Editorial Guidelines

These are the standards every page on Taste Korean Food is held to. They cover who writes the site, how a page is researched and checked before it is published, what we will not publish, and how to tell us when we get something wrong.

Who writes this site

Articles are published under the byline James Lee, a pen name. Behind it is a practicing surgeon with more than twenty years of clinical experience, working with a small editorial and photography team based in Korea. The same byline is used across the site so that readers and search engines can hold one consistent voice accountable.

Medical and wellness material — the K-Beauty Kitchen hub, and the Doctor's Nutritional Insight notes that appear on recipe and ingredient pages — is written and evidence-checked by that physician. It is general information about food. It is not a diagnosis, not a treatment plan, and not a substitute for care from your own doctor.

How a page gets made

Every page moves through the same five stages, and a page that fails a stage does not advance to the next one.

1. Research. Facts, history, and health claims are traced to primary sources before anything is drafted. If the sourcing is not there, the article is not written.
2. Structure. The page is planned so it answers one question well, rather than competing with our own pages that already answer it.
3. Writing. The draft may only use facts and links established in stage one.
4. Independent review. The draft is checked by reviewers working separately from the writer, who do not see the writer's reasoning — only the finished text. They cover accuracy, medical evidence, natural English, search quality, and images. Reviewers report findings; they do not quietly edit the page.
5. Publication checks. Automated checks run before and after publishing: internal links must resolve, images must carry real descriptions, structured data must match what the page visibly says.

Recipes are cooked in our kitchen

We cook and photograph every recipe we publish. The prep, cook, and total times and the yield on a recipe page are what we measured, not figures copied from elsewhere. The photographs are of the food we actually made.

Where a step is hands-off — an overnight freeze, a long ferment, a rest in the fridge — we say so, because a total time that hides eight hours of freezing is not useful to anyone planning dinner. Nutrition figures, where shown, are calculated estimates rather than laboratory measurements, and are labeled as estimates.

Evidence standards for health content

Food-and-health claims carry a higher burden than the rest of the site, because getting them wrong can affect what someone eats while recovering from surgery or managing a condition.

• A health claim needs at least two sources from peer-reviewed literature or a recognized medical institution — PubMed-indexed journals, the NIH, Cochrane, or academic medical centers such as Mayo Clinic, Cleveland Clinic, or Harvard Health.
• Content farms, personal blogs, and supplement sellers are not acceptable sources for a health claim, no matter how confidently they are written.
• Where the evidence is preliminary, in animals, or in small trials, the page says so in the same sentence as the claim — not in a disclaimer at the bottom.
• When the honest answer is that the research does not exist, we write that. A missing answer stated plainly is more useful than a confident one that is not supported.

How we check a citation

A link that resolves is not the same as a source that supports the claim. Every citation on a health page is checked on four levels, because we have found real failures at each one.

1. The source exists and the identifier resolves to a real record.
2. The journal and year match what the page claims.
3. The title matches the subject. Study identifiers sit next to each other in sequence, so a single-digit slip can land on a paper published in the same journal on the same day about something entirely unrelated. Only checking the title catches that.
4. The number belongs to the population being described. A figure from a subgroup, or from a secondary analysis, is not a result for everyone in the study. Abstracts sometimes blur this; we check which table a number came from.

What we will not publish

• A star rating that is not built from real reader reviews. Where no one has reviewed a page yet, it says so instead of showing an empty score.
• A source, statistic, or internal link we have not verified ourselves.
• A superlative we cannot support. "The best in Korea" is a claim about the world, and we treat it as one.
• Paid placement presented as editorial judgment. If a page ever carries an affiliate link, gifted product, or sponsored element, that is disclosed on the page itself, near the top.

Corrections

We publish corrections rather than quietly editing pages. When a factual error, a mis-stated measurement, or a misapplied source is confirmed, the page is fixed and the change is dated and described in our public corrections log.

If something on this site looks wrong, tell us — a recipe that did not work as written, a restaurant detail that has changed, a study that does not say what we said it says. Write to editor@tastekoreanfood.com with the page address and what you found. Reports that identify a real error are acted on, and health-related corrections are treated as urgent.

Updates to these guidelines

These standards change as we find new ways to be wrong. When a check listed here was added because something slipped through, that is the reason it exists.