What module(s) are you working on?
pnCommerce - online shop
pnunit/"Icarus Syndrome" - unit tester
pnGalaxia - workflow engine (new for fall/winter)
Where do you live?
Munich, Germany - The world capital of beer.
What is your real-life job?
What is the real-life? I took the red pill and I entered the wrong door. No seriously, I make my money with the development of websites and dynamic web apps. together with Klaus Lamberty from mediagourmet.de. Nothing very uncommon ;-)
Tell me about your postnuke "career".
A 24 hrs deadline for evaluation & implementation of a website and Postnuke could did it; this was Spring 2002. The Rest is History until now, a good history as far as i can see.
I prefer it because it worked that day and it still does. Its documented ;-) Now were using it for most customer projects.
When did you start working on your own module?
Spring 2002, developing of a mass customization application for http://www.eckerl.de to make website customer able to configure individual sunblinds. The interesting thing is that pncommerce takes a direction where such a effort can be integrated in the shop.
What is your development like? Do other people help you? How do you work together? How big is the impact of the community on your development?
Depends, some things are underdeveloped ideas, others like pncommerce are more real modules.
To take pncommerce as a example its just the direct combination of contract work and OS development. And i see my self here in the position to help people a bit instead of people helping me. So my position in the quality assurance is a nice choice The unit tester is a direct result of this work and will be used soon. more infos see the pncommerce interview from the pncommerce team
I was unsatisfied with PostCalendar beginning of this year and i started using the CalZone module . I found the authors site Closed and got the files in the end from a Italian webpage (thanks so far). I left it unpublished because there seems to be noone interested. Additionally I don't see much sense in costing people more time because they have to test 2 or 3 Calendars to find the one with less bugs. Through the fact that we already refactored it for Smarty it was a nice learning by doing example, but the point that PostCalendar has most of my features (but better) keeps me away from releasing that code.
Sebastian told me that CalZone is not really ready for release, but that everyone who needs it could get it via email. It can be see at www.keller-club.info
Just One year Ago i heard the first time of extreme programming and test first development. The chance to get this technique into postnuke was fascinating and seemed to be a chance to reduce debugging times. The nice module API design from postnuke seemed to fit into that concept and i took a first try with PHPUnit from the pear repository.
A second try (called "The Icarus Syndrome") is close to be finished and published, now using the simpletest classes and a selfmade pnAPIwrapper object. The work which is still left is just to write a bunch of tests for the postnuke core distribution, so that a practical approach can be taken. Additionally I am using just a subset of the functions av. because my own knowledge is limited here. But test first design is addictive.
As more I started hacking modules the topic workflow got into the spotlight of requirements. The WFMC documents
describing a workflow-engine gave me a direction to think, the implementation into tikiwiki Groupware as Galaxia Workflowengine is the source of code. Its still aanalytical work, but with the deeper look into smarty which is used by tikiwiki to, makes a integration possible. One person joined me here. I think when the fall & winter will bring more results than screenshots ;). The Galaxia concept fits into postnuke which has all needed elements, and waits to be ported. Anybody who likes to help is welcome here. As soon as a working candidate can be presented it can be used by the Postnuke Team for evaluation.
The other (dark?) side:
I am using Postnuke for most customer Projects (we all must eat) but I try to publish the small and bigger working results of my daily jobs. The goal is a double usage of the work.
The Project complexity grows steady, so i am happy to do this projects with a group of people.
What is the biggest difficulty in your development? And why? Is it a Postnuke inherent problem?
My Own Bugs. Nothing is as unlimited as my own st.... G***: has nothing tondo with pn.
Missing Smarty implementation.
Really we have no direction (@pncommerce) what will be implemented how and why into postnuke. As I have seen in the pncommerce developer meeting. I think a simplification of the the application design could be achieved with Smarty. But as more people experiment with this technique, as more different implementations are done.
An important way to control the program flow. But i don't feel that registering a session var is a good way to cope with this problems. Finish the announced pn API Functions, please ASAP ;)
Project external Communication & Guidance
Postnuke is on the border to a next level with .8. Much things could be done. But if Postnuke expects people to invest more time in more serious projects, id request more guidance. Timelines would be one thing id appreciate very much. Even the Xanthia port is undercommunicated. I know that management is a bad word in OS Projects: but 120 people in the
project need more guidance and people who do communications for them. The criticism on personalities helps the product in no way. We all must understand that. ;)
What features should the Postnuke .8 core have to simplify your work?
1. Exceptions, again
2. Postnuke wrapper Object for easier integration of OO Frameworks
Which route will Postnuke in your opinion go in the future?
I don't know, I'll follow the path as long as Postnuke isn't a limitation to my work. The way to a "Product" with a promoted name could be a good way.
Anything else you always wanted to say about Postnuke/your module?
1. Where is the Postnuke Europe developer meeting ?
2. Work together (!!!) please, accept that other peoples are eventually have other cultures because they simply live in other timezones and parts of the world. Integrate & communicate ! ;)
3. Love and Light ;-)