The address and the date matter more than they look. Six months later, ‘where did this come from’ is the question you cannot answer from the file alone, and it is the one that matters for anything you intend to cite or reuse.

Chapters and timestamps

Chapters are generated from timestamps written in the description, in the form 0:00 Introduction. They are not a separate data structure — the platform reads them out of the text.

Which means copying the description captures them automatically, and they remain useful next to a downloaded file: your player will not show chapter markers, but a text list of times tells you where to skip to.

For a long recording this is the difference between a usable archive and three hours you will never navigate. A four-hour conference talk with a timestamp list is searchable; the same file alone is not — long streams are where this matters most.

Some players and media servers do support embedded chapters, but adding them to a file means re-encoding or remuxing with extra tooling. For most purposes the text file is sufficient and considerably less work.

Archiving properly: what to keep

If you are saving something you expect to need later, a short checklist prevents the regret.

• The video file, at a quality that suits how you will use it.
• The description as text, expanded, with the links intact.
• The video's address and the date you saved it.
• The channel name — handles change, and a link to a renamed channel leads nowhere.
• The thumbnail, if it matters visually.
• Anything in the comments you rely on — corrections and clarifications frequently live there rather than in the description.

That last one is the least obvious and the most often lost. A pinned comment correcting an error in the video is part of the content, and it is nowhere in the file.

For creators: your own metadata is work

Descriptions, tags, thumbnails and chapter lists represent real effort, and they are harder to recover than the video.

Recovering a video from a platform is possible, if degraded — covered here. Recovering a description you wrote three years ago, from a deleted post, generally is not: nobody archived it and no tool retrieves it.

Most platforms offer a data export that includes this material, and it is worth requesting periodically rather than at the point of leaving. It takes days to arrive, which is exactly the wrong timing if the account is already closing.

The simplest habit: write descriptions somewhere before pasting them in. A document of your own descriptions costs nothing to maintain and survives anything that happens to the account.

Why automated tools rarely help here

Tools exist that fetch a description alongside a video. They are useful and they have a specific weakness worth knowing.

Descriptions are unstructured text. A tool can retrieve the block, and it cannot reliably tell which parts are links to sources, which is sponsorship, which is boilerplate repeated on every video, and which is the pinned correction. Sorting that out is reading, and reading is what you were going to do anyway.

For a handful of videos, copying by hand takes less time than finding a tool that does it properly. For hundreds, a command-line archiver is the right instrument — that category handles metadata thoroughly and is built for exactly this — where installed software genuinely wins.

What no tool does is decide what mattered. That judgement is the actual work in archiving, and it is the part that cannot be automated regardless of the instrument.