• Eduardo NunesEduardo Nunes, over 7 years ago (edited over 7 years ago )

    I remember answering this, and the results are very interesting. Thank you for sharing!

    If I may, I would like to add one more observation to your conclusions, which I think is at the core of how we interpret on-line conversations: it's not just that messages inside speech bubbles make it look like there's more text, it's also that chat messages mimic natural speech in a way that longform text (e.g. e-mail) doesn't (not most of the times, anyway). We think of chat messages as more spontaneous, and more heavily influenced by the other's replies (or lack thereof). Several speech bubbles in quick succession appear more desperate partly because they hint at the author's degrading confidence when faced with the absence of feedback, while a long-form text comes across as a stream of thought written in a single sitting — giving the impression that the writer knew beforehand what the point of the communication was, and understood exactly how the first couple of paragraphs relate to the last. This is in contrast to chat messages, where it seems like everything after the the first couple of messages was written on a whim. Of course, the conclusion here is pretty much the same as yours, since the only meaningful difference between all three versions you tested is formatting.

    4 points
    • Chantal Jandard, over 7 years ago

      Thanks for your response Eduardo! I agree. That's something that I should add into the article. The typing style was the same for each condition; in Full Conversation, multiple messages in quick secession appear similarly to one longer message. It was a UI different (and a design decision) to call attention to each distinct message.

      It would be super cool to repeat the study with all chat-style UIs (like Hangouts vs Facebook Chat vs iMessage, maybe?) and see how the different visual designs transform the perception.

      1 point
  • Marc Olivier LapierreMarc Olivier Lapierre, over 7 years ago

    I didn't know there were multiple versions, that was quite interesting. Thanks for sharing :)

    1 point
  • Chantal Jandard, over 7 years ago

    Hi guys! Awhile back, I posted a survey, which a whopping 1320 kind souls filled out! Here are the results; thanks for helping! :) I hope this is useful.

    1 point
    • Andrew ZimmermanAndrew Zimmerman, over 7 years ago

      I remembered the DN post as I was reading this one. Thanks for the info and the well-reasoned conclusion.

      Glad I thought Alex was a guy - I love validation!

      0 points
  • anthony thomasanthony thomas, over 7 years ago

    This isn't an issue with the UI, it's an issue with the user misusing and misinterpreting two entirely different UI functions.

    Chat is for real-time interactions like a conversation. Email is for more detailed thoughts like a letter. One requires more urgency and a quicker response time than the other. If you treat chat like email or email like chat, of course you're going to look socially incompetent.

    The guy in the experiment sent 4 separate messages in the FB chat without even getting a response back. He didn't even make sure she was there or even listening. That's like having a conversation with yourself. Of course, you're going to look socially incompetent.

    If the guy sent 4 separate emails like he did with the messages, he would look just as socially incompetent if not more. Nobody sends emails like they're in a chat. If the user did that, would you say there's a problem with email UI now?

    The problem isn't the UI design, it's with the user. If FB changed their chat to function like email, then it wouldn't be a chat anymore. It would defeat the whole purpose of chatting. They have messages for email-like communication already. Why would they change their chat to work like email just because one user doesn't understand how chatting works?

    I totally disagree with the author's conclusions. It just goes to show that you can do good research, but come out with the wrongest conclusions. The research doesn't matter if you don't have the reasoning ability to interpret the results correctly. What's the point of doing research if you draw false conclusions? Reason will always be above research. Research is just supplementary.

    0 points
    • Chantal Jandard, over 7 years ago

      Hey Anthony!

      Thanks for your thoughts! I agree that the conclusion isn't wholly supported by the research; a follow-up study would be needed to clarify that.

      That said, email wasn't a part of this study. This compared Facebook Chat with Facebook Full Conversation. For users of Facebook Chat, any content they send would be visible to the recipient in both formats. Notably, Facebook Full Conversation can be used in async, but can also support Chat-style conversations. (Actually, I never use Facebook Chat; I prefer to do any chatting on Facebook in Full Conversation view!) I typed the same in both instances, pressing enter after each messages, so really, there should be no differences of expectation.

      While one could argue that Full Conversation is more like email than Chat, since it supports chat-style conversation and shares content with Chat, I think it is fair to compare them as different UIs to a similar context.

      0 points
  • Laurens SpangenbergLaurens Spangenberg, over 7 years ago

    The medium is the message.

    That's why I hated Facebook for killing chat heads which made conversations feel more human and replace it with cold and sterile Messenger.

    Great research though.

    0 points