Computing has a great advantage over other sciences. This advantage is defined by its level of accessibility. The level of accessibility is defined by the cost of the tools required to experiment and the minimum amount of knowledge needed before you can even try to be taken seriously. Computers are readily accessible, they can be bought for a relativley low cost at your local electronics store. Once installed you can download, buy or write software fairly easily. Try doing this with any science and the first sticking point is where do you buy the stuff? You also need to get a fair amount of maths under your belt and then you can start learning about the subject. You can't get a job in a science field without at least a Masters degree.
Since you have to learn so much before you can do anything in most science fields, it has the effect of scaring many people off (the attitude of many people in the field doesn't alway help either). I feel that this has the effect of reducing the amount of unorthodox ideas that can enter the field, and thus slowing down the rate of progress. Because IT has the advantage of it accessibility, almost anyone get take their crazy idea and try to make it work. Try doing the same thing in sciences.
Sure you can see knowledge as an upside down pyramid, getting built up one brick at a time. What happens when there are bricks missing at given levels because orthodox thinking makes it difficicult to see what is missing. This is where I like to use my description of a pink elephant in multi-faceted sphere. Conventional thinking will look at the elephant through one of the facets and see just a trunk or just a leg. Without the ability to look at the elephant from other angles, then it is easy to conclude that what you see is a trunk. You could also just see two points of view, a trunk and a leg and not understand how they are related. You can then have people thinking that there is nothing is no link between the two. Of course there will always be the small group that believes that there is more than meets the eye, but they tend to be shuned because no one can see what they are getting at. Like brute force computing where you have many computers trying different combinations, science could have a huge advantage if more people were able to participate. Sure there the possibility of a lot of noise, but like any forum, you find ways to deal with this, for you realise that sometimes it there is an advantage in numbers.
What could be done to make science more accessible I am not sure. Maybe one method is to take advantage of the accesibility of computers and make advanced science applications easy enough for any fools to use and understand.