loveyudan-blog

[loveyudan-blog]

Image from Multispecies Cat’s Cradle, Nasser Mufti, 2011

String and Storytelling

Storytelling, also the ability to read it are a kind of markers of civilization, great stories can have great lives in the oral tradition. Mules and Men by Zora Neale Hurston, was a collection of folklore that lived in the oral tradition. The same can be The Odyssey. But we privilege reading and storytelling because they allow us to communication directly and transparently with people who live very far away from us, and they also allow us to kind of hear the voices of the dead. Stories are about communication.

We didn’t invent grammar so that your life would be miserable in grade school as you attempted to learn what the mark preposition is. We invented grammar because without prepositions, we couldn’t describe what it’s like to fly through a cloud, or jump over a puddle. If I’m doing my job, and you’re doing your job, you aren’t thinking about the fact that I’m contorting my mouth and tongue and vocal chords to create sounds that then exist as ideas in your brain. It’s just happening. But if my language gets confusing: if I made incorrect word order use, then  I made a barrier between you and me.

Storytelling at least good storytelling is an outgrowth of that urge to use language to communicate complex ideas and experiences between people. And that’s true whether you’re reading Shakespeare or bad vampire fiction, reading is always an act of empathy, it’s always an imagining of what it’s like to be someone else. So when Shakespeare uses iambic pentameter, or Salinger uses a red hunting cap, they aren’t doing this so that your English teachers will have something to torture you with. They’re doing it, at least if they’re doing it on purpose, so the story can have a bigger and better life in your mind. But for the records, the author is not that important. Whether an author intended a symbolic resonance to exist in her book is irrelevant. All that matters is whether it’s there because the book does not exist for the benefit of the author. The book exists for the benefit of reader. If we as readers could have a bigger and richer experience with the world as a result of reading a symbol and that symbol wasn’t intended by the author, we still win. Reading is a conversation between an author and a reader, but give yourself some power in that conversation. Go out there and make a world.

It’s extremely hard to get other people to feel what we are feeling. Using the techniques of hyperbole and metaphor, to try to describe the things that are happening insider of character. Because they are not using particularly compelling or original figurative language, readers may struggle to empathize with. They are trying to  communicate far more complicated and nuanced experiences and emotions, trying to talk to strangers, some of whom may live very far away and in fact live centuries after your death. They can’t hear the author intonation, that is the challenge author faces. Understanding language, you will have a fuller understanding of lives other than your own, which will help you to be more empathetic, although more importantly reading critically and attentively can give you the linguistic tools to share your own story with more precision. That gives us better tools to explain corporate profits and connects us to each other. Some stories, we know what it’s like to be outside in the evening, staring off into the distance at a future that may never be ours. And from the story, we learn more about those around us, those who came before us, and we learn more about ourselves.

[loveyudan-blog]

Image from Liquid Traces, by Charles Heller and Lorenzo Pezzani

Digital Narrative and Witness: Artworks and Ethics 

The topic of this week is digital narrative and witnessing. After studying a couple of artwork videos and a related article “Digital Narratives and Witnessing: The Ethics of Engaging with Places at a Distance” by Nishat Awan, combining with sort of my previous experience, I want to talk about how does digital technologies apply in the issue report with places that are at a distance from us and how we might have an ethical engagement with it.

Today's reporting artwork is not limited to traditional reporting approaches. 3D scan and simulation, power-topologies, location tracing, drones, virtual reality and various high-tech visualization methods have appeared. Nishat Awan claimed that visual material of distant places in crisis created by digital technologies is important in the context of humanitarian action. News reporting is no longer a simple map text video presentation, digital technology provides access to places in crises and makes news more objective and immersive. Virtual reality film Clouds over Sidra is successful in digital storytelling. Through VR context and narration, we are shown the emaciated child crying at the pain of hunger, or the harsh realities of life in a desert refugee camp, to provoke a response from us at an emotional level. Thus, through digital technologies, we are provided not only a real, immersive place simulation but also more experience of social issues away from us.

When talked about current digital technologies of crisis visualisations, Nishat Awan mentioned about “The Digital Saviour Complex” in the article. In an age of new media, no longer reliant on the mediation of newsroom editors and professional journalists in the field, we have to trust and rely on the technological. This work places the burden of proof on the refugee, in this case, a twelve-year-old girl, who has to show us her destitution and her will in the face of it; she has to perform it, and faith in the technological. It reminds me of “Digital humanitarian”, which is a term defined by Patrick Meier in 2010, and later he has written a book “Digital Humanitarians: How Big Data is Changing the Face of Humanitarian Response” to explain. The book begins with praises to how transparent, objective and powerful of technologies from professors, think-tank directors, heads of humanitarian organizations, policy advisors and media experts, based on the notion that digital, computerized engagement with crises is going to completely change the lives-saving industry, which holds the same point as the “Digital Saviour Complex”.

We hardly realize that the majority of high-tech tools are handled by the capitalist. There is an authenticity and immediacy associated with materials captured by digital tools, but at the same time they are easily exploited, misinterpreted, and hijacked by powerful actors. it seems that “Digital Saviour Complex” equals to “Colonial Saviour”, the colonial figure is alive and the mode is now digital. The project Dronestagram shows how seeing through digital technology can produce a different practice of witnessing, what the U.S. government was claiming in terms of the number of casualties and the accuracy of the bombs was a far cry from the reality on the ground. Due to the most issues attribute, we are difficult to touch and access to the original context, the places most of us will never see. We do not know these landscapes and we cannot visit them. Also making sense of the sheer amount and often shocking nature of these original images is difficult.

[loveyudan-blog]

Virtual and digital

[loveyudan-blog]

Black Feminist 

Amid the surging waves of the feminist movement in the 1970s, black feminism emerged from African American women’s long-cherished intellectual and activist traditions as a distinctive school of feminist thought.

[loveyudan-blog]

When Computational Art as a Technology

[loveyudan-blog]

Face Cage by Zach Blas, 2013

False Idea of the “Democratisation” of AI (2)

Last Thursday we took a guest lecture from Dr Lea Laura Michelsen. She talked about aesthetic practice of Biometrics and the renaissance of physiognomy, and introduced the concept of Zach Bas.

Zach Blas is a visual artist and writer, whose main interest is technology and politics. One of his most famous works is Fag Face masks, which shows strong complains and outrage for Biometrics. Facial Weaponization Suite is his initial works of this series in 2011-2014. He made an irregular plastic mask from combining thousands of facial data of participants to resist biometric identification. Face Cage (2013-2016) is his another artwork to expresses the unreasonable and unfair of biometric identification technology for human. This work simulated the principle of biometric identification. All participants were divided into different classes through some crude and simple approaches, then their facial biometric maps were transferred to physical data and were generated to masks which do not match on their face. Participants were painfully constrained and straggled in the cage made by their biometric information.

According to the explanation from Wikipedia, Biometrics is the technical term for body measurements and calculations. It succeed and popular because it is able to measure the basic human individual characteristics through machine as an pretty objective, accurate and effective. Hence, it has often served commercial, government and military by following a series of identification principles and classes. The information generated from it is regulatory and stereotypical, like a huge cage. The identification principle define who we are and what means normal. However, these data references just came from very shallow perceptions and current stereotypes about every person of the developers. Non normative ages, genders and races are often unsuccessfully identified.

Biometric technology provides more convenience and efficiency. On the other hand, it is a superficial tool but seriously impacts on our physical and psychological. At present, in the environment that the majority of personal information and private data are transparent and digitized by governments and related authoritative institutions, we have to have the critical thinking and basic protection awareness towards new technologies especially algorithm and AI. Meanwhile we have to pay more attention to personal information and identity.

[loveyudan-blog]

The painting generated by AARON

Generative Art and Computational Creativity

Tools: co-evolving with human

With everlasting pursuit of dream making and immersive experience in creative and entertaining industry, an increasing number of computing techniques and approaches are emerging. Huge requirements for creativity, individuation, real-time interactions and virtual environments results that non-linear medias and algorithms are popular. Traditional software such as Adobe series could not provide effective help anymore. Therefore, generative art by computational algorithms is emerging on art.

Generative art

It usually creates arts (painting, music, poems) relies on one or more non-human independent autonomous system, which is part or totally random for their production. Philip Galanter suggested that generative art refers to any art practice where the artist uses a system, such as a set of natural language rules, a computer program, a machine, or other procedural invention, which is set into motion with some degree of autonomy contributing to or resulting in a completed work of art.

Generative art includes a series of rules, and randomly generate artworks in this set of rules through algorithms, data or biological imperatives. In other words, it generally works as autonomous mechanism, which reduces subjective influence of artists. Outcomes are random in a specific range based on restrictive condition, parameter and setting, but the result is unpredictable. Thus, generative art is not a specific art form or movement. It is an approach to creating art without intention and concept.

Computational creativity

Computational creativity also known as artificial creativity or mechanical creativity, which is the academic scientific field devoted to the study of computational processes for autonomous creative tasks. Same as artificial intelligence, computational creativity emphases algorithms and intelligence.

According to Herbert Simon (1960), AI is the science of having machines to solve problems that do require intelligence when solved by human. AI is good at rational problem solving, especially for cases that well defined with complex solutions. On the contrary, computational creativity emphases “creativity”. It is hard to find an optimal solution for creative tasks because optimality is fake and ill-defined to creativity.

Nowadays, an increasing number of generative paintings, music and performance artworks practices are inventing. When we investigate the creative process, the initial questions would be:

What is creativity?

Can we and how can we understand creativity?

How can we build modelling and simulation through algorithms?

Practices in computational creativity

Computational painting is one of the initial practices in this field. The first computational painting is AARON system designed by Harold Cohen, is a long-term project since 1973. This drawing system connects software intelligence and a drawing robot, making original drawing in the physical and mechanical aspect without reference. In recent 40 years, the drawing skill of AARON is gradually revising to apply colors, 3D space and abstract patterns.

The majority of drawing principles of the kind of mechanical machines are realized by pixel. Generators usually recognize and locate each pixel on the reference pattern through coordinate, then robot arms draw specifically also based on pixel. In other words, creative ability, or unpredictability of drawing machine depends on the data of machine learning. The style, patterns and elements are quantitative, which shows unpredictability of potential painting outcomes created by machine. However, RobotArt claimed that the skills required to effectively paint are intrinsically human – graceful movement, sense of touch and pressure, ability to experience colour and value. Machine drawing field is gradually dissatisfying to simulate human painting, rather than apply all drawing process and behaviour of human, create new paintings from human, be more human-like.

Therefore, from this perspective, the core creativity of generative art is the process of drawing and unpredictability. 

How to drawing like human instead of a copy machine, and how to create new in the process art is keys of process. 

For unpredictability, what is the boundary of predictable and unpredictable, and how to realize the unpredictability (using physical way or algorithmic language like “noise” or “random”) are considerable points.

[loveyudan-blog]

Availability and Feasibility of Intelligent Products are Giving Rise to a False Idea of the “Democratisation” of AI

The readings this week is more about invisible parts behind AI system, including the entire eco-system, whole industrial supply channel of AI as well as some potential troubles.

Currently, AI has almost been fully commercialized and become a popular trend. We are gradually relying on its conveniences and efficiency. However, commercial, labor, resources and data, the whole industrial supply channel hidden each intelligent product are hard to know. Each small moment of convenience requires a vast planetary network, fueled by the extraction of non-renewable materials, labor, and data (Kate and Vladan, 2018). During the producing, artificial intelligence is not independent, it does not separate from human, but running based on the a large quantity of valuable resources and labor forces on the earth.

iPhone 11 sold more than 200 million in just two months. Smart phone has become an irreplaceable link connecting individual with the world in the modern society, but no customer exactly know where did each component embedded in such a tiny intelligent device come from, how they physically shipped from supplier sites and how they assembled from thousands of individual facilities of different countries, that becomes a complex structure of supply chains within other chains, in a sprawling network. Contemporary capitalism conceals the histories and geographies of most commodities from consumers. Consumers are usually only able to see commodities in the here and now of time and space, and rarely have any opportunities to gaze backwards through the chains of production in order to gain knowledge about the sites of production, transformation, and distribution (Kate and Vladan, 2018). Cost of one intelligent product includes millions of suppliers, shipped materials and workers before it assembled on the line, but the majority of them are ignored or invisible to users. Despite analysing all costs are impossible, it is increasing important that we develop and awareness of how those technical infrastructures be produced and run.

Besides, we don’t know our role and distance to intelligent systems. As the meaning of “algorithms” for the broader public is a level that unattainably complex, anything beyond the physical or digital interfaces such as data processing, running principles and cloud storages are out of control to users. It causes invisible and unequal status of customers.

They are a part of the iteration system who contributes data and test results, not the end of the product chain. It is difficult to place the human user of an AI system into a single category, the user is simultaneously a consumer, a resource, a worker, and a product (Kate and Vladan, 2018). Almost every phase of Internet device is obscured, we are difficult to image the whole algorithmic process of a single product. On the contrary, we are visible in interaction with the platform. We are always being tracked, quantified, analyzed and commodified. AI is a kind of black box, functions and user interfaces are shining and controllable on the elegant stage, but plenty of sensors monitoring and recording data could not see in the backstage.

However, opacity and information inequivalence are not root cause. Relying on intelligent products, internet platforms, even establishing own systems is trade nowadays. We do not need and we are impossible to control all details in the hidden system. Transparency wouldn’t help much – without forms of real choice, and corporate accountability, mere transparency won’t shift the weight of the current power asymmetries. The core algorithmic logic and processes, as well as training data are only accessed and controlled by a few companies. The key is how they process these data.

As mentioned in the last journal, algorithms are never politically neutral and trying to restrict potential perceptions and behaviors of human. Superposition of data samples, iterations and deeper layers of learning result in unrestrained thirst for new resources and fields of cognitive exploitation. The process of quantification is reaching into the human affective, cognitive, and physical worlds. Every form of biodata – including forensic, biometric, sociometric, and psychometric – are being captured and logged into databases for AI training (Kate and Vladan, 2018). As the result, these data became definition, category and prediction of human being, multiple cognitive economies and commodification of trust and evidence through cryptocurrencies occurred based on that and increasingly impact on the distribution of opportunity, wealth and core techniques towards monopoly.

The quantifications and assumptions about human by machine learning results are usually narrow and full of stereotypes. Datasets in AVA primarily shows women in the ‘playing with children’ action category, and men in the ‘kicking a person’ category. The cognizations of AI from training system is most stereotypical, projecting a normative vision of the human past into the human future. It never breaks out of the historical restriction, but it is being applied on a large scale without filter.

In addition, AI system became a weapon for the few to defend their own power and to lead public opinion. Social media platforms are increasingly occupying human life and are gradually changing their daily behaviors. Nevertheless, public platforms silently monitor and collect personal data of each user and control it by the companies. They analyze data from user accounts and user posts. They can decide if show the post in the public area, or which position to show. They even advertise and attribute new user group they want by specific current user information, whatever if it is fake. We see a new form of extractivism that is well underway: one that reaches into the furthest corners of the biosphere and the deepest layers of human cognitive and affective being (Kate and Vladan, 2018). In the future, major number of labor in the public system is augmented, and more and more centralized, privately-owned corporations will be emerging. They are using public data to generate enormous wealth for the very few.

[loveyudan-blog]

The Boundary of Human and Artificial Intelligence

This week we read and discussed two chapters of a 2007 book called Human-machine reconfigurations: Plans and situated actions.

When we talk about AI, initially, the definition and property of 'human' should be clearly. What is the humanlike, what is the difference between human and machine, and how to distinguish human and other life entities? Automata is one of the most important attributes of life entity, and it is the primary purpose to create robots. Meanwhile, subjective initiative and autonomous actions distinguish human-being with others. These characteristics result human enabling to subjectively judge, act and interact, called ‘agency’. According to Strathern, physical person is not a pre-existing entity but an object of the regard of others and an objectification of the relations that constitute her. The agency, in turn, is the one who acts with those relations as cause and reference.

Three necessary elements for humanness in contemporary AI standards: embodiment, emotion, and sociality.

Appropriate actions in a respect remains primary for humanness embodiment. Cognition is as an embodiment of actions, while actions rely on the physical environment. Same as human, the embodiment of AI is generating more cognitions through interactions with the context. In other words, AI has to possess a basic perception to something and the capacity of output, then gain various related input, transfer to new output. However, this learning capacity also be restricted in a specific cultural context. 

Emotion is a unique feature of human, it was positioned as the missing ingredient for full machine participation in the human world. Currently, emotions already were standardized and quantified. It became a pattern written in the language of the biological elements, can be not only recognized and deduced, but also simulated by AI. For instance, Pepper from Softbank, as service semi-humanoid robot, is versed in emotion detecting and semantic analysis, then give active feedback through voice, facial expression or body movements. 

Results from embodiment and emotion, AI also required the social ability with others (including human and other machines). Sociality of AI develops and iterates its cognition and functions, human and machine each alternately has served as a model for the other. Meanwhile, AI is influenced, even will gradually generate its own agency from the training and interactions of the environment, especially its designer. The book clarified that robot’s reliance on the performative capabilities of its very particular ‘human Estrade caregiver’. Therefore, It reflects and picks features of the designer. Moreover, deep learning is one of the results of sociality. AlphaGo keeps learning and growing by its sociality. Apart from fundamental principles settled in the early stage, all skills were input through practicing with others.

The three elements determine which part and level of human are simulated by AI, and the distinction between AI and general machine interfaces. AI is endowed own agency so it can be regard as an independent object. Thus, interface conceived very differently than as the meeting of a human and a machine, each figured as a self-standing entity possessed of pre-established capabilities. Rather, effective encounters at the computer interface are those moments of moving complicity between persons and things achieved through particular, dynamic materialities and extended socialities.

Meanwhile, initiatives in the fields of situated robotics and artificial life as indicators of more profound shifts in the human sciences and in contemporary society.

Before we explored AI, firstly a series of standards and rules to humanity were established. The book claimed, AI and robotics involve a kind of doubling or mimicry in the machine that works as a powerful disclosing agent for assumptions about the human. The generating and existing of AI is not politically neutral. It delimited characteristics of human-being, means it restricted potential perceptions and behaviors of human. AI brings various efficient exploitation and driving force in contemporary society, while, it as a mirror, criticizes human and the society.

[loveyudan-blog]

figure from: Shawn Wang

“Algorithm” What Do We Need to Concern

This post talks about my week 2 reading, Tarleton Gillespie’s 2014 essay Algorithm.

What is Algorithm? Gillespie claimed that this term has different definitions and meanings to different communities. For technical communities such as software engineers and designers, algorithms are simple approach to communicate with the computer. For social scholar group, algorithm means a macro and qualitative explanatory power. Also, for the broader public, the algorithm might represents a level that unattainably complex.

Algorithm as a technical solution to a technical problem.

From the literal aspect, algorithm means a set of logic and rules to train data and solve technical problems. As the essay stated, algorithms as a part of mechanisms that introduce and privilege quantification, proceduralisation and automation, itself is functional and fair. Same as YouTube, a series of video-sharing websites have an intelligent function, which recommends specific videos for each user according to their preference. We often find interested videos from that function and then never stop watching. The function realized by a kind of algorithm, searching videos with similar keywords or categories based on our favourite and history, that is why we always be recommended for interesting videos. Algorithm is a technical solution which provides intelligent experience accurately and effectively.

Algorithm as synecdoche.

Nowadays The “algorithm” be most concerned about by the public is not the algorithm itself, that in fact is an algorithmic system. “algorithm may serve as an abbreviation for the socio-technical assemblage that includes algorithm, model, target goal, data, application, hardware - and connect it all to a broader social endeavour.“ When algorithm be indicated as synecdoche, a whole system, or a business blockchain behind it. It is not natural and objectively anymore, but be a part of the algorithmic system procedure. Algorithmic system includes not only the physical algorithm solution, but also commercial stakeholders (providers, managers, designers, and users).

As a result, informatics design, data preparation and selection are more important than the “algorithm” design. Debating the models, cleaning the training data, tuning the parameters, and deciding on which algorithms to depend on in which context are keys to realize the high quality functions and impacts of an algorithm. All of them also influence the value, assumptions and experience of the final outcome. One solution can be realized by various algorithm approaches, but the standard of measuring an algorithm is not the coding itself, but the comprehensiveness, efficiency and accuracy of the solution.

Algorithm as talisman.

Nevertheless, we do need to clearly recognize the multiple meanings of the algorithm. We do need to clarify outcomes done by which part of the algorithmic system. As the example from essay, the deletion of pornography contents from a social platform, which is sometimes processed automatically by algorithms and sometimes performed manually by managers. An algorithmic system exists plenty of financial or political interventions while providing services for the public. Providers have subjective purpose during researching, designing and managing the algorithm, meanwhile, it also be exploited based on users aims and preference, that results in unfair and uncontrolled of  algorithmic system, and algorithm as scapegoat for corporate responsibility.  

Algorithm we talk about, judge and even criticize is a part of business activity based on algorithmic system. Although the algorithm be positioned as an objective tool which can provides accurate and no bias conclusions, it is also the talisman used by corporates to hide benefits and unnatural positions behind it. Therefore, how to correctly treat and concern about the algorithm is important. It is better to distinguish the algorithm program from the organizers behind it, even if they are connected closely and dependent on each other.

[loveyudan-blog]

What is “New” Media Arts

This post shows my afterthoughts on the book New Philosophy for New Media written by Mark B.N. Hansen. I have not finished it yet, but it can be one of my enlightening books on computational arts. Plenty of inspirations and thoughts have been gained from this book on the function and forms of computational arts.

This book continues the topic of Walter Benjamin’s “Work of Art” essay, detailed argues a shift in the function and medium of art in the technological period. It based on the philosophy of matter by Henri Bergson, mainly explores the relationship between mediums and viewers’ perceptions in new media art to explains what is new media art, and what is “new”.

Bergson believes that it’s impossible to separate the physical human body and external objects. They interact and coordinate together through memories. In a new technical space, media as the perceptor, sustained influences people to perceive information through the memory matters.

As the book claimed, stands as a beacon of hope that media can continue to matter in the digital age. The function and significance of media are more crucial in the computational arts. That can be divided in two parts:

1.     Aesthetic aims have to be realized by medial interfaces, meanwhile, aesthetic is a supplemental action, not restricted by materials or data.

2.     New media art emphasises the interaction with the audience. It transforms data to perceptive interface or context to realize its function as art. Besides, “mentary motivation between the image interface and the digital data” be applied in communication and resonation with audiences. Thus, as the book mentioned, the audience participates in the process through which the mediated digital data is transformed into a perceivable image.

In the past, as Descartes supposed, art was a praise of “spirit”. It resonates with the viewer from vision and matters, which is a single and sustained experience. However, with the flexibility brought by digitization, art in the digital age can be described as: the shift from the framing function of medial interfaces to physical. Therefore, computational arts emprises the diversity of the physical support. The selection of media and expression are more and more important, and functions also is more significant. It makes new media art “new”.

[loveyudan-blog]

 A Fish Can’t Judge the Water

This post shows my thoughts and reflection on the journal A Fish Can’t Judge the Water, which wrote by Femke Snelting in May 2006.

At the first, it uses a wood spoon example from a life scene, as the title, to figure out the prevalent necessity through imperceptible influence of tools, and prove it also in software. This article mainly mentions three concepts of software, that are current status, property and influence.

“We practice software until we in-corporate its choreography.”

As it is claimed in the article, software gradually becomes a natural habitat at the modern society. Software attracts users through its innovative functions, contains users through providing fresh scenarios and better experience. Mastering a new software is equal to possess more resources, that is why people flock to it. Software becomes a necessity, changing consciousness and behaviors of people in special fields.

“Software is never politically neutral.”

“Software produces culture at the same time as it is produced by culture.”

This reminded me of the Essay for 4x4: Beyond Photoshop with Code by Golan Levin. Software, as a kind of solution, usually created by anonymous engineers, not real users. The objective of software depends on the cultural context, policy or commercial requirements, but does not mean that all the needs in that field can be met.

When we used to relying on tools, we often lose the ability to create tools ourselves, and we fall into intangible limitations by specific shapes, medium or regulations. Almost artists in digital graphics use Adobe, that means, Adobe is the only approach to process digital graphics. It seems to be a consensus or a belief that once the need had been met, then no one would doubt that.

As a result, a new culture representing the progress of human civilization was formed. Specific software became a pretty necessary part to work. At the same time, it gradually becomes the cornerstone of development in one area. New cultures are being formed to satisfy more requirements through expanding new functions based on current principles and operations.