I worked on Twerq roughly three years ago now and it has been one of the biggest projects I’ve undertaken to date. At its core Twerq enables users to very easily search and find multiple items all in one go. It was one of the first services to “mashup” content, pulling together videos, images and text to help the user to find all the information they could possibly need on a specified search term. In addition to regular searches it also parses rss feeds on some of the worlds most popular news sources such as the BBC, Reuters and CNN.

It features repeat searching, bulk searching, mashups, rss feeds. For me the biggest challenge in this project was creating it so it was both scalable and as feature packed as it could possibly be. Search engine API’s were used a lot in the construction of the project, Google, Yahoo etc. For those items that did not feature API’s I had to use a technique called page scraping or “data mining” to retrieve the desired information for processing. To make this process even more efficienct I implemented caching to reduce bandwidth and overall load on the server.

Unlike FetchMP3, a more recent project, Twerq was not put through a formal closed beta test phase, instead we thoroughly tested at each stage of development to ensure that all was functioning as should be before a public beta.

This website required a combination of skills including:

• XHTML/CSS (WC3 valid)
• PHP/MySQL
• Additional Server Module Installation for PHP and Apache etc
• Javascript/AJAX (jQuery library)

Twerq has so many features it would be impossible to explain them all here so check it out at http://www.twerq.com.

Related posts:

1. FetchMP3.com
2. About Me
3. City Coupons

My latest and greatest project is a culmination of many many hours of both research and programming. Originally launched at the beginning of December, 2008, it has recently reached version 3. It includes the use of ajax, caching, php, mysql and intergration with external scripts such as ffmpeg and custom bash scripts.

FetchMP3.com in a nutshell is an audio ripping website. It has the ability to perform searches on some of the most popular video sites in the world, download a video that the user selects and extract the audio. This is especially useful for listening to speeches, audiobooks etc on iPods. Of course this can already be done using offline desktop tools but the process can often be arduous and time-consuming. FetchMP3.com has the ability to allow users to search a video site through FetchMP3′s built-in search engine, select a video and within a minute have it converted into a fully functional mp3 file.

The client in this project was a company called PapaFace, through constant contact with the client a service has been created that meets above and beyond the initial specification.

I have learnt a lot during the course of this project, I have developed good understanding of bash scripts, maintaining and installing additional *NIX modules and of course troubleshooting.

To sum it up I have used the following skills in the creation of FetchMP3:

• PHP/MySQL
• Javascript/Ajax (in particular the jQuery library)
• Bash scripting
• Script configuration/installation (ffmpeg, json, mcurl)
• Script customisation (mcurl)
• Server administration (installation of various modules, modifications to system configuration)

Some thumbnails are below:

Related posts:

1. Twerq
2. Bash, Bad Interpreter
3. About Me