I think that's a decision to be based off of what it is. If that's a concern, and if open source has limited inherent value to you or the project, perhaps it shouldn't be.
For me, i highly doubt my currently projects will make any money. Primarily home automation, and home cloud-lite stuff, Storage, etc. Nothing ground breaking, and where it might be "ground breaking" (though it's not), i specifically strive to not have it be ground breaking - a pillar of my work lately has been simplicity, ease of use and ease of maintenance - being side projects and all, i don't want anything i write to confuse me in two years.
So yea, i have no concerns about it being open source - and so i default to OSS. If it gains any adoption, i get contributors, testers, UX requesters, etc, all things i want. So OSS is purely a boon for me, in these cases.
edit: Plus, i like being able to add a trail of work on my resume/site/etc that i'm proud of. For few side projects that i have left closed source, i split off as much code as i could making them into OSS libraries and such. Even if the core of it can't be OSS, some of it can i'm sure :)