Hi Kevin,

That query will fail.

You must specify a query along with the source: operator to get any
results. We realize this does not allow for a full result set of
tweets from a source, but this limitation is in place to not crush the


On Wed, Aug 26, 2009 at 6:12 PM, Kevin Mesiab<ke...@mesiablabs.com> wrote:
> One could get started gathering these metrics by analyzing search
> queries in the vein of:
> feed://search.twitter.com/search.json?q=source:tweetdeck
> On Wed, Aug 26, 2009 at 7:03 AM, Shannon Clark<shannon.cl...@gmail.com> wrote:
>> Hi all,
>> This isn't specific to the app I'm building at the moment, but the recent
>> thread on how to determine who is using your application reminded me of a
>> general question I have about the APIs.
>> Is there is an API call to return information about updates done via a given
>> application? (i.e. the information which is available via the website about
>> which application was used to post a given status update). Ideally I could
>> see utility for queries of this form via multiple of the API's - as a filter
>> on the streaming API's for example or as an option to filter upon via other
>> API's calls or just as metadata inherent with each update which an app could
>> choose whether or not to use in some manner.
>> Ideal would be options to both positively filter and negatively filter -
>> i.e. for an app to offer a blacklist of applications your users do not wish
>> to see updates which were posted by those apps (but might want to see some
>> aggregated information about what you have negatively filtered - i.e.
>> @rycaut has 3 recent updates from PlaySpymaster which aren't displayed etc.
>> At scale I could also see useful data for the developer community about
>> activity & usage patterns of our applications - both raw usage (i.e. # of
>> status updates) but also diversity of usage (# of unique users, % of those
>> users' updates per app type, etc). Potentially as well Twitter might offer
>> aggregated data about usage patterns (perhaps only as relative usage w/o
>> specific data) which could include patterns of usage from even accounts set
>> private (without revealing anything about those accounts  just adding their
>> data into aggregated totals - and again if the specific data isn't shown
>> then certain attacks on privacy could be avoided)
>> Anyway, perhaps there are already ways to access this data, if so I'd
>> appreciate a pointer to them, if not, I hope this sparks some discussion.
>> Shannon
>> Founder, Nearness Function - strategic consulting, brand advertising &
>> sponsorships
>> Twitter - rycaut
>> Blogs: Slow Brand - http://slowbrand.com
>> Searching for the Moon - http://shannonclark.wordpress.com
> --
> Kevin Mesiab
> CEO, Mesiab Labs L.L.C.
> http://twitter.com/kmesiab
> http://mesiablabs.com
> http://retweet.com

Reply via email to