Thanks Peter. Any pointers on general docs on what the heck spritzer
and garden hose is?
The public timeline api says this:
statuses/public_timeline
Returns the 20 most recent statuses from non-protected users who have
set a custom user icon. The public timeline is cached for 60 seconds so
requesting it more often than that is a waste of resources.
If I pull this into an RSS feed:
feed://twitter.com/statuses/public_timeline.rss
I refresh it a few seconds later, I get new tweets. Is that 60 second
cache an out of date note in the docs?
The API also states this is rate limited. To get the data I am after,
I am going to be hitting this think pretty hard. If there is no
cache, it will be more than 60 seconds in frequency, more like as soon
as the script is done working, I will request it again. Pretty much
perpetual requesting.
Or is this treated more like the search API, and is rate limited very
liberally?
Thank you for your help this Saturday.
--
Scott * If you contact me off list replace talklists@ with scott@ *
On Oct 10, 2009, at 2:06 PM, Peter Denton wrote:
Hi Scott,
Since it seems you are looking for a sampling situation, you might
want to
poll the public timeline and check for 1st tweet, (created at and
1st update
timeframe are same/near day).
Also, you could expand your sample size and look into accessing the
spritzer or garden hose and again running some best guess scenario of
signup and first tweet.
Cheers
Peter
On Sat, Oct 10, 2009 at 1:19 PM, Scott Haneda <talkli...@newgeo.com>
wrote:
I'm not sure this is possible, I'm trying to avoid a local data
store to
make it possible.
I would like to get a small bit of data from a tweet, but only the
first
tweet, ignoring that user from that point forward.
I can of course grab their username and disregard, but my list will
grow
quickly.
If my goal is to get stats on what time of day most people are
joining
Twitter and posting their first tweet, what would be the best place
to query
for that data?
I just need the time, and perhaps location fields. Would the search
API be
good? I hear about this firehose thing, but don't know what that is.
I'm thinking the public timeline may be best, but there may be way
too much
data for me to deal with.
Suggestions on methods appreciated.
--
Scott
Iphone says hello.
