0

Most modern smartphones nowadays include a feature which predicts the next word to be typed when the user inputs some text onto the screen. This functionality is useful because it can save the user some time by suggesting the words they might want to add next, as seen below:

enter image description here

I have been trying to think of the best way to implement such functionality into a web application, particularly in a textbox. So far I've come across two implementations:

  1. WordPrediction

enter image description here

  1. Google Scribe

enter image description here

Are such approaches best design practice? Are there any other approaches that I should be aware of? What other considerations should I have?

Improve this question
3
  • I think you might find that, except for people who are non-fluent, having to switch from typing to deciding whether to pick (maybe none of the choices is appropriate) is more annoying than helpful because it derails thought.
    – MMacD
    Commented Jan 15, 2017 at 11:37
  • @MMacD thanks for your reply. Can you suggest any ways around it? Just for context, we wanted to implement this feature into a product to help people who would normally find it difficult to structure resumes. An example use case: someone wants to express their experience as an event planner on their resume. So they type out: "Assisted in planning..." and then the app would recommend recommends the following words as the next word: "and, events, development, themes".
    – methuselah
    Commented Jan 15, 2017 at 15:22
  • One possibility would be to list the various occupations, let the person pick (e.g. "event planner"), and provide a sample sentence for their use. The US Labor Department has a fairly comprehensive list available, so it wouldn't mean having to research all the occupations yourself.
    – MMacD
    Commented Jan 16, 2017 at 15:52

2 Answers 2

Reset to default
1

This instantly reminds me of a basic functionality you see in most code editors.

enter image description here

It helps when you don't remember the right syntax for a function. Or the properties of an object. Also it speeds along coding. In the code editor I use I often type a few letters and then just hit tab knowing it will autofill it for me.

For normal textediting, especially on a desktop with a keyboard, I can't really think of an use for such a feature other than a thesaurus. But you're the one with the idea.

Hope this commit helps.

Improve this answer
1
  • Hi Paul, many thanks for your response. Maybe some context will help. We wanted to implement this feature into a product to help people who would normally find it difficult to structure resumes. An example use case: someone wants to express their experience as an event planner on their resume. So they type out: "Assisted in planning..." and then the app would recommend recommends the following words as the next word: "and, events, development, themes".
    – methuselah
    Commented Jan 15, 2017 at 15:14
2

So I go by this understanding. You want to know about the best practice for text auto suggestions for an application.

From a user experience perspective, the auto suggestion feature should support the user and not assert the suggestion.

For example, I am an Android user. I had a chance to use an iPad(first version) when I got it as a gift. My interaction with Apple product was bare minimum. The text auto suggestion was something I got confused with(because of my prior interaction with Android). Because when I type "inter" I get the following suggestions.

  • inter

  • interested

  • interesting

When I hit the space button, the word interested will be auto selected. Now that is not what I wanted to use. This is a case where the auto suggestion asserts it's suggestions.

In case of Android, the options are displayed and gives the freedom to the user to choose which suggestion suits her/his context.

Always provide the power of choosing to the user. It will help him stay in your application longer since he/she feels secure and in command.

Improve this answer
1
  • Thanks for the comment! How about a situation there the text becoming auto-suggested needs to be useful for non-native English speakers, so it becomes a matter of context. For example, if a user wants to express their experience as an event planner on their resume, they might type in something like “Assisted in planning”. The app could recommends the following words as the next word: “and, events, development, themes”. How would that work in the context of a web app?
    – methuselah
    Commented Jan 15, 2017 at 19:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.