Ask HN: API limitations for Web services
Sometimes I have ideas of Web services that require the Twitter Search API. But then, after seeing the number of requests you can do by day (150 searches for example), I keep thinking: "well, this app could always be better written by them, on their servers. They could even copy my idea and do a much better implementation because for them there are no API limits."
With 150 (or 1000...) requests per day when your app got minimally successful it would also die.
Is this thinking correct or I'm missing something? (Not a web developer here...) The thing with Twitter is, that they don't really want to kill their app ecosystem, but really just want to keep growing because of their app ecosystem, they only want to be the transport layer so it is highly unlikely that they would come out and kill your project by duplicating it. A perfect example is the completely under-powered lists feature, the API is super powerful but twitter's implementation is as barebones as it can be, possibly because they want to encourage other people to build on top of their stack. Have you looked at the Streaming API for your ideas? You can also get full firehose access if it really does become successful. Obviously, sharecropping on someone else's farm is never a brilliant idea but if you're going to do it, Twitter may not be such a bad place to do it... I'll take a look at the streaming API. "You can also get full firehose access if it really does become successful (...)" Sorry, I don't understand what "firehose access" means in this context. Care to explain? :) Firehose gives you all the tweets in almost realtime. Its the big brother of the gardenhose. Its not publicly available. I would like to know what the conditions are to get it. Is just asking nicely enough? Twitter announced at LeWeb in December that they would open the Firehose to all developers in “early 2010”: http://mashable.com/2009/12/09/twitter-firehose/ Twitter has kinda relaxed limits for search. I have done 1 req every few seconds, and that was fine.
Their normal API has more constraints, but it's supposedly not that hard to get your limit lifted to 20K queries/hour.. Depending how often your searches update, you might be able to get away with using the RSS feeds from search.twitter.com.