This is just an example of some of the kinds of things you can do with semantic formats, specifically XML in this case (because XML is the least time consuming for me to make examples with). These exampleas are EXTREMLY superficial as all I had to work with was javascript and flat files. I do not recommend you use javascript for this, or you should at least take advantage of the available libraries, which I did not. The script that runs this page is openly available for you to look at if you want to be certain I didn't 'cheat': magic.js
The Jaccard index is an index of similarity used to compare the species present at two different locations. The wikipedia description of the calculation can be found here. Two example XML files, speciesList_ex.xml and speciesList_ex2.xml, will be used to calculate the index. Both files use the speciesList.xsd schema. This is meant to simulate data integration from two different servers. Each server would have species list files for various locations.
This portion of the example simply reads in an XML file that implements the Phylogeny Schema and displays it as a tree. The "Current As Of: " section is also passed in the XML and would indicate when the phylogeny was read from the server (assuming the server is kept current). This example would be able to handle trees of arbitrary size, but I can't get the CSS quite right. The Phylogeny Example XML file, which is what this script reads in, is also available for you to look at. TOLweb also outputs data in a format similar to this, but you can't use XMLHttpRequest to make calls from different servers (so, obviously, you won't be using this for Semantic Web applications that involves pulling data from more than one server). The TOLweb format is described here.
I'm far from an expert on the subject of the Semantic Web, but I believe in it's underlying principles. Being able to write programs that take advantage of the vast stores of information available online and in 'deep web' databases would be a giant leap forward in efficieny. As an ecology major, I'd love to see my field benefit from this revolution. I could easily see programs that perform phylogenetically independent contrasts by pulling data (body temperature or femur length, for example) about a specific taxon from databases around the world and then using a phylogeny pulled from, say, TOLweb to compute and plot the contrasts. Any data that has a standard method of collection or storage, from point-quarter data to the huge amount of stored by the INSDC (which actually CAN be passed out in XML), could be passed out in semantic formats and combined with data from other sources. In particular, biological collections in museums and college biology departments around the world would be even more useful if certain measurements were available to federated queries. My ideas are definitely aren't the best around, but just getting people talking about this is sure to stir up other people with better ideas. My clumsy javascript and flat files could be replaced with a world wide network of computers passing out semantically formated biological data. On this same note, I'd like to be clear that these examples are VERY bias to what I'm interested in. There are many, many applications that I haven't touched on or even thought about. The point is to get people thinking and to provide resources, some of which are available on the next page, that make implementing these powerful technologies easier.
<< Conclusions Resources >>
'The Semantic Web' by Conrad Williams is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
E-mail the Author