3 ms·
Actually, I think they could get traction, but for that to happen that would need to release an API to check if user "donated" to my blog/service/etc. Therefor
by rarestblog 16y ago
Actually, I think they could get traction, but for that to happen that would need to release an API to check if user "donated" to my blog/service/etc.
Therefore I can create a "premium" zone accessible for a smallest fee ever. That would create a need for user to create Flattr account. You pay to see some premium content. Also, you pay the same fee to get premium content on dozens, potentially, hundreds of sites.
Otherwise you're just hoping to sign up users to pay 10EUR(?) monthly as a kind of "permanent monthly donation". Even though it's not a lot of money, but still most people would avoid that: 1. it's a semi-permanent obligation 2. as practice shows - relying on people giving a donation for something available for free is a bad idea. (Let alone MONTHLY donation for same thing)
You need to provide something in exchange for their money. Give us the API to find out that user donated and we'll give him a reason to sign up with Flattr (premium content).
- fina 16y agoWhile seemingly similar that would be an entirely different concept. What flattr does is enabling small donations for things you already like. One of the nice things with flattr is that the users are already "sold" and have already payed the "fee". If they like what you've done they are going to flattr you, not thinking "is this really worth x". There are a lot of potentially nice things that differs from other revenue models, like making it harder for "link-bait". I can't really explain entire concept (as I see it) in a good way, you'll have to "get into it" yourself. You also have to keep in mind what the founders are coming from. I don't think it would be a very good fit for Peter Sunde to push for "paywalling" of the Internet with a subscription network. Finally, the monthly minimum amount is currently 2€.
- joeyh 16y agoI do wonder how many of the users currently on Flattr are there because they hope to manage to get enough Flattrs of their own to come out ahead. It would be interesting to run some stats: What percentage of Flattr users have created at least one Flattr button?