Language consistency would be helpful.

• API Representation uses "favorites", but the current Twitter interface uses "likes".




• In the documentation for "Account Activity API" notes "likes", while in "Search Tweets" notes as "favorites".




• Perhaps not all of us are aware that "favorites" is a precedent of "likes".

Twitter could include a list of approved Twitter libraries for different programming languages in the documentation. This would equip developers with robust resources and help them know which resources they can rely on to use the Twitter API.

To enable one to send a tweet to a group

Provide a way to programmatically unhide a previously hidden Tweet, given appropriate user consent.

As hiding criterias tend to change over time (eg global grey-list of unwanted commenters to hide), I wish my app would be able to retroactively access (and hide/unhide) historical replies to historical tweets of subscribed users with a depth of eg at least a year from now.

At least, to access the list of replies my app had previously hidden.

Allow a free-plan developer to serve 10Ks-100Ks of users subscribed for a "hide replies" allowing to hide tweets nearly real-time as they happen (provided that an app is capable to deal with such a firehose).

This at least requires easy access to a stream of direct and indirect replies to the users subscribed.

The following is written on this page.
"The stream leverages some of our content filtering technology where we exclude high-confidence noisy content."
https://developer.twitter.com/en/docs/labs/filtered-stream/faq

This is a different filter than the one I created.
To make sure that my filters work properly, I need to turn off content filtering technology.
I want to create an option that does not use content filtering technology.
It is up to each user to get the data to determine which tweets are noisy.

Thank you.

Add more trending stations/places, like for north Indians its only Delhi.

Twitter could build a more affordable version of the enterprise Engagement API, which would provide developers with Tweet metrics for owned Tweets.

Engagement data returned by the API includes number of Likes, Retweets, replies, impressions, and engagements (for example, number of link clicks). This API would enable developers to better understand their reach and impact on Twitter.

I would like the Twitter API to use OAuth 2.0 instead of OAuth 1.0. OAuth1.0a is fairly specific to the Twitter APIs at this point in time. It's complicated to implement, and there is a dwindling number of developer tools that support it. OAuth 2.0 is less complicated and easier to build into an app than OAuth 1.0.

Is there any plan to allow access to the cropping rectangles that are associated with each media image? These are called focusrects and are nested under the originalinfo property of extended_entities.media and could be passed along with the rest of the media info in the API request.

While I understand the motivation for the separate "expansions" in the GET Tweets endpoint, I wonder if you have considered using GraphQL to allow the user to determine the shape of the tweet data returned?

In adding support for the labs endpoint to twarc [1] I found that I wanted to knit back together the data that is separated out into tweets. If GraphQL was used I don't think this would be necessary.

I would be curious to know if Twitter Labs considered switching to GraphQL for fetching tweets and users.

[1] https://github.com/docnow/twarc

Gnip PowerTrack has a "Profile Geo" enrichment [0].

Adding this to the filtered streams API would make migrations easier.

[0] https://support.gnip.com/enrichments/profile_geo.html

From the operator reference [0], the filtered streams appear to lack the proximity operator, that existed in Gnip PowerTrack.

Having this in the filtered stream API would make it easier to migrate from Gnip.

Additionally, in PowerTrack the maximum proximity value is 6. Increasing this limit would make the operator extra useful.

[0] https://developer.twitter.com/en/docs/labs/filtered-stream/operators

It appears that the user-auth rate limits in the labs API is shared among all users.
I could see however that app-auth limits are separate, so there is distinction between app-auth vs user-auth, just not working as expected

At www.contexto.io and www.graphext.com we provide our customers the possibility of mapping and expanding Twitter communities leveraging the social graph. This allows us to get really contextual, relevant and different insights that most social media analytics tools offer (they mostly count keywords that matches a boolean query).

We currently use the free API following/followers endpoints for this right now but with the new policy starting this month we cannot make more than 100K requests/day which prevent us from scaling our product. We would be very happy to pay for a premium endpoint like this!

I currently use the statuses/sample endpoint to stream 1% of Tweets and related compliance events for those Tweets. Having the deleted Tweet events helps me know when Tweets I've received are no longer available so that my database and application stay compliant with Twitter's policies.

Please give me a similar compatible and convenient way to keep the data received from Sampled stream compliant.

There can be a huge upfront effort to sift through the volume of Tweet data in order to get a data set that is interesting and valuable. It would be great to add more upfront filtering options to reduce the amount of data needed for statistically relevant analysis.