Lotus Petal Architecture: Link Building with Parametric Geometry


Lotus Petal Architecture combines parametric geometry with hyperlinks to render thousands of clickable Uniform Resource Identifiers (URIs) visible on a single phone, tablet, or computer screen at the same time, with visual cues providing metadata, context, and navigational hints. The result is an interactive, 3D data map, accessible from any browser or mobile device. They require no external software and are compatible with most REST and JSON API’s.

Any type of filter may be applied but the Lotus Chart structure is well suited for machine learning flags and identifying growth trends across a data ecosystem. Applications include securities trading, scientific literacy, epidemiology, real estate, e-commerce, entertainment, market research, online dating, human resources, security, and fraud detection: in short, any occupation or pastime where humans are asked to scan through and make decisions at-a-glance on hundreds or thousands of extremely similar data points each day.

The clickable URIs are referred to as geometric links. Geometric links typically take the form of lines, curves, or geometric shapes. The Information Discovery Module is a system capable of generating a large number of geometric links in 3D space. The module arranges these links to resemble shapes and patterns found in nature, abstract design, and human-made objects, scenes, symbols, and environments. The Information Discovery Module uses specific protocols to interact with representations of an information or content resource over a network, typically the World Wide Web. Use of the Uniform Resource Identifier (URI) convention permits interaction across networks and transforms the Information Discovery Module from a static display into a dynamic navigation system.


A geometric link format confers several advantages:

-Text links are an inefficient use of screen “real estate.” Most search engines and database queries only display between 5 and 50 results at one time. Several thousand unique links can be displayed on-screen at one time when generated and rendered through parametric geometry.

- Geometric links rather than text-based links require no language translation.

- Geometric links can resize easily to fit various aspect ratios and mobile device screens.

- Geometric links can display additional qualities and attributes of the data referenced through visual cues, making it possible to quickly and intuitively make at-a-glance assessments.


At any one time, only a subset of geometric links will be visible, highlighted, and/or clickable. However the user interface of the Information Discovery Module allows the user to easily toggle the selection of active geometric links and preview link content before navigating to a new destination. The user interface may be a graphical user interface containing predefined searches and filters, or it may include a search tool or query builder. Any given query may result in a different number and position of geometric links being activated on the screen. These links can then be previewed via mouseover or automatic scrolling to deliver more detailed information about the link destination.



Lotus Petal Architecture can be viewed in action with two live implementations:

- Technical analysis of altcoin prices at http://dev.lotus.fm/B_U_/examples/index.html (L-DOH)

- Top 400 music videos from Portland, Oregon at https://lotus.fm (Lotus.fm)



Definition of Parametric Geometry:

Generation of geometry representing a parametric surface. (Source: https://threejs.org/docs/#api/en/geometries/ParametricGeometry) Parametric geometry is a subfield of computer graphics in which complex geometric shapes and scenes are generated solely using mathematical equations and parametric inputs. Geometry may be 2D or 3D. Tools used include three.js (JavaScript library for client-side 3D modeling), Grasshopper, and CAD. Lotus Petal Architecture encompasses implementations of parametric geometry, parametric design, procedural modeling, and generative modeling that result in algorithmically generated geometric links. Node.js integration enables real-time, asynchronous update of interactive data maps with no screen refresh required.



The Information Discovery Module may be compared to a front-end for big data. The module allows systems performing complex data analysis and machine learning operations to output a far wider range of results in a format that will be intelligible and relevant for humans. The structure helps users discover the most relevant information, and then takes them where they want to go. The Information Discovery Module is well suited to exploring online content properties, such as music, movies, books, and social media. Because geometric links can be shown more compactly than text links, the Information Discovery Module enables the user to browse through and become aware of a far greater range of content. The technology expands user choice and provides exposure to niche and emerging artists as well as blockbusters and celebrities. Visual cues and filters allow at-a-glance sorting through a more sophisticated selection criteria than simply “likes” and “number of plays.” They enable users to see what is trending in a given geographic area, amongst critics and experts, other artists, and their friends.


Use Cases:

Lotus Petal Architecture takes its name from an Information Discovery Module in the shape of a three-dimensional lotus flower. In this implementation, geometric links are rendered using parametric geometry as three-dimensional curves (“chart lines”) arranged to resemble veins on the surface of a lotus petal. The height and placement of each chart line conveys information about its relative popularity according to a system-defined scoring system that may include such factors as social media follows and “likes,” total number of plays, and other criteria. Color is used to identify active links. Placement of a geometric link on one of the outermost petals indicates lower relative popularity than links situated in the middle and inner petals. A highlighted link may indicate a content property with a strong growth trend. In this invention, machine learning may be combined with parametric geometry to flag and identify geometric links or provide metadata and context for geometric links and other elements of the Information Discovery Module system. For instance, machine learning (artificial intelligence) may be used to flag a hyperlink or hyperlinks pointing to content properties identified as a potential risk for click fraud and/or fake likes and followers.


Additional Use Cases:

“See the forest for the trees.” In this implementation, a three-dimensional computer-generated landscape is populated by parametrically generated trees of different colors, at different heights on a terrain in which some of the coordinates (x, y, and z-axis) are determined at random and others correspond to system data variables. Values such as a tree’s color, bark texture, leaf shape, and overall appearance (“skin”) are also determined by system variables. System variables correspond to location-based environmental data such as rainfall, pollution indexes, and population. Trees are geometric links to libraries of information. The goal of this implementation is to identify the human communities and natural areas most at risk from climate change. Rather than linking to geospatial coordinates on a map, “trees” link to online URIs or collections of URIs containing information relevant for specific communities and ecological zones. By eliminating the primary constraint of geography, the user is able to view climate information across national boundaries and observe how cities, towns, countryside, and natural areas on opposite sides of the earth are impacted by the same planetary trends.

Search Engine Results. In this implementation, the module is represented by a geometric solid, such as a sphere or a circle. Search results are organized into color-coded sectors. Colors may reflect one or more brand colors of the sponsoring organization, and represent a given type or category of search results (e.g web links, images, news stories, social media, or e-commerce products). Placement of search results towards the center or periphery of the shape may indicate relative popularity. Users can mouse over the different categories and see previews of different URIs. They can also turn and rotate the module representation and zoom in or zoom out. If the module representation is three-dimensional rather than two-dimensional, the user will be able to preview hundreds or thousands of link results at one time rather than five or ten. All this is possible without requiring a single screen refresh.

Fractal parametric design of an Information Discovery Module, in which dynamic updates and a fractal design structure are used to order a limitless amount of data in a finite 3D space.

Virtual reality and augmented reality implementations of the Information Discovery Module.

Modifications to Information Discovery Modules and specialized controls for the purposes of accessibility, in particular accessibility for low-vision and color-blind individuals.

Information Discovery Modules designed and generated by artificial intelligence systems.


Room 104
Sunday, March 8, 2020 - 16:30 to 17:30