Why we're starting a blog for Sygal

Sygal just went live on Mac and Windows, the site has been running for a few days, and the first user feedback is starting to come in. What was missing was a place to write about what we're learning while building the tool, not just about what it does.

Why now

We could have filled this space with generic articles like "how to convert a PDF to Markdown." That wouldn't have added anything: that content already exists everywhere, and it doesn't help anyone decide whether Sygal is the right tool for their actual problem.

What we do have is data we verify ourselves by actually testing conversions inside the app: what OCR really costs depending on the provider, what happens to your documents when you choose local-only processing over the cloud, and what we find when we compare real token costs across formats, including when the result doesn't match what we expected going in. That's the kind of verified content worth publishing, so that's what we're starting with.

What you'll find here

Three kinds of articles, depending on what brought you here:

If you're preparing documents for an AI or RAG pipeline: technical comparisons, real token-cost numbers by format, and notes from integrating different OCR providers.

If you handle paperwork day to day (invoices, contracts, scans): practical guides for turning piles of documents into usable text without burning hours on it.

If you work in a field where data can't leave the machine (legal, healthcare, finance): honest explanations of what actually happens to your files, and why local-first isn't just a marketing line.

One simple rule

Every article published here is checked against Sygal's actual code or against cited public sources, never against a hunch. If a number changes later (an OCR provider adjusts its pricing, for example), we update the article instead of letting it quietly go stale.

If there's a topic you'd like us to cover and we haven't yet, tell us. That's usually the best source for the next article.