imagiware.Check a page
← All guides

GUIDE 01 / HTML BASICS

What an HTML checkup can tell you.

An automated check is useful when it helps you ask a better question about your page. Here is how to turn the findings into practical improvements.

Start with the source.

A browser can make a page look reasonable even when its source has problems. It may repair missing elements or tolerate unexpected markup. That is convenient for visitors, but it can hide mistakes from the person maintaining the page.

HTML Checkup inspects selected patterns in a page’s source. You can paste HTML for a local check, or provide a public URL for a server fetch. The result is a list of observations to review, not a score that proves your page is correct.

Give the document a clear identity.

Start with a doctype, a document language, and a descriptive title. A language declaration helps software interpret the page, while a useful title gives context outside the page itself.

<!doctype html>
<html lang="en">
  <head>
    <meta charset="utf-8">
    <title>How to repot a fern | Greenhouse Notes</title>
  </head>
  <body>
    <h1>How to repot a fern</h1>
  </body>
</html>

This is an illustrative starting point. Set lang to the actual language, and write a title that describes the page’s subject rather than a generic label such as “Home.” The HTML Standard explains the underlying elements and attributes.

Review alternative text in context.

An automated check can notice a missing alt attribute. It cannot reliably decide what the image means to a reader. Ask what information or function the image contributes.

  • For an informative image, describe the relevant information.
  • For an image that acts as a link or button, describe its purpose.
  • For a purely decorative image, an empty alt="" can be appropriate.
<!-- Informative image -->
<img src="fern.jpg" alt="Fern roots tightly circling the old pot">

<!-- Purely decorative flourish -->
<img src="flourish.svg" alt="">

Use the W3C alternative-text decision tree to work through the context. Passing an attribute-presence check does not prove the text is useful.

Give headings and controls a purpose.

Headings should help a reader understand the page’s sections. Labels should make it clear what a form control expects. Repeated IDs can make links and label associations ambiguous. When a finding mentions one of these details, inspect the actual content and behavior before editing.

A checker sees markup; a visitor experiences the whole page. Test whether keyboard focus moves sensibly, whether the interface still works when zoomed, and whether instructions make sense without relying on color alone.

Read the severity, then use judgment.

Treat a finding as a prompt to investigate. A missing title and an unusually long title are different kinds of concern. One is an absent element; the other may be intentional and needs editorial judgment. The most useful fix is the one that improves the page for its audience.

What this tool does not do: HTML Checkup is not a full HTML conformance validator or an accessibility certification. It does not execute your page’s JavaScript, inspect visual layout, test every link, or evaluate every possible markup rule.

A practical next step.

  1. Check the source of a representative page.
  2. Fix concrete omissions and review context-dependent suggestions.
  3. Re-run the check and confirm the page still behaves correctly.
  4. For a broader review, combine source checks with a standards validator, keyboard testing, and appropriate assistive-technology testing.

For complete HTML conformance checking, the Nu HTML Checker is a useful separate resource. RxHTML does not currently run that engine.