“The Semantic Web” is actually a fairly old (in Internets years) term that basically refers to computers and computer programs accessing Web sites, instead of human eyeballs.
The long and the short of it is that with the Semantic Web, machines can extract data piecemeal from other machines, then bend, fold, spindle and mutilate it into a form that can be subsequently presented to human eyeballs.
Newsfeed aggregators are probably some of the oldest and best-known examples of the Semantic Web. These are sites that reach out to other sites through protocols like RSS, and get data in a machine-readable format. They then reformat this data into human-readable form, and slap it all together before presenting it to you with a pink bow tied around it.
RSS is supported natively by many sites. It is usually indicated by an orange icon that looks like this:
This tells you that, if you have an RSS reader, or you are operating a site that can read RSS from other sites, that a semantic interface is available to the site.
The BMLT Gets Semantic
Now, the BMLT also has the ability to allow other sites or programs to read the data it produces. It doesn’t use RSS. Instead, it specifies a simple way to load a Web page that will return machine-readable data.
In particular, you can send a URL to a BMLT root server that will execute a tightly-focused custom search, and then return the results to you in the following Six formats:
- CSV (Comma-Separated Values)
- This is the simplest file type. You can download these files directly into spreadsheets like Microsoft Excel.
- The general format is that each line of the file contains the data for one meeting. The file will contain the results of the search, which could be any number of meetings.
- The NAWS Export is a CSV file, with a format dictated by NAWS.
- Also, it is easy to create “junk” JSON if you aren’t careful (but we are always careful, so our JSON is breathtakingly beautiful).
- The BMLT tends to use JSON for most of its interactions with the BMLT satellites.
- XML (eXtensible Markup Language)
- XML is a rock-solid data transport language. It tends to be more unwieldy than JSON (which is a big reason that many Web developers don’t really like it).
- The single biggest advantage that XML gives, is XML Schema. XML Schema is a “contract.” It spells out exactly what is and is not allowed and/or required of an XML document. If it’s in the schema, then it may be in the document that you receive. In some cases, it is guaranteed to be in it. In other cases, it may be optional. However, you’ll never get anything that isn’t in the schema, and the schema can also map out things like default and enumeration values.
- Google KML (Keyhole Markup Language)
- This is a language used by Google Earth. It is a geographic specialization of XML. You can drop a KML file directly into Google Earth, and immediately get a map full of markers.
- KML support tends to be mainly around Google services and applications, but those services make up the lion’s share of GPS applications on the Web.
- GPX (GPS eXchange Format)
- This is another XML-derived language that is supported a bit more widely than KML. Google apps will generally accept GPX data. Many GPS devices and software programs will accept GPX files.
- POI CSV (Point Of Interest Comma-Separated Values)
- This is a very simple CSV format that many GPS devices understand.
- It can be easily loaded into a spreadsheet.
- You can also easily create a custom Google map with a POI CSV file.
Basically, semantics give the BMLT truly awesome power. You can use these commands to extract pretty much any data that you can imagine (as long as the Root Server has it), in almost any format that you want. It’s a bit technical, but very cool.
In the pages under this one, we’ll explore these options more intimately.