<?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 Lico Takahashi on Medium]]></title>
        <description><![CDATA[Stories by Lico Takahashi on Medium]]></description>
        <link>https://medium.com/@licota?source=rss-9a14543fcb5e------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>Stories by Lico Takahashi on Medium</title>
            <link>https://medium.com/@licota?source=rss-9a14543fcb5e------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 16:05:57 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@licota/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[Why your design is biased]]></title>
            <link>https://uxdesign.cc/why-your-design-is-biased-f25b300f8559?source=rss-9a14543fcb5e------2</link>
            <guid isPermaLink="false">https://medium.com/p/f25b300f8559</guid>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[ui]]></category>
            <category><![CDATA[human-centered-design]]></category>
            <category><![CDATA[diversity]]></category>
            <category><![CDATA[user-research]]></category>
            <dc:creator><![CDATA[Lico Takahashi]]></dc:creator>
            <pubDate>Wed, 21 Nov 2018 17:02:21 GMT</pubDate>
            <atom:updated>2018-11-23T13:19:12.251Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ui0_oYxAgOyLQL2ax53aSg.jpeg" /></figure><p>“And the next one is a game project,” the middle-aged lecturer said and opened a new slide on the screen. The slide showed an overview of the usability test for an online game. I woke up from my daydreaming, as I am a gamer and interested in game development. But the next thing the lecturer said was not about designing a great game.</p><p>“We recruited the test participants from avid gamers, and it was painful,” he made a face. “You know, working with those <em>nerds</em>. So hard to communicate, speaking gibberish — oh, I don’t want to do that again!”</p><p>It could have been meant to be a joke, as a few of the students laughed. But I didn’t understand. What was he, a visiting lecturer from some usability company, talking about? Was he trying to ridicule the test participants, who had bothered to come to the usability lab to give valuable feedback? What kind of message was he communicating to the young designers?</p><p>I know sometimes it is hard to communicate with people from fields you are not familiar with. Learning jargons takes time and effort. But what if the test participants were middle-aged managers who wore suits and worked for a big company in an industry that use lots of technical terms? Would the lecturer still have laughed at them?</p><h3>Social categorization and othering</h3><p>We tend to prefer people who are similar to us, and sometimes hate people who are different from us. In social psychology, such process is called <a href="https://link.springer.com/referenceworkentry/10.1007%2F978-1-4614-5583-7_414"><em>othering</em></a>. Othering happens when we categorize people based on some features, such as gender or ethnicity, and distinguish groups that we belong (ingroup) and don’t belong (outgroup). We see ingroup people positively and outgroup people negatively, because we want to have a positive self-image.</p><p>Distinguishing ingroup and outgroup is a very common psychological process. Categorization is an essential cognitive skill for human being, and favoring ingroup people doesn’t mean you are a bad person. Everyone does that. But you must be careful if you are a designer, because such categorization affects your understanding of others. People not only favor ingroup members, but also process information more deeply for them than outgroup members. Ingroup members seem more diverse to us, while outgroup members all seem alike (<a href="https://journals.sagepub.com/doi/10.1177/1368430207075153">Boldry et al., 2007</a>). We can remember detailed information about individual ingroup members, while remember less detailed or positive information about outgroup members (<a href="http://psycnet.apa.org/record/1981-07959-001">Howard &amp; Rothbart, 1980</a>). In other words, our understanding of outgroup people is often shallow and stereotypical.</p><h3>Are tools neutral?</h3><p>People think technologies are neutral. <em>C’mon, your devices and apps are just tools, how can tools be biased?</em> But it is a little naive to see any tools as neutral. Designers project their views of the world into the design of things, whether intentionally or not. For example, <a href="https://medium.com/hh-design/the-world-is-designed-for-men-d06640654491">Kat Ely</a> describes how everyday things are designed by men and for men, marginalizing women and minorities. Seat belts, medicine, work environments, power tools — they are all traditionally created by men and for men, making it harder for other people to use. It happens because of the imbalance in the technology and design industry, Ely points out, as most tech and design positions are still held by males. Gender is a basic social category and, if you are a male, you tend to understand males better than people of other genders.</p><p>Designers have traditionally used so-called I-methodology, projecting their own preferences and skills onto the target users. It could easily reproduce biases of the designers. Human-centered design was born as an alternative to such methodology; it incorporates people’s perspectives instead of designers’, putting “human” needs at the center of the design.</p><p>But even human-centered design can be biased, unless it is used with care and knowledge. First you need to carefully choose the sample population in the user research and user testing. Even if you choose proper and diverse people, you could misunderstand the needs of people who are different from you. In the processes such as interviews and focus groups, you might fail to let them speak up. You might see your projection instead of the real people. And in the persona method, which is very popular among UX people, you might lose meaningful details of the user research findings and fall into stereotyping. Actually, personas are <em>meant to be</em> stereotypical; <a href="https://www.amazon.com/dp/0672326140/">Alan Cooper </a>explicitly indicated “stereotypical personas are more effective,” and emphasized how “shooting for believability, not diversity” is important.</p><h3>How technology gets biased</h3><p>Researchers began to explore the issues of social bias in the interaction design. It is often discussed in the context of gender, as it is one of the most salient biases found in the technology field. Gender-based digital divide seems to persist, even in the young population with higher education (recently I conducted a survey with university students, which showed surprisingly stereotypical gender differences). Besides, looking at gender issues is a good start to reconsider the traditional design practice, as the philosopher Martha Nussbaum says in her discussion on social justice:</p><blockquote>Looking at women’s lives helps us see the inadequacy of traditional approaches; and the urgency of women’s problems gives us a very strong motivation to prefer a non-traditional approach. — <a href="https://www.tandfonline.com/doi/abs/10.1080/713678045">Women’s Capabilities and Social Justice</a></blockquote><p><a href="https://onlinelibrary.wiley.com/doi/abs/10.1002/9781118976005.ch4">Breslin and Wadhwa</a> introduce four main “gendering” problems in human-computer interaction.</p><ul><li><strong>Disparate numbers</strong>: tech positions are filled by white or Asian men.</li><li><strong>The I-methodology</strong>: designers tend to reproduce the norms of the group they belong.</li><li><strong>Design stereotypes</strong>: designers tend to use stereotypes when they design for people who belong to the other groups than themselves.</li><li><strong>Focusing on the difference</strong>: researchers reinforce traditional binary categorization by focusing on the differences based on the grouping.</li></ul><p>These four factors are intertwined with each other, encoding specific norms and values into technology in our everyday life. Sometimes it is visible (e.g. stereotypical pink colors for girls apps), but often it is embedded deep enough and not easily noticed. For example, <a href="https://www.researchgate.net/publication/28552207_Configuring_the_User_as_Everybody_Gender_and_Design_Cultures_in_Information_and_Communication_Technologies">Oudshoorn et al. (2004) </a>revealed how the designers of the Digital City Amsterdam created products that were only usable for people with specific learning style, which was found more often in men than in women. Such design issues often remain unnoticed, steering people with different characteristics away from the usage and thus depriving them of the opportunities they deserve (which, in turn, might enhance stereotypes against those people and create more disparity).</p><h3>How can we reduce bias in design?</h3><p>The following are a few suggestions to reduce bias and stereotypes in design. (For more details, see <a href="https://onlinelibrary.wiley.com/doi/abs/10.1002/9781118976005.ch4">Breslin &amp;Wadhwa, 2018</a> and <a href="http://www.informatik.uni-bremen.de/soteg/gict2009/proceedings/GICT2009_Bath-geloescht.pdf">Bath, 2009</a>.)</p><h4>1. User involvement</h4><p>Communicating with diverse people helps to prevent the unintended biases and stereotypes. ISO guidelines for human-centered design recommend that we involve users throughout the process of design and development. A design project should start with a thorough requirement analysis of the intended users, using direct methods such as interviews, observation and focus groups. Also, user testing should be conducted with diverse users if your product is targeted for diverse population. But as we have discussed, user involvement does not automatically assure bias-free design.</p><h4>2. Recognizing scripts and metaphors</h4><p>When we design things, we use certain <a href="https://www.researchgate.net/publication/312487882_The_De-Scription_of_Technological_Objects">scripts</a> explicitly or implicitly. An obvious example is when we use personas and user scenarios. We create imaginary characters and write scripts about what they think, do and feel. Such scripts can reflect our biases and wrong assumptions, even when they are based on the user research.</p><p><a href="https://www.researchgate.net/publication/228735397_Narrative_Transformation_Designing_Work_Means_by_Telling_Stories">Narrative Transformation</a> and<a href="https://journals.sagepub.com/doi/abs/10.1177/0162243911401633"> Mind Scripting</a> are proposed as tools for designers to notice and understand their own values and assumptions in the design activities. <a href="https://www.amazon.com/Computation-Human-Experience-Learning-Doing/dp/0521386039">Critical Technical Practice</a> would also be helpful to make hidden metaphors in the everyday design practice visible. In addition, <a href="https://vsdesign.org/publications/pdf/non-scan-vsd-and-information-systems.pdf">Value Sensitive Design</a> is proposed as a tool to actively inscribe desirable values into technology and products.</p><h4>3. Designing accountably</h4><p>Design can not include everything. Some user needs are always cut out during the design process. In the complex digital society, design success can be evaluated only within the <a href="https://link.springer.com/article/10.1007/BF00749282">specific situations</a> in which the product is designed for. What designers can do is to be aware of their design decisions, to be accountable for the “cuts” they made in the design process. Did your design decision exclude certain group of people? Would it harm or hinder particular groups? Can you justify your decisions?</p><h3>Design is never neutral</h3><p>Human-centered design took the power back from systems to humans. Most designers today recognize the importance of human needs as well as system requirements. The next step is re-examining the scope and assumptions of “human” in human-centered approaches, to distribute the benefits of technology to a broader population.</p><blockquote>The wants and needs of young, healthy, middle-class people with connections and a reasonable amount of spare cash are overrepresented among Start-up City’s priorities…Structural social injustice and systemic racism are harder to tackle — and that’s where the tech sector has, until recently, thrown up its hands. (<a href="https://www.newstatesman.com/laurie-penny/2014/04/tale-two-cities-how-san-franciscos-tech-boom-widening-gap-between-rich-and-poor">Laurie Penny on a tale of two cities</a>)</blockquote><p>Design is never neutral. Each of your design decision enables or restricts the ways in which people interact with the technology and information. It forms people’s <a href="https://en.wikipedia.org/wiki/Performativity">performances</a> in everyday life. It creates meanings, reinforcing or altering the existing norms about who we are and how we live.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fupscri.be%2Ff%2F50d69a%3Fas_embed%3Dtrue&amp;dntp=1&amp;display_name=Upscribe&amp;url=https%3A%2F%2Fupscri.be%2F50d69a%2F&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=upscri" width="800" height="400" frameborder="0" scrolling="no"><a href="https://medium.com/media/05d5fd32eda31cbd1b83287606744532/href">https://medium.com/media/05d5fd32eda31cbd1b83287606744532/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f25b300f8559" width="1" height="1" alt=""><hr><p><a href="https://uxdesign.cc/why-your-design-is-biased-f25b300f8559">Why your design is biased</a> was originally published in <a href="https://uxdesign.cc/">UX Collective</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Data basics for UX people]]></title>
            <link>https://uxdesign.cc/data-basics-for-ux-people-6a798b41693b?source=rss-9a14543fcb5e------2</link>
            <guid isPermaLink="false">https://medium.com/p/6a798b41693b</guid>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[user-experience]]></category>
            <category><![CDATA[user-research]]></category>
            <category><![CDATA[design]]></category>
            <category><![CDATA[usability]]></category>
            <dc:creator><![CDATA[Lico Takahashi]]></dc:creator>
            <pubDate>Fri, 13 Apr 2018 13:01:38 GMT</pubDate>
            <atom:updated>2018-04-28T08:17:21.220Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Eo8TP3HJ0PXUQwHfJFvxcA.jpeg" /></figure><h3>Why Data, Anyway?</h3><p>UX designers have numerous methods to improve their design, such as user interview, focus group, diary study, persona, storyboard, task analysis, customer journey map etc. Many of the methods are intuitive and powerful; they speak a lot about user needs and stories. These <em>qualitative</em> methods are driven by the urge to understand the users, to empathize them to create better solutions.</p><p>However, qualitative methods aren’t always the best ways, especially when it comes to evaluating the prototypes and products. We often face <a href="http://journals.sagepub.com/doi/abs/10.1177/154193129804201902">evaluator effects</a><em> </em>when we conduct usability tests. We think we observe something objectively, but in fact we might be just seeing what we <em>want</em> to see. As the Yale psychology professor Paul Bloom <a href="https://opinionator.blogs.nytimes.com/2015/06/06/imagining-the-lives-of-others/">says</a>:</p><blockquote>People are often highly confident in their ability to see things as others do, but their attempts are typically barely better than chance.</blockquote><p>That’s why we need to step back and take a different perspective to understand the users. That’s where <em>quantitative</em> data and statistics come in.</p><h3>Quantitative Metrics in UX Research</h3><p>Quantitative research is not as difficult or expensive as one might think. If you conduct a qualitative usability test, such as think-aloud, then you can just measure some data during the test session, or/and add a short questionnaire after the session.</p><p>Here are some metrics that you can use to measure effectiveness, efficiency and satisfaction.</p><h4>Metrics for Effectiveness</h4><p>Can the user complete the tasks successfully? You can measure Task Success, Errors, Issues-based Metrics (usability problems found in the test), or Self-reported Metrics.</p><h4>Metrics for Efficiency</h4><p>How easily and quickly can a user complete the tasks? You can measure Time on Task, Efficiency Metrics such as page count and click count before completing the tasks, Learnability Metrics such as Task Time across Trials, or combination of those metrics. Behavioral metrics such as Eye-Tracking are also useful to measure efficiency.</p><h4>Metrics for Satisfaction</h4><p>Is the user satisfied by the interaction with your product? You can use Self-Reported Metrics such as System Usability Scale (SUS), Computer System Usability Questionnaire (CSUQ) and Net Promotor Score (NPS). Additionally, you could use behavioral and physiological metrics such as heart-rate, skin conductance or face expression, to estimate stress-levels and emotions that participants are not willing to say (some participants are too nice to speak their negative emotions!).</p><p>Which exact data you should collect depends on the goals of the users, the goals of your product, and conditions such as project schedule, budget and other resources.</p><h3>The 4 Types of Data You should Know</h3><p>Understanding your data is critical when analyzing the research results, because different types of data require different types of analysis. Below are the 4 types of data that you should know to do some statistics.</p><h4>Nominal Data</h4><p>Nominal data are groups or categories that are not ordered by numbers. For example, if you want to compare the performance between different user groups, such as “male vs. female” or “frequent user vs. non-frequent user”, the groups are not ordered and therefore are nominal data. (Of course you can put a number for each group for convenience, but that doesn’t mean you can compare the groups as numbers, as it is only arbitrary coding.) The examples of nominal data include user characteristics, locations, gender or groups with different expertise.</p><h4>Ordinal Data</h4><p>Ordinal data are ordered groups or categories. For example, when you ask users how often they use your website by choosing from “very often”, “often”, “sometimes” and “rarely”, then the acquired data are ordinal. In this case, “very often” means more than “often”, and “often” is more than “sometimes”. They are comparable but the distance between each rank is meaningless. In other words, you don’t know how much more often the user who chose “very often” uses it compared to a user who chose “sometimes”. Therefore, you can’t analyze the data using an average rank or other statistical methods.</p><h4>Interval Data</h4><p>Interval data are continuous data where the distances between each value are meaningful but there is <em>no true zero point</em>. An example is temperature data in Fahrenheit or Celsius. The distance between 10 degrees and 20 degrees are meaningful, but the zero point for temperature is only arbitrary. In UX researches, subjective rating data are often treated as interval data. Let’s look at this scale.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*aFO6-O9auwTYs1BTKOXQWQ.png" /></figure><p>It can be treated as ordinal data, but if the distances between each point are same and meaningful, then it can be treated as interval data. You can judge if it’s ordinal or interval by asking if the halfway point between two dots makes sense.</p><h4>Ratio Data</h4><p>Ratio data are almost same as interval data, but they have a true zero point. For example, when you measure task complete time, the acquired data are ratio data because <em>there is an absolute zero point</em> (zero seconds means no time at all, unlike when you say zero Celsius degrees where it’s just a certain point in the scale). Other examples of ratio data include age, height, weight and number of tasks completed.</p><h3>What kind of Statistical Test should you use?</h3><p>Now we understand the differences between 4 data types, let’s think about some statistical methods to analyze the data.</p><h4>For Nominal Data — Frequency, Chi-square</h4><p>Simple descriptive statistics can be used for nominal data. For example, you can count the numbers for each category and make a frequency table. To compare nominal data, you can use a statistical test called <a href="https://en.wikipedia.org/wiki/Chi-squared_test">chi-square</a>. Chi-square test is used to decide if two categorical data are related or unrelated (it’s called “dependent” or “independent” in statistics). For example, if you want to compare the preferences between different user groups, then you conduct chi-square test to see if the difference between user groups are significant or not. If you use Excel, you can use the function “CHITEST” to do the chi-square test easily.</p><h4>For Ordinal Data — Frequency, Chi-square, Wilcoxon Rank Sum Test</h4><p>Looking at frequencies is a common way to analyze ordinal data. For example, 20% of users rated the design excellent, 40% of users rated good, and so on. The differences between each order are meaningless, so <em>you can’t calculate average ratings</em>. For more advanced analysis, there is a method called Wilcoxon Rank Sum Test which is used to compare ordinal data. Wilcoxon test can be done using Excel, but it would be easier if you use a programming language such as R.</p><h4>For Interval data — All descriptive statistics, T-Test, ANOVA, Correlation</h4><p>Interval data allow you to use wide range of descriptive statistics, such as average and standard deviation. Also, you can use some inferential statistics to derive a general conclusion that applies to larger population, not limited to your test participants. One of the most common ways to analyze interval data is comparing the means(averages), using T-test or ANOVA. T-test is used to compare two samples; ANOVA is used to compare three or more samples. If you use Excel, you can simply use the function “TTEST” for T-test. It’s also possible to do <a href="http://www.excel-easy.com/examples/anova.html">ANOVA with Excel</a>, using Analysis ToolPak add-in.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rkp1tnAJZlMqEaiFwNyxWw.png" /><figcaption>Scatterplot with trend line</figcaption></figure><p>Another useful way to analyze interval data is looking at relationship between different variables. The chart above is an example of scatterplot with trend line, showing the correlation between two variables. You can calculate <a href="https://en.wikipedia.org/wiki/Correlation_coefficient">correlation coefficient</a> to see how the two variables are correlated. Excel has “CORREL” function to calculate correlation coefficient, as well as chart functions to draw scatterplot with trend line, including <a href="https://en.wikipedia.org/wiki/Coefficient_of_determination">r-squared</a> value that shows how strongly the values are correlated (r-squared is simply the square of the correlation coefficient).</p><h4>For Ratio Data — All descriptive statistics (including geometric means), T-Test, ANOVA, Correlation</h4><p>There are not much difference between interval data and ratio data, and all the statistics that are used for interval data can also be used for ratio data. One difference is that you can use <a href="https://en.wikipedia.org/wiki/Geometric_mean">geometric mean</a> for ratio data. Geometric mean is another way to calculate average, which is useful in measuring differences in time.</p><h3>How many Participants do you need?</h3><p>The number of participants needed for a research depends on the goals of your research and your tolerance for a margin of error. Generally, you need less participant in the first stages of the design and development, while you need more participant in the later stages to find remaining issues.</p><p>For a quantitative research, what you need to consider is how much statistical errors you can tolerate. When you do the research with fewer participants, your data tend to contain more statistical errors. When you do the research with many participants, your data tend to be closer to the true population. That’s why <a href="https://en.wikipedia.org/wiki/Confidence_interval">confidence interval</a> is important. Confidence interval is an estimate of a range of values that includes the true population value for a statistic, such as a mean. You decide what level of confidence you need, such as 90% or 95%, and calculate the confidence interval to show how accurate the measures actually are.</p><p>This article quickly introduced the basic ideas about data and statistics, and some terms might have sounded too technical. But you don’t need to do all the intimidating statistics, of course. You can start by gathering simple data, such as counting task success or asking one question (e.g. “how easy was the task?”) while doing qualitative usability test. Try them out and iterate the process to make them work better. Quantitative data and statistics are not very intuitive for most people, but they can help a lot to improve usability and user experience.</p><p>(Reference: <a href="https://www.amazon.com/dp/0124157815">Measuring the User Experience</a> by Tom Tullis and Bill Albert)</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fupscri.be%2Ff%2F50d69a%3Fas_embed%3Dtrue&amp;dntp=1&amp;display_name=Upscribe&amp;url=https%3A%2F%2Fupscri.be%2F50d69a%2F&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=upscri" width="800" height="400" frameborder="0" scrolling="no"><a href="https://medium.com/media/05d5fd32eda31cbd1b83287606744532/href">https://medium.com/media/05d5fd32eda31cbd1b83287606744532/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6a798b41693b" width="1" height="1" alt=""><hr><p><a href="https://uxdesign.cc/data-basics-for-ux-people-6a798b41693b">Data basics for UX people</a> was originally published in <a href="https://uxdesign.cc/">UX Collective</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The German Tradition of Holistic Design and its Relations with Today’s UX Culture]]></title>
            <link>https://blog.prototypr.io/german-tradition-of-holistic-design-and-its-relations-with-todays-ux-culture-346ee1c00a90?source=rss-9a14543fcb5e------2</link>
            <guid isPermaLink="false">https://medium.com/p/346ee1c00a90</guid>
            <category><![CDATA[germany]]></category>
            <category><![CDATA[education]]></category>
            <category><![CDATA[design]]></category>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[ux-design]]></category>
            <dc:creator><![CDATA[Lico Takahashi]]></dc:creator>
            <pubDate>Mon, 18 Dec 2017 09:12:20 GMT</pubDate>
            <atom:updated>2017-12-26T08:00:33.975Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*q5StViSjl4qba3AId8h6gw.jpeg" /></figure><p>Studying UX at a graduate school is not always a comfortable experience. After I finished the hectic first semester, I was having a kind of mental breakdown. <em>What am I studying here? What kind of skills have I learned? I’m doing interviews, creating personas and storyboards, analyzing tasks, making primitive paper prototypes, and testing them with users— but can I offer a “design” solution by doing such things? Am I maybe in a totally wrong place?</em></p><p>Everything I study here seemed overly basic and general, lacking tangible knowledge or skills. Interdisciplinary approaches seemed to generate confusions rather than solutions. Even those colorful Post-Its seemed so lame to me (I admit I watched <a href="http://99u.com/videos/55967/natasha-jen-design-thinking-is-bullshit">the talk by Natasha Jen</a> more than once). I could imagine myself, a few years from now, suffering from a serious <a href="http://mentalfloss.com/article/75699/what-imposter-syndrome-and-what-can-you-do-about-it">imposter syndrome</a>. Everyone in the UX community seemed to have cool graphic design skills, and to know how to create sophisticated user interfaces using a variety of expensive tools. After all, that’s why they are called UX <em>designers</em>, isn’t it?</p><p>I was in this mist of uncertainty when I came across the name of Max Bense, a German philosopher who taught at the <a href="https://en.wikipedia.org/wiki/Ulm_School_of_Design">Ulm School of Design</a>, a short-lived but highly influential school which stemmed from the famous German design school <a href="https://en.wikipedia.org/wiki/Bauhaus">Bauhaus</a>. Learning about Bense’s innovative approaches opened me up to a new perspective on design.</p><h3>Ulm School and its Holistic Approach to Design</h3><p>Ulm School of Design was founded in 1953 in the city of Ulm, located in southern Germany. Founded by former students of Bauhaus, the school had a unique and clear vision of design, that was “no ornaments, no styling, few color, simple geometrical forms and a very puritan way of designing products, visual communication and buildings” (<a href="http://www.idemployee.id.tue.nl/g.w.m.rauterberg/conferences/CD_doNotOpen/ADC/final_paper/025.pdf">Buerdek 2002</a>). The <a href="https://www.bauhaus100.de/en/past/works/design-classics/ulmer-hocker/">Ulm Stool</a>, designed by one of the founders Max Bill, embodies the idea perfectly.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/450/1*viJYh_p8VwxMkxWzvcXGZg.jpeg" /><figcaption>Ulm Stool, a simple and sturdy furniture suitable for various purposes. (image from NOVA68.com)</figcaption></figure><p>Although the Ulm School of Design’s most renowned departments are product design and visual communication, there existed a small but influential department that contributed to the unique education of the entire school. It was called <em>the Information Department</em>.</p><p>To understand the origin of the Information Department, it is useful to look at some pre-history of the school.</p><blockquote>The plans for the Ulm School were driven by an anti-fascist, democratic impetus. Inge Scholl, the later initiator and founder of [Ulm School of Design], and her younger siblings Hans and Sophie had initially been engaged in the Hitler Youth in Ulm. However, with the beginning of World War II, Hans and Sophie Scholl started to turn away from Nazism and eventually became active resistors in the ‘White Rose’ circle.(<a href="http://www.proceedings.blucher.com.br/article-details/the-information-department-at-the-ulm-school-of-design-8554">Oswald 2012</a>)</blockquote><p>For the founders, it was an urgent mission “to rescue the the concept of <em>Industriekultur</em> from Nazi corruption by regrounding it with humanist tradition of social responsibility and moral education” (<a href="http://www.jstor.org/stable/1511852">Betts 1998</a>). That’s why they included politics and journalism in their holistic sets of faculties.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*D_O-fhJ6Od3JyuFzLTGTRA.png" /><figcaption>A diagram that illustrates the school’s holistic approach to design. (source: Scholl, I. &amp; Aicher, O. ca. 1951. School of design — Research Institute of Product Form. Undated script)</figcaption></figure><h3>Bense’s Ambition to Measure the Beauty</h3><p>The Information Department was originally supposed to teach journalism and advertisement. Max Bill, a hugely influential designer and one of the founders of Ulm School of Design, asked the lecturers to teach “sharp advertising copy and slogans” (<a href="https://www.amazon.de/dp/3518802305/">Schmidt 2008</a>). However, the lecturers refused to do it. Arno Schmidt, an avant-garde writer who was expected to lead the department, quit after a hefty quarrel with Max Bill, practically saying “Do not even think I will sell your nicely designed toilet lids” (Oswald 2012).</p><p>From around 1955, Max Bense began taking clear initiative of the department. Though he was only a part-time department head, since he was working as a full professor of philosophy at Stuttgart University, he took responsibility for the curriculum and published his ambitious plans for the department. He proposed 30 subjects including:</p><ul><li>Conversion of natural languages and artificial languages into precise languages.</li><li>Experiments on grid systems, shortening techniques and montage techniques.</li><li>Concentration and dispersation of form and topics.</li><li>Syntactic and semantic shortening, compression, distortion, lengthening, alienation.</li><li>Accidental and attributive descriptions, phenomenological reduction and deflation of meaning.</li></ul><p>What Bense tried to do was to bring science into aesthetics of visual and verbal communication. In particular, he believed that the value of a text should be evaluated scientifically, by precise analytic means such as statistics, logic, and syntactics. “Just like artistic painting was regarded useless for visual communication, an artistic approach to text was considered useless for contemporary verbal communication” (Oswald 2012).</p><p>Bense was too busy to teach all the subjects because of his job in Stuttgart, so he recruited his colleagues to teach in Ulm. Bense’s colleagues began teaching not only in the Information Department but also in shared subjects for students of all departments of Ulm School of Design. Oswald argues that it was this influx that formed “the schools tendency towards a greater integration of scientific methods into the design process”.</p><h3>Is Design Art or Science?</h3><p>Considering the time background, it would be reasonable to assume that Bense was highly influenced by Claude Shannon’s <a href="https://en.wikipedia.org/wiki/Information_theory">Information Theory</a>. Shannon’s landmark paper, <a href="https://en.wikipedia.org/wiki/A_Mathematical_Theory_of_Communication"><em>A Mathematical Theory of Communication</em></a>, was published in 1948, shortly before Ulm School of Design was founded.</p><p>Information theory is essentially the mathematics behind the digital revolution. Computer science, modern telecommunications and cryptography all derived from the theory. A physicist and philosopher James Owen Weatherall describes the impact of information theory as follows:</p><blockquote>“Shannon’s groundbreaking idea was that you could study the information itself — the stuff that’s carried by the light waves from objects in the world to your retinas, or the stuff that passes from one person to another when they speak — independently of the waves and the words. It is hard to overstate how important this idea would become.” (<a href="https://www.amazon.com/Physics-Wall-Street-Predicting-Unpredictable/dp/0544112431/ref=sr_1_1?ie=UTF8&amp;qid=1513506761&amp;sr=8-1&amp;keywords=physics+of+wall+street">Weatherall 2012</a>)</blockquote><p>In Shannon’s information theory, the amount of information is evaluated statistically. The basic idea is that “the amount of information carried by a signal has something to do with how easy it is for the receiver to decode, or in other words, on how unpredictable the signal is” (ibid.). An information-sparse signal can be decoded easily even when it is delivered in a lot of noise, because it is highly predictable (e.g. conversation about today’s weather, or complaints about traffic). On the other hand, an information-rich signal can not be decoded or predicted easily (e.g. details of a new film the receiver hasn’t seen, or an explanation of a new scientific discovery).</p><p>You don’t have to understand the details of Shannon’s theory, but the important thing is that he changed the basic understandings of information and communication. Information became independent from its containers, and it became measurable by means of mathematics. And that was basically what Bense aspired in the context of design — breaking down the design into pure mathematical elements, and measuring them numerically and statistically.</p><p>Did Bense succeed? Well, we can’t say that he solved design problems by his information approach. Oswald says, “the proposed quantitative analysis of art and design artefacts did not show relevance, neither for ‘beauty’ nor for what now is called usability”. However, the stir he caused in design education has spread beyond disciplines, in and out of the country. His unique approach attracted a lot of controversy about the relationship<em> </em>between design and science. “The result is valid until today: Design cannot be completely absorbed in science, and not at all in ‘artistic intuition’” .</p><h3>Today’s UX Design and Education</h3><p>Today, in Germany and in other countries, UX design and Human Computer Interaction are taught in highly interdisciplinary ways, just like Ulm School of Design. In the context of user experience, the word “design” has become to be used in a much broader sense than its visual expressions.</p><p>Still, there are confusions and misunderstandings that the outputs of a designer’s work should be visual expressions, such as web designs or (visible part of) user interfaces. We need to make it clear that graphics is not the core of design; it is just an instrument.</p><blockquote>If we really want to educate <em>communication </em>designers, as opposed to <em>graphic </em>designers, then we should deal with all aspects of communication — visual and verbal. Otherwise design will continue to offer superficial auxiliary services. (Oswald 2012)</blockquote><p>We design communication itself, not the vehicles of it. We design for people, instead of designing for materials. It doesn’t matter if you use paper, graphic softwares, cardboards, plastics, or even <em>words</em>. As long as you try to create better communication, and tackle the problem with interdisciplinary ways rather than purely aesthetic (or purely scientific) ways, you are absolutely a designer.</p><p>In the age of new user interfaces such as Amazon Echo and Google Home, Oculus Rift and Microsoft Hololens, designing communication and experience is becoming more and more important compared to conventional graphical approaches. Bense might have been too early to come — today his concept of designing information seems to make much more sense. Design theory and practice have evolved since then, and we have gotten large variety of tools from interviewing and observation to testing and evaluation. Each tool might seem fragile, as it is mainly based not on science but on human aspects. Nevertheless, by combining tools from different disciplines, we can design a rich and comprehensive user experience in virtually any context. And that is why I study human-centered methods and processes every day.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fupscri.be%2Ff51076%3Fas_embed%3Dtrue&amp;dntp=1&amp;url=https%3A%2F%2Fupscri.be%2Ff51076%2F&amp;image=http%3A%2F%2Fapi.screenshotlayer.com%2Fapi%2Fcapture%3Faccess_key%3Dfe59908dad3baab69ffab249a2224b03%26viewport%3D1024x612%26width%3D1000%26url%3Dhttps%253A%252F%252Fupscri.be%252Ff51076%253Fscreenshot&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=upscri" width="800" height="400" frameborder="0" scrolling="no"><a href="https://medium.com/media/b85dfbb5286d8a25cf2e754b9462cf45/href">https://medium.com/media/b85dfbb5286d8a25cf2e754b9462cf45/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=346ee1c00a90" width="1" height="1" alt=""><hr><p><a href="https://blog.prototypr.io/german-tradition-of-holistic-design-and-its-relations-with-todays-ux-culture-346ee1c00a90">The German Tradition of Holistic Design and its Relations with Today’s UX Culture</a> was originally published in <a href="https://blog.prototypr.io/">Prototypr</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to Listen to Your Users like a Psychologist]]></title>
            <link>https://blog.prototypr.io/how-to-listen-to-your-users-like-a-psychologist-5bfe01f466aa?source=rss-9a14543fcb5e------2</link>
            <guid isPermaLink="false">https://medium.com/p/5bfe01f466aa</guid>
            <category><![CDATA[design-thinking]]></category>
            <category><![CDATA[user-experience]]></category>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[psychology]]></category>
            <dc:creator><![CDATA[Lico Takahashi]]></dc:creator>
            <pubDate>Sat, 11 Nov 2017 17:17:25 GMT</pubDate>
            <atom:updated>2017-11-13T10:32:39.179Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xJRLdfKzeq3WU20CHUnHUw.png" /></figure><p>Empathy is the key to a successful Design Thinking process. If you are a designer, you are probably familiar with user interviews and other user-centered methods. User interviews give you many valuable insights, especially when you are trying to specify user needs and requirements for a new product. However, listening to users is not as simple a task as it may seem. Remember the last time you talked to your boss about your concerns, or you called customer service to complain about poor performance of a product. Did they understand your point? Did you feel like you were really heard?</p><blockquote>Empathy is at the heart of design. Without the understanding of what others see, feel, and experience, design is a pointless task. — Tim Brown</blockquote><p>Listening is not an inherent ability, but a skill that you must practice and learn. It’s like singing or acting. You might be able to sing without practice, but you must practice if you want to be a good singer. And just like singing, there are some techniques that help you improve your listening skills. That’s why psychologists are better listeners than most of us. They are trained to listen properly, through education and practice.<strong> If you listen like a psychologist, you will be able to empathize better and understand the deepest needs and requirements</strong> that users are trying to communicate.</p><p>I am not a psychologist, but I studied clinical psychology for 4 years and received bachelor’s degree from a top university. Here are some knowledge I learned from prominent psychologists about professional listening.</p><h3>1. Accept Radically</h3><p>A user interview is an unnatural situation to be in. You are a stranger who comes in and asks about a user’s work and life. It’s no wonder that he or she becomes protective and says only nice things. To find out true feelings and opinions, you should show that you are accepting the user radically.</p><p><em>Radical acceptance</em> is the word often used in the context of mindfulness. You accept the people and things as they are, without any judgement at all. Expectation is the biggest obstacle for radical acceptance, because it would cause joy or disappointment according to the results. Users will notice when you are disappointed in your heart. So, don’t expect particular answers, be curious and open to any kind of answers. After all, you are here to find out user’s point of view, not to prove your hypothesis.</p><h3>2. Be Open to Wonders</h3><p>You never know what other people think. You can’t see what they are experiencing in their mind. It’s a mystery for you. However, we all tend to forget the fact and jump to false conclusions that match your assumptions. Suppose you are interviewing an old lady, and she says she doesn’t like Facebook.</p><blockquote>“I don’t use Facebook a lot. It’s not for me.”</blockquote><blockquote>“I understand. My mom says social media makes her crazy.”</blockquote><blockquote>“Well, it’s just…the communication there. It’s not quite my style.”</blockquote><blockquote>“Ah, I know. Actually, I prefer talking in person, too.”</blockquote><blockquote>“Oh, I don’t like talking in person, I just prefer Twitter over Facebook!”</blockquote><p>In this case the lady is very honest and tells you what she really thinks, but in real life that doesn’t happen a lot. The user will think you are a bad listener and give up talking about it.</p><h3>3. Don’t Ask “Why”</h3><p>You might have been taught to ask “why” a lot. While it is crucial to understand users’ motivations, it could be better if you ask differently. “Why” is a difficult question and not many people have clear answers to it. Besides, some users might find the question a little bit offensive, like they are being criticized. For example:</p><blockquote>“I don’t like school.”</blockquote><blockquote>“Why don’t you like school?”</blockquote><blockquote>“I don’t know. I just don’t!”</blockquote><p>Let’s think about a different way to ask. Instead of asking “why”, you can use another question word, such as “when” and “where”.</p><blockquote>“I don’t like school.”</blockquote><blockquote>“When do you hate school most? In what situation?”</blockquote><blockquote>“Lunch time. I don’t want to eat alone.”</blockquote><p>In this way you will get much more information, and hopefully be able to find a path to “why”.</p><h3>4. Listen to Meta-Message</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*aOoO4fxze-Qe9AeKgJ-XlA.png" /></figure><p>A users’ physical voice sometimes tells you different stories from the words she is speaking. One might say “I like it” in a flat tone, or in a cheerful and excited voice. Such non-verbal information is called <em>meta-message</em>. When you conduct an onsite interview, the volume of information is much larger than when you read an E-mail or a written questionnaire. That is a great benefit of onsite user interviews, and you should take advantage of it.</p><p>There are several channels of meta-message along with physical voice. Facial expressions, gestures, posture and pace of speech are all telling you something about the user’s mind. Observe carefully and try to understand the true meaning behind the user’s words.</p><h3>5. Summarize Briefly</h3><p>Summarizing is a great way to check your understanding during an interview. It is also good for reassuring the user. Talking to someone who just nods and asks next question is sometimes quite uncomfortable. The user will be relieved if you give her a short summary of what you’ve just heard. It will give the user the confirmation that you understand correctly, and you are actually listening. And you can give her an opportunity to correct you, if you misunderstood something. Summarization will clarify what you understood, how you understood it and where you paid attention while listening.</p><p>Try to keep the summary very short and simple. The user is here to talk, not to listen to you. Keep in mind the summarization is for the purpose of listening.</p><h3>And…Let’s Practice</h3><p>As I’ve already said, listening is a skill that needs to be practiced and developed. You can’t become a good listener without listening, just as you can’t become a good singer without singing a lot. So let’s practice listening. You can listen carefully, not only when you conduct user interviews, but also in office and at home. Listen to your colleagues, friends and family members. It will help to improve your personal relationships too!</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fupscri.be%2Ff51076%3Fas_embed%3Dtrue&amp;dntp=1&amp;url=https%3A%2F%2Fupscri.be%2Ff51076%2F&amp;image=http%3A%2F%2Fapi.screenshotlayer.com%2Fapi%2Fcapture%3Faccess_key%3Dfe59908dad3baab69ffab249a2224b03%26viewport%3D1024x612%26width%3D1000%26url%3Dhttps%253A%252F%252Fupscri.be%252Ff51076%253Fscreenshot&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=upscri" width="800" height="400" frameborder="0" scrolling="no"><a href="https://medium.com/media/b85dfbb5286d8a25cf2e754b9462cf45/href">https://medium.com/media/b85dfbb5286d8a25cf2e754b9462cf45/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5bfe01f466aa" width="1" height="1" alt=""><hr><p><a href="https://blog.prototypr.io/how-to-listen-to-your-users-like-a-psychologist-5bfe01f466aa">How to Listen to Your Users like a Psychologist</a> was originally published in <a href="https://blog.prototypr.io/">Prototypr</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AI and Human-Centered Design: What’s the Future?]]></title>
            <link>https://uxdesign.cc/ai-and-human-centered-design-whats-the-future-5c88f523c07a?source=rss-9a14543fcb5e------2</link>
            <guid isPermaLink="false">https://medium.com/p/5c88f523c07a</guid>
            <category><![CDATA[future-of-work]]></category>
            <category><![CDATA[design]]></category>
            <category><![CDATA[human-centered-design]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Lico Takahashi]]></dc:creator>
            <pubDate>Sat, 28 Oct 2017 14:23:36 GMT</pubDate>
            <atom:updated>2017-11-07T01:01:22.980Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EtfGsMe22tlI_1rwSd-vhQ.jpeg" /></figure><p>IDEO, one of the world’s most influential design firms, <a href="https://medium.com/ideo-stories/design-for-augmented-intelligence-9685c4db6fbb">announced</a> that it acquired a Chicago-based data science company <a href="https://datascopeanalytics.com/">Datascope</a>. You might have wondered: will computers replace designers? Will Human-Centered Design give way to Computational Design? As an MSc student majoring in Usability and Human-Computer Interaction, I’d like to discuss how AI would transform Human-Centered Design in the near future.</p><h3>Adobe Sensei blows our minds</h3><p>During the Adobe MAX 2017 conference, Adobe showed off some future visions of Sensei, an Artificial Intelligence built into the company’s most popular design tools such as Photoshop. The demonstration was amazing; Sensei looked at a rough sketch, then automatically found photos that match the topic of the sketch, with faces sorted by direction and ready to be combined into a professional-looking movie poster. The speaker, David Nuescheler, concluded his presentation by saying that they are trying to make “<em>innovation that lets you focus on the creative work you love, without the technology getting into your way.”</em></p><p>His words capture the core of design thinking. After all, human should not be struggling with inessential tasks such as hunting for images that fit your topic or aligning and cropping images so that they look nice and tidy. Technology should make people happy, instead of wearing them out.</p><h3>Human-Centered or AI-Centered?</h3><p>Adobe Sensei will liberate designers from miscellaneous tasks and let them focus on more creative work. And the story doesn’t end there. In the near future, AI will enhance designers’ workflow and assist the decision-making process itself. Designers will not have to struggle with tiresome activities of endless prototyping and testing (let’s be honest…it is exhausting, isn’t it?) Instead, they can focus on the most “human” aspects of design.</p><p>AI does not kill Human-Centered Design; on the contrary, it will make design much more human-centered than it already is.</p><h3>AI and Elements of Human-Centered Design</h3><p>International Organization for Standardization(ISO) defines Human-Centered Design as follows.</p><blockquote>Human-centred design is an approach to interactive systems development that aims to make systems usable and useful by focusing on the users, their needs and requirements, and by applying human factors/ergonomics, and usability knowledge and techniques.</blockquote><blockquote>(<a href="https://www.iso.org/standard/52075.html">ISO 9241–210</a>:2010 Human-centred design for interactive systems)</blockquote><p>In a nutshell, Human-Centered Design is about creating what people really want, in a way people want to use. To achieve that goal, we designers employ four main activities.</p><ul><li>Understanding and specifying the context of use</li><li>Specifying the user requirements</li><li>Producing design solutions</li><li>Evaluating the design</li></ul><p>Let’s look at the activities one by one, and think about what AI can bring to each activity.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/784/1*k2nvHQ4z7Qb5EeWthKOv_A.png" /><figcaption>Human-centred design activities (ISO 9241–210)</figcaption></figure><h4>1. Understanding and specifying the context of use</h4><p>This activity is about grasping people’s needs in the real world. For example, you observe the workplace to find out what’s happening there, and conduct interviews to find out what the workers want or need.</p><p>AI will help this activity by providing rich data which is automatically tagged and organized in a way designers can understand and handle with ease. Suppose you are replacing a communication software in a big company. Instead of observing and analyzing the usage of current software, you can just provide bunch of data to AI, and let AI predict what people want. AI will analyze people’s common behaviors and mistakes, and provide possible ways that can improve user experience.</p><h4>2. Specifying the user requirements</h4><p>After understanding the context of use, you specify the functional and other requirements for the product or service. This is about translating people’s needs into specifications, such as features, functions and required performance criteria.</p><p>AI will help this activity by automatically extracting possible requirements from research reports, past cases and relevant user interface knowledge. It will also detect and resolve potential conflicts between user requirements, and create all the test cases instantly.</p><h4>3. Producing design solutions</h4><p>Once you specified user requirements, you create design solutions that would fulfill all the requirements. This activity includes following tasks:</p><ul><li>designing user tasks, user-system interaction and user interface to meet user requirements, taking into consideration the whole user experience</li><li>making the design solutions more concrete (for example making use of scenarios, simulations, prototypes or mock-ups)</li><li>altering the design solutions in response to user-centred evaluation and feedback</li><li>communicating the design solutions to those responsible for their implementation</li></ul><p>AI will help this activity by auto-generating user interface designs according to the requirements, making prototypes for testing, and at the same time delivering documents for communication. Interface will be generated by combining known best solutions, and also by inventing whole new solutions — just like <a href="https://www.nature.com/nature/journal/v550/n7676/full/nature24270.html">AlphaGo Zero</a> found brand-new strategies, without any help of human data, in the traditional game of Go.</p><h4>4. Evaluating the design</h4><p>Evaluation is essential for Human-Centered Design, because it allows us to get user-feedback and to improve our design solutions accordingly. Human-Centered Design is an iterative process at the core, and hence fit well together with AI and machine learning.</p><p>AI will help this activity by monitoring user behavior and collecting rich data, not only during the testing phase but also after the product / service is dispatched. It will detect a user’s level of satisfaction and efficiency from interaction pattern, face expression, eye movement and other signals, and modify the design on the spot. Design modification will become more subtle and continuous, and the iteration cycle will never cease.</p><h3>You dream, AI makes it real</h3><p>Design is decision-making. AI will help design process in many ways, but at the end of the day it’s humans who make decisions. Why? Because design is not for computers, but for humans.</p><p>AI will generate prototypes effectively, test them, collect feedback data and suggest bunch of brilliant ways to improve usability. It could even transform user experience with real-time adjustment and whole new strategy for human-computer interaction.</p><p>Still, AI needs you. Because you are the one who have vision. AI might be able to try and learn much better than you, but you can <em>dream</em> much better than AI. You can decide what you want to create, whom you want to satisfy, and how you want to satisfy them. Using the analogy of driving, your job is not driving a car, but deciding where to go and how to experience the journey.</p><p>And that should be the heart of design.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fupscri.be%2Ff%2F50d69a%3Fas_embed%3Dtrue&amp;dntp=1&amp;display_name=Upscribe&amp;url=https%3A%2F%2Fupscri.be%2F50d69a%2F&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=upscri" width="800" height="400" frameborder="0" scrolling="no"><a href="https://medium.com/media/05d5fd32eda31cbd1b83287606744532/href">https://medium.com/media/05d5fd32eda31cbd1b83287606744532/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5c88f523c07a" width="1" height="1" alt=""><hr><p><a href="https://uxdesign.cc/ai-and-human-centered-design-whats-the-future-5c88f523c07a">AI and Human-Centered Design: What’s the Future?</a> was originally published in <a href="https://uxdesign.cc/">UX Collective</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What I learned in the 4-week Deep Learning course on Coursera]]></title>
            <link>https://becominghuman.ai/what-i-learned-in-the-4-week-deep-learning-course-on-coursera-1c142c2ce40a?source=rss-9a14543fcb5e------2</link>
            <guid isPermaLink="false">https://medium.com/p/1c142c2ce40a</guid>
            <category><![CDATA[online-education]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[deep-learning]]></category>
            <dc:creator><![CDATA[Lico Takahashi]]></dc:creator>
            <pubDate>Mon, 02 Oct 2017 13:44:16 GMT</pubDate>
            <atom:updated>2017-10-02T16:39:13.309Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1T8Vc4hWQ9-k2RvHqR1Vjg.jpeg" /></figure><p>Andrew Ng, the AI Guru, launched new Deep Learning courses on <a href="https://www.coursera.org/">Coursera</a>, the online education website he co-founded. I just finished the first 4-week course of the <a href="https://www.coursera.org/specializations/deep-learning">Deep Learning specialization</a>, and here’s what I learned.</p><h3>My background</h3><p>I happen to have been taking his previous <a href="https://www.coursera.org/learn/machine-learning/">course on Machine Learning</a> when Ng announced the new courses are coming. So it was a natural sequence that I enrolled in the Deep Learning courses. But I was by no means a data science person — actually, I didn’t even know how to multiply matrices or vectors until 2 months ago.</p><p>It was just a curiosity that persuaded me to start machine learning course in the first place. The summer holidays began, I didn’t have much things to do, and an online course seemed to be a good way to spend time during the cool German summer days in a sleepy town in the middle of nowhere.</p><p>Thanks to Andrew Ng’s great teaching skill and the carefully designed assignments, I completed the machine learning course in 5 weeks, with 100% grade. And at that point, I was quite attracted to this area of computer science.</p><h3>What is Deep Learning anyway?</h3><p>Nowadays, people are talking about Deep learning everywhere. But before I started these courses, I had only a vague notion of what it means. So, it’s about computers recognizing cute kittens, isn’t it?</p><p>Well, it might sound a little bit bland, but deep learning is basically a re-branding for the algorithm called <em>neural network</em>. Neural network is originally inspired by human brain, but implemented in a much smaller and simpler way.</p><p>The basic idea is combining many simple units (or <em>neurons</em>) to perform rather complex tasks. One unit calculates something, another unit uses the output from the first one and performs similar calculations, the next one uses the output from the second one, and so on. A neural network is formed in layers, with multiple units in each layer. When there’s only one middle (or <em>hidden</em>) layer, it’s called “shallow” neural network. And when there’s more than two middle layers, it’s called <em>deep neural network,</em> or simply deep learning.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/907/1*R9Vd96dBUU2fJO9dnK0igA.png" /></figure><h3>Week1: Introduction to Deep Learning</h3><p>In the first week of the course, you learn why deep learning is so hot these days. “AI is the new electricity,” Ng says, because it is transforming many fields from the car industry to agriculture to supply-chain. What caught my attention was the fact that deep learning is a highly iterative process, with lots of trial-and-error to fine-tune the models. It could be fun, caring for your models and grow them up. Sounds a bit like a simulation game.</p><h3>Week2: Neural Networks Basics</h3><p>In the second week, you learn how to build a logistic regression model, structured as a shallow neural network. It would be helpful if you have a basic understanding of calculus and/or machine learning, but the lectures are for beginners and not so hard to follow. I like the way Ng says “If you don’t understand the calculus, don’t worry!”</p><p>They also give you a tutorial of python/numpy, and some exercise of <em>vectorization</em>, which is a must in the deep learning era, to speed up the calculation. And the best part is the programming assignment — you implement a logistic regression model to classify <em>cat vs. non-cat</em>, with bunch of lovely cat pictures.</p><h3>Week3: Shallow Neural Networks</h3><p>This week you implement a shallow neural network. You initialize parameters, do forward-propagation, compute cost, do backward-propagation to get all the derivatives, update parameters gradually, put all the functions into a neural network model, and see how it works with rather complex classification tasks. If you don’t understand any of these words, don’t worry about it, you will learn pretty quickly with the friendly lectures and practical hands-on assignments.</p><h3>Week4: Deep Neural Networks</h3><p>It’s time to embark on deep neural networks. You implement all the functions of the deep learning, and train your models for the <em>cat vs. non-cat </em>image classification. Yes, lots of cute cats again. After successfully trained your deep neural network model, you can try it with your own cat picture. I was so happy when my deep neural network recognized my beautiful cat!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/667/1*gtGQFOGoTXFSdwCH86WUUQ.png" /><figcaption>My Deep Neural Network knows this is a cat!</figcaption></figure><h3>What’s Next?</h3><p>Now I learned how to implement a deep neural network, but only in a very basic way. I don’t know how to prevent over-fitting, how to choose hyper-parameters, or how to optimize my models effectively. These practical aspects come in the second course of the Deep Learning specialization.</p><p>Before proceeding to the next course, I got curious if I can implement a similar project by myself. So I decided to make a image classifier, with some pictures other than cats. I’ll write about my first attempt to make my own deep-learning project in the upcoming posts.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fupscri.be%2F8f5f8b%3Fas_embed%3Dtrue&amp;dntp=1&amp;url=https%3A%2F%2Fupscri.be%2F8f5f8b&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=upscri" width="800" height="400" frameborder="0" scrolling="no"><a href="https://medium.com/media/c43026df6fee7cdb1aab8aaf916125ea/href">https://medium.com/media/c43026df6fee7cdb1aab8aaf916125ea/href</a></iframe><figure><a href="https://becominghuman.ai/artificial-intelligence-communities-c305f28e674c"><img alt="" src="https://cdn-images-1.medium.com/max/255/1*2f7OqE2AJK1KSrhkmD9ZMw.png" /></a></figure><figure><a href="https://upscri.be/8f5f8b"><img alt="" src="https://cdn-images-1.medium.com/max/255/1*v-PpfkSWHbvlWWamSVHHWg.png" /></a></figure><figure><a href="https://becominghuman.ai/write-for-us-48270209de63"><img alt="" src="https://cdn-images-1.medium.com/max/255/1*Wt2auqISiEAOZxJ-I7brDQ.png" /></a></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1c142c2ce40a" width="1" height="1" alt=""><hr><p><a href="https://becominghuman.ai/what-i-learned-in-the-4-week-deep-learning-course-on-coursera-1c142c2ce40a">What I learned in the 4-week Deep Learning course on Coursera</a> was originally published in <a href="https://becominghuman.ai/">Becoming Human: Artificial Intelligence Magazine</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>