Export Formats That Outlive Software

An export is the thing that carries your record across a decade of tool changes. Most people have never opened one.

For a tool-based version of the same discipline, review time reporting tools. The aim is to preserve dependable evidence while keeping the ordinary recording action simple enough to repeat.

What a complete export contains

Every entry, not the last year. Some tools export a default range and the default is not "everything".

For a neutral companion reference on privacy-aware tools, see Mozilla privacy resources; it helps separate the choice of format from the conclusions drawn from the record.

Start and end times, or start plus duration, with the date unambiguous.

The category, as text rather than as an internal identifier.

All tags, in a parseable form rather than concatenated into a sentence.

And the free-text note, which is the field most often dropped.

If any of these is missing, the export is not a backup.

The traps

Internal identifiers instead of names. A column of numbers referring to a category table that is not exported.

Durations as formatted text — "1h 30m" — which every subsequent tool has to parse and some parse wrongly.

Local dates with no year, or the ambiguous month-day-year form, which makes a decade of data unreliable at the boundaries.

Truncated notes, which happens silently at a character limit.

And a separate export per year that you have to request individually, which people do once and never repeat.

What good looks like

One file. Every row. Plain CSV or JSON.

Dates as year-month-day.

Durations in minutes as an integer.

One row per entry, with tags in a single delimited field.

A file like that will still be readable in twenty years by anything.

Checking it

Open it in a text editor. Read three rows.

Count the lines and compare against what the tool says you have.

Check the first entry date — does it match when you actually started, or has it silently begun at some later point?

Search for a note you remember writing.

Four checks, five minutes, once a year.

If the export is inadequate

Some tools offer an API where the export function is poor. A short script run monthly is a reasonable answer and is worth the hour it takes to write.

Some offer nothing useful, in which case the honest assessment is that your record is not portable, and that should weigh on whether to keep using the tool for something intended to last.

Storing exports

Dated filenames, so you have a sequence rather than one file overwritten.

Keep a few years of them. They cost nothing and they let you detect a corruption or an omission that appeared at a particular point.

And keep one outside the ecosystem you use daily.

The conversion you will eventually do

At some point you will move the record into a new tool or a script of your own.

That job is easy from a clean CSV and painful from anything else, which is the whole argument for caring about this before you need to.

What to check

Open your most recent export now — does it contain your notes?

Does it start from your first entry or from some later date?

Are the dates unambiguous?

And do you have exports from more than one point in time?

The point

Open your export and read three rows.

Most people have never done this, and an export that has silently been producing empty files is discovered at exactly the wrong moment.

Worth adding

Dated filenames, kept as a sequence rather than overwritten.

They cost nothing and they let you detect a corruption that appeared at a particular point.

Additionally

Check the export four ways: open it, count the lines, check the first entry date, search for a note you remember writing.

Five minutes, once a year.

Finally

A good export is one file, every row, plain format, dates year-first, durations as integers.

A file like that will still be readable in twenty years by anything.

In summary

The traps: internal identifiers instead of names, durations as formatted text, ambiguous dates, truncated notes, and per-year exports you have to request individually..

Last word

The practice this points at is unglamorous: a small daily act, a short weekly correction, an annual read, and a written page of conventions beside the data. None of it is difficult and all of it is easy to skip, which is why most records do not reach the point where they start paying.