Ledger

Original Article ↗ Hacker News Discussion ↗

Plaintext double-entry accounting tools like Ledger, hledger, and Beancount attract programmers and power users who value transparent data, strong querying, and the ability to script or version-control their finances. Commenters compare these tools with spreadsheets, QuickBooks, GnuCash, and bank apps, arguing that while setup and data import can be tedious, the long‑term benefits include accurate multi-account tracking, better reporting, and easier bulk changes. Many highlight specific workflows—Emacs integration, CSV/QFX import pipelines, mobile capture, and web UIs—as well as the educational value of learning core accounting concepts such as the accounting equation and transaction balancing.

Ledger vs hledger and other plaintext tools

  • Ledger and hledger are broadly compatible; many use the same journal with both.
  • hledger: better documentation, saner but less powerful query language, fewer bugs, chronological ordering of transactions by default, good CSV importing, web UI (hledger-web), Haskell API usable from a REPL.
  • Ledger: more powerful and flexible queries, assertions and checks (e.g., time-period constraints), works well with Emacs, extensive docs, but default file-order sorting can be surprising (can be overridden via sort options or config).
  • Some users run both: hledger for imports/validation, Ledger for advanced reporting.
  • Alternatives mentioned: beancount (+ Fava UI, praised docs and investment handling), PTA (numbered chart-of-accounts, one-line journal), beancount/ledger comparison docs.

Workflows and tooling

  • Common pattern: keep a single journal in Git; mostly use a few balance/report commands bound in editors (especially Emacs ledger-mode).
  • Some recommend starting small (one “zone” like monthly bills) and iterating.
  • Examples:
    • hledger-ui for interactive use.
    • Drafts + TextExpander + Dropbox for quick mobile entry.
    • Makefiles, Python/Lisp/awk scripts to convert CSVs to ledger; some summarize yearly.
    • hledger-flow for multi-account management.
    • Combining ledger output with Jupyter for custom charts.

Bank data access and automation

  • Many banks only offer CSV or even PDF, making import painful; regional standards (e.g., CODA, differing by bank) add friction.
  • Some European users have APIs (FinTS/HBCI/PSD2), but often only via third-party aggregators, raising privacy concerns.
  • Tools: ledger-autosync (QFX → Ledger), ofxstatement (proprietary → OFX), custom ML for auto-categorization.

Accounting concepts and learning

  • Thread dives into double-entry basics: accounting equation, debits/credits as left/right rather than “positive/negative.”
  • Ledger’s negative-number model and “transactions sum to zero” is praised as conceptually clear.
  • Recommended learning resources: Ledger manual sections on accounting principles, Kleppmann’s “Accounting for computer scientists,” GnuCash tutorial, beancount docs.

Use cases, benefits, and limitations

  • Used successfully for: personal finance, small businesses, multi-entity/multi-currency setups, investment tracking, tax reporting, shared expenses, and long-term net-worth history.
  • Strengths: plain text, easy bulk recategorization with text tools, strong querying, privacy (no SaaS), composability with scripts.
  • Limitations: initial setup and ongoing manual entry can be tedious; some users track only selected categories or eventually revert to GUI tools/spreadsheets when import/cleanup effort feels too high.
  • Debate over simpler tools (Excel/SQL, bank apps, budgeting apps) vs the rigor and flexibility of double-entry plaintext systems.