[Air-L] Monitoring tags on Tumblr, Instagram, Pinterest etc

Shulman, Stuart stu at texifter.com
Thu Oct 17 04:19:23 PDT 2013


The publishers (Tumblr or Twitter for example) face a serious challenge
trying to monetize the data beyond just advertisements. The rate limits on
public APIs are designed to allow folks to taste and test within the
constraints of a black box. Authorized re-sellers like Gnip, Datasift, and
DiscoverText offer the full fire hose and enhanced metadata.

In the case of Twitter, the full history is available and you can use
powerful filtering rules to pull very specific slices from moments or
periods in Twitter history:

https://docs.google.com/forms/d/1c_I_BV7s2sTjb4P6FpLqtp5tojzy3I3aAS-PfngdVj8/viewform


On Wed, Oct 16, 2013 at 10:47 PM, Alex Leavitt <alexleavitt at gmail.com>wrote:

> You can query the API, but you only get the latest 1000 notes (really a
> bummer...).
>
> Alexander Leavitt
> PhD Student
> USC Annenberg School for Communication & Journalism
> http://alexleavitt.com
> Twitter: @alexleavitt <http://twitter.com/alexleavitt>
>
>
>
> On Wed, Oct 16, 2013 at 5:34 AM, Jen Jack Gieseking <jgieseking at gmail.com
> >wrote:
>
> > Thanks, Stuart.
> >
> > And beyond GNIP / 30 days? Any other ideas? Thanks, all!
> > JJG
> >
> >
> >
> > --
> > Jen Jack Gieseking, Ph.D.
> > New Media and Data Visualization Specialist, Postdoctoral Fellow
> > Digital and Computational Studies Initiative, Bowdoin College
> > jgieseking at gmail.com
> > www.jgieseking.org
> > www.spatiallyinclined.org
> > @jgieseking <https://twitter.com/jgieseking>
> >
> >
> > On Wed, Oct 16, 2013 at 8:18 AM, Stuart Shulman <
> stuart.shulman at gmail.com>wrote:
> >
> >> Yes. Gnip serves that data and you can experiment with collecting it via
> >> the DiscoverText 30-day free trial:
> >>
> >> https://app.discovertext.com/Home/SignupContactTrial
> >>
> >> We provide 1000 gratis 'credits' where each item and the associated
> >> metadata count together as one spent credit. You can run a 1000-item
> pilot
> >> test of the Gnip Tumblr stream, or spread your credits across the fire
> hose
> >> for Twitter, Disqus & WordPress.
> >>
> >> This is sample metadata from a Tumblr item using a Power Track operator
> >> to exclude reblogs, but if you don't include the rule, you get all the
> >> reblogs and you can use the "Sifter" engine to deduplicate and cluster
> >> around reblogging inside of DiscoverText.
> >>
> >> actor id:   tag:gnip.tumblr.com:2012:blog/29604921
> >>   id:   tag:gnip.tumblr.com:2012
> >>
> :post/60837734713/update/9dc81eb1beb4540312e1b20d251b98b7ff0cdcf2d0442b176ca40c5739a49448
> >>   link:
> >>
> http://crash--test.tumblr.com/post/60837734713/ferrari-430-scuderia-by-lyon1845
> >>   note count:   40
> >>   object type:   photo-album
> >>   posted time:   9/10/2013 2:41:22 PM
> >>   rule match:   -is:reblog flickr
> >>   tag:   430
> >>   tag:   cars
> >>   tag:   F430
> >>   tag:   Ferrari
> >>   tag:   Scuderia
> >>   target display name:   crash--test
> >>   target link:   http://crash--test.tumblr.com/
> >>   target object type:   blog
> >>   tumblrreblogkey:   4VCo4kLi
> >>   tumblrtype:   photo
> >>   url:
> >> http://www.flickr.com/photos/60710285@N05/9691950494/in/photostream/
> >> Show top meta  verb:       update
> >>
> >>
> >>
> >>
> >> On Wed, Oct 16, 2013 at 8:02 AM, Jen Jack Gieseking <
> jgieseking at gmail.com
> >> > wrote:
> >>
> >>> Hi all,
> >>>
> >>> To add to this convo, has anyone been able to grab the notes per post
> >>> (reblogs, comments, likes) associated with a specific hashtag on
> Tumblr?
> >>> I'm guessing not given the previous answer but one never knows.
> >>>
> >>> Thanks,
> >>> Jack
> >>>
> >>>
> >>>
> >>> --
> >>> Jen Jack Gieseking, Ph.D.
> >>> New Media and Data Visualization Specialist, Postdoctoral Fellow
> >>> Digital and Computational Studies Initiative, Bowdoin College
> >>> jgieseking at gmail.com
> >>> www.jgieseking.org
> >>> www.spatiallyinclined.org
> >>> @jgieseking <https://twitter.com/jgieseking>
> >>>
> >>>
> >>>
> >>> On Fri, Aug 30, 2013 at 1:58 PM, Alex Leavitt <alexleavitt at gmail.com>
> >>> wrote:
> >>>
> >>> > There is currently no way to track ALL tags on Tumblr unless you want
> >>> to
> >>> > shell out money for full access to GNIP (and even then you have to
> >>> specify
> >>> > terms and get the data moving from the current data forward; you have
> >>> to
> >>> > pay extra for historical data). Via the API, you can get posts per
> >>> tag, and
> >>> > if you have some programming experience, you could set up a cron job
> to
> >>> > continually run through the API to collect data over a long period of
> >>> time.
> >>> >
> >>> > I'm not familiar with Pinterest or Instagram's APIs (I assume they
> >>> operate
> >>> > similarly), but you might check out
> >>> > http://www.cc.gatech.edu/~sbakhshi/gilbert.chi13.pinterest.pdf to
> see
> >>> how
> >>> > they collected data.
> >>> >
> >>> > As far as cross-network, there is no service (to my knowledge) that
> >>> allows
> >>> > you to do this.
> >>> >
> >>> > Alexander Leavitt
> >>> > PhD Student
> >>> > USC Annenberg School for Communication & Journalism
> >>> > http://alexleavitt.com
> >>> > Twitter: @alexleavitt
> >>> >
> >>> >
> >>> >
> >>> > On Fri, Aug 30, 2013 at 6:06 AM, Shulman, Stuart <stu at texifter.com>
> >>> wrote:
> >>> >
> >>> > > Tumblr, Twitter and Facebook are part of the DiscoverText
> capability:
> >>> > > http://www.discovertext.com/import.html
> >>> > >
> >>> > > We support Gnip's "Power Track" operators for four streams of data:
> >>> > > http://www.discovertext.com/Gnip.html
> >>> > >
> >>> > > There is a 30-day free trial with access to Facebook, Twitter,
> >>> WordPress,
> >>> > > Tumblr, and Disqus:
> >>> > > https://app.discovertext.com/Home/SignupContactTrial
> >>> > >
> >>> > > This is an example of how a sophisticated research group uses DT &
> >>> Gnip:
> >>> > > http://vimeo.com/55175059
> >>> > >
> >>> > > Stu
> >>> > >
> >>> > >
> >>> > >
> >>> > >
> >>> > > On Fri, Aug 30, 2013 at 6:49 AM, Deller, Ruth A <
> >>> R.A.Deller at shu.ac.uk
> >>> > > >wrote:
> >>> > >
> >>> > > > Hi all
> >>> > > >
> >>> > > > Does anyone know of an easy way of keeping track of tags over
> time
> >>> in
> >>> > > > Tumblr, Pinterest or Instagram. I'm especially interested in
> >>> numbers of
> >>> > > > tags over time and particular spikes in their popularity, and, if
> >>> > > possible,
> >>> > > > how manh unique users are associated with them and other tags
> often
> >>> > used
> >>> > > in
> >>> > > > conjunction.
> >>> > > >
> >>> > > > Also, is there any way you know of of being able to see how often
> >>> > things
> >>> > > > have been crossposted across different social media platforms
> (ibc
> >>> > > flickr,
> >>> > > > twitter, facebook etc)?
> >>> > > >
> >>> > > > Ruth
> >>> > > > _______________________________________________
> >>> > > > The Air-L at listserv.aoir.org mailing list
> >>> > > > is provided by the Association of Internet Researchers
> >>> http://aoir.org
> >>> > > > Subscribe, change options or unsubscribe at:
> >>> > > > http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org
> >>> > > >
> >>> > > > Join the Association of Internet Researchers:
> >>> > > > http://www.aoir.org/
> >>> > > >
> >>> > >
> >>> > >
> >>> > >
> >>> > > --
> >>> > >
> >>> > > Dr. Stuart W. Shulmanhttp://people.umass.edu/stu
> >>> > >
> >>> > > Founder and CEO, Texifterhttp://texifter.com
> >>> > >
> >>> > > LinkedIn: http://www.linkedin.com/pub/stuart-shulman/10/351/899
> >>> > > Twitter: http://twitter.com/#!/StuartWShulman
> >>> > >
> >>> > > Director, QDAP-UMasshttp://www.umass.edu/qdap
> >>> > >
> >>> > > Editor Emeritus, JITPwww.jitp.net
> >>> > > _______________________________________________
> >>> > > The Air-L at listserv.aoir.org mailing list
> >>> > > is provided by the Association of Internet Researchers
> >>> http://aoir.org
> >>> > > Subscribe, change options or unsubscribe at:
> >>> > > http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org
> >>> > >
> >>> > > Join the Association of Internet Researchers:
> >>> > > http://www.aoir.org/
> >>> > >
> >>> > _______________________________________________
> >>> > The Air-L at listserv.aoir.org mailing list
> >>> > is provided by the Association of Internet Researchers
> http://aoir.org
> >>> > Subscribe, change options or unsubscribe at:
> >>> > http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org
> >>> >
> >>> > Join the Association of Internet Researchers:
> >>> > http://www.aoir.org/
> >>> >
> >>> _______________________________________________
> >>> The Air-L at listserv.aoir.org mailing list
> >>> is provided by the Association of Internet Researchers http://aoir.org
> >>> Subscribe, change options or unsubscribe at:
> >>> http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org
> >>>
> >>> Join the Association of Internet Researchers:
> >>> http://www.aoir.org/
> >>>
> >>
> >>
> >>
> >> --
> >>
> >> Dr. Stuart W. Shulman
> >> people.umass.edu/stu
> >>
> >> Editor Emeritus, JITP
> >> jitp.net <http://www.jitp.net>
> >>
> >> Director, QDAP-UMass
> >> umass.edu/qdap <http://www.umass.edu/qdap>
> >>
> >> Founder and CEO, Texifter
> >> texifter.com <http://www.texifter.com>
> >>
> >> LinkedIn: linkedin.com/pub/stuart-shulman/10/351/899<
> http://www.linkedin.com/pub/stuart-shulman/10/351/899>
> >> Twitter: twitter.com/#!/StuartWShulman<
> http://twitter.com/#%21/StuartWShulman>
> >>
> >>
> >
> _______________________________________________
> The Air-L at listserv.aoir.org mailing list
> is provided by the Association of Internet Researchers http://aoir.org
> Subscribe, change options or unsubscribe at:
> http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org
>
> Join the Association of Internet Researchers:
> http://www.aoir.org/
>



-- 

Dr. Stuart W. Shulmanhttp://people.umass.edu/stu

Founder and CEO, Texifterhttp://texifter.com

LinkedIn: http://www.linkedin.com/pub/stuart-shulman/10/351/899
Twitter: http://twitter.com/#!/StuartWShulman

Director, QDAP-UMasshttp://www.umass.edu/qdap

Editor Emeritus, JITPwww.jitp.net



More information about the Air-L mailing list