The actual vision to create a flexible, XSLT and Java based web framework and CMS started out in early 2002, when Thomas quit his job as technical director at Orange Interactive. His former employer was building a proprietary CMS system which has already targeted the same goals but had substantial design flaws and was generally over-engineered.
In autumn 2004 :torweg has won the pitch to create a new media management website for Pelikan, a German manufacturer of stationary products. A framework was still needed and nothing on the market seemed appealing. So we took the chance to build a limited framework prototyping many of the ideas we already had in mind.
When the last localisations of the media database had been in place in early 2005, we started to think about kicking off the real project and pick a name: pulse. A local CVS server was set up and we started to build the new framework from scratch.
In the first year, many changes were made, technologies adopted and frameworks exchanged.
In 2009 we started to become more serious about publicising the project and finally made a first official release (pulse 0.4) on March, 19th. A few days later we graduated from the enterprise-incubator (namely on March, 25th).
Development continued swiftly and in August pulse 0.6 beta saw the light of day, followed by three maintenance releases all to be published until October 2009.
Also in October 2009, the first preview release of the new 0.7 series had been released with the downloads already hosted at Sourceforge. The 0.7 series streamlined the API while adding more and more functionality. This was also the time when we really started to do some PR activities other than being there, out in the open.
After many performance problems at dev.java.net, we finally decided to move the whole project over to Sourceforge in March 2010. We were quite baffled how much this simple move increased the interest in pulse. So we had probably been wasting a lot of time with the initial decision on where to publish the project.
With the first release candidate of pulse 0.7, we've officially reached a production/stable level. The project should now be easy enough to understand to be seriously used by other people than ourselves.
Currently we are already working on the next release - pulse 0.8 - which will further consolidate the API, remove some relicts from the past and present a new administration interface furthering our ideas to create a simple and intuitive to use web application. For those of you who are interested in out future plans: take a look at our development road map.