That would reduce our requests by ten-fold.

Jesse

On Wed, Jan 21, 2009 at 11:16 AM, iematthew <[email protected]>wrote:

>
> Perhaps a leaner version for requesting a user's followers and friends
> would help? Say, a method that only returns the ID and screen name for
> the user's followers or friends, but in lots of a thousand or ten
> thousand at a time.
>
> On Jan 21, 12:19 am, Jesse Stay <[email protected]> wrote:
> > Alex, you and I have discussed this, but I still think this is a bad
> > decision until some sort of better method is placed around getting the
> list
> > of followers of a user.  This basically limits how big any application on
> > your platform can get.  Right now it takes 400 requests alone to get
> Robert
> > Scoble's followers.  It takes 350 requests to get Guy Kawasaki's
> followers.
> > It takes similar to get Chris Pirillo's followers.  Does this mean we
> just
> > exclude allowing them on our apps now?  Why develop for the Twitter
> platform
> > any more if we know we can only grow to your limit?
> >
> > Jesse
> >
> > On Tue, Jan 20, 2009 at 4:48 PM, Alex Payne <[email protected]> wrote:
> >
> > > Up until now we've allowed users and IPs on our whitelist an unlimited
> > > number of requests per hour.  When our whitelist was in the tens and
> > > low hundreds, this made sense. Now that we have more developers on the
> > > whitelist than we can reasonably maintain close communication with, we
> > > need to put a ceiling on the number of requests per hour whitelisted
> > > accounts and IPs can make.
> >
> > > Starting later this week we'll be limiting those on the whitelist to
> > > 20,000 requests per hour. Yes, you read that right: twenty THOUSAND
> > > requests per hour. According to our logs, this accounts for all but
> > > the very largest consumers of our API. This is essentially a
> > > preventative measure to ensure that no one API client, even a
> > > whitelisted account or IP, can consume an inordinate amount of our
> > > resoures.
> >
> > > If you run one of the services that routinely exceed 20k
> > > requests/hour, please get in contact with us ([email protected]) as
> soon
> > > as possible. Chances are good that you'll simply need to slow your
> > > crawl rates, implement more caching on your end, and limit requests to
> > > only active accounts. We're happy to work with you to find solutions.
> >
> > > --
> > > Alex Payne - API Lead, Twitter, Inc.
> > >http://twitter.com/al3x
>

Reply via email to