>What’s interesting to me of course is that even though they don’t have a convenient API for all of the data, all of the data (except passwords they said) is public. So if you and I can see it, Facebook, e.g. could easily see it too, and tell what everyone is doing just the same.
If true, this sort of nullifies their claim not to sell your data. No one needs to buy it; it's just sitting there!
Let's be realistic here. The extent of that information is the users' profile, which currently consists of a 160 character bio, a human readable name (which can be anything) and any external links they choose to supply; their followership information; and whatever they choose to include in their posts.
Is there really enough salable information in there to make it worth the full time effort that would have to be put in working against the bot detection measures, the inevitable IP blocks and account suspensions? Is that really something you could base a business on?
I imagine the Ello social graph could be useful in an online anthropology study, but I can't see anyone getting much mileage out of it beyond that.
(Although to contradict myself, CircleCount.com does maintain a database of Ello user statistics. They did ask the Ello team nicely about this, and I imagine there is some kind of exception made here.)
If the data were worthless it would also make the promise not to sell pointless, paralleling my original point.
But I actually think there's plenty there, even if you just include the posts. What turns out to be valuable can be surprising. I remember when "interests" was a plain text input field on Facebook where people put a comma delimited list. And I loved it; I even put little jokes in there (e.g. coffee, chocolate, and mint, concurrent or consecutive). Then one day, they turned every one of those list items in the text field into, essentially, a fucking ad.
If true, this sort of nullifies their claim not to sell your data. No one needs to buy it; it's just sitting there!