Comments on: Semantic “SEO”: You’ve Had It Wrong All Along

By: h3sean

h3sean — Fri, 02 May 2014 01:50:00 +0000

I would have to deal with the design first before I comment on the meat of this entry (since I’m sure Brent knows his stuff on SEO).

I find the font too small and ugly – both the headings and the paragraph fonts. It would make it much more pleasant to read if it was tweaked in such a way that it would be very reader-friendly.

By: Scott Fillman

Scott Fillman — Thu, 01 May 2014 15:05:00 +0000

The reasoning being built into the resource description framework will use advanced ontological logic to identify data object properties and relationships, and attempt to understand the nature of an entity. The goal is to eventually be able to aggregate properties from and analyze relationships between different entities in order produce results that are more accurate. The goal of RDF is to be syntax agnostic in identifying resources, so that schematic data and semantic markup can provide structural details about resources and granular metadata to describe the nature of the data contained in the structured markup. Reasoning languages such as OWL or SWRL can use the description of the nature of the data to determine what the entity is and index it accordingly via SPARQL. Theoretically, this would reduce the occurrence of irrelevant results. This is undoubtedly the goal of search providers and the future of search.

By: Kristine S

Kristine S — Thu, 01 May 2014 10:24:00 +0000

Also to add right now it is very poor at search for any search that is not easily translated into a thing.

A thing can be an object, person or place. An info nugget of data, however like today search for WP-Property Importer (an translatable thing) pulls up many results yet a search for WP-Property High Rise Importer (not translatable as a thing in the current algo) you get a myriad of results surrounding Import – only a very few that have WP-Property in the results at all. Where the old algorithm would have used signals to determine what it was I was searching, now it tries to translate it to a thing, however if that thing does not exist it just throws back a kitchen sink or irrelevant results.

The article has a lot of good points. Just have differing views on some of the finer points

By: Dixon

Dixon — Thu, 01 May 2014 07:00:00 +0000

That’s a long post if a morning.

I think I agree with most of that, except that most web entities won’t have schematic data. Does not make that object bad… It means you need to take signal from places that do have them and flow those signals to nearby objects. Links still provide that conduit and nothing else will do.