Thanks John. I'm mostly interested in using a single keyword at present, but it would be nice to not have rate-limiting issues with multiple keywords.
I've read the Streaming API page, and I guess I get the ideas there, although I'm not familiar with the Twitter API terminology yet. I was hoping to avoid the complexities of dealing with an ongoing persistent stream of data. I'm also disappointed if there is no other way to simply get a count of tweets every hour or day etc. without having to transfer all this content I would simply throw away after counting. It seems like it would be a waste of bandwidth and CPU on my end and yours. I can Google a keyword and get an estimated count of webpages without having to view every result. It would be nice if Twitter could do something similar. I'm also reading the thread on "estimating tweets per day" and I see that Twitter may want to keep that information confidential. I also find that disappointing because I'd like to compensate my data for the growth in Twitter data. Regards, Mike. On Oct 14, 4:56 pm, John Kalucki <[email protected]> wrote: > The track parameter on the Streaming API will give you all statuses > that match a keyword. When rate-limited, you'll get an accurate count > of the limited statuses. You'll only get accurate results for sets of > keywords if your stream is not limited.
