First of all I wish to thank Sebastian who gave me the invite to Flattr to begin with, a few months ago. This was not only helpful to me, but also allowed me to have a bit of insight on the whole evolution of the service over these.
Now, I haven’t really written much about it before, mostly because I didn’t want to stress its presence — if you read my blog through syndication on Planet Gentoo or Gentoo Universe, the fact that the posts have a flattr button is hidden from you altogether. But since there has been a lot going on in the past month, I thought it might have been the right moment to write down some notes about it.
You might or might not have heard the fact that Debian developers got into bickering about what is and is not allowed on Planet Debian, mostly about Flattr, with a number of very different ideas about why it is bad for their developers to have Flattr at all – ranging from calling it a scam for the 10% they get from the poured money, to calling the “social networking buttons” webbugs (including Feedburner), to simply saying that it’s a commercial use of the Debian infrastructure – whatever the outcome, I don’t think Flattr stands to gain many fans here.
But what I found interesting in that outcome is the consideration that the Flattr platform itself might not be faring too well; the reason why that’s said is that after the initial money inflow of the registration, just like the “real world economy”, the consumers pay, and the producers get the money. In my case, I haven’t really had the need to add any means since I registered, and while I feel bad about it, I don’t think that will change anytime soon (mostly, because I’m actually going through a patch rough enough that I cannot afford to do so realistically; even though I promised myself not to withdraw funds anytime soon at least, I ended up doing so to pay for the SSL certificate from StartSSL, as well — so thanks to all of you who contributed!).
There are also a few issues with the technical interface of the service: when I decided to implement support for flattr in my fsws (which I have opensourced already, but I haven’t had time to document or in general write about), I started finding a few issues with the resulting widget not being properly usable within XHTML, but those were thankfully fixed already in the newest generation of Javascript code. On the other hand, when they decided to update their widgets, the change went live to all the users without warning, dang!
The one thing that it seems to lack here, though, is support for the same metadata API used by Facebook: OpenGraph — right now you have to provide the URL to flattr, the description, the tags, the type and so on so forth whenever you provide a flattr widget. This means that any of the index pages of my blog is repeating the same content many, many times. On the other hand, if it was using OpenGraph, maybe slightly extended (even Facebook has a fb_admins value that is not documented there), it would only take the URL, then the rest of the service would be able to access the details from there.
Having such a situation would also make it nicer to have “Flattr anything” extensions: right now they really cannot work, because you do not know who owns a given webpage, which kind of webpage it is, which tags to use, and so on; and you cannot either declare a canonical URL to use, so the same page could be submitted twice, if it is reachable by two different URLs (like most of this blog is); and this is considering we have two, not one, methods to declare these canonical URLs.
Finally, I have to say that from at least one side, Flattr has proven itself profitable to me; in the past two months, where its usage surged, I received around €20/month with it, which is definitely not a lot (at least considering the expenses Tinderbox has caused me in the same time), but it’s not little. Once again, thank you all.
Also it has been funny to see some of my older posts being flattred from time to time, I guess some people actually ended up finding them with google, I hope they found what they were looking for.