The P-language of search engine optimization
Posted by Michael Martinez on December 20, 2007 in Advanced SEO
Have you ever heard of P-Language? There is more than one and they really have nothing to do with each other. “P-Language” is a descriptor for a poosited language, a pseudo language, or a placeholder language in a family of languages. The languages may be programming languages, specification languages, or human languages.
You define a P-Language when you need a special language to describe or explain something. For example, suppose you’re analyzing a document in an ancient language and you find traces of two idioms (an idiom is a mode of expression, a way of talking — the familiar euphemisms and expressions you grew up with are your idiom).
You’ll call one idiom idiom A and the other idiom idiom B. When you have clearly documented the portions of your ancient document that can be attributed to either idiom, you have attributed those portions to “writer A” and “writer B”. Drawing upon other linguistic principles you conclude thath writer A is older than writer B and that writer A probably was writing in another language.
At this point, having shown an older language was used, you find you cannot identify the language. So for lack of a better name you call it a “P-Language” (or “A Language”, “C Language”, “Hypothetical Language 1273″, etc.). You begin to investigate the “P-Language” in the hope of understanding its grammar, syntax, vocabulary, and idiom.
Now, what if you’re not a linguist studying ancient documents? Maybe you’re developing a computer programming language, call it “Michael”, and you want to get it into production quickly. You’ve added all sorts of neat programming commands and structures to the language but explaining them in detail to a computer is tedious. So what if instead you design a second language that is much more simple but which can be used to describe your “Michael” language? You could call that second language a “P-Language”.
In the old days, computer scientists designed what are called “high level” or “human readable” programming languages like COBOL, BASIC, etc. and then wrote compilers that translated commands in those languages into machine-understandable language. But writing a good compiler takes a long time.
So someone eventually came up with the idea of creating an intermediate language that was pretty simple but which was still too complex for a computer to understand. They wrote compilers for the intermediate language pretty quickly however because it was so simple. From that point onward, more complex “human readable” programming languages could be translated into the simpler “P-Language”.
The programmers’ P-Language concept has been adapted in other technical fields, including various World Wide Web technologies. Complex document architectures can be described in terms of a “P-Language” that explains how aspects of Web documents are handled. A P-language gives Web technologists a common language to work with as they develop new concepts.
And there is no reason for why search engine optimization cannot be described in terms of a “P-language”. The idea of the P-language is to simplify what we are looking at in lowest common denominator terms. But the lowest common denominator is not required for human-to-human communication. Rather, it’s required for human-to-machine communication.
And that communication has to work with virtually any machine, not just a specific type of machine that has been designed to work with a specialized command set. Some of you may see where this is going. Anyone who has looked under the hood of a Web document has surely taken note of some basic HTML elements in widespread use:
- Title elements
- Keywords meta tags
- Description meta tags
- Language declaratives
- Robots meta tags
- A HREF elements
- IMG elements
- BOLD or STRONG tags
These and a few other markup commands constitute a significant part of the SEO P-Language that we use to communicate with search engines, browsers, and SEO tools. It’s not a very complicated P-Language at all and most competent SEOs have a pretty firm grasp of this subset of the SEO P-Language.
But there is another subset of the P-Language that most SEOs don’t understand very well. Those are the common query operators that function across the major search engines. Although there is no well-defined universal query language, there are several types of query structures that all the major search engines honor (they may return different results for various technical reasons).
The most commonly used query is the FIND ALL mode query. That is a query where you type in one or more terms (keywords) without any special punctuation or command operators. Term1 Term2 Term3.
Most people expect a FIND ALL mode query to favoir documents that place keywords in close proximity to one another but the search engines don’t actually do that. Instead they return documents that have the best intersections of relevance scores. In other words each document is scored for each term in the list. It occasionally happens that you get a document which is highly relevant for one term but lacking one of the others showing high in the search results.
You can modify the FIND ALL mode query by requiring every term in the list. To do so, you add a plus symbol ( + ) immediately before each term. Keep the spaces between the terms, however, or else you’ll be using a different type of query.
An EXACT FIND search is usually constructed by placing double quotes ( ” ) around your terms. You can also connect all your terms with plus symbols ( + ) (unlike above, you leave no spaces between the terms). The EXACT FIND mode query requires that the terms be found in the documents in the precise order entered. Punctuation in the document may be ignored.
You can combine FIND ALL and EXACT FIND mode queries to create a hybrid or complex query structure.
You can also impose filters on your query by using the negation operator — the hyphen ( - ) — in front of any term that you don’t want found in the documents. Negation can also be used in conjunction with query operators or directives. The most commonly known operator among SEOs is inanchor (ironically, most SEOs still don’t know how this works).
You use the inanchor query operator to list only documents that have the specified anchor text pointing to them. Most of the explanations of inanchor that I have found on the Web still get it wrong. If you specify allinanchor then everything after the operator has to be found in anchor text pointing to the document in order for the document to appear in search results.
Other query operators that SEOs have used include the valueless link (aka linkdomain on Yahoo!), title, and inurl but you’ll seldom see SEOs put these operators together to create new query functions.
For example, if you know a unique word or expression is found only in outbound link anchor text on a domain and you want to see if that word or expression occurs in a sitewide link, you could do a “site:example.com UNIQUE EXPRESSION” query but you could expand your query to all .com, .edu, .org, .info, .cc domains pretty easily.
On the other hand, just using the site operator doesn’t restrict your results to pages that have the expression in their on-page text. You can, however, use intext (inbody on Yahoo!) to filter out all pages that are associated with the expression only through inbound link anchor text.
You’ll get a much better idea of how many sites in Google’s index are linking to a domain this way than by looking at SEO Tools or by checking Yahoo!’s backlink report.
You can give yourself a competitive advantage by creating new query formats through experimentation, evaluating the results, and adjusting the query structures until you’re satisfied. Learning how to perform useful, informative queries helps you evaluate SEO tools better (and you’ll begin to see why I don’t trust most SEO tools).
The SEO Method (experiment, evaluate, adjust) helps us in many ways. It’s not just a mantra for creating Web pages and throwing them out to see what happens. It’s a fundamental principle of the discipline. Experiment with everything. Evaluate everything.
By compiling for yourself a list of HTML commands and query operations that you use in evaluating every Web site, you’ll begin to communicate through the P-Language of search engine optimization. You’ll start to see things at a new level. You’ll empower yourself in ways you hadn’t thought of before.
You speak to search engines in two ways: through your documents and through their interfaces. The combination of commands and directives that you have at your disposal gives you increased reach and flexibility.
Comment
Log in or Register to post a comment.