Friday, October 2, 2015

Do you like it? Part 1

It's a well known wisdom, that asking early whether people like the groundbreaking product you're working on or not, has a tremendous potential.

Potential to destroy that said product, damage your brand, or even kill off your company; depending of its size.

Did that get your attention?

Good, because it should. The critical feedback in building new products (vs. copied), is the long term one. As the name implies, it takes longer to form compared to the short term one. People have to use your product for months instead of hours. There's no shortcuts, no silver bullets.

Quickly dashing around for likes, opinions, ideas and suggestions is an open invitation for a disaster. You're only chasing popularity and trends. Meanwhile, important values and product opportunities are drifting away - never to be seen or achieved again. It's a great way to inflict potentially irreversible damage to everyone in the value chain.

Alright, let's give that some time to sink in.

In the next part, I'll explain why short term feedback sucks.

Thanks for reading and see you in the next post. In the meantime, agree or disagree, debate or shout. Bring it on and spread the word.



  1. <sarcasm>
    That kind of reminds me of the latest release of my fav mobile OS: SailfishOS.
    It seems like the creators of said OS has been influenced by some random focus group, that has only tried the OS for hours, not months, and thereby judged it by comparing it to the two major players.

    1. Hi Thomas and Kimmo,

      Here's the link to the the most extensive user study about SfOS 1.0 UX

      The most challenging thing to learn in SailfishOS, was the best thing it can offer. The study concluded that learnability is the biggest challenge, and should be addressed. Not with tutorial alone, but in-context tips and helps. In other words, using the most familiar learning method of all times: learning by doing.

      There was a small set of user interviews done to select the most desireable design for 2.0 update, but nothing as extensive as the linked study.

      Thanks for the sarcasm. I need my daily dose :)

  2. If you want to make others learn something you have to make it desirable. And sadly SFOS isn't the so damn pretty girl/boy (SFOS aesthetics) that you will tolerate that it snores (SFOS missing features).

    Overall SFOS is a good release and improves things. Even the things that i was a bit critical before using the OS in my phone (ie. status bar and favorites on the home screen) don't bother me. However the two things that enrage me everyday are: a) Unlocking the screen with a horizontal slide b) Locking the screen requires one more action (swipe and tap).

    Especially for a) there was no reason to break the action that all the users were used to. You could just add a different wallpaper to the lock scree to make it "separate" and call it done.

    1. Yep, there needs to be incentive to stick around. I know some people lost part of that with 2.0, while others gained some. There might be improvements to the screen locking behavior at some point. It will take time for the design team to find solutions to the feedback we're getting.

      Personally, I was waiting that unlocking with side edge gesture would allow peeking to home from lock screen - just like you can do from an app. Sadly that didn't happen yet, but in theory it should be possible. Also, using one gesture to get to home is better for long term use (muscle memory). Using only edge swipe leaves up/down/left/right flick gestures for something else.

      Thanks for commenting :)

    2. The bad/funny thing is that the design team is slow and the people will get used to the new behavior.

      Maybe the early releases could be used as a testbed for stuff like that if some UX stuff are easy to implement. Ie i can't imagine its that difficult to change the unlocking from sideways to vertical.

    3. Some user interview sessions were used as the testbed for these changes.

      Maybe things would've been better if the vertical flick still unlocked the device. Currently it does nothing, which feels kind of rude towards existing users.

    4. Was the Pulleymenu "(2.0) fix" decision made by user interview session? It's really frustrating that Jolla is using resources to fix something that isn't broken. Peek, Pulleymenu, Statusbar, Cover swipes, Locking device, Ambiance change, Lock screen Pulleymenu etc. Only thing which is better now (2.0) is notification screen and that's because it's prettier now, but now it's missing online indicator. So... make your conclusions.

    5. Yes, we tested pulley menu visuals with end users. WIth old and new users alike. Changing visual style can be a good way to add some refreshing feel to the product, when for example no new functionality can be introduced.

  3. And to add to the beauty comment above HW design (not talking chips inside) also plays a big role to this.

  4. Is there any hope for future SFOS versions having official support for reverting UI changes made in 2.0?

    - Horizontal swiping for unlocking would be okay, if the visual feedback for it wasn't such a cluster WTF.
    - Checking the time used to be pleasant, now its a pain.
    - Swipe from the bottom edge for events was way more useful for me than app grid (because my favorite apps' icons were always front and center).
    - Icons for favorite apps front and center where they should be.
    - Cover actions *sigh*. How about a setting to have swiping from the edge rotate the carousel and leave horizontal swipe on cover to do cover action.

    I can appreciate the fact that any new OS hoping to gain market share has to feel familiar to users of the leading established OS, but come on. If Sailfish looks, feels and blows like Android, I could just as well get a nicer phone with less clunky UI and billions of fart apps.

    1. I doubt that there's going to be any 180 degree turns in design direction, but let's see. Our design team lead has been looking into the feedback, and is setting priorities on those with the rest of his team. Here's some info to points you made:

      - I asked earlier about the visual feedback of the unlocking gesture, and it's intentionally different from the standard peek feedback.

      - Because the 2.0 toolbar labels and icons are smaller now, lighting conditions have much bigger impact on their legibility. But, since the tablet is used mainly indoors (more stable lighting/less distractions), it's not considered a critical issue.

      - When no apps are running, the empty home screen is probably a bigger design imperfection, and there's ongoing work to find a solution. What makes this a challenging task, is the separation of backgrounded apps from the installed ones. We cannot show the launcher when no apps are running, since it makes accessing events _really_ annoying. One option is to duplicate functionality from launcher, to have something on the home screen, when no apps are running.

      - I don't think non-tappable cover actions are coming back. Now when they're buttons, people use them more. I guess they're less intimidating. From the gathered feedback, people either didn't care (didn't use cover actions) or didn't understand how to use them.

      I can't really comment anything in more detail as I'm not anymore working on the OS main screen designs actively. We're rotating tasks so that every designer gets an opportunity to design different areas.

      Anyways, even if these are already known issues to us, I appreciate your comments. Thanks :)

    2. The new cover actions have another problem beside breaking an established/known action. If you tap to select a minimized app there is a big chance -depending on cover place on the grid and finger reach- to hit the button and have the button action instead of just maximizing the app. With swipe covers this was impossible.

      The first time this happened it was a nice WTF moment.

      A proposed solution -in TJC- that i liked was to make the lower part of the cover swipeable.

    3. Whenever there's multiple touch targets in close proximity, that are interested in the same interaction event (tap), these are bound to happen. Can't really help it now, it's gone. I would just make peace with it.

      Changing the cover action area to work with gestures might not really help, as a horizontal flick gesture is not as accurate as a tap (that's why it originally used an entire cover). If you miss the tiny action area, you're interacting with home carousel, and that is equally frustrating. And when you casually play with the home carousel, you could accidentally trigger a cover action.

      The only sane way is to make the carousel work with edge swipes and leave on-screen gestures fully for covers. No other way around it with the carousel design.

  5. About gesture tradeoffs. ( I do consider tradeoff as trade-in and trade-out ) So we lost some gestures because of partner space. In India users will get snapdeal? But what about rest of us? Users might be happier if get some trade-in too. Maybe Verkkokauppa, or something unlike, H&M, IKEA? :)

    1. Yes, I'm sure there are endless possibilities. At least when it's ready. User customization needs a lot of work still.

      Finally, let's hope people can adjust their daily device use in ways that makes room for various partner spaces :)