Software projects have a very ugly tendency to be delayed – most often very delayed. Microsoft is delayed with Vista and even the ultra-lean “getting real” phenomenon, 37Signals, seems delayed with their new CRM’ish software solution, Sunrise.

What I’m trying to do is of course to try and build a good defense – before publicly admitting that we have made some huge mistakes and basically done exactly what we hoped never to do; be un-realistic (fools).

We promised to release Headlight™ (the software formerly known as Project Drangonflux) early 2006. It’s now becoming mid/late 2006. We know that many companies are desperately waiting – and have done so for over a year.

First I want to say that we are very sorry – in more than one way. We’re very sorry to disappoint our customers and also sorry to disappoint ourselves; It’s mentally very hard constantly pushing self-defined deadlines … and obviously delays also mean lost sales – so it’s something we take deadly serious.

To be or not to be – it’s not the question, actually. It will be, that I promise. The question is when and perhaps also what the hell happened!?

Why is Headlight delayed for more than 6 months?
Right now I don’t really know. The main focus is getting the job done. But one thing which stands clear is that working with new and fairly un-tested technology – Flex 2.0 / AS3.0 / AJAX / .NET 2.0 / Analysis Services 2005 and trying to integrate with Beta APIs (e.g. Google’s API for Adwords) have some pretty steep learning curves.

With a fully loaded suitcase of experience from designing and developing StatLynx v1.0 / v2.0 / v3.0 we didn’t think for minute that our launch would be delayed. We simply underestimated everything, more or less.

When is Headlight ready for its first public beta?
Honestly, I can’t say for sure – and even if I thought I knew exactly spot-on “when” then I would probably hesitate announcing it (nor would I never ever again print in on a t-shirt and send to a 1000 of our customers for Christmas). It can be anything from a few weeks to a few months. If we get lucky and unscramble some of our current issues it will be ready pretty soon (BTW > We need a DB-VLDB Developer).

We receive about 10 new beta requests every day and a lot of mails from existing StatLynx customers. We are as eager as anybody about launching – but we have a strict protocol – in contrast to our commucation strategy – not to release anything before our core technology is 100% stabile. It simply isn’t yet and we will never sacrifice performance / security for the purpose of reaching a deadline. We are trying to do the responsible thing here… We focus on product quality and on delivering a great out-of-box experience. Maybe in the past we would have just gone ahead but now we’re not going to do that.

Interested in working on probably one of the most demanding database-centric web applications around?

Check it out (in danish): http://www.traceworks.dk/df_careerOps.asp#dba

Take a look at this new job opening: http://www.traceworks.dk/df_careerOps.asp (in danish)

So if you know anybody that knows somebody that… ask them to stop by.

PS: more openings to come – Europe, Sweden – soon

When you work with charts and graphs, something like this immediately grab your attention. It’s not a particularly useful discovery, but sort of fun when you are a CSS geek. It’s a java applet that analyze the content and links of your site and show a series of connected dots colour coded for which HTML tags it encounters. blue: for links, red: for tables, green: for the DIV tag, violet: for images, etc.

This is how the Dojo looks. Pretty ain’t it? You can try it here