Tabalyst
ENFR
Get started
ENFR

Tabalyst ProductsComing soon

From the big picture to the records that matter.

Tabalyst Report summarizes and profiles a file. Tabalyst Explore is the next step: filtering and investigating it. It is coming soon: there is nothing to install or open yet.

Get started with Tabalyst ReportFollow the project on GitHub

The idea

Overview first, then investigation

Report summarizes and profiles. Explore filters and investigates. Two ways of looking at the same data, one after the other.

  • TODAY / TABALYST REPORT

    The overview

    What a file contains: types, missing values, distributions, ambiguous or rare values. A report you can share and a JSON profile for your tools.

  • NEXT / TABALYST EXPLORE

    The investigation

    Which part of the data is concerned: narrowing a dataset to a population, then reading the records behind a number.

Direction

Facets that suggest themselves

A facet is a dimension you filter by: a category, a period, a price range. Building every facet by hand is slow. The direction for Tabalyst Explore is to propose the relevant facets from the data itself, and to suggest what is worth looking at, so that you start exploring instead of configuring.

A category column
A facet with counts per value
A date column
A period
A price column
A range

An illustration of the direction, not a specification: the real choices are decided when Tabalyst Explore is built.

Direction

Filters and counts

The aim is to narrow a dataset with filters and see the counts follow, then open the records that remain. Which filters exist, how they combine and how missing values or text search are handled will be stated when they are built and tested, not before.

Examples

Questions it is meant to help with

Scenarios for the future Tabalyst Explore, one per domain. They show the kind of investigation it is meant to support. They are not an interface and not a promise of questions asked in natural language.

  • Insurance: among the customers whose birth date is ambiguous, which products do they hold?
  • CRM: which contacts share an email address, and where do they come from?
  • E-commerce: which products have no price, and in which categories?

Scope, not hype

Local-first, and honest about limits

  • Local-first is the aim. The goal is to explore on your own machine, like Tabalyst Report. How it will be delivered is not decided.
  • Large datasets are a design goal. No volume, speed or latency is promised.
  • The direction is reading and diagnosis, not correction of your files.
  • The first release will state its limits and its formats, as Tabalyst Report does today.

Next step

What you can use today

Everything below runs on your own files today.

PRODUCTAvailable

Tabalyst Report

The overview: interactive HTML reports and JSON profiles for CSV, Excel, JSON and JSONL.

TOOLSAvailable

Tools

Tabalyst Inspect, Tabalyst Scan and Tabalyst Sample, in the same package.

Today

Start with the overview.

Tabalyst Report is available now. Install Tabalyst and run it on a file: everything stays on your machine.

Get started with Tabalyst Report