<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Mark Hazlewood on Medium]]></title>
        <description><![CDATA[Stories by Mark Hazlewood on Medium]]></description>
        <link>https://medium.com/@mhazlewood?source=rss-c541d5a9d889------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*IGwTnsgekHQW5P9p2yJhHQ.png</url>
            <title>Stories by Mark Hazlewood on Medium</title>
            <link>https://medium.com/@mhazlewood?source=rss-c541d5a9d889------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 15:09:58 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@mhazlewood/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[On interfaces (and CS education)]]></title>
            <link>https://medium.com/@mhazlewood/on-interfaces-and-cs-education-c97591fd4921?source=rss-c541d5a9d889------2</link>
            <guid isPermaLink="false">https://medium.com/p/c97591fd4921</guid>
            <category><![CDATA[opinion]]></category>
            <category><![CDATA[education]]></category>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[computer-science]]></category>
            <category><![CDATA[design]]></category>
            <dc:creator><![CDATA[Mark Hazlewood]]></dc:creator>
            <pubDate>Mon, 12 Jun 2017 17:13:21 GMT</pubDate>
            <atom:updated>2017-06-12T17:13:21.348Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2CWezoF60_DnL8ngcHmbFg.png" /></figure><p>Let’s talk about the “UI”. It’s an innocent two-letter acronym that, like many, seems to be abused a lot. Many times we hear “the UI” used dismissively, referring <em>merely</em> to that unimportant part of a system that happens to be visible and tangible for that system’s users.</p><p>I find that for many of the people building software systems there is a concept of their work as a whole bunch of complicated and interesting programming, doing complicated and interesting things to some data in complicated and interesting ways.</p><p>OK that’s fine.</p><p>This complicated and interesting software then has some arrangement of physical widgets thrown on top of it (the UI) that just happens to be the part people are manipulating to make that interesting part do its thing. It’s an afterthought. Clearly it’s not the point of the thing. Not the interesting part.</p><p>But listen: Real life isn’t a hackathon. The world is not full of people who think like engineers. In an <strong>interactive system </strong>where the <strong>reason your software exists</strong> is to provide some kind of value to humans, the ways in which the system interacts with humans is <strong>the most important interface in the system</strong>.</p><p>Really.</p><p>No, <strong>really</strong>.</p><p>It’s the job of all <strong>entire rest of the system</strong> to serve <strong>that</strong> interface. Directly. In real-time (or at least real-time enough to be <a href="https://www.nngroup.com/articles/response-times-3-important-limits/">perceptible</a>).</p><p>If you think about that for a minute I hope you’ll realize there are some significant implications. You might also realize that it’s not a world-view shared by some (many?) of the people actually building these systems, at least in my experience.</p><p>To put it another way: If the design and functioning of all the programming that makes a thing work isn’t <strong>serving the human interaction primarily</strong>, then that system has failed. Period. All you’ve created is, at best, an interesting science project. At worst and much more likely, you’ve created yet another piece of unusable software only of use to those who made it (and maybe, if you’re lucky, to those relative few who happen to think the way you do).</p><p>All too often I’ve seen systems that have fundamental, inherent limitations in usability and optimal interactive experience due to <strong>avoidable</strong> limitations of the backend system. <strong>This works directly against the entire purpose of the system</strong>, which is to provide concrete and tangible value to <strong>people</strong>.</p><p>So. All that being said. I do, of course, live in reality.</p><p>We need engineers. We need nerds. We need people who are obsessed with technology and how to make it <strong>do things</strong>. We need people with their minds living in that space. I even go there myself sometimes.</p><p>I’m being a bit hyperbolic as a way to make a point. It’s obviously just as important that the software <strong>do something</strong>. If it doesn’t do its complicated and interesting thing then there’s a good chance it will serve exactly nobody.</p><p>Of course.</p><p>And yet …</p><p>We don’t seem to be teaching an appreciation for the <strong>human interface</strong> as a primary concern in technology education. It’s wildly irresponsible of us as an academic field and tech industry not to be doing this for up-and-coming engineers. Looking at some of the <a href="http://csmajor.stanford.edu/Requirements.shtml">top</a> <a href="http://guide.berkeley.edu/undergraduate/degree-programs/computer-science/">CS</a> <a href="https://www.cs.cornell.edu/undergrad/csmajor">programs</a> in the <a href="http://catalog.mit.edu/degree-charts/computer-science-engineering-course-6-3/">country</a>, exactly <strong>zero</strong> of them include core required courses on HCI or usability (from what I can tell … the poor information architecture of university sites is a topic for another day).</p><p>Yes there are electives and tracks, but is that enough? These aren’t disconnected fields with separate concerns. Being an objectively <em>good</em> computer scientist necessitates understanding the practical applications and implications of the systems you’re building. Is that even debatable?</p><p>If you, as a practicing engineer, don’t want to think primarily about the human interface that’s fine. Really. But understand that it <strong>needs</strong> to be considered by <strong>someone </strong>for a designed system (to say nothing of a marketable product) to be effective and successful. Not haphazardly or by accident based on whatever is the easiest “UI” to build after all the complicated and interesting parts are done. But by thoughtful and deliberate consideration based on the psychology and behaviors, goals and motivations, of the real-life people who will be the <strong>primary</strong> and most important interface in your interactive system.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c97591fd4921" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[At least one reason why command line interfaces are terrible]]></title>
            <link>https://medium.com/@mhazlewood/at-least-one-reason-why-command-line-interfaces-are-terrible-295924a2e95c?source=rss-c541d5a9d889------2</link>
            <guid isPermaLink="false">https://medium.com/p/295924a2e95c</guid>
            <category><![CDATA[cloud-computing]]></category>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[usability]]></category>
            <dc:creator><![CDATA[Mark Hazlewood]]></dc:creator>
            <pubDate>Mon, 08 May 2017 02:05:11 GMT</pubDate>
            <atom:updated>2017-05-08T02:05:39.685Z</atom:updated>
            <content:encoded><![CDATA[<p>As you may recall, the <a href="https://aws.amazon.com/s3/">Amazon Web Sevices (AWS) Simple Storage Service (S3)</a> experienced a major outage event in late February of this year. The outage began around 9:40 AM PST and lasted several (long, excruciating) hours that day. Amazon was eventually able to fully recover the service by 2 o’clock that afternoon.</p><p>Huge portions of the modern web are dependent on S3. The impact of the outage was widespread and, consequently, has been <a href="http://blog.catchpoint.com/2017/03/01/aws-s3-outage-impact/">well</a> <a href="https://techcrunch.com/2017/02/28/amazon-aws-s3-outage-is-breaking-things-for-a-lot-of-websites-and-apps/">documented</a>. Suffice to say things at AWS were probably a bit tense for a while.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/480/1*t96chMNkg4jNEvJwWpNFyw.gif" /></figure><p>Shortly after this event <a href="https://aws.amazon.com/message/41926/">Amazon published a detailed breakdown</a> of the problem, how it occurred, and how they fixed it. The root cause they identified basically boiled down to those two little words that get blamed for so much in our modern interactions with technology: human error.</p><p>Specifically (emphasis added):</p><blockquote><em>At 9:37AM PST, </em><strong><em>an authorized S3 team member</em></strong><em> using an established playbook </em><strong><em>executed a command</em></strong><em> which was intended to remove a small number of servers for one of the S3 subsystems that is used by the S3 billing process. Unfortunately, </em><strong><em>one of the inputs to the command was entered incorrectly</em></strong><em> and a larger set of servers was removed than intended.</em></blockquote><p>As those of us in the field of user experience have been touting for years, human error is never really the root cause of an event like this. Yes people make mistakes and (even engineers) are far from infallible. But that’s exactly the point! The fact that our squishy brains are so fallible is a well known and fixed variable in any system.</p><p>The inability of a designed system to predict and cope with basic aspects of human psychology is what’s at fault here. The system is what failed. It failed in its interface with the user. In this case, that interface seems to have been a command line program.</p><p>So in short: command line interfaces are terrible. I’ll go put on my flame armor and elaborate.</p><h4>Recognition and recall</h4><p>At the risk of greatly oversimplifying complex cognitive processes that professionals spend their lives studying: There are basically two paths for information to become activated and brought from memory into attention. <strong>Recognition</strong> makes use of contextual cues in the external environment. It relies on the strength of association between those cues and knowledge you’ve previously stored in memory. <strong>Recall</strong> refers to the raw retrieval of that information from memory.</p><p>Our brains are constantly using both of these systems in tandem to process and make sense of the world. Top-down and bottom-up together. You might call it … <em>middle-out</em>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/480/1*Hrlee69fr8jGpCMNZUrjPA.gif" /></figure><p>… anyway the big point here is that recognition is simply a better system to rely on when you’re talking about performing cognitive-heavy tasks.</p><p>Tasks, for example, like executing a program that could potentially take down your entire cloud based hosting platform for half the country by typing in its name from memory followed by a series of (invariably) poorly named and (usually) poorly documented arguments.</p><p>If you’re trying to perform a task without any external cues then you’re relying completely on recall. And that can be bad news. Leveraging recognition is faster. It’s less error-prone since it relies less on our biased and fallible memory. It, by definition, incurs less cognitive cost.</p><p>The many advantages of using recognition as a knowledge retrieval system are why we have graphical interfaces in the first place. We make use of visual metaphors, spatial memory, contextual semantic meaning, etc. all to provide that association that’s needed for easier activation and retrieval of <strong>accurate</strong> information. Remember how everyone loved multiple choice tests in school? Recognition is the reason.</p><h4>There’s a lot more to talk about</h4><p>In <a href="http://www.jnd.org/books/design-of-everyday-things-revised.html">The Design of Everyday Things</a>, <a href="https://twitter.com/jnd1er">Don Norman</a> delves deeply into this topic. He frames it through the lense of <strong>knowledge in the head</strong> vs. <strong>knowledge in the world</strong>, but at the core it is the same conversation. If this is something that piques your interest I’d highly recommend reading it.</p><p><a href="https://twitter.com/rbudiu">Raluca Budiu</a> also has a great article written on this over at the <a href="https://www.nngroup.com/articles/recognition-and-recall/">Nielsen Norman Group site</a> (which I referenced heavily for this quick and shallow rant). Go read that one too.</p><p>And on the specific subject of the historically bad usability of command line interfaces, I always highly recommend <a href="http://web.mit.edu/~simsong/www/ugh.pdf">The Unix Hater’s Handbook</a> for a lighthearted yet apt read.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=295924a2e95c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Iterations and relativity]]></title>
            <link>https://medium.com/@mhazlewood/iterations-and-relativity-6ab87d9892f?source=rss-c541d5a9d889------2</link>
            <guid isPermaLink="false">https://medium.com/p/6ab87d9892f</guid>
            <category><![CDATA[bad-sketching]]></category>
            <category><![CDATA[physics]]></category>
            <category><![CDATA[product-management]]></category>
            <dc:creator><![CDATA[Mark Hazlewood]]></dc:creator>
            <pubDate>Tue, 04 Apr 2017 22:40:50 GMT</pubDate>
            <atom:updated>2017-04-04T22:40:50.967Z</atom:updated>
            <content:encoded><![CDATA[<p>I got thinking about relativity, specifically frames of reference, and how it relates to an iterative development process. Many teams’ operation is built around agility and quick iterations, rapidly responding to user needs and requests (and demands and whims). But these methods are chaotic and wildly ineffective without <strong>something</strong> in place to provide context for the work.</p><p>Call it whatever — success criteria, goal setting, product vision, a north star, guardrails. It’s something to measure progress against. It’s a point of reference in the void.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/508/1*NxGZZ984gtR2jTqZJzea_Q.png" /></figure><p>Simply iterating, without some way to measure how successful each iteration was, means you’ve only got a single point of reference — where you are right now. You have no idea where or how far you’ve progressed in relation to a measurable goal. Any goal. You may as well be going nowhere.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/906/1*d98Wr24K6QyLwSy3Yo0DlQ.png" /></figure><p>Yes, at times you’ll be going backwards or sideways or slantways. Sometimes you’ll learn something new and branch off and try a new frame of reference just to see if it sticks. But at least you won’t be spinning in the dark.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6ab87d9892f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A review of Equil Smartmarker™]]></title>
            <link>https://medium.com/@mhazlewood/a-review-of-equil-smartmarker-d37e8eb3173f?source=rss-c541d5a9d889------2</link>
            <guid isPermaLink="false">https://medium.com/p/d37e8eb3173f</guid>
            <dc:creator><![CDATA[Mark Hazlewood]]></dc:creator>
            <pubDate>Sun, 07 Jun 2015 00:20:41 GMT</pubDate>
            <atom:updated>2015-06-07T00:20:41.970Z</atom:updated>
            <content:encoded><![CDATA[<p>I recently had an opportunity to try out Equil’s <a href="http://www.myequil.com/smartmarker/">Smartmarker</a>. The Smartmarker is a hardware and software system for digitizing and streaming whiteboard drawings in real-time, as well as digital archival of those drawings.</p><p><strong>Here’s a quick summary of my thoughts followed by a more detailed review.</strong></p><h4>Pros</h4><ul><li>Polished device experience, great hardware design</li><li>Huge writing area — you’d be hard pressed to run into the edges in a typical whiteboarding session</li><li>Easy to set up and genuinely fun to use (except for switching colors)</li><li>Equil Note software is nice for real-time streaming and drawing archival</li></ul><h4>Cons</h4><ul><li>Pen’s signal to the receiver is lossy and low-resolution</li><li>Swapping color rings is difficult and clumsy</li><li>Some issues with buggy Bluetooth connectivity</li><li>Erasing is hit-and-miss</li></ul><h4>Bottom Line</h4><p>While it’s clear that Equil spent time on designing the physical experience to be a pleasing one, I was disappointed by the performance of the system overall. This disappointment is punctuated by a hefty $700 price point. The Smartmarker is on a good path to facilitate a mixed analog+digital world, but it is not yet technically ready for adoption by a wide audience.</p><h4>The Smartmarker use-case</h4><p>My company recently went through a physical relocation of our headquarters that will have a large(er) segment of employees working remotely more often. I’m part of a small internal team to research and evaluate technologies to help facilitate remote collaboration. As a member of the UX team I’m focusing on use-cases related to design activities, which include <strong>freehand whiteboard drawing</strong>. We evaluated the Smartmarker as a potential solution for digitizing, one-way streaming, and archiving of those drawings.</p><p>That brings up one important point. As a system that tracks and digitizes physical whiteboard drawings, the Smartmarker system does <em>not</em> enable <strong>two-way</strong> collaboration on the <strong>same artifact</strong>. That’s not necessarily a point against it, but just a clarification. The digital versions of the drawings can be modified within Equil’s software but those modifications won’t show up on the original.</p><p>The one-way aspect might seem obvious but it’s important to stress the niche that Smartmarker is serving.</p><h4>What’s in the box?</h4><p>The system is composed of two main components: a sensor bar and a pen enclosure. There are of course a few various and sundry cables for charging and data transfer, as well as a handy case to keep all primary components in one place.</p><p>There have only been a handful of unboxing experiences I found so pleasant that I wanted to capture them. The first was my Kindle DX (the big one, from the way way back when Kindle first came out). The Smartmarker was another. It’s got very well designed packaging, showcasing the components nicely while also being functional.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Lx4DHVDpPX--sOOYmq5JRg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*o1J6EKGsJs-fZof-6lLgBA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pqpc0yxQh9TFCgNmgpow6w.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*d3RZKPqf46KRlr0wkQWYeg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lbVrHTKjEF6v32GR0GLu0g.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*r6eKHAIsPtGmKUYxDWQjrg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*d3GLygrb6YwuBi-iOh5xyw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JLiaoDg4GtVbWBuVyJK7dw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1lBW6WDLnF0FkMg3PfQkEA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uDjrMs8_rRn7G68YVe0f5Q.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bRl0bvQxgvAty8T35suUmA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*52fLdiZFtazSYyQnabM2xg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*aEOofgWnvTO2XTiNgYShrg.jpeg" /></figure><h4>Use</h4><p>After initial charging, which took about 2 hours for me, setting up the pen for use is straightforward. For either real-time viewing or streaming to others you’ll need the accompanying Equil Note software. If you don’t need the real-timey aspects right away you can always sync your drawings from the sensor for archival later. Mount the sensor to a spot on your whiteboard, turn it on, (optionally) bluetooth sync with Equil Note, and start drawing. The system allows capturing several colors with the use of “color rings” (see in the image above). These rings are manually switched via the pen enclosure; black is installed by default with additional red, blue, and green rings provided.</p><p>There’s not really much more to it and the simple instructions Equil provides were very easy to follow.</p><h4>Performance eval round 1, solo test</h4><p>During my first solo test I was up and running in just a couple minutes, drawing and streaming to Equil Note. I asked a couple colleagues to download the software, and they were able to connect to my streaming session fairly easily. From there they could watch me draw and annotate the drawing (which, again, doesn’t modify the original). I kept the drawings relatively simple— large rectangles, some large text, a few swirly scribbles to test performance.</p><p>Here’s two small snippets thatI kept from this first test, with lines noting where I tried to locate the edges of the drawing area.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/771/0*4wP0sZeY2GSSxszW.png" /><figcaption>Testing some limits</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/878/0*B2fPWx_tBH5R28sU.png" /><figcaption>The yellow was added by viewers of the stream</figcaption></figure><p>During this test I began to have some doubts. I was not overwhelmed by the digital quality of my drawings or the usability of Note. I ran into a Bluetooth connectivity issue where I had to completely shut down the whole system (sensor + Note software) and restart.</p><p>And let’s talk more about swapping those color rings. This was a big con for me and really stood out given the rest of the hardware experience. Once you’ve opened up the pen enclosure (be sure to remove the pen because it’s going to come flying out at you or the floor anyway) jimmying the old ring off then properly aligning and pushing the new one on can be frustrating. Maybe it’s my giant sausage fingers. Between this test and the next one, I swapped colors like this a few times but then just gave up and decided to keep everything black.</p><h4>Performance eval round 2, group test</h4><p>A week later we ran a group test. I won’t go into much more detail here. Suffice to say, for this session there were several of us in a room attempting to draw finer-resolution UI sketches on the whiteboard. This would include panels, buttons, text, and some icon images.</p><p>Performance was OK. Connectivity was OK. Streaming and drawing archival was OK.</p><p>Honestly the biggest issue for us, and why we eventually decided to return the unit, was the quality of the resulting digital drawings. The signal appears to be lossy with resolution and quality greatly under what I would expect. This results in the drawings being choppy and unreadable, which is unacceptable for our team’s use-case.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/714/0*-F4SoQuG3Z6iFjBl.png" /><figcaption>Frequent gaps in lines</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/677/0*K5lFFrcyVr0BgtgB.png" /><figcaption>Unreadable text</figcaption></figure><p>In their defense, it could be that we just had a poor operating environment. There is a whole section of the Equil Note software dedicated to debugging both the infrared and ultrasonic signal quality of the system.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/614/0*mDJip9N63PbaFuIr.png" /><figcaption>Shown disconnected from the system.</figcaption></figure><p>But to be frank I had neither the time nor inclination to troubleshoot the signal quality. I needed something that just works.</p><h4>Wrap up</h4><p>So as I said up front: I think the Smartmarker is on a great track but not quite ready for general adoption.</p><p>In my mind a 99% reliable signal, with high quality and high resolution drawings, is the bare minimum for a system like this. We were in a typical office environment without any large sources of interference and yet still had problems there. So that’s the first thing to work on.</p><p>I’d also love to see the color rings replaced with something like a selector or button toggle on the outside of the pen enclosure. This needs attention. Fumbling with those rings detracts greatly from an otherwise pleasant hardware experience (again, especially at this price point).</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d37e8eb3173f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Be Deliberate]]></title>
            <link>https://medium.com/@mhazlewood/be-deliberate-6536cfeb7d4f?source=rss-c541d5a9d889------2</link>
            <guid isPermaLink="false">https://medium.com/p/6536cfeb7d4f</guid>
            <dc:creator><![CDATA[Mark Hazlewood]]></dc:creator>
            <pubDate>Mon, 15 Dec 2014 17:09:20 GMT</pubDate>
            <atom:updated>2014-12-15T17:09:20.073Z</atom:updated>
            <content:encoded><![CDATA[<blockquote>Every object tells a story if you know how to read it.</blockquote><blockquote>- Henry Ford</blockquote><p>I had need to use a baby changing table recently inside of a Target men’s room. This label caught my attention and I snapped a picture.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/500/0*fJfBppf0-sWxy5Z2.jpg" /></figure><p>Sometimes I notice things like this — I’m sure I’m not alone — and I’m trying to make a point to capture and share them more.</p><p>Anyway, take a look at those words for a second. How easily this could have read “DO NOT LEAVE CHILD UNATTENDED” in big bold letters with imagery conveying some unlucky infant colliding with the tiled floor. This text struck me because it’s certainly a message that has been done before, and it <em>could </em>have just gone with that default, but it didn’t.</p><p>What’s the difference? Both messages meet the requirement: Place a safety label so we don’t get our asses sued if some numbskull drops their kid on the floor. Maybe it was just a random choice of words that happened to meet that requirement.</p><p>But remember that every choice made in the design and presentation of a Thing <a href="http://www.wanterfall.com/Communication-Watzlawick&#39;s-Axioms.htm#111">conveys something</a>. Somebody was deliberate with the presentation of this message. It doesn’t presume that I am an idiot or a neglectful parent. Somebody took care in choosing those words, in that presentation, with the ellipsis giving it both an informal, passive, <a href="http://blog.codinghorror.com/making-considerate-software/"><strong>considerate</strong></a><strong> </strong>tone (it also feels more within modern vernacular, but maybe that’s just me).</p><p>I’m sure this probably doesn’t consciously register with most users of this table. It is discrete and subtle. In fact many good UX design decisions <em>should </em>be invisible in that way. But make no mistake that it gives an impression. It tells the user something about how the maker of this thing <em>thinks about them (</em>or doesn’t).</p><p><em>Originally published at </em><a href="http://mshazlewood.tumblr.com/post/83628099112/be-deliberate"><em>mshazlewood.tumblr.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6536cfeb7d4f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Good Design vs. The Chain of Command]]></title>
            <link>https://medium.com/@mhazlewood/good-design-vs-the-chain-of-command-8b6d9668579a?source=rss-c541d5a9d889------2</link>
            <guid isPermaLink="false">https://medium.com/p/8b6d9668579a</guid>
            <dc:creator><![CDATA[Mark Hazlewood]]></dc:creator>
            <pubDate>Mon, 15 Dec 2014 14:23:46 GMT</pubDate>
            <atom:updated>2016-04-15T13:42:16.860Z</atom:updated>
            <content:encoded><![CDATA[<h3>Good Design vs. The Chain of Command — Social context matters in good design</h3><p>So I recently lost a debate on the front-end design for a particular UI capability. This isn’t anything new, I lose a lot of debates on user experience, but this particular case was interesting because it’s a good example of how <strong>social context </strong>can play a very important role in design decisions.</p><p>Without going into too much detail, the purpose of this widget was to manipulate <strong>angular values </strong>for a circular area in a geographic context. This circular area can have a slice cut out of it, and the angular extents for that slice are configurable. For the purposes of this discussion let’s call this thing the <strong>Ground Pie.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/300/0*WPkxGbgTwwyro5Mh.png" /><figcaption>Ground Pie over Niagara Falls with a slice removed</figcaption></figure><p>Being a student of Interaction Design and specifically a firm believer in Goal Directed Design, I came into this discussion with a fundamental idea:</p><blockquote>The less abstraction between your user’s conceptual model for the thing they are trying to manipulate and the UI to manipulate the thing, the better.</blockquote><p>Put another way, <strong>less UI. </strong>Often in the simplest and most direct means of manipulating something is also what’s most easily understood by your users.</p><p>Traditionally in our applications this kind of widget has been presented to users as a pair of numeric entry fields, spatially separate in the layout from the visual rendering of the map and Ground Pie. That is, you specify a start angle from 0 to 360 and and end angle from 0 to 360 in a form on the sidebar or another panel.</p><p>So here is where we get into a discussion of <em>abstraction, </em>or maybe a better term would be <em>level of indirection. </em>Let’s start with what I believed the user goal to be.</p><blockquote>User Goal: I want the Ground Pie cover a different geographic area on the ground</blockquote><p>With that goal in mind, we can look at the steps necessary for a person to make use of numeric entry fields and how well they allow the user to achieve that goal:</p><ol><li>Figure out desired start and end angles visually to achieve desired geographic result</li><li>Mentally convert desired visual/geographic coverage to angular values, in degrees</li><li>Shift visual focus away from Ground Pie to angle entry fields</li><li>Type in new values</li><li>Shift visual focus back to Ground Pie and wait for changes to take effect to verify results (hopefully this feedback is immediate!)</li></ol><p>Clearly this workflow can be optimized, right?! There is a mental shift that needs to happen to convert a desired geographic coverage for the Ground Pie to numeric angles. Some people will be able to perform this conversion more efficiently than others. For engineers, this is no issue — but engineers are not the target end users for this application.</p><p>This workflow also includes two separate shifts of visual attention. Again, something that software engineers and intermediate/expert computer users shouldn’t have a major problem with, but still it is a shift in context and contributes to overall cognitive load.</p><p>So my suggestion was to move away from these numeric fields and use a straight click-and-drag solution instead. The user would see their Ground Pie on a map, where they wanted it placed, and would be able to click and drag handles to change the angle. Here’s a simple mockup:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/300/0*PsnpQSjQ4r59sA50.png" /><figcaption>Proposed Ground Pie with “improved” design by direct manipulation</figcaption></figure><p>Our above workflow could now be simplified:</p><ol><li>Figure out desired start and end angles <strong>visually</strong> to achieve desired geographic result</li><li>Click and drag handles to achieve desired coverage, observing simultaneous immediate feedback to verify</li></ol><p>Makes sense right? From 5 steps down to 2, and no shifting attention. Score one for UX!</p><p>Well, maybe not so much …</p><p>It turns out I had the user goal wrong. In this particular application, with this particular set of target users, the goal was actually something more like</p><blockquote>User Goal: Configure the Ground Pie to use the angles my commanding officer tells me to use</blockquote><p>You see, I was informed that in this application users would not be changing these values on the fly all the time. They would be setting them once, when the system is initially set up, and based on written directions from someone above them in the chain of command.</p><p>Oh right, did I mention this is a system used by military personnel? Yeah … that’s important.</p><p>There is a social context here that acts as another layer of abstraction. The user of the system doesn’t necessarily know or care anything about the geographic area covered by Ground Pie, at least with regard to configuring it. For the configuration, he only cares that he’s able to configure it to the values given by his CO.</p><p>That’s not to say numeric entry fields in a completely separate panel are 100% ideal, but they’re certainly much closer to the user’s mental model than I had previously thought. I could think of a design that combines these two; providing no shift in attention with immediate visual feedback <strong>AND </strong>numeric entry. But as always there is a question of cost versus gain and if we can just use the cheapest solution (i.e., standard widgets available in the Java Swing library) and be “OK” from a UX perspective … small victories.</p><p><em>Originally published at </em><a href="https://hcizen.wordpress.com/2013/07/11/good-design-vs-the-chain-of-command/"><em>hcizen.wordpress.com</em></a><em> on July 11, 2013.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8b6d9668579a" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>