On Jul 17, 1:57 pm, CreativeEye <creativv...@gmail.com> wrote:

> 1) Get Twitter Public timeline repeatedly.

My understanding is that this does not give you all tweets, just a
random selection.

> 2) Get follower network - user profiles and get their statuses.

You would reach the API limit quickly, I'd expect.

I don't remember the robots.txt definition very well, but I think
twitter also disallows classic web crawlers: http://twitter.com/robots.txt

> I do know Firehose is an option, but that would again be something
> like Approach 1. right?

Firehose is only an option if Twitter allows you to use it.

> Please guide me how to proceed.

I think there is no reliable way to get ALL tweets, though I would be
pleased to learn otherwise. (with the exception of the Firehose, which
I suppose one can not plan for).

Maybe by being "sneaky" about it one can get a lot of tweets. For
example by getting people to use your service to access twitter, so
that you are using up their API limits, not your own. Or at least get
the service whitelisted so that you can make lots of requests (I doubt
they would be enough to get ALL tweets, though).

Reply via email to