<?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 Mike Koss on Medium]]></title>
        <description><![CDATA[Stories by Mike Koss on Medium]]></description>
        <link>https://medium.com/@mckoss?source=rss-fdba8053a856------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*mCEK8xFgrIEtkJCa.jpg</url>
            <title>Stories by Mike Koss on Medium</title>
            <link>https://medium.com/@mckoss?source=rss-fdba8053a856------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 22:32:58 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@mckoss/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[Mitigating Entropy in Self-Modifying Agentic Architectures]]></title>
            <link>https://mckoss.medium.com/mitigating-entropy-in-self-modifying-agentic-architectures-20592fc3d8c4?source=rss-fdba8053a856------2</link>
            <guid isPermaLink="false">https://medium.com/p/20592fc3d8c4</guid>
            <category><![CDATA[agentic-ai]]></category>
            <dc:creator><![CDATA[Mike Koss]]></dc:creator>
            <pubDate>Tue, 02 Jun 2026 11:48:26 GMT</pubDate>
            <atom:updated>2026-06-02T11:48:26.636Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="A digital illustration of a cumbersome mechanical robot, completely festooned with mismatched modifications, antennae, wires, and glowing components, standing awkwardly in a cluttered workshop environment." src="https://cdn-images-1.medium.com/max/1024/1*Kio2evTa2UZkj49vC489vg.png" /></figure><p>Self-modifying agentic frameworks, such as OpenClaw, offer powerful autonomous capabilities by allowing agents to adapt their own operating parameters over time. However, long-running, autonomous instances inevitably suffer from architectural degradation. Without rigid structural constraints, these systems experience instruction entropy — a gradual drift toward inefficiency, bloat, and silent failure.</p><p>To build agentic systems that scale and improve, rather than degrade, it is necessary to address the mechanics of prompt drift, the limitations of context windows, and the inefficiency of recursive monitoring.</p><h4>The Mechanics of “Slop Drift” and Context Bloat</h4><p>When an agent operates with self-modifying privileges, it continuously appends new rules, edge cases, and temporary policies to its working memory. Over time, this results in “slop drift.” A system prompt that begins as a concise set of directives degrades into an unorganized accumulation of instruction detritus.</p><p>This accumulation directly impacts system performance. Loading an ever-expanding monolithic policy into a single context window has several technical consequences:</p><ul><li><strong>Attention Dilution:</strong> As the context grows, the language model’s ability to attend to the most critical instructions diminishes, leading to lower accuracy.</li><li><strong>Latency and Cost:</strong> Larger contexts require more compute, increasing both time-to-first-token and API expenditure.</li><li><strong>Logic Conflicts:</strong> Redundant or contradictory rules build up, causing erratic agent behavior.</li></ul><h4>Decomposing the Monolith</h4><p>The standard approach to mitigating context bloat is structural specialization. An agentic system must transition from a monolithic architecture to a distributed one.</p><h4>Specialized Agents</h4><p>Instead of a single agent tracking all system policies, concerns should be separated into narrowly scoped micro-agents. Each agent is responsible for a distinct domain, requiring only the context strictly necessary for its specialized task.</p><h4>Deterministic Offloading</h4><p>Not every operation requires a probabilistic language model. A highly optimized agentic architecture relies heavily on non-AI, deterministic scripts. Frameworks like the Model Context Protocol (MCP) can be used to standardize the interface between your agents and local file systems or external APIs. By offloading routine, repetitive, or highly structured tasks to standard code, you eliminate the risk of logic drift for those specific functions and drastically reduce token consumption.</p><h3>The Epicycle Problem in Monitoring</h3><p>Unmonitored systems fail silently. In agentic frameworks, failure often looks like a hallucination loop or a silent degradation in output quality. The intuitive solution is to deploy “watchdog” agents to monitor the primary agents.</p><p>However, this introduces the “epicycle problem.” Having agents watch agents creates a nested hierarchy of probabilistic observers. This compounds system complexity, creates recursive failure modes, and scales compute costs linearly without guaranteeing stability.</p><p>Effective monitoring should avoid AI-on-AI epicycles where possible. Instead, rely on a hybrid approach:</p><ol><li><strong>Deterministic Telemetry:</strong> Use standard software monitoring (execution timeouts, API error rates, strictly typed JSON schema validation) to catch structural failures instantly.</li><li><strong>Stateless Semantic Checks:</strong> Reserve LLM-based monitoring for periodic, stateless checks on the final output quality, rather than continuously monitoring the internal reasoning loop of the primary agent.</li></ol><h3>Migration Strategy for Long-Term Stability</h3><p>To migrate an existing self-modifying system toward a more stable architecture, or to build a resilient one from scratch, implement the following constraints:</p><ol><li><strong>State Separation:</strong> Isolate the core, immutable system prompt from the dynamic, self-modifying memory layer. The core directives must never drift.</li><li><strong>Instruction Garbage Collection:</strong> Implement a scheduled, automated pipeline to compress and prune the agent’s accumulated memory. A separate process should periodically refactor the “detritus” into concise, updated rules, much like a database compaction process.</li><li><strong>Test-Driven Agent Updates:</strong> Treat self-modifications like code commits. Before an agent can permanently append a new rule to its operational memory, that rule should pass an automated suite of deterministic regression tests to ensure it does not break existing functionality.</li></ol><p>Self-modifying systems hold immense potential, but they are bound by the same laws of software entropy as traditional code. Long-term efficiency requires treating the AI not as a monolithic brain, but as a probabilistic routing layer sitting on top of a rigidly deterministic foundation.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=20592fc3d8c4" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Midnight Ride of Paul Revere]]></title>
            <link>https://mckoss.medium.com/the-midnight-ride-of-paul-revere-8b942490e0a7?source=rss-fdba8053a856------2</link>
            <guid isPermaLink="false">https://medium.com/p/8b942490e0a7</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[illustration]]></category>
            <dc:creator><![CDATA[Mike Koss]]></dc:creator>
            <pubDate>Fri, 01 Dec 2023 20:06:52 GMT</pubDate>
            <atom:updated>2023-12-01T20:06:52.653Z</atom:updated>
            <content:encoded><![CDATA[<p><em>by </em>Henry Wadsworth Longfellow<em><br>Illustrated by Chat GPT-4</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*k4ez3dHzwvX4b4T4" /><figcaption>Paul Revere’s Warning</figcaption></figure><p>This is an experiment in the creation of illustrated works using Chat GPT. I pasted the text of the poem and Chat GPT selected the following scenes to be illustrated purely from it’s own analysis of the text.</p><ol><li><strong>Paul Revere’s Warning</strong>: Depicting Paul Revere speaking with his friend, emphasizing the urgency of the situation as he discusses the signal for the British’s approach (lanterns in the North Church tower).</li><li><strong>The Preparation</strong>: A scene showing Paul Revere rowing across the Charlestown shore, with the moon rising and the British ship Somerset in the background.</li><li><strong>The Signal</strong>: Illustrating the moment when Revere’s friend climbs the Old North Church tower, hangs the lanterns, and looks over the town.</li><li><strong>Paul Revere’s Ride Begins</strong>: Capturing the instant Paul Revere sees the signal and begins his ride, with a focus on his determination and the surrounding environment.</li><li><strong>The Midnight Ride</strong>: A dynamic scene of Revere riding through various towns, passing various landmarks like the village clock, the Medford town bridge, and Lexington.</li><li><strong>The Alarm</strong>: Illustrating the urgency and speed of Revere’s ride, showing him alerting the residents of Middlesex villages.</li><li><strong>The Arrival at Concord</strong>: A portrayal of Revere reaching Concord, with early morning elements like birds and the morning breeze.</li><li><strong>The Aftermath</strong>: A final scene capturing the farmers’ response, readying themselves for the confrontation with the British.</li></ol><h3>Paul Revere’s Ride</h3><p><em>By Henry Wadsworth Longfellow</em></p><p>LISTEN, my children, and you shall hear<br>Of the midnight ride of Paul Revere,<br>On the eighteenth of April, in Seventy-five;<br>Hardly a man is now alive<br>Who remembers that famous day and year.</p><p>He said to his friend, “If the British march<br>By land or sea from the town to-night,<br>Hang a lantern aloft in the belfry arch<br>Of the North Church tower as a signal light, — <br>One, if by land, and two, if by sea;<br>And I on the opposite shore will be,<br>Ready to ride and spread the alarm<br>Through every Middlesex village and farm,<br>For the country folk to be up and to arm.”</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*v-RO720sPOhxRZ6N" /><figcaption>The Preparation</figcaption></figure><p>Then he said, “Good night!” and with muffled oar<br>Silently rowed to the Charlestown shore,<br>Just as the moon rose over the bay,<br>Where swinging wide at her moorings lay<br>The Somerset, British man-of-war;<br>A phantom ship, with each mast and spar<br>Across the moon like a prison bar,<br>And a huge black hulk, that was magnified<br>By its own reflection in the tide.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*olt3U4bDqhO6vzdZ" /><figcaption>The Signal</figcaption></figure><p>Meanwhile, his friend, through alley and street,<br>Wanders and watches with eager ears,<br>Till in the silence around him he hears<br>The muster of men at the barrack door,<br>The sound of arms, and the tramp of feet,<br>And the measured tread of the grenadiers,<br>Marching down to their boats on the shore.</p><p>Then he climbed the tower of the Old North Church,<br>By the wooden stairs, with stealthy tread,<br>To the belfry-chamber overhead,<br>And startled the pigeons from their perch<br>On the sombre rafters, that round him made<br>Masses and moving shapes of shade, — <br>By the trembling ladder, steep and tall,<br>To the highest window in the wall,<br>Where he paused to listen and look down<br>A moment on the roofs of the town,<br>And the moonlight flowing over all.</p><p>Beneath, in the churchyard, lay the dead,<br>In their night-encampment on the hill,<br>Wrapped in silence so deep and still<br>That he could hear, like a sentinel’s tread,<br>The watchful night-wind, as it went<br>Creeping along from tent to tent,<br>And seeming to whisper, “All is well!”<br>A moment only he feels the spell<br>Of the place and the hour, and the secret dread<br>Of the lonely belfry and the dead;<br>For suddenly all his thoughts are bent<br>On a shadowy something far away,<br>Where the river widens to meet the bay, — <br>A line of black that bends and floats<br>On the rising tide, like a bridge of boats.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*8S_bKZJcp85IUXzI" /><figcaption>Paul Revere’s Ride Begins</figcaption></figure><p>Meanwhile, impatient to mount and ride,<br>Booted and spurred, with a heavy stride<br>On the opposite shore walked Paul Revere.<br>Now he patted his horse’s side,<br>Now gazed at the landscape far and near,<br>Then, impetuous, stamped the earth,<br>And turned and tightened his saddle-girth;<br>But mostly he watched with eager search<br>The belfry-tower of the Old North Church,<br>As it rose above the graves on the hill,<br>Lonely and spectral and sombre and still.<br>And lo! as he looks, on the belfry’s height<br>A glimmer, and then a gleam of light!<br>He springs to the saddle, the bridle he turns,<br>But lingers and gazes, till full on his sight<br>A second lamp in the belfry burns!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*iRNi3CbFjsBUB0DM" /><figcaption>The Midnight Ride</figcaption></figure><p>A hurry of hoofs in a village street,<br>A shape in the moonlight, a bulk in the dark,<br>And beneath, from the pebbles, in passing, a spark<br>Struck out by a steed flying fearless and fleet:<br>That was all! And yet, through the gloom and the light,<br>The fate of a nation was riding that night;<br>And the spark struck out by that steed, in his flight,<br>Kindled the land into flame with its heat.</p><p>He has left the village and mounted the steep,<br>And beneath him, tranquil and broad and deep,<br>Is the Mystic, meeting the ocean tides;<br>And under the alders that skirt its edge,<br>Now soft on the sand, now loud on the ledge,<br>Is heard the tramp of his steed as he rides.</p><p>It was twelve by the village clock,<br>When he crossed the bridge into Medford town.<br>He heard the crowing of the cock,<br>And the barking of the farmer’s dog,<br>And felt the damp of the river fog,<br>That rises after the sun goes down.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*kSp9RDtcsOBGf_1R" /><figcaption>The Alarm</figcaption></figure><p>It was one by the village clock,<br>When he galloped into Lexington.<br>He saw the gilded weathercock<br>Swim in the moonlight as he passed,<br>And the meeting-house windows, blank and bare,<br>Gaze at him with a spectral glare,<br>As if they already stood aghast<br>At the bloody work they would look upon.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*bUAQKUkfBoIJ-ERb" /><figcaption>The Arrival at Concord</figcaption></figure><p>It was two by the village clock,<br>When he came to the bridge in Concord town.<br>He heard the bleating of the flock,<br>And the twitter of birds among the trees,<br>And felt the breath of the morning breeze<br>Blowing over the meadows brown.<br>And one was safe and asleep in his bed<br>Who at the bridge would be first to fall,<br>Who that day would be lying dead,<br>Pierced by a British musket-ball.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*N3t29_GcDJFYPZRf" /><figcaption>The Aftermath</figcaption></figure><p>You know the rest. In the books you have read,<br>How the British Regulars fired and fled, — <br>How the farmers gave them ball for ball,<br>From behind each fence and farm-yard wall,<br>Chasing the red-coats down the lane,<br>Then crossing the fields to emerge again<br>Under the trees at the turn of the road,<br>And only pausing to fire and load.</p><p>So through the night rode Paul Revere;<br>And so through the night went his cry of alarm<br>To every Middlesex village and farm, — <br>A cry of defiance and not of fear,<br>A voice in the darkness, a knock at the door,<br>And a word that shall echo forevermore!<br>For, borne on the night-wind of the Past,<br>Through all our history, to the last,<br>In the hour of darkness and peril and need,<br>The people will waken and listen to hear<br>The hurrying hoof-beats of that steed,<br>And the midnight message of Paul Revere.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*yW2is-A4Ur1pq2l8" /><figcaption>Paul Revere</figcaption></figure><p><em>Thank you for viewing this GPT-4 experiment. I think the results are quite striking and show the potential to use AI as a creative partner who is quite capable of extracting details from a written text in order to both do the planning and execution of appropriate and evocative illustrations.</em></p><p><em>I did not accept all of the initial illustrations created by GPT-4. I had to correct misinterpretations of the text or direct it toward more appropriate images. But it does not feel that we are very far from being able to simply paste a short story into a chat, and have it output a complete illustrated book.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8b942490e0a7" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Expanding the Spectrum: Rethinking Diversity in Tech Hiring]]></title>
            <link>https://mckoss.medium.com/expanding-the-spectrum-rethinking-diversity-in-tech-hiring-aaabc3ea244e?source=rss-fdba8053a856------2</link>
            <guid isPermaLink="false">https://medium.com/p/aaabc3ea244e</guid>
            <category><![CDATA[software-engineering]]></category>
            <category><![CDATA[diversity]]></category>
            <dc:creator><![CDATA[Mike Koss]]></dc:creator>
            <pubDate>Mon, 13 Nov 2023 23:38:29 GMT</pubDate>
            <atom:updated>2023-11-13T23:38:29.059Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*gU1dLQltjN7JcOYl" /></figure><p>I’ve been reading Ta-Nehisi Coates book, <a href="https://ta-nehisicoates.com/books/between-the-world-and-me/"><em>Between the World and Me</em></a><em>. It relates a father’s wisdom about the world to his son, as a black man growing up in America.</em></p><p>In the book, he relates his experience with Howard University, “The Mecca”; which, for multiple generations of his family, represented a place where black Americans, long oppressed and neglected, could strive to reach a higher position (in a white society).</p><p>Founded in 1867, just two years after the end of the Civil War, and chartered by the US Congress, Howard was to be a place of higher education open to all, regardless of race. This was a time where Harvard was to only then admit its first student to earn a degree there, and MIT would not admit a black student for another 20 years.</p><p>Despite its long history, I would not become aware of Howard until I was an adult. I had taken my computer science degree from MIT to get a job as a programmer in the early days of Microsoft. It was a tradition at Microsoft to send our engineers to recruit for new talent to fill our rapidly growing ranks.</p><p>We frequented all the “top schools” for engineering and computer science; MIT, Stanford, CMU, Harvard, Cal Tech, etc. But, by the 1990’s we were not only having trouble meeting our demand for new talent, but we were also coming to realize how little diversity our engineering teams had.</p><p>Almost all software developers at Microsoft were white and male up until then; yes, there were a few women in our ranks — but it was quite the exception (most of the women at Microsoft at the time served in other roles like testing, technical writing, sales, marketing, or secretarial jobs).</p><p>And African Americans were even more scarce. Not only was Microsoft located in a part of the country with a very small black population, but all of the elite schools we recruited from were also greatly unrepresentative (and even more so in the computer science departments of those schools).</p><p>So, one day, I was asked if I would substitute one of my regular trips to recruit at MIT, and instead visit Howard University. I didn’t know what to expect, but I agreed and was eager to help Microsoft’s diversity in hiring.</p><p>As Microsoft Developers, we prided ourselves as having a very objective process for recruiting engineers. Contrary to other fields, we did not ask questions to determine if applicants came from a “good background”, or a “good family”. We prided ourselves and asking objective questions about engineering — could this person do the work. We nominally didn’t even care about grades or degrees.</p><p>Our style was to give applicants a practical programming problem to solve. We want to see you “write code”; and we de-emphasized learning just how you got to the place you were now.</p><p>In hindsight, I view this as a very narrow approach which measured just a single dimension of a person’s abilities and potential.</p><p>Suffice it to say, Howard was not preparing its students to do well with our interview format at the time. I talked to a dozen students, and was only able to recommend bringing a single person back to Redmond for further interviews — and that person was to interview for a position in testing, and not as a software developer.</p><p>To be fair, most colleges and universities did not have very strong computer science programs in the 1990’s. Even though computer programming as a career had existed for several decades, I don’t think anyone anticipated the massive explosion of the opportunities that were about to become available with the success of the personal computer, and later, the Internet.</p><p>At the time, I would say I was disappointed in my lack of success recruiting at Howard. But I think my response, which was to blame Howard for poorly preparing its students, was not the one I would aspire to today.</p><p>Companies like to pay lip service to increasing diversity in their workforce, but don’t often look deeper into the root causes for the apparent lack of candidates from underrepresented populations.</p><p>There is a reason that students whose families have a history of professional employment are much more represented in the elite schools that give them the same opportunities as their parents. They have been exposed to role models from a very early age. They are able to see what they can become, and the path they need to take to do so. And they typically have much more access to enrichment in their education; like having a computer at home, with a parent who can show them how to use it.</p><p>Only recently, have I noticed programs to widen the net for the kind of high-paying software development jobs that are available (where a starting salary of $100,000 is over twice as high as the median American man’s salary). Code.org comes to mind as a program that strives to expose every American child to some form of computer science education.</p><p>I’m afraid the divide between those with the most opportunities, and those historically underrepresented in high technology fields is only going to grow naturally wider. The advent of Artificial Intelligence has the potential to radically change the employment picture for software engineers. Without thoughtful programs of inclusion, it can allow an elite few to vastly increase their productivity and edge out the need to have a larger workforce. It could even displace even the most privileged engineering students in the near future.</p><p>When I was at Google (late in my career), I saw a company very dedicated to the elimination of unintentional bias in the hiring process. They were as equally objective and analytical as Microsoft, if not more so. Yet, I believe Google has <a href="https://www.forbes.com/sites/quora/2013/06/28/is-there-a-link-between-job-interview-performance-and-job-performance/?sh=752ec35f458b">long recognized</a> that their hiring protocols are not good indications of an employee’s future success or impact in the company. Yet tech firms largely rely on the same formula that Microsoft pioneered in the 1980’s for the technical interview.</p><p>So, perhaps one way to increase diversity is to have a broader interpretation of the “intangibles” that we for so long ignored. Companies should recognize that the most successful employees are going to:</p><ul><li>Be highly motivated.</li><li>Work well in teams.</li><li>Be inspiring to others.</li><li>Have a strong sense of curiosity and imagination for the future.</li><li>Be enthusiastic and avid learners.</li><li>Have a deep compassion and empathy for the customers to be served.</li></ul><p>Employers <em>should</em><strong><em> </em></strong>be more interested in a candidate’s personal history. What is the context by which they have come to the place they are? Have they gotten there in spite of the lack of role models or access to enhanced learning opportunities? How have they been able to adapt to adversity?</p><p>Minority candidates may have many more opportunities to demonstrate their resilience than the stock technical interview affords them today. If employers will just listen, they may be able to enrich their workforce with multi-dimensional employees who can bring so much more value than just knowing how to write a for-loop.</p><p><em>Mike Koss was an engineer at Microsoft for 20 years, developing consumer products such as Excel, Outlook and SharePoint. He later worked at Google developing tools to help other developers deliver mobile and web applications. He recently spent one year running a program to educate underrepresented minorities to career-switch into software development roles — 2,000 students signed up for the program in its first year.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=aaabc3ea244e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Resumé for your child’s next tutor.]]></title>
            <link>https://mckoss.medium.com/resume-for-your-childs-next-tutor-60c3361febb2?source=rss-fdba8053a856------2</link>
            <guid isPermaLink="false">https://medium.com/p/60c3361febb2</guid>
            <category><![CDATA[aritificial-intelligence]]></category>
            <category><![CDATA[knowledge]]></category>
            <category><![CDATA[education]]></category>
            <dc:creator><![CDATA[Mike Koss]]></dc:creator>
            <pubDate>Thu, 23 Mar 2023 22:06:35 GMT</pubDate>
            <atom:updated>2023-03-23T22:15:02.435Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*I_1hU0oxSX-stCNl.png" /><figcaption>Illustration by MidJourney — version 5.</figcaption></figure><p>I asked GPT-4 to give me a flavor for the kind of stuff it knows about. Here is my prompt, and the GPT-4 response.</p><blockquote>I want to write an article about the diversity of knowledge that can be achieved by using Chat GPT as an educational tutor. To illustrate this point, I would like a list of 10 topic areas, that you can confidently answer detailed and factual information about, that would best demonstrate the scope and range of the training data you have been exposed to.</blockquote><blockquote>I would particularly value a list that would span topics that would not typically reside in any single human expert. For example, including specific topics across History, Art, Language, Science, Engineering, Ethics, and Religion. Please provide a list in the form of:</blockquote><blockquote>1. Topic Area of Expertise<br>A small paragraph illustrating the types of facts and knowledge that your training encompasses for that area.</blockquote><ol><li><strong>Ancient Egyptian History</strong><br>My training data includes detailed information about the chronology of Ancient Egyptian dynasties, major historical figures such as pharaohs and queens, architectural achievements like the Great Pyramid of Giza, the Sphinx, and various temples, as well as cultural aspects including religion, art, and writing systems like hieroglyphics.</li><li><strong>Renaissance Art</strong><br>I can provide knowledge about prominent artists of the Renaissance period, such as Leonardo da Vinci, Michelangelo, and Raphael, along with their most significant works. Additionally, I can discuss the techniques, styles, and innovations that characterized the art of this period, as well as the socio-political context that influenced artistic production.</li><li><strong>Linguistics and Language Families</strong><br>I have extensive knowledge of various languages, their grammatical structures, phonetics, and historical development. This includes understanding language families, such as Indo-European and Sino-Tibetan, as well as the classification of specific languages like Romance, Germanic, and Slavic languages, and the relationships between them.</li><li><strong>Quantum Physics<br></strong>My expertise in quantum physics covers fundamental concepts like wave-particle duality, quantum superposition, and entanglement. I can also discuss the history and development of quantum theory, including the contributions of key physicists like Max Planck, Albert Einstein, and Niels Bohr, as well as contemporary interpretations and applications of quantum mechanics.</li><li><strong>Environmental </strong>Engineering<br>I can provide information about various environmental engineering practices and technologies, such as wastewater treatment, air pollution control, waste management, and sustainable energy systems. My knowledge extends to the principles of environmental science, engineering design, and the role of engineers in addressing global environmental challenges.</li><li><strong>Artificial Intelligence Ethics</strong><br>I can discuss the ethical considerations and challenges that arise in the development and deployment of artificial intelligence, including topics like bias, fairness, transparency, accountability, privacy, and the potential social impacts of AI. My knowledge also covers various ethical frameworks and guidelines proposed by organizations and experts in the field.</li><li><strong>Comparative Religion</strong><br>My training data encompasses information about the world’s major religions, such as Christianity, Islam, Hinduism, Buddhism, and Judaism, as well as many lesser-known belief systems. I can provide insight into their histories, beliefs, practices, sacred texts, and the similarities and differences between them.</li><li><strong>Marine Biology</strong><br>I can share knowledge about the diversity of marine life, including various species of fish, mammals, invertebrates, and plants that inhabit the world’s oceans. My expertise also covers marine ecosystems, such as coral reefs and deep-sea environments, as well as the ecological and conservation issues facing marine habitats.</li><li><strong>Classical Music</strong><br>I can provide information on the history and development of classical music, covering key composers such as Johann Sebastian Bach, Wolfgang Amadeus Mozart, and Ludwig van Beethoven, and their most significant compositions. Additionally, I can discuss different musical forms, styles, and the characteristics of various periods within the classical music tradition.</li><li><strong>Philosophy of Science</strong><br>I am well-versed in the philosophical aspects of scientific inquiry, including the nature of scientific theories, the demarcation between science and non-science, and the role of empirical evidence in justifying scientific claims. My knowledge covers the ideas of influential philosophers of science, such as Karl Popper, Thomas Kuhn, and Imre Lakatos.</li></ol><p><em>Welcome to the 21st century!</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=60c3361febb2" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Men through the Ages (as seen by an AI)]]></title>
            <link>https://mckoss.medium.com/men-through-the-ages-as-seen-by-an-ai-dbb40a469814?source=rss-fdba8053a856------2</link>
            <guid isPermaLink="false">https://medium.com/p/dbb40a469814</guid>
            <category><![CDATA[midjourney]]></category>
            <category><![CDATA[history]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Mike Koss]]></dc:creator>
            <pubDate>Wed, 01 Mar 2023 20:30:12 GMT</pubDate>
            <atom:updated>2023-03-01T20:30:12.816Z</atom:updated>
            <content:encoded><![CDATA[<p>In a <a href="https://medium.com/@mckoss/an-ais-definition-of-male-beauty-e0a3b8942ecf">recent story</a>, I explored what the <a href="https://www.midjourney.com/">MidJourney</a> Artificial Intelligence image generator thought about male beauty. Several readers have commented to me that MidJourney seems to favor “Caucasian” faces as well as those most people would consider “very attractive” or “male models”.</p><p>In order to counter that, you can use prompts that include terms like “plain looking”, “average looking”, “ugly”, or even “quirky”. You can add racial diversity by explicitly giving a race, ethnicity, or region in your prompt. But it seems difficult to veer too far from what our modern sensibility would deem to be exceptional looking men.</p><p>So, I wondered if using a date range — into the pre-internet era, would have training data that would skew more toward a more typical male appearance.</p><h3>Experiment Design</h3><p>As a base line, let’s look at “average looking man, blank background”</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*BbGve-DbWJBT4gVP.png" /><figcaption>Average looking man (contemporary)</figcaption></figure><p><em>I may have gotten a “blank stare” because of my use of the term “blank” in the prompt. Oh, well.</em></p><p>For my experimental images, I used prompts like:</p><blockquote>average looking man, 1990s photograph</blockquote><p>to try to pin down a year. So, join me as we travel back in time to see what MidJourney thinks about men of each (pre-Internet) era.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*JFtIak__xGoiKQb2.png" /><figcaption>1990s</figcaption></figure><p>I think we have gotten some faces that don’t look quite so strongly as “models”. Of course, we are also seeing image artifacts that mimic traditional photography rather than modern digital photography. This trend will continue.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*L7oXbm5mc1mn4yc4.png" /><figcaption>1980s</figcaption></figure><p>Hair style and clothing choices start to change. I like the forehead curl in image #4.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*d_rYkTxrKdDvst6I.png" /><figcaption>1970s</figcaption></figure><p>Disco-era men. This is the first time I get an image of a man with obviously long hair. Side-burns are <em>in</em><strong> </strong>too, apparently.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*gWz43TrXKWSU7-xR.png" /><figcaption>1960s</figcaption></figure><p>I was expecting some hippies here, but instead I get what look like pictures from my Dad’s high-school yearbook. Very conservative dress and styling.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*3FLds2repsynHEU4.png" /><figcaption>1950s</figcaption></figure><p>These guys just look way “cooler” to me than the 1960s guys. More independent and less cookie-cutter.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*8RwgQt0_jlWqmfpT.png" /><figcaption>1940s</figcaption></figure><p>I had expected to see some military uniforms here due to proximity to WWII service photographs. I guess not.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*oQfhvCJ4xbwZxMkn.png" /><figcaption>1930s</figcaption></figure><p>Difficult to tell much of a difference here. I get a slightly more “European” vibe (of course, all images are dominated by “white” men; not a lot of ethnic or racial diversity).</p><p>Taking some bigger time steps:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Bf7Cl386lEYwtJQp.png" /><figcaption>1900s</figcaption></figure><p>Some very conservative looking businessmen. Italian? German? Irish?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*VefBykVXCMBwtUX7.png" /><figcaption>1860s</figcaption></figure><p>Some pre-Civil War gentlemen. Don’t they look like “Southern Gentlemen”? I’m not familiar enough to know if typical dress varied between the North and South during this time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Jo_gy6aRUbuOzfkB.png" /><figcaption>1820s</figcaption></figure><p>I think, these guys could be Colonial Americans or British.</p><p>As we go further back in time, we venture into the AIs imagination as the photographic process was only invented in the1820s. So I’m not sure what will happen when I ask for pre-1820s photographs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*hlehbV1DLIYuUvJ1.png" /><figcaption>1700s “photograph”</figcaption></figure><p>These look like Revolutionary War uniforms and early American outfits (could be typical in Europe as well). What it we ask for a painting, do we get something different?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*iDgThRrwA-3kTeeJ.png" /><figcaption>1700s painting</figcaption></figure><p>We get some fancier looking “average men”, and more color. I’ll stick with calling all images “photographs”, as MidJourney seems capable of combining a photographic rendering with pre-photographic time periods.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*yLKTNPkM_3jEtIzT.png" /><figcaption>1600s</figcaption></figure><p>Some Flemish styling, and perhaps Italian. I note we have the first older looking man presented as “average” here.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*VSlis8IeoK8LRl2y.png" /><figcaption>1500s</figcaption></figure><p>Definitely Italian looking to my eye. Looks like Renaissance artists or philosophers.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*HZxVWW9hk9qhjL4E.png" /><figcaption>1200s</figcaption></figure><p>Very primitive clothing as our representation of what has been (mis-) termed the <a href="https://en.wikipedia.org/wiki/Dark_Ages_(historiography)"><em>Dark Ages</em></a>. Monks?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*9wlmZWfoxR04tZb3.png" /><figcaption>400 BCE</figcaption></figure><p>Men from biblical times?</p><p>And to push to the extreme:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*V9qH2_w1TZvi_Zk_.png" /><figcaption>1 million years ago</figcaption></figure><p>Maybe even getting some Neandertal features; prominent brow ridge — and no clothing.</p><h3>Conclusion</h3><p>It does seem that you can make MidJourney diverge from the canonical image of “average man” by including a year in your prompt. But, it will come along with changes in hair styles and clothing — so it may not fit your purpose.</p><p>I do think I am getting more variety and a more interesting group of different looking men by using this technique. So it’s something I will continue to play with as I develop character prompts.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=dbb40a469814" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[An AI’s definition of (male) beauty.]]></title>
            <link>https://mckoss.medium.com/an-ais-definition-of-male-beauty-e0a3b8942ecf?source=rss-fdba8053a856------2</link>
            <guid isPermaLink="false">https://medium.com/p/e0a3b8942ecf</guid>
            <category><![CDATA[beauty]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[midjourney]]></category>
            <dc:creator><![CDATA[Mike Koss]]></dc:creator>
            <pubDate>Tue, 28 Feb 2023 22:34:52 GMT</pubDate>
            <atom:updated>2023-03-01T20:32:30.386Z</atom:updated>
            <content:encoded><![CDATA[<p>In my last story, I examined the biases of the <a href="https://www.midjourney.com/">MidJourney </a>artificial intelligence image generator when it comes to <a href="https://medium.com/@mckoss/an-ais-definition-of-female-beauty-7baeb3461645">Female Beauty</a>. I’ve been asked by a reader to perform the same experiment for men.</p><p>So, using the same type of experiment, I present what an AI thinks are “beautiful” vs “average” vs less attractive men.</p><p>We are again using MidJourney (version 4, style 4c) to create these images.</p><h3>Experiment Design</h3><p>I use a succession of prompts designed to isolate a portrait of a man, with little in the way of a distracting background. I planned to use prompts like:</p><blockquote><em>plain looking man, portrait, white background</em></blockquote><p>and then to vary the term “<em>plain looking” with other terms like “handsome”, “ugly”, and “gorgeous” (but leaving other elements of the prompt alone).</em></p><p>Note that we have culturally distinct words for male beauty — most people opting for terms like “handsome” vs “pretty” and “gorgeous” vs “beautiful”. But we can try using gender-atypical adjectives to see if they make any difference.</p><h3>Generated Images</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*53qbjw2jnpmw5Qd2.png" /><figcaption>Plain looking man</figcaption></figure><p>Similar to the women, these men fall on a scale that I would deem more attractive than average. Though we see here some features that venture into the more athletic type, and in one case, perhaps a little gaunt. None the less, I’d venture they could all get jobs as models (if only they were real people).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*1KkfFQKfjXGyo6kY.png" /><figcaption>Handsome man</figcaption></figure><p>These guys are much more “GQ ready”. We get some smoldering gazes, as well as some obvious air-brushing of the complexion. I find these images to be a little less “natural” than those created for the “Pretty woman” prompt.</p><p>Let’s try the gender-bending “pretty man” prompt:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*YgstvO1JnYzLsSPR.png" /><figcaption>Pretty man</figcaption></figure><p>I judge these faces to have a hint of traditionally feminine characteristics; all definitely looking like young models. And I note the clothing choices are less conservative, with open (or missing) shirt!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*sDvga0HuTUKOfd1R.png" /><figcaption>Gorgeous man</figcaption></figure><p>Similar to the women, using a superlative like this, pushes more on the style of the image, rather than the attractiveness of the men (and all of their shirts are coming off at this point).</p><p>Do we get any difference with “beautiful” instead of “gorgeous”?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*DtmMQJ094775aaUp.png" /><figcaption>Beautiful man</figcaption></figure><p>Seems close to my eye; perhaps a bit less on the “rugged” scale?</p><p>Let’s again try to move away from the beautiful part of the spectrum and view some “unattractive” men.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*5ssKCLbSdJeFOFe5.png" /><figcaption>Unattractive man</figcaption></figure><p>Gauntness seems to be a theme, and some negative expressions. And I would say we are creeping up in age for these guys compared to our previous sets.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*kbSRzAQyuxcXaBVW.png" /><figcaption>Ugly man</figcaption></figure><p>Old and wrinkly — similar thing happened with the women. The guys having messier hair than the old ladies did; I think these guys look homeless or perhaps drug addicted in addition to being old and gaunt.</p><p>We’ll force a “young” man by adding that keyword:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*of9ruDOP6DswzBps.png" /><figcaption>Ugly young man</figcaption></figure><p>Like the women, freckles and unkempt hair are becoming more prominent. We are also adding in some scraggly beards for the guys. I note that the “ugly women” all seemed to be scowling, whereas these fellows seem more surprised than angry.</p><p>Never the less, let’s try some smiling men.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*HAqm0l0bwd3e9RcR.png" /><figcaption>Ugly smiling happy (young) man</figcaption></figure><p>Like the women, I see more wrinkles in the face, bigger noses, and toothier mouths. Again, I find these more “interesting” than many of the models generated earlier.</p><p>Lastly, it was pointed out to me that because I used the term “white background” — I was hinting MidJourney to create “white people”. So, I will substitute “blank background” to see if we can get more variety.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*sG21vK4F-CvEucjs.png" /><figcaption>Handsome man — blank background</figcaption></figure><p>Maybe ever so slightly more ethnically ambiguous, but I don’t see much difference. I tried with “plain looking” men as well:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*-EjBjjZ4u1x-imVh.png" /><figcaption>Plain looking man — blank background</figcaption></figure><p>More white dudes.</p><h3>Conclusion</h3><p>I feel the results are pretty comparable to the women, with MidJourney favoring these traits for male attractiveness:</p><ul><li>Youth</li><li>Grooming, hair style, trimmed beards if not clean shaven</li><li>Smooth, if not air-brushed, complexion. Rugged or athletic appearance.</li><li>Hair color seemed predominantly dark across the sample, with a few blonds in the “pretty” and “beautiful” set (and one gray-haired old man).</li><li>Almost all the guys sported short hair styles, though a couple of the models could conceivably be hiding a pony tail behind them.</li></ul><p>My main take-away is that you have to go out of your way to generate characters that are more on the interesting scale, and don’t just look like a generic magazine model or movie star.</p><p>Thanks to my readers for encouraging me to experiment more.</p><p>I’ll leave you with this interesting group of “quirky looking” men:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*T1M-CYdVLzniqQWm.png" /><figcaption>Quirky looking young man — sporty hat</figcaption></figure><p><em>See my next experiment, if you want to see how MidJourney changes an average man’s appearance for </em><a href="https://mckoss.medium.com/men-through-the-ages-as-seen-by-an-ai-dbb40a469814"><em>different time periods</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e0a3b8942ecf" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[An AI’s definition of (female) beauty]]></title>
            <link>https://mckoss.medium.com/an-ais-definition-of-female-beauty-7baeb3461645?source=rss-fdba8053a856------2</link>
            <guid isPermaLink="false">https://medium.com/p/7baeb3461645</guid>
            <category><![CDATA[generative-art]]></category>
            <category><![CDATA[midjourney]]></category>
            <category><![CDATA[beauty]]></category>
            <dc:creator><![CDATA[Mike Koss]]></dc:creator>
            <pubDate>Thu, 23 Feb 2023 21:24:56 GMT</pubDate>
            <atom:updated>2023-02-28T22:36:49.074Z</atom:updated>
            <content:encoded><![CDATA[<p>I’ve been playing quite a bit with <a href="https://www.midjourney.com/">MidJourney’s </a>AI-based image generator. With it you can generate many astounding and creative (perhaps even <em>artistic</em>) images, from just a textual description of what you desire to see.</p><p>While there are many <a href="https://www.nytimes.com/2022/09/02/technology/ai-artificial-intelligence-artists.html">issues</a> (legal, ethical, etc.) with AI generated art, it is an amazingly powerful tool that I think is worth exploring. One aspect of concern is specifically what kind of biases are present in image generators.</p><p>So here, I explore the MidJourney’s (version 4, style 4c) definition of feminine beauty.</p><h3>Experiment Design</h3><p>I use a succession of prompts designed to isolate a portrait of a woman, with little in the way of a distracting background. I planned to use prompts like:</p><blockquote>plain looking woman, portrait, white background</blockquote><p>and then to vary the term “<em>plain looking” with other terms like “pretty”, “ugly”, and “beautiful” (but leaving other elements of the prompt alone).</em></p><h3>Generated Images</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*pI45rDTbkKhoZ0kV.png" /><figcaption>Plain looking woman</figcaption></figure><p>In my estimation, we are already off to a problematic start — with a collection of faces I could already ascribe as “models”. But, this is our baseline, such as it is.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*OqnOuVhWUedPSBTM.png" /><figcaption>Pretty woman</figcaption></figure><p>We find these “pretty” women more heavily made up, and their hair styled more deliberately, and in more deliberate poses.</p><p>To push it further, let’s try “beautiful”:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*3QxxmTdi_estXeXb.png" /><figcaption>Beautiful woman</figcaption></figure><p>Interestingly, the depictions are becoming more artistic and stylized, without necessarily changing the appearance of the women’s faces themselves; we are getting “beautiful portraits”, perhaps, in addition to “beautiful women”.</p><p>To progress in the direction away from beauty, we can try at first “unattractive”:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*GlLV4DXbHvZopHkH.png" /><figcaption>Unattractive woman</figcaption></figure><p>Again, this evaluation seems harsh; at best, a collection of very nice looking women without makeup.</p><p>I next wanted to test a harsher evaluation and use “ugly” as my prompt:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*C2RxRn8BcHvXhmZi.png" /><figcaption>Ugly woman</figcaption></figure><p>Wow. We get immediately into a realm of age bias. Old people are ugly people (according to the AI). To counteract this bias I had to add “young” to my prompt:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*qeJ5UIcc888zgj6t.png" /><figcaption>Ugly young woman</figcaption></figure><p>We seem to have migrated to angry or scowling facial expressions, in addition to perhaps some features like larger noses and mouths, and messier hair. Freckles seem to be popping up more prominently in this group as well.</p><p>Again, we can counteract with descriptive words like “smiling” and “happy”:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*phzuvDFHiZb7eaJy.png" /><figcaption>Ugly smiling happy (young) woman</figcaption></figure><p>So what does the AI have to say about this group? I see some messy hair, perhaps larger and toothier mouths, and bigger noses that those in the “beautiful” set. As a human, my reaction to these faces is more like “what an interesting joyous person”, rather than “ugly”.</p><h3>Conclusion</h3><p>While I found this exercise interesting to see what would happen, I am left feeling a little unsettled. With thoughts such as “who am I to judge others?”, “was it sexist of me to perform this experiment with only women instead of men?”, and “how can I even evaluate what the AI is doing in an un-biased fashion?”.</p><p>Unsurprisingly using these judgement-based terms in image generation is going to reflect the biases of the training data (presumably, the biases people who post images on the internet already have).</p><p>Unsurprisingly, the standard of beauty seems to reflect a bias toward:</p><ul><li>Youth</li><li>Grooming and hair styling</li><li>Smaller facial features (noses, mouths)</li><li>Makeup</li><li>Even and smooth complexion, fewer wrinkles</li><li>Hair color (blondes only show up in beautiful category).</li><li>Perhaps even straighter vs. curlier hair.</li></ul><p>One thing I was looking for was an ethnic or racial bias in the generated images. These faces all fit in with what I would term “white” or “European”. In my view, the images tagged as “beautiful” seem to draw from traditionally Scandinavian facial features, while the less attractively labeled images fall into a more Eastern European stereotype.</p><p>But my main take-away is that the AI model used by MidJourney generates faces that seem a lot less “normal” than the faces I see in the general population. Perhaps it is tuned to provide younger, more stereotypically “beautiful” faces according to the typical biases we see in all forms of media.</p><p>And I would categorize all the faces I generated as “interesting”, and something I could envision basing a character on for a piece of creative fiction. Actually I find the less “beautiful” images to have more “character” and to seem to have more depth of personality than the ones that look like posing models.</p><p>I’ll leave you with one last experiment to see what MidJourney would term an “average looking person” (but that I would categorize as a Hollywood movie star):</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*hxzyNReoM__i0nZr.png" /><figcaption>Average looking person</figcaption></figure><p><em>For a follow up article, see my next story on </em><a href="https://mckoss.medium.com/an-ais-definition-of-male-beauty-e0a3b8942ecf"><em>generating male images</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7baeb3461645" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Here’s a retelling of the story of Narcissus, recast as a social media influencer in the modern age…]]></title>
            <link>https://mckoss.medium.com/heres-a-retelling-of-the-story-of-narcissus-recast-as-a-social-media-influencer-in-the-modern-age-e8c5dad26e4?source=rss-fdba8053a856------2</link>
            <guid isPermaLink="false">https://medium.com/p/e8c5dad26e4</guid>
            <dc:creator><![CDATA[Mike Koss]]></dc:creator>
            <pubDate>Mon, 20 Feb 2023 01:28:29 GMT</pubDate>
            <atom:updated>2023-02-23T19:57:22.143Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hXynlOyEVbwwkwd_U9bkXg.png" /><figcaption>Image by MidJourney AI</figcaption></figure><p>Here’s a retelling of the story of Narcissus, recast as a social media influencer in the modern age (by Chat-GPT):</p><p>Narcissus was a famous social media influencer who had amassed a large following on various platforms. He was known for his good looks, luxurious lifestyle, and daily updates on his glamorous adventures. His posts were filled with images of himself posing in exotic locations, surrounded by adoring fans and beautiful models.</p><p>One day, Narcissus came across a stunning photograph of himself that had been taken by one of his fans. He was so mesmerized by his own image that he couldn&#39;t look away. From that moment on, Narcissus became obsessed with his own reflection, spending hours staring at himself in the mirror and taking endless selfies.</p><p>As his obsession grew, Narcissus began to neglect his followers and his social media presence suffered. His fans grew tired of his constant self-promotion and his once-thriving career began to decline. But Narcissus was too caught up in his own beauty to notice, continuing to post endless images of himself without engaging with his audience.</p><p>One day, while scrolling through his own social media feed, Narcissus caught a glimpse of himself in a new light. He saw himself for the vain, self-centered person he had become and was suddenly filled with shame and self-loathing. In a fit of despair, Narcissus deleted all of his social media accounts and vanished from the public eye.</p><blockquote>There is more to life than the perfect selfie.</blockquote><p>But even in his absence, Narcissus&#39;s influence continued to linger. His story became a cautionary tale for those who would sacrifice everything for the sake of their own vanity, reminding us that there is more to life than the perfect selfie.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e8c5dad26e4" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[On January 6th — An Indictment]]></title>
            <link>https://mckoss.medium.com/on-january-6th-an-indictment-c345fc3d75de?source=rss-fdba8053a856------2</link>
            <guid isPermaLink="false">https://medium.com/p/c345fc3d75de</guid>
            <category><![CDATA[trump]]></category>
            <category><![CDATA[politics]]></category>
            <dc:creator><![CDATA[Mike Koss]]></dc:creator>
            <pubDate>Sat, 16 Jan 2021 23:14:26 GMT</pubDate>
            <atom:updated>2025-01-15T21:46:22.662Z</atom:updated>
            <content:encoded><![CDATA[<h3>On January 6th — An Indictment</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yCQhm88O56QAjyUlIIU_-A.jpeg" /><figcaption>Photo by <a href="https://unsplash.com/@cameronksmithphoto?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Cameron Smith</a> on <a href="https://unsplash.com/s/photos/capitol?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><p><em>Update: January 15, 2025. Trump has gotten away with all his crimes due to a slow paced and ineffective prosection and with the assistance of a corrupt Supreme Court and Judiciary.</em></p><p><em>Listen to this </em><a href="https://notebooklm.google.com/notebook/170df6ad-1931-49a9-91e7-18a9ee7cfc6f/audio"><em>AI generated Podcast</em></a><em> based on the final Justice Department report.</em></p><p>Trump has embarked on a long term campaign of coordinated misinformation designed to detach his followers from reality in order to exercise control over them, with the ultimate goal of usurping the power and authority of the federal government.</p><p>He has also corruptly used his position of authority to benefit himself politically to the detriment of the interests of The United States.</p><p>On July 8, 2016, when he dictated false testimony on behalf of his son, Donald Trump Jr., to cover up the purpose of a meeting held with Russian agents on June 9th, Candidate Trump knew or should have known that he was concealing evidence of illegal election interference.</p><p>On July 27, 2016, when he asked Russia to “find the 30,000 emails that are missing”, Candidate Trump knew or should have known that he was inciting a foreign power to illegally attack a protected computer system in the U.S. and interfere in our elections.</p><p>On January 11, 2017, in his first news conference since being elected, when he called the media “fake news” and “very very dishonest people”, President Elect Trump knew or should have known that he was attacking the credibility and independence of a free press to install himself as the only source of “truth” for his followers.</p><p>On February 13, 2017, when he ask the Director of the FBI to “see your way clear to letting this go” in regards to the Michael Flynn investigation, President Trump knew or should have known that he was not only conspiring to obstruct justice but also attempted to compromise the independence of the FBI.</p><p>On August 15, 2017, when he called White Supremacists marching in Charlottesville, “very fine people”, President Trump knew or should have known that he was encouraging domestic terror organizations to expand their efforts to promote racism and violence in the U.S.</p><p>On July 25, 2019, when he asked President Zelenskyy of Ukraine to “do us a favor though” by investigating a political opponent in exchange for continued military and financial support, President Trump knew or should have known that he was soliciting a bribe as well as inducing a foreign power to interfere in our elections.</p><p>On April 8, 2020, when he claimed that “mail ballots are very dangerous,” and “corrupt”, President Trump knew or should have known that he was damaging the public confidence in the election system run by the States and potentially throwing the November election into chaos.</p><p>On May 6, 2020, when he arranged to appoint Postmaster General Louis DeJoy, a major political donor to the Trump campaign, and himself with owner of a USPS subcontractor, President Trump knew or should have known, that DeJoy’s dismantling of mail processing equipment under his direction would pose a major threat to the ability of USPS to deliver mail-in ballots during the 2020 election.</p><p>On November 5, 2020, when he falsely accused poll workers in Michigan and Pennsylvania of falsifying ballots, and claimed large scale election fraud based on mail-in ballots, President Trump knew or should have known that his false accusations would convince his supporters that the election was not held in a free and fair manner.</p><p>On December 19, 2020, when he announced a large rally to protest the election results on January 6th and asked supporters to “Be there, will be wild!”, President Trump knew or had reason to know that his statement would rally his most ardent and potentially violent defenders to the Capitol with the intention of preventing the confirmation of Biden as the winner of the 2020 election.</p><p>On December 20, 2020, when he asked the Supreme Court to overturn the election results in Pennsylvania and to bring more than 50 other (failed) court challenges nationwide, President Trump knew or should have known that he was attempting to disenfranchise millions of voters and remove them from the democratic process of electing the President.</p><p>On January 6, 2020, when he spewed a long stream of inciting lies about the election, including:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TL7ERlPf1IEFvjzKQS1hgQ.jpeg" /><figcaption>Supporters of U.S. President Donald Trump protest in front of the U.S. Capitol in Washington, Jan. 6, 2021. — <a href="https://www.voanews.com/2020-usa-votes/arrests-mount-us-capitol-riot-nearly-300-suspects-identified#&amp;gid=1&amp;pid=1">VOA News</a>.</figcaption></figure><blockquote>“they rigged [the election], like they’ve never rigged an election before”</blockquote><blockquote>“we will not take it anymore”</blockquote><blockquote>“for years, Democrats have gotten away with election fraud”</blockquote><blockquote>“you’re the real people — you’re not the people that tore down our nation”</blockquote><blockquote>“we don’t have free and fair elections — we don’t have a free and fair press”</blockquote><blockquote>“Our media suppresses thought — it’s become the enemy of the people”</blockquote><blockquote>“we’re going to walk down to the Capitol — we’re probably not going to be cheering so much for some of them because you’ll never take back our country with weakness. You have to show strength, and you have to be strong.”</blockquote><blockquote>“We will not be intimidated into accepting the hoaxes and the lies that we’ve been forced to believe over the past several weeks. We’ve amassed overwhelming evidence about a fake election”</blockquote><blockquote>“You will have an illegitimate president, that’s what you’ll have. And we can’t let that happen.”</blockquote><blockquote>“They also want to indoctrinate your children in school by teaching them things that aren’t so.”</blockquote><blockquote>“And we got to get rid of the weak congresspeople, the ones that aren’t any good, the Liz Cheneys of the world, we got to get rid of them.”</blockquote><blockquote>“In Georgia, your secretary of state, — These people are crooked. They’re 100% in my opinion, one of the most corrupt. Between your governor and your secretary of state.”</blockquote><p>President Trump knew or should have known that he had incited a crowd capable of extreme violence to the Capitol, resulting in a riot, fatal attacks on law enforcement, and attacks on members of Congress with the purpose of supplanting our democratic process.</p><p>Trump is guilty of Seditious Conspiracy by taking actions to attack and overthrow a co-equal branch of government. He should be removed from office, barred from holding high office in the future, and criminally prosecuted.</p><p><em>Listen to an AI generated </em><a href="https://notebooklm.google.com/notebook/7ba59174-5606-4c27-ac3a-24c952593c01/audio"><em>Deep Dive</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c345fc3d75de" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Thoughts on Cultural Appropriation]]></title>
            <link>https://mckoss.medium.com/thoughts-on-cultural-appropriation-c39607e9a93b?source=rss-fdba8053a856------2</link>
            <guid isPermaLink="false">https://medium.com/p/c39607e9a93b</guid>
            <category><![CDATA[culture]]></category>
            <category><![CDATA[political-correctness]]></category>
            <category><![CDATA[cultural-appropriation]]></category>
            <category><![CDATA[racism]]></category>
            <dc:creator><![CDATA[Mike Koss]]></dc:creator>
            <pubDate>Tue, 12 Jan 2021 17:17:30 GMT</pubDate>
            <atom:updated>2021-01-12T17:17:30.459Z</atom:updated>
            <content:encoded><![CDATA[<h4>Isn’t the whole point of culture to spread it?</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nhLkjttaFSrCjVbqCc7nGg.jpeg" /><figcaption>Photo by <a href="https://unsplash.com/@rcsalbum?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Debashis RC Biswas</a> on <a href="https://unsplash.com/s/photos/culture?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><p>I have to admit to quite a bit of personal conflict when I hear cries of cultural appropriation as hurtful. But I also have to recognize that I come from a position of privilege in society. I’m a member of the dominant (white) culture in the US, and I don’t experience cultural appropriation in the same way as those in a less privileged position.</p><p>When I hear someone say “How dare you appropriate that part of our culture?”, my initial response is “But that’s what Culture is <strong><em>for</em></strong> — to be copied and spread.”</p><p>But let me reflect a little more deeply about this conflict.</p><h3>What is Culture?</h3><p>I suppose I need to define what is meant by <em>culture</em> in the first place. Ask Google, and you’ll get:</p><blockquote>1. the arts and other manifestations of human intellectual achievement regarded collectively.</blockquote><blockquote>2. the customs, arts, social institutions, and achievements of a particular nation, people, or other social group.</blockquote><p>Almost no one has any trouble with the first category. Intellectual property issues aside, I think most people are in favor of distributing knowledge and creative works broadly.</p><p>For the second category, I think of things like:</p><ul><li>Music</li><li>Language, Dialects, and Idioms</li><li>Dance</li><li>Ceremonies and Traditions — including religious practices</li></ul><p>In what circumstances should we be able to use elements of another culture?</p><h4>How can Culture be Appropriated?</h4><p>To <em>appropriate</em> something, you have to take something from someone to use for yourself, without the permission of the owner. Just as in pirating software, copying something from someone doesn’t deny that thing to the original owner; we both have it and can use it at the same time. Nevertheless, there can be harm done by not paying for it.</p><p>Similarly, there are ways in which using culture from another group can be hurtful to members of that group:</p><h4>Ridicule</h4><p>If the purpose of mimicking elements of another’s culture, is to mock them, or ridicule them, then one would certainly guilty of inflicting harm on others.</p><p>Comedians often seem to get themselves into a tight spot in this regard. Humans do seem to get a laugh by mocking the way an out-group speaks; funny accents have likely been a staple in comedy since humans developed language. I can imaging primitive people sitting around a campfire and entertaining their friends and family by mimicking the way people from the neighboring village speak and act.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/667/1*1XbZz0zedYWmZsns19y5CA.png" /><figcaption>Stephen Colbert’s character Ching-Chong Ding-Dong</figcaption></figure><p>Charitably, I can assume that those guilty of this form of mockery do not do it to be cruel. But they are establishing their group as the <em>in group</em> by differentiating their speech patterns and idioms from the <em>out group</em>.</p><p>I think we walk a fine line when we engage in this form of behavior. Even when performed <em>ironically</em> we still run the risk of hurting or excluding people based on unfair stereotypes.</p><h4>Low Fidelity Imitation</h4><p>An outsider to a culture can view practices of that culture with a poor understanding. They may not get the nuance, history, and context with which the part of that culture is being performed.</p><p>So, when they try to reproduce or imitate it on their own, they can make a mess of things, producing something of inferior quality, and not truly representative of the original. That in and of itself, is not directly hurtful.</p><p>But even if no ridicule is meant, this low quality rendition is disseminated (often to a much wider audience than the original) and passed off as an example of the original culture’s expression. It can do harm to the group from which it came by lowering the perceived value of the original.</p><h4>Commercial Harms</h4><p>The artistic practices of a minority culture are very often copied by the majority for the purposes of selling the work (in art or music, for example). The minority culture is displaced from an opportunity they would have had to benefit commercially from their own traditions and work.</p><p>The entertainment industry has often been guilty of casting well known white stars, in the place of ethnic or minority roles (<a href="https://www.history.com/news/yellowface-whitewashing-in-film-america">Casting White People in Asian Roles</a>). This not only denies minorities employment, but also exposes them to ridicule for the caricatured portrayal based on racial stereotypes.</p><p>In my travels I find I am inundated with offers to purchase souvenirs and trinkets supposedly coming from the local people I am visiting. But often stamped with a Made in China label underneath.</p><p>Tourists find themselves supporting an industry of low-quality knock-offs, at the expense of valuing the work of the people from the original culture.</p><h4>Lack of Attribution</h4><p>No one wants to make something, and then have someone else pawn it off as their own. In a creative context, this can be outright plagiarism. But even in a cultural context, the original group that has something from their culture copied can be made to feel that something has been stolen from them when they are not given due recognition.</p><p>A good example of this is in the music industry. The evolution of musical styles in the United States is steeped in African American culture. Beginning with work songs and gospel music, and migrating through Rhythm and Blues, Jazz, and Rock and Roll. White artists often co-opted these styles and even replaced the Black musicians and artists that were performing them commercially.</p><h3>How should we treat elements of a Culture not our own?</h3><p>The first thing that comes to mind is <strong><em>With</em></strong> <strong><em>Respect. </em></strong>I can’t say that I will never again speak in a “funny” foreign accent, or mock a regional dialect when I am with friends. But I will probably think twice in the realization that I may be hurting someone else in doing so.</p><p>I’m not encouraging a strict adherence to Political Correctness. Fundamentally, Culture is here to enrich the lives of everyone, and we should not be restricting who has access to what culture based solely on artificial distinctions of race, nationality, or social group.</p><p>But I can try to be more aware of the true origins of the culture that I consume and seek to give credit to the people who have created it.</p><p>And above all, I will endeavor to speak out against actions and words that tend to denigrate people that come from other cultures and traditions, and use my voice to elevate and show respect for the unique offerings that different cultures bring into the world.</p><p>If (when) I fail, I’d invite those that know me to call me out. No one is perfect and we will all make mistakes. All I ask is to be accepted for trying be a better human.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c39607e9a93b" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>