<?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 Joyce Wang on Medium]]></title>
        <description><![CDATA[Stories by Joyce Wang on Medium]]></description>
        <link>https://medium.com/@wangjinghan2020?source=rss-456c682c8ca9------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*gib4LlAmqEKrvGFZ.jpg</url>
            <title>Stories by Joyce Wang on Medium</title>
            <link>https://medium.com/@wangjinghan2020?source=rss-456c682c8ca9------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 02:45:29 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@wangjinghan2020/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[Classical Chinese Garden Multiverse]]></title>
            <link>https://medium.com/@wangjinghan2020/classical-chinese-garden-multiverse-46640aaada81?source=rss-456c682c8ca9------2</link>
            <guid isPermaLink="false">https://medium.com/p/46640aaada81</guid>
            <category><![CDATA[generative-design]]></category>
            <category><![CDATA[grasshopper-3d]]></category>
            <category><![CDATA[computational-design]]></category>
            <dc:creator><![CDATA[Joyce Wang]]></dc:creator>
            <pubDate>Sat, 11 May 2024 03:07:53 GMT</pubDate>
            <atom:updated>2024-05-11T03:07:53.350Z</atom:updated>
            <content:encoded><![CDATA[<h4>Optimizing field of view for classical chinese garden configurations</h4><p><em>Yilin Wang, Heqiao Meng, Joyce Jinghan Wang, Mingyue Zhang</em></p><p><em>Columbia GSAPP | Generative Design | Spring 2024</em></p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FDvykq8dIWjo%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DDvykq8dIWjo&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FDvykq8dIWjo%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/edeb008b7f6517cb8ac626841318b4aa/href">https://medium.com/media/edeb008b7f6517cb8ac626841318b4aa/href</a></iframe><blockquote><strong><em>Introduction</em></strong></blockquote><p>Classical Chinese gardens have their own distinct aesthetic interest. Unlike the patterning of western gardens, Chinese gardens emphasize on the sense of experience and artistic conception. One’s own journey during the exploration creates the intricate visual and emotional changes that multiply the meaning of the garden culture, since there is no single way of exploring a garden. Also, we want to challenge the subjectivity of scenery creation by using objective generational tools to obtain more possibilities. Therefore, we would like to seek a pivot point for landscape designers between the subjective creation and the objective aesthetic principles, designers can use it to generate more possible layout configuration and then use genetic algorithms to find the optimal one.</p><blockquote><strong><em>Methodology</em></strong></blockquote><p><strong>Design Space Model</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Rdk5n3wEz0lkd5cTeD9RWQ.png" /><figcaption>Neutral Plan</figcaption></figure><p>Although this is a method that could be applied to all garden designs, we selected the Humble Administrator’s Garden for the experimental study, because its size and topography are typical for classical Chinese gardens. We removed the existing architecture and paths, then generalized the landscape including water and mountains. We kept the old trees, which would be our basis for selecting good views.</p><p><strong>Input Parameters</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VL6lIjv9wvT412lZ9yL_vQ.png" /><figcaption>Generative Logic | part 1</figcaption></figure><p>We divided the site into two areas, where we input two types of architecture on them, because the architecture in classical Chinese gardens has various typologies based on its locations and relation to the site. We input three types of pavilions of panoramic views within the area adjacent to the water, and two types of buildings of partial views to the other area along the bounding wall.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*24p4loc0AyjjktCn6gIifA.gif" /><figcaption>Architecture Schemes</figcaption></figure><p>We use “populate 2d” and randomly distribute the architectures for each area.</p><p><strong>Performance Metrics</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7aevuuwUFj6afJfJ44Okmg.png" /><figcaption>Generative Logic | part 2</figcaption></figure><p>After 20 schemes are generated, we set up criterias for selecting the scheme with the optimized field of view and experience. We used “Isovist” as the main tool to define the field of view, where a good field of view is the one where people see more water and the lengths of sight lines are varied.</p><p>Therefore, we set the Criteria/Fitness Objective 1 to maximize the overlapping area of the Isovist and the water. In each scheme, there are around 20 pavilions, each pavilion has a value for the overlapping area, we calculated the average value of the overlapping areas.</p><p>We set the Criteria/FO 2 to maximize the variance of sight lines. We divide the Isovist perimeter into 30 equal lengths, and connect the division points into the center of the pavilion and form sight lines. Then we calculate the variance of 20 building’s sight lines.</p><p>The Criteria/ FO 3 is the variance of the distances between each two pavilions/buildings among all the architecture. A higher variance means there are more diverse experiences meandering from one pavilion to the next in a path.</p><p><strong>Optimal Scheme</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ap4tb-8ckUWg3AfNXztXxQ.jpeg" /><figcaption>Generative Logic | part 3</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4gLDiBP3Na0PJgeJ9o_CgA.gif" /><figcaption>Wallacei Process</figcaption></figure><p>The optimal scheme of configuration of layout of architecture distribution in the master plan is generated through Wallacei (Multi-Objective Genetic Optimization Algorithm), where the original generator of all those schemes is the Gene in this algorithm to further direct fitness value optimization. A Radar Diagram with 3 above-mentioned Criteria/ Fitness Objectives for each configuration is generated.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/792/1*73K8wIi3_so1IIZAJLJCng.png" /><figcaption>All Schemes</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bHjA9z5fxzUct32Zbrfb9Q.png" /><figcaption>The Optimal Scheme</figcaption></figure><p>Through 10 Genes and 5 individual configurations per Gene, an optimal scheme out of 50 Configurations by the Average Fitness Rank is selected.</p><p><strong>Path</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cAx_oG_gW-6xUksIC9XVMg.gif" /><figcaption>the Optimal scheme with ISOVIST Path</figcaption></figure><p>The path is first discovered and guided through sight, thus we test the optimal configuration with isovist sight from each pavilion.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*BC5RL-bk-EwI_iqwjCXHsA.gif" /><figcaption>PedSim Heatmap</figcaption></figure><p>Through PedSim, a heatmap of the tendency of pedestrian stimulation is generated in the optimal scheme. As a Chinese proverb by the Writer Xun Lu, “There is no path, but when many walk on it, it becomes a path.” Path is a result of the stimulation of Tendency to walk by the people.</p><blockquote><strong><em>Results</em></strong></blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7-FW67_EaucP4g0Ftxt_Kg.jpeg" /><figcaption>Original Plan v.s. Generative Plan</figcaption></figure><p>As a result, we imagined a scheme from a designer’s perspective, drew it based on the optimal architectural configuration and the path tendency simulated by PedSim Heatmap. We define it as a design tool and at the same time, a starting point to assist designers to develop a more detailed plan. We acknowledge that everyone would have different interpretations of the result, so we believe that the tool could help generate more possibilities.</p><blockquote><strong><em>Reflection</em></strong></blockquote><p>There are some design moments that have reached our expectations. For example, the pavilions along the water are good locations that have good views; and the adjacency of certain pavilions create an opportunity to form an architectural cluster that enriches the typology. However, we reckon the optimization has certain limitations. For instance, it has lost the nuance of the orientation of pavilions and the connections between pavilions and the paths. In addition, the paths stimulated by PedSim are more towards efficiency than meandering. Garden is about getting lost, but it is not the tendency of algorithms. Even so, we think such conflicting scenarios are the potential and the value of this project.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=46640aaada81" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>