WordPress Search Engine Optimization

I asked MA.TT a question – of course it’s complicated, but it definitely has something to do with the three-letter-acronym (TLA) called SEO:

As I indicated in my question, this has nothing to do with Google and/or similar proprietary (“secret”) algorithms. I was talking about WordPress as a search engine.

IMO Matt’s response / answer was excellent. It underscores how challenging such a project might be. Yet Matt also emphasized (during the interview with Om) that text will continue to be of central importance to the way the World-Wide Web works (and will continue to work). As an addendum to Matt’s insight, let me note that any kind of artificial intelligence initiatives will always be a matter of pattern recognition, and the patterns that they attempt to recognize will always be text, a written representation of something (e.g. spoken language), etc.

Since my time for asking such a question was limited, I couldn’t go into many details. Even here, I do not want to bore you with the pros and cons of hitchhiking along on other people’s websites. As I have written in previous posts, the foundations for quite satisfactory information retrieval (aka “search”) are already present on the so-called (WP) PLATFORM.

At this point, I would like to try to elucidate at least one part of what I was talking about when I mentioned that an information-retrieval system need not necessarily be a one-size fits-all solution. People who have been following my writing for a long time know that this is a “pet peave” of mine, which I mention time and again. Note also that during Om Malik’s interview with Matt Mullenweg, the two also discussed McDonalds (TM). I find the use of brand names such as McDonalds or Google or Tupperware or whatever intriguing – is this a good or a bad thing? IDK… – but I digress.

The point I want to underscore especially emphatically here is that WordPress isn’t simply a “software” or an “app”. It is also a community which also involves a lot of people who have a great deal of natural intelligence. There are many points of view. There are many approaches. And just as there are many roads to Rome, so too there are many possible solutions to information retrieval (“search”) technology which are possible, viable, etc.

Let me give a couple examples beyond the ones I mentioned when I asked Matt the question in Paris (at the 2017 WordCamp Europe conference). The examples I mentioned of how “every website is a search engine” both came from the wordpress.com website. There are certainly many more examples possible from this website, but I chose to highlight discover.wordpress.com and wordpress.com/tags as premier examples. In my opinion, these two search options show two different kinds / levels of community engagement.

Likewise, there are also many search capabilities available to self-hosted wordpress sites. The most straightforward of these is undoubtedly the search widget (a “search box” in which the search text is entered and then searched). This is a very simple algorithm, and it is primarily useful for “known item” searches – for example, if you already know the title of a post or a string of words inside the post content. Normally, this search box does not search many other fields, such as the the tag field or comments. However, a site’s registered users can search these fields via the “backend” portion of the WordPress software. In this sense, each site has more search functions (and functionality) available to registered members… and therefore also offers higher levels of capabilities to more engaged users.

Note also that such different levels of capabilities are also part and parcel of the distinction between tags and categories (with respect to the primary function of information storage – which is the basis for later information retrieval capabilities): while in a standard WordPress implementation many users are able to create tags for posts, only a rather limited set of users are capable of creating categories for posts. Of course expert site administrators can configure such settings (which may be good, but having reliable standards also makes learning how to use WordPress easier for new users).

In my opinion, community engagement is also probably also the crux of WordPress search engine optimization across sites. For example, it might be important to distinguish between different meanings for the same string – e.g. “development” might mean very different things in different settings / contexts (such as with respect to software, economics and psychology). I think it might make a lot of sense for WordPress to provide support for the development (no pun intended 😉 ) of sub-communities within the greater WordPress community – and thereby to enable people to share, exchange and review each other’s ideas, to set these ideas into their appropriate contexts, etc. Indeed, there is a long tradition of abstracting and indexing in scientific literature – and learning from decades (if not even centuries) of experience and insights might be a very good thing to do.

One recurring theme I heard repeated throughout the WordCamp Europe conference in Paris time and again was the notion of how WordPress functions as a community of engaged people. This aspect of community engagement is definitely a very strong advantage of WordPress with respect to search engine optimization. Another is the very strong foundation of “open source” ethics – which I also described in my previous post.