It sounds from the explanation that was provided that the free users for your product contribute not only to customer acquisition but also create value through their contributions to the site. In this case, free may make sense for your software, as only you can weigh your infrastructure and support costs vs. the gain in marketing and usefulness of the product.
But your experience may not be common to many other businesses. If free users aren't contributing to the product other than marketing, it would come down to a decision of whether or not that goodwill was worth the added cost. It can be hard to measure that cost, but perhaps Twitter Analytics and web referrers can provide an idea.
The main point I was probably trying to make with regard to our particular model is that it is a viable alternative to offering a free plan as a trial version. If you're using "free" as a "trial", you're doing it wrong. There are plenty of opportunities to utilize a freemium model without resorting to it being Shareware. Ours is but one example.
Chargify is a great example, as are others that have startup/education plans. The point is that if you want to have free as an option, you should own it and promote it and make it work for you. Having a free option as a subtitle is simply a dumb idea. If you want to charge people, charge them. Don't be half-arsed about it. Having a free plan because you're not sure people will pay is the fastest way to find the unemployment queue.
Most importantly, if you have a free plan, there should be a REASON for having that free plan. Having a free plan is sensible when there is a natural progression to a paid plan (e.g. Chargify), there is benefit to your business by fostering a community (e.g. UsabilityHub) or there is potential to make use of micro-transactions (e.g. Xbox Live). The point is, that there are plenty of options to be explored by all businesses, not just those that leverage the input of the community like we do.
Yeah I agree, utilising free users in this way isn't common but that's precisely the point. Most startups that employ a freemium model see free users only as potential upgrades down the track. If those users instead had to contribute back to the system in some way in exchange for having a free service they then cease to become a burden and become an asset instead.
But your experience may not be common to many other businesses. If free users aren't contributing to the product other than marketing, it would come down to a decision of whether or not that goodwill was worth the added cost. It can be hard to measure that cost, but perhaps Twitter Analytics and web referrers can provide an idea.