Yes, it's true that many users will choose the paid app, but as Thom explained there are also a lot of them who will choose the free alternative at all costs.
So, the idea behind this post is to find a way to get them to and try to earn some money from them (by conversions into full versions, in-app purchase or ads - this will be the next post tomorrow), instead of ignoring them. This is especially important for individual developers and small teams that do not have means to push the apps in other ways such as advertising.
Secondly, the store is already full of limited, crippled or badly designed apps (and some of them are paid for) so some of the users like to test before buying. Therefore, to earn money an app has to be good, and having a free alternative is just a support mechanism for those not blown away by your screenshots and description. The same goes for recommendations – if it’s good you’ll share it with your friends and when they see your recommendation on fb or tweeter, more people are likely to install it if it’s both free and good (since you recommend it) than if it’s only good (at least that’s how I see it:)
@thom I’ve talked with some of developers who brought the “marketing expense” issue up. Some of them were just starting their companies and without references they could not start working for clients. For them, money for that project was not the main aim – but it’s far better to say to a client that you have a free app which was downloaded many times, than just having a paid apps which is having average sales.