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.