Tuesday, November 25, 2014

Why the status bar has to go

The small black stripe at the top of the smartphone of your choice. Home for various tiny icons. Through subtle changes in them, we can decipher what's going on under the hood.


To better understand the status bar we have today, we must look at the desktop computing environment where the convention came from. The following image illustrates how different common desktop environments have solved the status bar. Top or bottom, (left or right. Always visible by default.


That design somehow felt like the only possible solution that anyone could ever come up with. Whenever someone started to design an operating system, they first drew that familiar bar across one of the display edges. Just like small kids default drawing the sun into one of the top paper corners. Only with the difference that kids move on, discovering other possibilities for sun placement.

Then came the advent of smartphones. Everything we got used to in the desktop environment, had to be crammed down to a smaller screen. So that we wouldn't mistake it as something else than a desktop </sarcasm>. A ceremonial bar was again crafted across the top screen edge, to give permanent residence for status icons. And after repeating that design pattern countless times, we should realize that the advent is now gone. It's no more, and here's some further incentive:

  • As a digital medium, software is dynamic in nature. A fixed or static layout is more a design decision, not a requirement. Displays also exist for dynamic content, and suffer from static one. If you haven't yet heard about screen burn-in, well now you have.
  • A small bar is a compromise in legibility. To not waste screen space, the bar height is kept tiny. This results in uncomfortably tiny icons. Some have made the bar automatically hide, to not distract user, but have still kept the bar and icons tiny. Sigh.
  • Lack of structure and meaning. On a small bar, all icons compete with each other for user attention. Since everything is visible all the time, a subtle change in one icon is easy to miss. All icons appear visually equal in importance, even if they rarely are.
  • Technical overhead. This concerns mostly app developers, but they're users as well. No discrimination, please. Better developer experiences are needed as well. Controlling status bar visibility and behavior is yet another thing to be mindful when creating your application. Also the OS owner has to maintain such complexity. Both sides lose.
  • Lost screen estate. Even if little, it all adds up. It's not really a full screen if something is reserving a slice of what would otherwise belong to your app. There is a dedicated full-screen mode in Android, further increasing the technical overhead and complexity, for both app developers and system maintainers.
  • Information overload and "over-notifying". We're bad at focusing on multiple things at the same time. Status bar at the top is screaming for attention and every time you take a glimpse at it, you need to refocus back to the whatever you did before. It's important information no doubt, but user decides when.

Even if mobile devices are almost identical to desktops as computer systems, smartphones are used in completely different way, than stationary desktops and laptops. Smartphone use is mainly happening in occasional brief bursts, instead of long sessions (desktop). User unlocks the device, goes into an app, locks it again and repeats.
It's important to understand that there's a reason for the user to do that. The device is not the center of your life, and is put aside all the time, just to be pulled out again when required.

And before the user reaches that app (or notification drawer/view), several opportunities present themselves to expose user to the system status without the need to make it persistently shown. Like making it part of the natural flow of things.

That is exactly what Sailfish OS does. It solves the aforementioned problem by showing important system information as part of the home screen content, resulting in:

  • Dynamic screen usage, behavior designed for displays
  • Superior legibility due to larger icons
  • More meaningful icons are emphasized, more layout possibilities
  • Less coding leads to faster app development
  • Single behavior is simpler to maintain from the OS side
  • All apps are full-screen by default
  • Less clutter, information is showed on demand

Don't blindly embrace a legacy design as an absolute truth. Make sure you define first what is the problem it solves. Rapid advancements in both software technology and mobile context understanding, can provide you great insight in finding alternatives that didn't exist back then.

And keeping in mind that mobile != desktop will alone carry you a long way. Remember that natural interaction in mobile context needs solutions that desktop didn't have to solve. Use your head.

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.

46 comments:

  1. What a great read & thoughtful design :)
    Glad to see a new UI which is different from mobile os "template"
    I think static status bar is practical for effortless network status diagnostic in an area with unstable network like where i live
    Wish you & sailfish success.. and jolla reach global availability soon :D

    ReplyDelete
    Replies
    1. Hello red7oker, thanks for the support and encouraging words :)

      Unstable connections are a real nuisance. We have them here as well. Rarely a status bar of any kind has been able to give enough data to locate the issue. It can be anything ranging from application, driver, hw, infrastructure, operator etc. To the user, things work or not. Seeing status information all the time only raises more question usually :D

      Great that you stopped by to comment. Have a great week!

      Delete
  2. Hi - This is well thoughtful design, challenging the status quo. Saves lot of space, even on desktop i set the panel as auto hidden on smaller displays. I love the current peek view that Sailfish offers. It lets me see stuff without leaving the current app. It so happens that I look for current Date while am using another app . I know, this is going to sound silly but am embarrassed to admit that most of the days I don't seem to recall the current Date. It will be nice to show time as well.

    https://together.jolla.com/question/81135/show-date-during-peek-view/

    ReplyDelete
    Replies
    1. Hey Anand, good to see you here again. The date information is currently very hidden and difficult to reach out to (lock screen/calendar app). It surely effects your experience, and we're looking how to make it better available.

      Thanks for being active in tjc, take care :)

      Delete
    2. Thanks for looking into this. Happy to contribute my small bit to such a wonderful Jolla team. Cheers!

      Delete
  3. Replies
    1. Glad you like the reference. Take care and thanks for commenting :)

      Delete
  4. Hello!

    I see this a little bit different. Let me explain.

    All smartphone operating systems have a status bar. I mean ALL (Firefox, Ubuntu, iOS, Android, WP, Bada, Tizen, Symbian, Palm, Webos, ...)!
    The question is: why.

    You basically say that no developer thinks about it and puts it into the OS by default. Just because everybody else does it.
    And then you list some disadvantages. You don't even care about possible advantages.

    Think about cars for a moment. Every single car in the world has a steering wheel. Really every one. For a time period of over hundred years every car had such a thing. Even the very first car had one. Man, this concept is really old! Sure it's old fashioned and belongs to the nursing home, or not?
    But in reality the reason why every car has one is: you simply can't do it better. Steering a car by a steering wheel is the best solution.

    I think, the status bar is so common because it is not possible to do it any better, just like the steering wheel. Many people think, if something is old, it is automatically bad and has to be replaced by something new. Even if it's really good and can't be done any better at all. This is a problem of our modern society which really pisses me off sometimes.
    By the way, did you know that the smartphone status bar almost killed the wrist watch industry?`Today everybody has a smartphone which constantly shows the time, therefore no need for a wrist watch.

    Now let me come to the disadvantages you listed.
    Static content, dispay burn in: I have a Nokia N8 for five years now, which has an AMOLED display. These displays are known for burning in. But there is no sign of burn in on my phones display. I haven't heard this of any other phone with AMOLED display either. Also are you aware that your Jolla phone has an IPS display? Those displays won't burn in.

    Small height, tiny icons: Have you ever used a phone with a status bar? This is just not a problem. The icons on my symbian phone are small (the display is only 3.5") but as they are icons one can identify them on a glance. In fact that is the very advantage of icons, they can be identified extremely fast and are even language independent.

    Lack of structure and meaning: This problem also does simply not exist. It is solved by showing only a minimal amount of icons in the status bar. Only icons which are important at the moment are shown. On my symbian phone I have normally 3 to 4 icons.

    Technical overhead: As I developed some apps for myself (Symbian, Sailfish, Android) I can say: that is no problem. The developer has two coices: app in full screen mode without status bar OR app in normal screen mode with status bar. This is just a simple decision.

    Lost screen estate: I don't really get your point here. Bascically see the above point.

    Information overload: This is the key point and in reality the basic and main advantage of the status bar. There are basically two types of informations: static and dynamic ones. The static ones are: time, battery level, mobile network state, if enabled wifi and bluetooth state. All of that are basic informations which the user should have at one glance without any further actions.
    Then the dynamic ones: new sms, missed calls, new/missed app notification (whattsapp, ...), alarm time set, slient mode activated. Those are also important informations which the user should have at one glance.

    Now, how is all of this solved in Sailfish? It's scattered all over the place. Some informations are even in between two screens (e.g. battery). I have to hold the display to see them. Whenever I want to see one of these basic informaions, I have to do actions (pull the display...). Is this comfortale for the user? I don't think so. I think it is a usability problem.

    Jolla says, they want to be unlike. That is ok for me. But for me, the unlikeness comes at the cost of simply not having features (as the status bar).

    The missing status bar is one of the reasons why I returned my Jolla after a week of testing.

    ReplyDelete
    Replies
    1. Hi Jens,

      Wow, what a good and thorough comment. I'm glad this topic is still raising discussion. Let me make few points clear before answering your comments.

      First, it was never about what do I personally like. My post was about clarifying topics that we went through during the early days of SfOS design.
      Second, you're comparing Jolla (and namely its software development function) to completely different leaque of organizations. Time and manpower might not be that much of an issue for them, as there's other forms of income to support RnD efforts. No hurry to ship, just look at Tizen.
      Third, we didn't remove the status bar as a functionality. It's available on demand, and actually easier to bring up than on Android (immersive mode), where you use top/bottom edge swipes to bring back system UI when immersed. Don't you think that the steering wheel removal comparison is a bit far fetched? :)

      "...Even if it's really good and can't be done any better at all. This is a problem of our modern society which really pisses me off sometimes."

      Everything can be improved upon. Especially SfOS 1.0, which has its own issues and problems, which will partly be corrected in 2.0. More of these a bit later.

      "By the way, did you know that the smartphone status bar almost killed the wrist watch industry?"

      Not the status bar alone, but because time being visible on lock/idle screen. It's the screen you see most frequently. At least for those who keep the device in their pockets/handbags.

      "I haven't heard this of any other phone with AMOLED display either. Also are you aware that your Jolla phone has an IPS display? Those displays won't burn in."

      Good to hear you haven't experienced any problems with your N8. I have first hand experience of seeing N9 screen burning in (real use and lab test environment). The SfOS design does not target/optimize for any specific screen technology. Staying on the safe side, because dynamic content should be harmless to all screens.

      "Small height, tiny icons: Have you ever used a phone with a status bar? This is just not a problem. The icons on my symbian phone are small (the display is only 3.5") but as they are icons one can identify them on a glance..."

      Yes, I have used quite a few. By the way, how good is your eyesight? Wait, are you actually using yourself to qualify what is legible? Please don't :) Icons are good, larger icons even better. I'm not questioning the advantage of icons at all. It's just that in a mobile context when you have limited time to see the information, and larger information helps a lot there. It's never an issue when sitting in an office or coffee shop.

      "Lack of structure and meaning: This problem also does simply not exist. It is solved by showing only a minimal amount of icons in the status bar. Only icons which are important at the moment are shown. On my symbian phone I have normally 3 to 4 icons."

      Showing information based on relevancy is a good idea, but difficult to implement in reality on an OS scale (lots of dependencies, modality and exceptions). Especially with limited resources. It felt easier to show all that we need on peek, instead of building a relevance engine to figure out what to show and when.

      "Technical overhead: As I developed some apps for myself (Symbian, Sailfish, Android) I can say: that is no problem. The developer has two coices: app in full screen mode without status bar OR app in normal screen mode with status bar. This is just a simple decision."

      Even if it's not a big issue for 3rd party, everything adds up; especially for the OS maintainer. Not only do you have to implement the functionality, but also support it in the future upgrades, different device configurations (phone, tablet etc), and finally offer a stable and secure API with good documentation. Looking at Android system UI documentation doesn't strike me as being simple :)

      ..continues in next comment..

      Delete
    2. So we skipped that part alltogether with peek (all apps are immersive by default), to get a concistent behavior without unnecessary exceptions.

      "time, battery level, mobile network state, if enabled wifi and bluetooth state are basic informations which the user should have at one glance without any further actions ... new sms, missed calls, new/missed app notification (whattsapp, ...), alarm time set, slient mode activated. Those are also important informations which the user should have at one glance."

      I agree, they are important. Lock screen is actually a great place for this information. It's the first thing you see when pulling a device out, and since you might not have an accurate grip to peek immediately, glance functionality similar to N8 and N9 (LPM) and lumias are the best there is. The SfOS equivalent feature, called Sneak Peek, got sadly delayed a lot and ultimately cancelled. 2.0 should have it at some point, though.

      "Now, how is all of this solved in Sailfish? It's scattered all over the place. Some informations are even in between two screens (e.g. battery). I have to hold the display to see them. Whenever I want to see one of these basic informaions, I have to do actions (pull the display...). Is this comfortale for the user? I don't think so. I think it is a usability problem."

      You're correct about SfOS design failing in that area. I've been painfully aware of it for a long time. There are three clear problems:

      1. Having three different ways to access the hidden status information (pull up in lock screen, pull down in home, swipe from app). If something is hidden, it should be dead simple to access (side swipe should've alsways done that to reduce learning curve and offer a smoother UX.
      2. lock screen had too little information about the system state, as it's the most important thing when it comes to deciding whether to unlock or put away (notifications, low battery, bad connection etc). Through animation and pacing, larger icons coul've been used to increase legibility also in outdoor conditions.
      3. third, you could not simply peek from lock screen to home, to see what your apps are up to, because the lock screen was treated as a page on the home vertical. I'm sorry it took us so long to separate the lock screen. SfOS 2.0 should serve you much better, although it still doesn't show status bar in apps, only on home and lock screens.

      "Jolla says, they want to be unlike. That is ok for me. But for me, the unlikeness comes at the cost of simply not having features (as the status bar)."

      Unlike doesn't mean being different just for the sake of being different. We wanted to make an OS because existing ones do not make justice to what makes us all human. The interface is just an necessary evil between you and what you use the device for. You buy it to do things. You make/receive calls, send texts, browse web and take pictures and so on. The graphical user interface is becoming less and less relevant as the technology improves. It needs to be there, but it's not why the product exists.

      It's Jolla's goal what's unlike and why we do this stuff every day. What we did, is what we could. Nothing more. Nothing less.

      ..continues in next comment..

      Delete
    3. "The missing status bar is one of the reasons why I returned my Jolla after a week of testing."

      We've lost many customers due to Sailfish OS not being nearly as feature complete as people are grown to expect. If you say that a static status bar would've been a feature that might've tipped the scale in favor of you staying, what is then the feature that you would've chosen for omission instead?

      As everyone can see, the OS is far from ready; even with us going in great lengths to deliver as slim and focused OS as we can, there's still a lot of work to be done.

      Finally, I want to thank you for your honest feedback, the time you've spent on commenting, and the support you've given to Jolla. I hope things work great for you with whatever OS you end up with. Maybe in a future we're grown big enough to provide you with a more feature packed user experience.

      Take care :)

      Delete
  5. I know I'm late for this post. But considering Sailfish 2.0, I have a question for Jolla: Where will go these legibility and uniqueness of the current status bar design?

    ReplyDelete
    Replies
    1. Hi AliNa

      Apologies for my late reply. A bathroom renovation and some traveling has kept me away from computers lately.

      Thanks for your comment. Information legibility is a very important thing on the go, as the time you have for looking at things is reduced or impeded by the environment.

      As I understand it, people didn't feel comfortable with the dynamic nature of the old SfOS status information design. That's why it was changed to be closer to iOS and Android counterparts in SfOS 2.0

      Personally, I have a good eysight, but I know that there are people not so lucky, and they will most likely find the status information change hard to understand. Time will tell how things work out.

      Take care :)

      Delete
    2. Thanks for your reply.

      I agree that most people don't feel comfortable with the dynamic status informations, however I like it.

      I'm eagerly waiting for SFOS 2.0, but based on the previews currently there is lots of free space between the status bar and app covers when peeking into Home from an app; I think this can be a room for bigger clock and bigger status icons.

      Delete
    3. It's the animation of SfOS 1.0 home screen (status information), that annoys people. A detail that could be easily solved.

      There's definitely free space when peeking, but I'm told it's too risky to offer anything outside iOS, Android and WP status area appearances. As in "if it's not broken, don't fix it."

      Well, let's see how things pan out in the end. Take care :)

      Delete
    4. "it's too risky to offer anything outside iOS, Android and WP status area appearances." - do you want an identity or be like everyone else? I thought Jolla was about being unlike. I guess that's why I'm disappointed with SfOS 2.0 so far.

      Delete
    5. Hi Jonas,

      Apologies for my late reply, and for the dissapointment that the design changes have caused. However, this is my personal blog and I wrote about my own observations, learnings and opinions.

      Things that I write here don't necessarily reflect features or their design that ships with Sailfish OS updates. My job at Jolla is to provide information about sustainable design alternatives, that can be used to make bigger decisicions how to compete in the mobile industry. My job is not to make those decisions.

      The status bar design, as seen on the 1.9.8.2, is an industry standard, and works well enough on tablets that are aimed to be used indoors. It's a little too small for phones with smaller screens. Maybe there will be adjustments for it in the future.

      Thanks for the feedback, take care :)

      Delete
    6. Hi Jaakko,

      "but I'm told it's too risky to offer anything outside iOS, Android and WP status area appearances. As in "if it's not broken, don't fix it."

      I'm kinda shocked to read such a thing. That's very sad :( Innovation is not something reserved for big companies.

      "My job is not to make those decisions."

      This is even sadder ! Let designers do their job ! I hope that's not too much demotivating for you and that you will stick with Sailfish OS design because I guess the OS owes you a lot.

      Thanks for being really transparent. Thanks for being so explicative.

      Delete
    7. Hello François,

      Don't be shocked. Early days of Jolla were much different from now. We're not the small startup anymore, and the work culture has changed with the size. That might slow down things at times, but if it keeps me from burning out, I should not complain.

      It's been a long and personally consuming war against expectations set by the mobile industry; and if a dynamic status bar is something that people just didn't accept, so be it. Maybe the time wasn't right.

      At least Jolla gets to fight another day. Thanks for your understanding and encouraging words :)

      Delete
  6. I have a couple of days now, install SfOS2.0 and I have mixed feelings about the status bar. I know this is my personal preference, therefore subjective.
    The status bar looks similar to other mobile OS, although it is minimalistic and is using larger symbols. I have to admit that since the information is located at the top of the screen, it is easier to get attention. e.g. network connection, rather that the old style that I had to scan throughout the screen to get the information.
    Notificaiton symbols, like email, messages, etc have really small icons, why? there is lots of screen space there.

    ReplyDelete
    Replies
    1. Yes, information being on a single row is always superior to multiple rows. Persistent location helps greatly over time, but an information array like SfOS 1.0 status area, is bad if you have a regular "laundry list" for information to be checked.

      Personally, I don't really have a preference for devices with larger devices, but for smartphones, that are used in dynamic and more challenging environments, smaller icons, even on a single row, is not optimal.

      I don't know if there's a specific reason why notification icons are made smaller. It's most likely related to general art direction and individual preference of area designer.

      Thanks for stopping by to comment. Much appreciated.

      Delete
  7. UX is getting worse.
    Do we next get small items to aim and poke like a sniper? I really hope not. If so, I will return my phone and tablet. I like more "Lucky Luke mode"

    ReplyDelete
    Replies
    1. I don't think there's going to be any more tiny buttons to tap on. Cover actions are probably the smallest. You should check TJC for topics about status area changes. It's the only way you're able to affect it, as I'm not anymore working on system UI / home screen designs.

      Delete
    2. So, what are you working on now?

      Delete
    3. Mainly working on finding time to reply comments :)

      Joking aside, I was occupied by packaging and warranty leaflet, some minor things here and there, but nothing really to write home about.

      There's quite a bit of work left from 2.0 designs, and developers will take some time to finish up with those.

      Maybe I see some more software design next year :)

      Delete
  8. You got an answer post Jaakko :D
    Why the status bar has NOT to go by our guest writer Fra

    ReplyDelete
    Replies
    1. Nice, I believe it's my first one so far. Thanks Simo for the heads up, Review Jolla for enabling it, and Francesco wor the substance itself.

      Keep it up :)

      Delete