One problem with charging is not the size of the fee, but the complexity in working out the license, who pays and how much. Right now as a developer in a medium sized software house (several hundred people) I'm free to download and use open source packages without any consultation. This is no longer the case once there is a pay license. Even if it's only a cent per 100 installs, not only do I have to try and understand the license, I have to get a lot of other people to do the same (many of them who won't even know what XSLT is). Even if I decided to fight the cost approval fight, and even if I won it, I don't know how we'd work out the numbers, there's absolutely no way anyone could predict how many machines it gets installed on even within the company during a dev/testing cycle. So open source generally wins out, even it's it's inferior, buggy and slow. Even if they have to spend *more* money in the end writing their own code, the company will tend do it, because the costs are easier to predict and track, and the costing infrastructure is set up to handle this approach. This isn't a criticism of the proposal, just an (unhelpful) observation that simplicity is a big reason that open source works, as well as zero cost. >There are lots of risks associated with this strategy - one of them being that users will simply stick with Saxon 9.1. Well, actually, that's not really a risk - it's not a problem to Saxonica if they do I think that's a loss. A large user base which sticks close to latest releases is a good way to validate the product.