WordPress 3.0 Ends the 'It's Just a Blog' Argument

For four years our platform advice has had a reliable fault line. WordPress for content-centric publishing, Drupal for structured complexity. When a project’s content model outgrew ‘posts and pages,’ the conversation moved platforms.

WordPress 3.0, released June 17 as ‘Thelonious,’ redraws that line. After a month of building against it, we are ready to say how much. The headline features are custom post types, custom taxonomies, and the WordPress MU merge that brings multisite into core. They aim precisely at the structured-content territory WordPress used to cede. They land well.

The ‘it’s just a blog’ argument, already wobbling since 2.7’s admin rebuild, is over. What replaces it is a more interesting conversation about fit, and this post is our updated map.

Custom post types: content modeling arrives

The feature that matters most, translated: WordPress content is no longer only posts and pages. A developer can now declare new first-class content types. Program, Person, Location. Case Study, Course. Each gets its own fields, admin screens, archives, and templates.

This is the modeling capability that made our Drupal picks write themselves for institutional sites. The university’s programs cross-referenced with faculty. The firm’s practice areas linked to attorneys and insights. The manufacturer’s products tied to specs and support content. In 3.0, a practiced builder makes that structure in WordPress, with the editorial experience content teams already love layered on top.

Custom taxonomies complete the picture. These are labeling schemes beyond categories and tags, like Department, Region, and Audience. They attach to any content type, can be queried in templates, and are managed by editors.

Field notes from our first month, because capability and craft are different claims. The machinery is registration-based. Types are declared in code, which we consider a feature. Content models belong in version control, not in a clickable interface someone can rearrange on a Tuesday. But it means this is implementer territory, not site-owner DIY.

The field-management tools are maturing fast around the new APIs, and conventions are still settling. Expect the next year of plugins to sand the rough edges. Build with restraint. Model what the institution actually has. Resist the type-for-everything enthusiasm. Do that, and the results are already production-worthy. We have shipped one. More are scoped.

Multisite: one platform, many sites, actual governance

The MU merge deserves equal billing for a specific audience: organizations running families of sites. The university’s departments. The health system’s service lines. The association’s chapters.

WordPress MU, the multi-site fork powering WordPress.com, is now simply WordPress. It switches into a network mode where one install serves many sites. Shared codebase. Shared upgrades. Per-site content and permissions. Central plugin and theme governance.

The governance half is the strategic half. The alternative is a dozen independent installs spreading across departments. That is precisely the unpatched, unowned sprawl our security postmortems keep finding. A governed network turns that mess into one upgrade, one security posture, one approved theme set, with departmental freedom inside the rails.

The caveats are real. Running a network is its own discipline. Plugin compatibility in network mode wants testing. A misbehaving plugin now misbehaves everywhere. But for the multi-site institutions we serve, this consolidation path is the release’s sleeper value.

The rest of the release, briefly

Custom menus give editors drag-and-drop navigation management. Overdue, welcome, and a small governance win, since navigation changes stop requiring template edits.

Twenty Ten, the new default theme, is a respectable standards-based foundation and a statement of intent. The default now demonstrates current practice rather than embarrassing it. And the release ships the usual pile of civilities, lighter admin refinements and bulk updates, that the platform’s pace keeps delivering. Three releases a year, each absorbable. The rhythm argument from our 2.3 coverage keeps compounding.

Upgrade and adoption notes

For sites already on WordPress, the 3.0 upgrade itself has been routine in our testing. The usual ritual applies: backup, staging, plugin verification. Anyone running the old WordPress MU should read the merge path carefully rather than optimistically.

The subtler advice concerns adopting the new features on existing sites. Custom post types are most valuable when the content model is designed around them. Bolting types onto a site whose content has lived in shoehorned posts for years is a real, if worthwhile, migration. Content gets audited, re-sorted, redirected. The works.

Our sequencing advice: adopt types at redesign time, when the content model is on the table anyway, rather than as a standalone retrofit. The exception is any site actively straining against the posts-and-pages ceiling today, where the strain itself is the business case.

And a naming note for implementers that will save future grief. Name types and taxonomies for the institution’s language, not the developer’s. Editors live in these screens, and ‘Programs’ beats ‘custom_content_v2’ for the decade the site will run. Small decisions, long shadows. The usual physics.

The redrawn line, stated honestly

So where does this leave the platform conversation? Our updated shorthand: the middle of the market just became genuinely contested, and that is excellent news for buyers.

WordPress 3.0 credibly serves structured institutional sites that would have been automatic Drupal picks a year ago. It brings a friendlier editorial experience and a gentler build curve.

Drupal keeps clear title where complexity deepens. Intricate content relationships and views. Fine-grained permission models. Workflow machinery. And the multilingual architecture our internationalization series has covered. Drupal 6’s language system still has no WordPress equal worth the name. Multilingual WordPress still means plugin assemblies we would scope cautiously. Community-scale features, likewise Drupal.

The evaluation method is unchanged. Describe your content as types and relationships on one page, then let the description choose. But the WordPress column of the worksheet just grew a lot, and several evaluations we would have called quickly last year now run the full exercise. Competition among good options remains the best buying environment there is. The options just got more competitive.

One prediction to close, since round-number releases invite them. Within two years, ‘WordPress or Drupal?’ becomes the default frame of most mid-market CMS evaluations in our sectors, with the commercial mid-tier squeezed between them. The economics point that way. After 3.0, so does the capability map. Filed, dated, checkable.