Status updates
33 views
Skip to first unread message
Santiago Perez
Jun 26, 2008, 3:03:50 PM6/26/08
to Twitter Development Talk
I'm interested in accessing all statuses as they are posted. I'm
trying to do so with the show API for statuses but given the rate of
status creation I see myself needing about 20 requests per second to
the API. I'm concerned to whether this could get me banned from
accesing the API, and I wanted to know if there's a more efficient way
to access all statuses for a given period.
Thanks,
Santiago
trying to do so with the show API for statuses but given the rate of
status creation I see myself needing about 20 requests per second to
the API. I'm concerned to whether this could get me banned from
accesing the API, and I wanted to know if there's a more efficient way
to access all statuses for a given period.
Thanks,
Santiago
e.p.c.
Jun 26, 2008, 3:27:41 PM6/26/08
to Twitter Development Talk
Read http://groups.google.com/group/twitter-development-talk/web/jabber-pubsub
and switch to XMPP/PubSub.
and switch to XMPP/PubSub.
jungle
Jun 26, 2008, 3:53:55 PM6/26/08
to Twitter Development Talk
E.P.C: jabber-pubsub has been suspended quite some time ago and will
take a
long time to come back, if ever.
http://groups.google.com/group/twitter-development-talk/browse_thread/thread/6f24e47cca5b68f6/85ba8ab70e7f87dc
So Santiago's question remains unanswered. I myself have been
speculating
with increasing the rate at which I poll the public timeline because
I'm
losing huge loads of status updates. I don't want to overload
Twitter's
servers, so I would like to know how many requests per second Twitter
is
willing to endure before considering it abuse.
Maybe even a "rate-limit-status"-like API call for public timeline
polling?
On Jun 26, 4:27 pm, "e.p.c." <epcoste...@gmail.com> wrote:
> Readhttp://groups.google.com/group/twitter-development-talk/web/jabber-pu...
> and switch to XMPP/PubSub.
Andrew Badera
Jun 26, 2008, 3:55:40 PM6/26/08
to twitter-deve...@googlegroups.com
It will come back, the question is when.
If it doesn't, that's the death knell for twitter.
Having just this week closed a round of funding however, I bet we see full XMPP service back in the next 30-60 days. Hopefully sooner.--
--
--Andy Badera
http://higherefficiency.net
http://flipbitsnotburgers.blogspot.com/
http://andrew.badera.us/
http://changeroundup.com/
and...@badera.us
(518) 641-1280
Google me: http://www.google.com/search?q=andrew+badera
If it doesn't, that's the death knell for twitter.
Having just this week closed a round of funding however, I bet we see full XMPP service back in the next 30-60 days. Hopefully sooner.
--
--Andy Badera
http://higherefficiency.net
http://flipbitsnotburgers.blogspot.com/
http://andrew.badera.us/
http://changeroundup.com/
and...@badera.us
(518) 641-1280
Google me: http://www.google.com/search?q=andrew+badera
jungle
Jun 26, 2008, 4:22:59 PM6/26/08
to Twitter Development Talk
I hope you are right, but in the meantime, how many requests per
second is acceptable?
On Jun 26, 4:55 pm, "Andrew Badera" <and...@badera.us> wrote:
> It will come back, the question is when.
>
> If it doesn't, that's the death knell for twitter.
>
> Having just this week closed a round of funding however, I bet we see full
> XMPP service back in the next 30-60 days. Hopefully sooner.
>
>
>
> On Thu, Jun 26, 2008 at 3:53 PM, jungle <jha...@gmail.com> wrote:
>
> > E.P.C: jabber-pubsub has been suspended quite some time ago and will
> > take a
> > long time to come back, if ever.
>
> >http://groups.google.com/group/twitter-development-talk/browse_thread...
second is acceptable?
On Jun 26, 4:55 pm, "Andrew Badera" <and...@badera.us> wrote:
> It will come back, the question is when.
>
> If it doesn't, that's the death knell for twitter.
>
> Having just this week closed a round of funding however, I bet we see full
> XMPP service back in the next 30-60 days. Hopefully sooner.
>
>
>
> On Thu, Jun 26, 2008 at 3:53 PM, jungle <jha...@gmail.com> wrote:
>
> > E.P.C: jabber-pubsub has been suspended quite some time ago and will
> > take a
> > long time to come back, if ever.
>
>
> > So Santiago's question remains unanswered. I myself have been
> > speculating
> > with increasing the rate at which I poll the public timeline because
> > I'm
> > losing huge loads of status updates. I don't want to overload
> > Twitter's
> > servers, so I would like to know how many requests per second Twitter
> > is
> > willing to endure before considering it abuse.
>
> > Maybe even a "rate-limit-status"-like API call for public timeline
> > polling?
>
> > On Jun 26, 4:27 pm, "e.p.c." <epcoste...@gmail.com> wrote:
> > > Readhttp://
> > groups.google.com/group/twitter-development-talk/web/jabber-pu...
> > > and switch to XMPP/PubSub.
>
> --
> --
> --Andy Baderahttp://higherefficiency.nethttp://flipbitsnotburgers.blogspot.com/http://andrew.badera.us/http://changeroundup.com/
> > So Santiago's question remains unanswered. I myself have been
> > speculating
> > with increasing the rate at which I poll the public timeline because
> > I'm
> > losing huge loads of status updates. I don't want to overload
> > Twitter's
> > servers, so I would like to know how many requests per second Twitter
> > is
> > willing to endure before considering it abuse.
>
> > Maybe even a "rate-limit-status"-like API call for public timeline
> > polling?
>
> > On Jun 26, 4:27 pm, "e.p.c." <epcoste...@gmail.com> wrote:
> > > Readhttp://
> > groups.google.com/group/twitter-development-talk/web/jabber-pu...
> > > and switch to XMPP/PubSub.
>
> --
> --
Andrew Badera
Jun 26, 2008, 4:24:12 PM6/26/08
to twitter-deve...@googlegroups.com
What does the request threshold-checking service call tell you you're allowed?
Stut
Jun 26, 2008, 4:26:16 PM6/26/08
to twitter-deve...@googlegroups.com
I can't speak to how what request rate is acceptable, but I pretty
much guarantee it's cached on Twitter's side so there's probably a
maximum rate that will be effective. Chances are it'll still result in
you missing tweets.
much guarantee it's cached on Twitter's side so there's probably a
maximum rate that will be effective. Chances are it'll still result in
you missing tweets.
Alex: Care to share the cache lifetime for the public timeline?
-Stut
Alex Payne
Jun 26, 2008, 4:27:20 PM6/26/08
to twitter-deve...@googlegroups.com
We cache the public timeline for sixty seconds at a time.
>>> Baderahttp://higherefficiency.nethttp://flipbitsnotburgers.blogspot.com/http://andrew.badera.us/http://changeroundup.com/
>>> and...@badera.us
>>> (518) 641-1280
>>> Google me:http://www.google.com/search?q=andrew+badera
>
>
--
Alex Payne
http://twitter.com/al3x
jungle
Jun 26, 2008, 4:49:21 PM6/26/08
to Twitter Development Talk
If there are 30 new tweets in one second, since I can't get more than
the last 20 tweets from the API, I would have to call the API more
than once per second. How does the time they are held in the cache
affect me?
> >>> Baderahttp://higherefficiency.nethttp://flipbitsnotburgers.blogspot.com/htt...
the last 20 tweets from the API, I would have to call the API more
than once per second. How does the time they are held in the cache
affect me?
Stut
Jun 26, 2008, 4:51:43 PM6/26/08
to twitter-deve...@googlegroups.com
Alex Payne
Jun 26, 2008, 4:52:49 PM6/26/08
to twitter-deve...@googlegroups.com
The web page may follow a similar caching scheme. It should not be
treated as a source of more current public updates.
treated as a source of more current public updates.
>>>>> Baderahttp://higherefficiency.nethttp://flipbitsnotburgers.blogspot.com/http://andrew.badera.us/http://changeroundup.com/
Alex Payne
Jun 26, 2008, 4:54:20 PM6/26/08
to twitter-deve...@googlegroups.com
We do not intend to provide a contiguous stream of all public updates
via our REST API. You can get a representative sample of public
updates, but that's the extent of it.
via our REST API. You can get a representative sample of public
updates, but that's the extent of it.
--
Alex Payne
http://twitter.com/al3x
Stut
Jun 26, 2008, 4:55:07 PM6/26/08
to twitter-deve...@googlegroups.com
So the simple fact is that requesting the public timeline more often
than once a minute is a pointless waste of resources. In fact a rate
of once every 50-55 seconds is probably best.
than once a minute is a pointless waste of resources. In fact a rate
of once every 50-55 seconds is probably best.
-Stut
Alex Payne
Jun 26, 2008, 4:56:52 PM6/26/08
to twitter-deve...@googlegroups.com
Yes, exactly.
>>>>>>> Baderahttp://higherefficiency.nethttp://flipbitsnotburgers.blogspot.com/http://andrew.badera.us/http://changeroundup.com/
Reply all
Reply to author
Forward
0 new messages