Windows 8 Blue Screen of Death

Posted by Amish Patel on Sep 14, 2011 in Touch, UI, UX, Windows8 | 0 comments

With developer preview builds of Windows 8 out in the wild, I can now comment on how some of the interesting changes in the release.

Most people have met the wonderful “Blue Screen of Death”, it plagues college kids around the world, reminding you to save your term papers every twenty minutes. Only the super nerdy actually know what it was is for or even found it useful. (I worked in Windows and still have no clue!)

In the majority of cases you got to the blue screen because of a) faulty hardware, b) hardware drivers have resulted in a crash or c) corrupted or buggy software is trying to touch parts of the OS it shouldn’t be.

The New BSOD

Not only is the new error screen less techy, but it expresses the new ‘personality’ and tone you will find throughout Windows 8.

What I find the most interesting is the language used in the error message,

“ Your PC ran into a problem that it couldn’t handle, and now it needs to restart.”

“It’ll restart in:”

Looks like Windows is trying to get some separation from their cheaply developed devices and hardware. Fingers crossed!

Mouse meets touch

Posted by Amish Patel on Sep 2, 2011 in Industrial Design, Interactive Languages, Microsoft Project, Touch, UI, UX | 0 comments

Touch is the latest input interaction weapon of choice for technology innovators.  Everyone in the core is aware that Apple has been supporting multi-touch tack-pads for a couple years now. Not to diminish their credit for making it mainstream, but the track-pad was an obvious candidate. A device where touch input was already mapped to traditional mouse cursor manipulation and its implementation was already that of a capacitive touch surface! No brainer! Thus, without any surprise, over the past few years we have seen the hardware industry starting to layer on touch surfaces (touch digitizers) onto our traditional mice.

Touch Mice

For the last couple years I was a touch interaction program manager for the Windows team, and I had the opportunity to partner alongside the Microsoft Hardware team in their research and development of the newly released Microsoft Touch Mouse. The goal was to create a touch optimized mouse, tailored for Windows7, which pushed industrial design (ID) and did not compromise on core mouse functionality and ergonomics.

Engineering a Touch Mouse UX

To be clear, I wasn’t on the hardware or research teams that build this mouse, but as a partner / consultant of sorts on the project there are a few design and software challenges that were interesting… worth sharing here at least.

As a UX program manager, the name of the game is to step back, analyze the requirements and figure out what the experience landscape of the project truly is. What I learnt quickly was that creating a touch mouse is not simple feat. When executing against any program like this there are many gears in motion to get it out the door. From a UX point of view, there are a handful of experience factors that need to be measured and balanced to deliver a desirable user experience. Each experience factor is tightly related with the next, and unfortunately in most cases, a direction in one factor will lead to a a large challenge or tradeoff from another… in the end you can only hope the customer agrees with the decisions you made.

Experience Factors

Industrial Design

It’s always nice to look good. The goal for the design team is to find the balance of brand, industrial design, and technology. When looking at the Microsoft Touch Mouse the goal seemed to be to drive the newly emerging ID brand family while pushing the boundaries on material and technology.

Technology

The early versions of Microsoft Touch Mouse included three types of digitizer explorations: camera-based, capacitive-sensing, and articulated (which used multiple existing mice linked together) // read more of the story here

Differentiated Design (Artist Edition)

Comfort and Ergonomics

Comfort and ergonomics has been the name of the game for great mice over the decades. When looking at a touch mouse there are some factors in particular that came into play. 

• The more ergonomic you make a mouse the less flat it becomes. As you start to curve the surface, the interaction language (in this case finger gestures) can change drastically. 
• The mouse will have a desired form which should compliment the desired primary target posture.
• The gesture language needs to be supported in such a way that does not interfere with users comfort in the target posture.
• It still needs to be a kick ass mouse at the end of it all.

When comparing the touch mouse to the magic mouse through the ergonomic lens, I feel that Microsoft succeeded over Apple in delivering a comfortable experience in the target posture.  Apple has been credited for a desirable design, however there have been many users complaining of developing wrist pains and finger fatigue.

“The Magic Mouse does not contour to the users hand” – MMFixed

The folks over a MMFixed have done a great analysis of the magic mouse and have gone so far to offer a add on solution to making it more confortable for everyday use. 

Performance

When I think UX performance I am thinking: end user performance, software reliability, power, accuracy, and more generally “this thing is worth having”.

As the form factor starts to take direction (through the balancing act between ID and ergonomics), the software and hardware engineers start prototyping everything from digitizer tech, to touch gesture engines. 

For the Microsoft Touch Mouse having the drive for a more shapely and thus ergonomic form, the challenge now comes from the fact that the users hand will be gripping, resting and constantly touching more of the surface area on the mouse. Which also means that it’s the same area ideal to perform comfortable finger gestures.

Performance UX Challenges 

Interactive Language

• Primary goal is to ensure gestures are comfortable to perform.
• User must perceive the gesture to be quick and avoids re-posturing to perform (a low initial cognitive tax to quickly get to habitual / instinctive is the goal).

Gesture Recognition on a Curved Surface

• Creating touch interaction and gestures engine against a flat surface is a relatively easy challenge today. (Phones, tablets, tables, track-pads all have established gesture recognition algorithms and techniques we can utilize).
• The challenge comes from creating a reliable gesture engine on a curved surface, where the differences in the users hand size, shape and posture will lead to more drastic changes to that of touch screen.
• Gesture like “pinch and stretch” became more fatigued and harder to recognized reliably due to the curved surface .
• Interactions like a “thumb-swipe” were gained thanks to the distinct curves of the mouse (try it, its awesome!).
  

Posture and Constant Touching

• Noise! Touch is a two state model (touching or not touching), how do you best recognize the users intent of gesturing rather than just a finger moved when the the mouse is just being used to manipulate the cursor?
• With an active capacitive digitizer, constant touches will draw power and the gesture recognizer will always be on to constantly deterring intent. (defiantly don’t want to sell a mouse you need to change batteries for every month).
• Below was my development hardware device the team used to work through all the above problems.

Tailored for Windows 7

The goal is to make a kick ass touch mouse, but more importantly a kick ass touch mouse for Windows7. A device created to amplify the Windows 7 experience. This means not just creating cool gestures that have no substantial mapping to the software it controls. Everyone one loves to pinch their fingers together to zoom. But is zooming really an essential Windows7 experience? Maybe new features like desktop window arranging and task switching should come to life. I don’t want to give it away, but they did a pretty good job not going completely overboard and over engineering the experience.

As the team from MS Hardware put it

• Enhances Windows 7 navigation
• Allows easy switching between tasks
• Easy to learn and fun to use
• Uses gestures to quickly scroll and pan, navigate, and manipulate content
• Helps you get more done in less time

 

You can find the gesture set and experience demo here (in this ridiculous video).

Cheers

A Little Windows Phone 7 Personality

Posted by Amish Patel on Mar 25, 2011 in TUI, UX, Visual Design, Windows Phone | 12 comments

Kudos to the Windows Phone 7 design team, a nice little touch of personality put into the Messaging tile UX. I didn’t notice it until a couple weeks back when getting off a plane and I finally saw a bunch of texts all come in at once…. I wonder what the emoticon is for 20, 50, 100 unread messages?

The Future of Touch Interaction @SXSWi 2011

Posted by Amish Patel on Mar 22, 2011 in Touch, TUI, UX | 108 comments

I was fortunate enough to speak again at SXSW Interactive this year. The topic at hand was to share our thoughts on touch interaction design, and propose answers to where we are now and how we  all may get to the next stage of true human computer interaction.

We booked Ballroom A and around 1000 people attended, so this was by-far the largest talk I have done so far. The talk should be shared publically sometime in the near future by the folks at SXSW. In the meantime you can view the slides and the imbedded videos below and please read an attendees ‘real-time’ summary and review of the talk.

Courtesy of Sheryl Yu Lin:

“I’ll admit I went into this session with a bit of doubt as to whether it would be another portfolio slide deck, or company plug. But my fears were soon put to rest, by this well thought out presentation by Microsoft. I was pleasantly surprised.

This session focused on discussing the language of touch, and where we are today, with some insights into what we need to to next to evolve it further”. Full Summary….

>> SXSW: Audio Stream <<

Staying Grounded

Posted by Amish Patel on Feb 7, 2011 in Touch | 42 comments

So I’ve been using the new Samsung line of Super AMOLED phones, best known to most as the Andriod powered Galaxy-S or Focus for Windows Phone 7. No doubt the screen is magnificently bright and beautiful, the phone is super light and thin…but it would be great if I could touch it!

So as many know, the current generation of touch devices use a capacitive digitizer bonded to the glass screen in order to sense your touch input. For those of you living in colder climates you see the barriers to this technology when you try to answer your phone with gloves on… inevitably leads to failure.

I don’t want to go into how capacitance in touch systems work, but its important to know that one of the fundamental characteristics of touch capacitance is that the device you are touching needs to be well grounded in order for you to conduct a charge on for the device to recognize. Many of you using iPhones don’t see the perils of this as the design of the iPhone accounts for this very well.

However, the same is not true about my Samsung Focus. Here’s the scenario that gets me every morning. I, as many, use my phone as my alarm clock. For years it has found its place as a permanent resident in corner of my bed. It has its home on the side of the bed and you always know where it will be!. At 5:30am your iPhone alarm goes off, you blindly reach over with your right arm, slide to unlock the screen in order to stop the incessant buzzing. So what happens to this scenario when you reach over and try to slide to unlock the Samsung?…. watch….

Why is this happening… I can only assume two things:

• The ground plate is inadequate
• The engineers at Samsung have made the decision that people will always be holding their phones with one hand and interacting with the screen with the other. Not the greatest assumption, I hope they don’t assume the same for 7”+ devices?

Flickr: Clever use of hover

Posted by Amish Patel on Jan 11, 2011 in UI, UX | 66 comments

For a while now I have been appreciating the 2010 UX update to Flickr’s photo viewing experience. Not only does it provide larger image viewing, a light box, an action menu and a structured navigation model (aided by subtle, yet much appreciated keyboard shortcuts); however, I especially enjoy the “photo story” control.

This control exposes metadata such as: contact information, date, camera, lens, views, comments, favourited, galleries included, and most notably the geo-location data for the image.

The reason I really enjoy this experience is the smart use of hover input over the map surface. When the map is idle and the cursor is not hovering over the map, the map shows the images location at a continent level view. As the cursor moves over the map and toward the center, the map changes from continent > country > city (vise versa). The one downer is that this doesn’t work on my touch pc, since touch != hover.

Experience it here: http://www.flickr.com/photos/amishpatel/5342069576/
Learn more @ blog.Flickr photo http://www.flickr.com/help/general/#1567038

What Kind of Friends Are We?

Posted by Amish Patel on Dec 20, 2010 in Social Communication | 12 comments

Do Not Pull

Posted by Amish Patel on Dec 12, 2010 in UX | 5 comments

Today I came across a “DO NOT PULL” button on Photojojo.com and even though (in my opinion) it doesn’t serve any advantageous purpose; I think it’s awesome! Now all they have to do is put something at the bottom of the page to get back to the top.

Designing Touch User Interfaces

Posted by Amish Patel on Nov 27, 2010 in Touch, TUI, UI, UX, Visual Design | 14 comments

When it comes to creating software there are several engineering processes and philosophies that take shape based on the type of product, release cycle, release vehicle, etc. … Regardless of the process, the following post is a quick glimpse into how I approach designing touch experiences. Below is a figure that illustrates the different factors which weigh in during my design process,  the areas are intentionally oriented adjacent to the elements that have the most direct impact based on how I think about designing / how they feed into my development process… read into it as you much or little as you like.

Functional Requirements

• What is the end user problem to be solved?
• What functionality does your design need to incorporate in order to provide a valuable solution to the problem?
• Not everything gets built, prioritize the functional components in order to best address the problem at hand.
• (The one we all hate) Based on the development platform what technical limitations need to be factored into your design?
• Once again ask… what is the value proposition to the end user? Can this problem be solved without this feature?
• Take a step back and ask, how does your feature relate to the product offering as a whole?
• Will the user see your design/feature as a “nice to have”, “makes sense” or “I don’t know how I would have lived without it”?

Visual Design

• Given the principles set forth by the visual design style how does this help or hinder your functional or interaction goals?
• How do we maintain a clean, simple visual style and create a design that does not take us down the route of Fisher Price UI (next post).
• Most touch devices today have limited screen real-estate, pair that with a large opaque interaction tool (your hand) how do we maximize functionality with the space that we have?
• How does the user know what possible before touching the screen?    

Interaction Design and User Experience

• There’s No Hover
  • There’s no touch hover state (well not on any mainstream hardware in the near future). Touch is fundamentally a two state interaction model, either the user is out of range of the device (“up”), or the user has made contact with the surface (“down”).  For indirect pointing devices such as a mouse or stylus, the hover state allows users to confidently target and explore the UI’s functionality without committing to an action. With touch, the user is essentially committing to an action on touch down; so how do we give them back that level of confidence?
• Leave the Mouse Behind
  • Let’s ensure our controls are truly optimized for the input method. Touch is not a single (x,y) pixel point, touch contacts are fat-opaque pointing tools that naturally have shape, size and even orientation.
  • Always design with the Fat Finger Problem in mind. Users’ fingers, hands and wrists will occlude the content beneath it.
  • Simultaneous / Multiple Input:  For decades GUI has been designed against a “one point of input, and one area of focus” model. With touch we need to explore the use of multiple and simultaneous input across controls, applications, screens and even devices.
• Touch with Confidence
  • How do we create something that not only is touchable, but more importantly designed such that the user perceives this UI element as confidently targetable?
  • Allow users to easily cancel or negate an action.
• Feedback is King
  • Always give responsive feedback: touch is direct, so when that finger hits the screen we should immediately be providing feedback.
• Be Forgiving
  • Touch input is direct, but imprecise and consequently users will accidentally tap or drag over unintended objects. Thus, users should be confident that their intended action can be reversed.
  • Require explicit user input to activate destructive functions or to cause large state transitions. It’s crucial to remember that touch is a very sloppy and users will accidentally touch objects, so there should be some sort of user action to confirm destructive actions like delete.
• Posture
  • What is the target form factor?
  • How will the user be sitting, standing, (driving?) when using this device?
  • What are the primary postures and orientations the user will be utilizing this device?
  • Does the user change their posture based on their confidence of targeting the elements on the screen?
  • For example: some mobile phones promote experiences designed against one-handed-single finger input, whereas tablets can target whole handled or two even handed interactions.
• Discoverability vs. Learnability 
  • Something may not be immediately discoverable, but is it learnable? I favour learnability over discoverability.  The challenge is how you use visual cues to teach the user what is possible.

Next….

NEXT: IS BIGGER ALWAYS BETTER? Why are designers cranking up the size of their controls to “make them touch-able?” There’s a better solution!

Blending the chrome …

Posted by Amish Patel on Oct 16, 2010 in UI, UX, Visual Design | 241 comments

Applications like Google Chrome have led the Windows browser UI “resurrection” by blending the window boarder chrome and the application chrome together. (The visible graphical interface features of an application are sometimes referred to as "chrome")*.

Google Chrome 7.0b

Zune 4

I decided to write this post after spending a week using Internet Explorer 9 Beta. One thing most people notice is the cleaner UI, its very "Google Chrome like”. As the week went went on there was one area of interaction that didn’t jive well with me at all.

Most, if not all Windows users have discovered and learned that double-clicking on the window chrome maximizes/restores the the application window.  I think it’s safe to say this works in 99% of all applications; so the question is: what do you users expect to happed when we start to blend the chrome together?

If a user has a pre-established, pre-learnt expectation between input and action, we have to be very delicate on how we change the UI around them. This doesn’t mean we should never change the behaviour of our UI to deliver better user experiences, but more importantly we need to ensure that the change is predictable and clearly understandable. 

Internet Explorer 9 Beta: Your chrome is not only confusing, its extremely frustrating….

I’m confident that this is something that will change as the IE team gathers beta feedback, but I want to call this out as a general problem we may all face sometime or another. The chrome surrounding Internet Explorer may look simpler, but looks can be deceiving. The images below illustrates how each area of the chrome reacts to a users double-click:

Green: maximize or restore application window  Red: close window  Blue: open new tab

If it looks like a duck?

Even though IE’s glass chrome is clean and continuous from the end user’s perspective, its far from that under the covers. Two different outcomes occur when performing the same action on what seems to the end user to be the same control. If the tip of your mouse pointer falls 1px into the blue area (across an invisible boarder), then instead of maximizing the window when double clicking you create a new tab (which is super annoying). What I find even more puzzling is the fact that the area to the left and right of the application controls also maximizes the window (little green boxes). It seems like the design is “when double clicking: if it looks like the window chrome it acts like windows chrome, except in an area of fixed height and dynamic width beside the address bar”. But what confuses me even more is that when the user drags anywhere on the chrome (blue+green) it moves the window (taking the approach of “if it looks like window chrome it acts like window chrome all the time”).  

To add temperament to my frustration, I use IE9 on a Windows Touch enabled device, so when your large and opaque finger is double tapping on what looks like a nice large open area of window chrome, its really two small areas and you have no confidence or predictability on what the outcome will be! Finger crossed for maximize!   

I’m a firm believer when it comes to UX design that a single interaction should perform the same outcome when afforded by similar or congruent visuals.

Here’s hoping they fix this one! 

Amish Patel

* http://en.wikipedia.org/wiki/UI_chrome