Search. Enter words or quoted strings into the Search/Inspect box, and Ontaria will find subjects which are described using that text, or otherwise closely related to it. (Current bug: only the first word is used, and the search does not notice word boundaries, so a search for "tim" will match "time".) See Help : Search-Inspect.
Inspect. Enter a URI, URI-Reference, QName (with a known namespace prefix), or a known namespace prefix (with trailing colon), to display information about the identified thing. See Help : Search-Inspect.
Browse. Lots and lots of links, making it easy to wander around to related information, while staying on site.
Multiple Views. Multiple views may be possible for a given subject. Currently they are listed on tabs. Clicking on a tab brings it to the front (selects that view); the stacking order is preserved during navigation by being encoded in the site URI.
Pre-Loaded Data. Every few days a set of RDF sites are harvested, and the data is gathered into a starting set. Users may request new sites be added.
Real-Time Harvesting. New URIs encountered in search text should be immediately tried on the web, to gather more information. URIs in returned material should also be followed, if they meet certain weighted-distance limits. Harvesting should obey robots.txt and use HTTP caching, per-host serialization, and conditional retrieval so it minimally impacts harvested sites.
Ontology Views.When viewing classes and properties, special tabs should be available which make it easier to conceptualize that part of the ontology.
Validation. Content should be examined for syntactic and semantic errors, which can shown to users.
Provenance. All information should be labeled and linked to the source of the information.
Inference. At least RDFS and partial OWL inference should be performed. Provenance on inferences should be explanations/proofs.
Optional QName Labels. There should be a toggle to control whether things are named with their rdfs:label (eg "Type") or their qname (eg "rdf:type").
More Front-Page Boxes. "Ten Random Subjects" is more a quick hack than a useful tool. At least "Ten Random Ontologies", and maybe "Most Recently Changed", "Most Widely Used", etc.
Shopping Cart. A way for users to indicate which items are of special interest to them, perhaps for aggregation and republication as a new ontology.
On-Site Authoring. Fix typos immediately, when you see them. Or even add new terms, descriptions, etc.
Trust Profiles. There should be different sets of rules saying how much to trust each source; users should be able to create them and select which ones they want to use.
More Metadata. Using an ontology about projects and the social uses of ontologies and, the site should display eg maturity and funding source.
Category-Based Ontology Browser. Like Yahoo! or the open directory project, for ontologies. Perhaps a front-page box like dmoz suggests (scroll down).
See also the often out-of-date SemWalker TODO list.