In showing my work to some people, I found that on the whole they said that it was lacking connections between the images shown. There wasn’t much of a narrative being shown for the stories. So I started experimenting with making the UI reflect the conversations its showing.
How Boris Johnson refused to fight the virus
^ Conversational
I made a prototype that prioritises the conversation of the story, rather than the article itself in a familiar interface.
[https://stripe-weak-aragosaurus.glitch.me/](https://s3-us-west-2.amazonaws.com/secure.notion-static.com/986aea85-53b6-4fa6-9026-dbea45600f6a/Screen_Recording_2020-06-01_at_02.28.25.mp4)
https://stripe-weak-aragosaurus.glitch.me/
It guides the reader to get more from each story by bring the first paragraph forward in the user flow and then encourages reading commentary from twitter underneath.
I redesigned it such that it still has the same flow of conversation but looks like Apple news, so I could compare them purely on the basis of information consumed.
https://obtainable-meowing-hovercraft.glitch.me/
To test this I loaded up Apple news on a phone and loaded the same stories with the links form the stories in apple news in my interface to do an A/B test to compare.
I found it was incredible difficult to test the knowledge and perspectives some had on a story before and after as they didn’t change much in either my interface or Apple News. So I focused more on general feedback on the interface.
User 1: was able to recall more details about a subset of the stories form apple news, but little form my interface. Felt the interface had far too much information on it.
User 2: Said the images got in the way a lot of the time. Once showed the previous interface above said they much preferred more words.
User 3: Too much content in one space. Felt the news was all form one source.
With the feedback on the homogeny of the stories, I made an interface using Google Trends to source stories and tweets from. Google Trends and headlines from Reddit has so far proven the best.