RadRace 2007 Wrap-up

RadRace 2007 Wrap-up

OutSystems has entered a rapid application development competition in the Netherlands www.radrace.nl. This was the first time we have entered a competition of this kind and our main objectives were:

- To understand if this type of competitions match real enterprise projects;
- To benchmark OutSystems Platform and the OutSystems Agile Methodology against other alternatives (This is often requested by those evaluating OutSystems);

Out of eleven contestants, we came out in fourth, with a practical tie with second and third positions.

Why haven’t we won RADRace and what can we learn from it?

Crossmarx, the clear winner, had a proprietary framework that had evolved from their previous experiences in this competition.

Compared to the teams placed in the first 3 positions, our team lacked the experience of being former runners in previous editions of RADRace. We have learned that winning this type of competition is much more than having a good development tool, development skills or methodology. It helps to understand what is valued by the Jury and to avoid “useless” effort on differentiators that we typically show in our demos and are not rated – only the tests broth by the jury were covered and leaving a module at 80% of completion means getting only 50% of the points.

OutSystems came before all Java and .NET and code generators alternatives and was shown as the perfect example that even in these competitions it is possible to end up with solutions that have excellent usability. OutSystems was also in the first position in terms of platforms that could deploy to both .NET and Java servers.

It is also relevant to notice that the average productivity of all the teams was considered twice a normal devevlopment team, in terms of number of function points built – tough competition.

So, do this RAD competitions match real project problems?

No, unfortunately they don’t. In real projects several “ilities” come into play: usab-ility, scalab-ility, maintainab-ility and reliab-ility concerns that are not addressed in these competitions. Integration, for instance, that tends to be the part of the project with the highest risk, wasn’t covered at all in any exercise. There was also no focus on performance and security, often crucial in enterprise applications.

Another difference of these competitions is that all requirements are written down. In real projects requirements are often fuzzy, and methodologies like the OutSystems Agile Methodology offer greater advantages. Of course not all things were clear (the problem was a complex photography site, we must confess we’re point-and-shoot kind of guys and don’t understand a lot about focal ranges and the like) and here an axiom of the Agile methodology stood out - you need clear communication with the business users - and that was not always the case as most clarifications were made in Dutch.

How does OutSystems stack up with other alternatives?

In an OutSystems project, the first demo to clients tries to show a broad range of functionality so that we can gather more feedback and, as a result of this, OutSystems was the only team that covered all exercises. Unfortunately this proved to be the wrong strategy for this contest because of the rating system. The effort spent by our team to create a robust application according to best practices was also not factored in. Next time we’ll know better how to hit this target :).

At the end of the second day, while OutSystems ended up with a built-to-change solution that could grow into a real project, others finished with quick-and-dirty throw-away prototypes. The platform built-in performance optimizations and out-of-the-box horizontal scalability meant that the OutSystems solution was the only one that could be deployed and used in a highly transactional website without any re-architecture.

What we would really like is to enter a competition where complex enterprise applications are compared against each other on how quickly they can be modified without risk. This being said, of course we would love to get the first place in this one also, and next year we’ll surely try! From what we have learned, it will not be hard to have a strategy next time that will guarantee us that lead. We’ll keep you posted.