The Birth of GrowSearch

When Paul and I started working on GrowSearch it began with a simple idea: Develop a site search and site map plug-in for web sites that was easy to install and style, easy to use, accessible and fast. So that's what we did.

There are a huge amount of site search tools out there. The problem is they are either too complicated to install, too complicated to style, don't return good results in a simple, meaningful way or add all sorts of branding and limitations to what should be a simple addition to a site. Even if you get a good tool, of which there are many, it still doesn't return a site map as part of the service so the designer still has to edit the site map everytime new content is added. Or not have one at all. What a pain.

What does it do?

GrowSearch does exactly what it says on the tin, and then a little bit more. All in a way that even my Mum, with her novice Internet skills, can do whilst thinking about her weekly shop. It searches the web site it's plugged in to and returns pages by search term density. Not only that, but while we were developing it we realised that if it was going to index the web site for the search results it might as well produce a site map at the same time. So we gave it that ability. One of the things we're most proud of is that it does all this on the fly. As soon as the content updates so does the site map and search results. Not only that but it's a breeze to install: One file returns both a search result or site map depending on what a user wants. Easy peasy for designers.

We needed to be able to ignore certain file types or directories so we gave it that ability too. Some information just isn't relevant for a content search. We wanted to display not only the file containing the relevant information, but the directory it lives in too. A contextual result, like breadrumbs so users could decide if that file in that location was probably going to give them what they needed. So we did that: Both the search results and the site map list files by directory. We needed to be able to style the output as well so users could see the difference between a directory and a page. It had to look good. So we made the output a simple XHTML list that anyone could style with basic CSS skills. We also added the ability to add any character like a forward slash after the directories in the results. If a web site was delivered with .php pages the search terms can be highlighted in the pages too with a simple class added to the stylesheet.

Our favourite bit

The best bit for us was making it cheap, easy and accessible. The search is powered by 3 files: The accessible XHTML form or include, the results page and the optional highlight include. The form and results page uses semantic mark-up which is accessible to all. The price is a pound and people can use it anywhere they like, as many times as they like. We invite anyone using it on commercial projects to donate a pound per time but its not mandatory; just a hope that if our fellow developers and designers like it, a pound might come our way from the project fees.

So there it is. The simplest, prettiest little plug-in that packs a mighty wollop for anyone who knows how useful site search and site map can be if done right. See thsi site's search and site map, or take a look at the demo and let us know what you think.

More articles