Which Email Formats Will Still Open in 20 Years

Which email formats last is decided by who can still read them, not by the disk. Files do not rot. Formats are abandoned. A mail archive becomes unreadable because the software that understood it stopped existing, not because anything happened to the bytes. That distinction decides which format you should be keeping things in.

The question worth asking is not which format is best today. It is which one will still open when the person asking is somebody you have never met, working on a machine that has not been built yet.

What Ends an Email Format

Three things decide whether a format is still readable in twenty years. None of them is about quality.

Age is evidence and it goes undervalued. A format that has already been readable for forty years has survived every technology shift since. One that is ten years old has survived none. Existing history is the only real information anybody has about the future here.

One Email Format That Died in Plain Sight

Doubt that a format can die inside one working career and DBX settles it.

In 2003 it sat on nearly every Windows machine in the world, because Outlook Express came with the operating system. No current mail program reads one. Not Outlook, not Thunderbird, not Apple Mail. The format did not fail and nothing corrupted. Microsoft stopped shipping the program and nobody else had ever implemented it.

1997   ships with Internet Explorer
2001   on most Windows machines
2007   Vista replaces Outlook Express
2009   Windows 7 has no mail client at all
2026   no mainstream program reads one

The people holding DBX files in 2003 had no reason to worry. It was the most common mail format on earth. Ubiquity is not durability. The two get confused constantly when somebody is choosing where to put an archive.

How Each Email Format Rates for Longevity

Against those three tests, the formats separate more sharply than their popularity suggests.

The top two rows are not there because those formats are clever. They are there because a message stored as text under a public specification is the one thing in mail that everything can read.

PST is better than its reputation and worse than the top two. The specification is public and other people's software reads it, which is not the position DBX was ever in. It is still binary and still one company's format. A binary index is also a single point of failure that a text file does not have.

The Email Format That Is Not a Format

Archiving mail as PDF comes up constantly and it answers a different question than the one being asked.

As well, never instead. Keep the message in a message format and generate a PDF where somebody needs a page. Doing it the other way round throws away everything the headers carry. Those are what answers most questions asked years later.

What Actually Goes Wrong to Email Archives

Format choice is one risk and not the commonest one.

Four of the five are about handling rather than about formats. Choosing well matters and it comes fifth, behind knowing where the file is and having more than one copy.

An archive nobody has opened is a claim rather than a fact. Once a year, open one, search it and find something specific. That habit catches four of the five rows above while they are still fixable.

Choosing a Format for an Archive You Will Not Maintain

1. Keep the message, not a rendering. EML for individual messages, MBOX for folders. Everything else is a copy of a copy.
2. Do not encrypt an archive with a password one person knows. Whoever opens it in ten years is not that person. Disk encryption with managed keys is a different thing and it is fine.
3. Write down what it is. A plain text file beside the archive saying what it holds, where it came from and when. Nothing else in the folder explains itself.
4. Keep the original alongside any conversion. Converting is safe when the source survives. Converting in place is a decision that cannot be revisited.
5. Open it once a year. Not check that it exists. Open it, search it, find one message. Anything less proves nothing.

The last one is the whole discipline. Every other item on this list is a decision made once. That one is a habit. It is also the only item that finds problems rather than preventing them.

Where an old archive is in a format nobody supports any more, reading it is the first step and converting it is the second. Our file viewers open several of these in a browser tab, so you can see what is there before deciding. The converters move the contents into something that will still open in twenty years.