Settings

The schedule

A CSV from the provider, or the schedule that came back from the single point of contact. It is opened by this tab and goes nowhere.

What was read

Numbers

Click a number to narrow everything below it to that number.

Who spoke to whom

Around a moment

The time of the offence, the last sighting, the alarm activation. Everything either side of it, with how long before or after.

Numbers on the same cell

Everything, in order

Take it with you

How it works

COMMS reads the CSV in your browser and works out the patterns: who called whom, when, how often, and where calls were routed through the same cell. It is careful about dates and time zones, because a wrong one changes the story, and it never guesses who a number belongs to. Nothing is sent anywhere.

The guide: Reading a schedule of call data

Questions people ask about COMMS

What kind of file does it take?

A CSV: the export a provider gives you, or the schedule returned through a single point of contact, or anything you have already pulled into a spreadsheet and saved as CSV. Quoted fields, commas inside a location column, both line endings and a byte-order mark on the first heading are all handled. You then match up which column is the calling party, the called party, the date and time, the duration, the direction, the event type, the cell reference and the cell location; only the calling party and the date and time are required.

Why does it make such a fuss about dates?

Because it is the mistake that does the most damage and shows the least. If a file written as month first is read as day first, every event moves by up to eleven months, every pattern and every window around a moment is wrong, and nothing on the page looks broken. So the whole column is examined before any row is read: any date with a first component above twelve settles the order for the file and the page says which reading it used and why. If every row reads both ways, it says so in red rather than choosing quietly.

What does the same-cell finding actually mean?

That two different numbers appear against the same cell reference within the window you set. That is all it means. A cell serves an area (hundreds of metres in a city, several miles in open country), cells are reselected for load and signal rather than movement, and a schedule records the cell serving the call, not where the handset was. It is a place to look, not a conclusion, and the page says so where the finding is rather than in a footnote.

Can it handle a large schedule?

Yes. The parsing and the arithmetic are done in the tab and a schedule of tens of thousands of rows is read in a few seconds. The tables on screen show the first few dozen rows so the page stays usable; the download has everything.

Is any of it sent anywhere?

No. The file is read by the browser’s own file reader, everything is worked out in the page, and the downloads are written by the tab. There is no account and no server involved at any point. The settings you choose are kept in this browser; the schedule itself is not kept at all, so reopening the page means opening the file again.

Does it say who a number belongs to?

No, and it never will. Attributing a telephone number to a person is a separate exercise resting on subscriber checks, handset examination, admissions and other evidence. This reads the schedule and shows you the patterns in it.

Doing this often? OBSCURA Case reviews communications in bulk, with threads and a who-wrote-to-whom view, and builds the chronology. Everything is US$5 a month or US$48 a year.