A cursor interaction investigation for Nemo Mobile

A few days ago, Qwazix and I were investigating on implementing a way to move text cursor in Nemo Mobile, something that today’s mobile operating systems have already in their store with more or less success and usability. So the question came: how to be original while keeping ease of use and clever user interaction?

  • The needs

Let’s see what the user need in term of cursor interaction. Basically, you’ve got a text that needs reformating and want to navigate inside it easily. It need to be quick, intuitive, and without false positive (i.e: you want the cursor to be at the end of the line while the cursor stubornly stays two letters before, no matter the number of attempts). This last one is very important since today’s offerings tends to get on user’s nerves, even on platforms such as iOS and Android that have large funds and top developers to overcome such issues.

  • The competition

So what’s in the shelves of the competition? Well, the most common cursor interaction is without a doubt the ‘long tap show lense’ method. If you’re familiar with Maemo/Meego handsets, you must already know this feature, since it seems to have gathered a consensus between major mobile firms. Indeed you can also find it in iOS, along with other major OS’s on the market.

ios_n9_cursor

While interesting and practical, bad implementation makes it difficult to use at times, and false positive are from experience far from being inexistant. You may have to be patient and try several times before placing the cursor in the right place. The good is that it is quite touch friendly and intuitive, and good looking too (particularly BlackBerry implementation, see below).

q5cursor2

/ a bit to the left, and the cursor move smoothly with your finger to this direction, with a nice colored feedback /

In the other hand, we have keyboard cursor handling, as seen in Android (at least with third party keyboards) and WebOS Community Edition.

android_cursorwoce_cursor

The later is, in my opinion, the most interesting approach. What we have is a stick-like cursor which behave like the little red stick from IBM laptop keyboards.

ibm_cursor

  • Finding the most suitable cursor interaction

Back to the question: what’s best for Nemo Mobile? Well, since the OS will primarily be used on handled phones, we need something that require the less movements possible from the hand/finger, meaning we don’t want the user to need to touch and slide along the whole screen just to reach ‘this particular piece of text’. Secondly, we want the user to find his way easily, in that he would not need hours and a good manual to find how to move his cursor along his texts. And last but not least, we need the less false positive as possible (none at all would be best, in fact).

We can already dismiss the magnifying glass-like implementation from iOS and Harmattan since it conflicts with our first issue. Also, we would not want something *too* original since it would need the user to learn a new trick in the already overcrowded world of user interaction. It may also conflict with our third and last issue since it would require testings and testings to remove as much false positive as possible (and since it would be completely new, we would be in unknown territory without previous trials).

So we figured out that the WebOS community and IBM (sadly, yes^^) were not complete fools when they proposed this stick idea, and that we may have something interesting here to implement as a cursor position switch: it’s intuitive in that it has precedent in OS interaction and doesn’t require much intellect to use, and by design it remove much of the false positive possibilities (the WebOS community has also proved that it works flawlessly when implemented well).

Now, behold gentleman. Our mighty proposition as a cursor interaction for Nemo Mobile:

nemo_cursor_props

In between the WebOS community proposition and the IBM red stick, it would answer the forementioned issues while being consistent with the Glacier UI design.

But since we know that the Nemo Mobile community will be the first to undergo this new feature if released, we want to know your opinion. So, do you think this is the right move, or would you prefer us to think harder and find something more original and less from the ‘physical world’?

9 thoughts on “A cursor interaction investigation for Nemo Mobile

  1. I think that the Trackpoint/nipple mouse idea’s great. It should solve the edge-case problems involved when trying to select text at the edges (pun not intended) of a screen.

    But, I think that precise feedback’ll be needed for this. Maybe a vibration after passing each character would be sufficient to let a user estimate his movements?

    • The haptic feedback is a needed feature for such a design. So yes, if we go this route, we need it badly! It feels half backed otherwise.

  2. Am sorry but I don’t think its ok. It maybe fuctional but looks out of place. And from the look of the very neat Glacier UI, that’s not very neat. A better approach should be considered.

    • You should try it, it’s already implemented into nemo and although it’s only a proof of concept right now it work really well. I never triggered side keys when using it on n9’s screen. And it could go along with a long tap on text to place cursor too ;)

      When done, it will be far more precise and fast than N9’s way.