Thanks for your great comments. Like I said on my post – the idea is here and needs to be fine tuned. Input welcome – always.

But I don’t think the money will come from fee charging in some form on the average user side of the gig. I can’t remember where I got this from but a big chunk of usage on twitter isn’t down to the big users, charging on a freemium model with SLA’s based on query volumes etc. I don’t think that’s the big buck deal.

There doesn’t seem to be a great impetus there to open up their raw fire hose of data, the XMPP access. They could obviously get into peering deals and let those “proxies” out how they will pay their twitter bills, pretty much as Florian says.

I think Twitter’s pay out will come in offering quicker than “real time” data with some real intelligent filtering and trending. Think Bloomberg Professional but on the editors desk in every serious news organisation. Sure, anyone can build that sort of client and back end but twitter will always have the updates first. If they just queued all the tweets in a sort of buffer on the public side there would be some real value in getting that time critical stuff before everyone else.

We really are talking minutes at most here, but I think there would be lots of super premium subscribers to such a service and for them there would be value in keeping access as unbridled as possible, more voices = more data points to filter and more chance of getting a tweet from someone that has something people will want to hear about.

But hey, what the hell do I know.

Twitter is one of the best network facilitation tools available but I think we all under value the data derived from tweets. In the retail marketplace, choosing a product is largely an emotional process. Moods affect consumption patterns. A tweet is an emotion, a mood, or an emotional reaction. Friends and their followers are micro-societies and these social groups have patterns

Facebook Connect already shares behavior patterns with their advertisers

I think Ev, Jack and Biz should be developing an analytical tool that can be used to trend these emotions and behavioral patterns derived from tweets. This emotional intelligence can then be marketed in the same way that agencies sell high priced strategic target marketing lists.

This business model would have the least amount of disruption to the twitter operation as it exists today.

“Just not so convinced you can chaRge for basic data IO.”

I do think they need to adopt an AdWords type model though where they pass on the expense to a long tail.

Not sure whether the Twitter API would be something worth charging for. Perhaps, but i just imagine another free API would come out and twitter would die.

I think POSTing and GETing basic data from sites needs to be free. There is value in algorithms prior to getting the data (not something Twitter does) and there is value is using data context to do something – such as relevant ads.

Just not so convinced you can change for basic data IO.