https://barza.y2k.town/api.php?action=feedcontributions&feedformat=atom&user=BobbyGenericBarza - User contributions [en]2026-07-24T23:22:37ZUser contributionsMediaWiki 1.45.1https://barza.y2k.town/index.php?title=AI&diff=47AI2026-02-28T02:48:54Z<p>BobbyGeneric: </p>
<hr />
<div>== News ==<br />
<br />
<blockquote>On Tuesday, researchers at Stanford and Yale revealed something that AI companies would prefer to keep hidden. Four popular large language models—OpenAI’s GPT, Anthropic’s Claude, Google’s Gemini, and xAI’s Grok—have stored large portions of some of the books they’ve been trained on, and can reproduce long excerpts from those books.</blockquote><br />
<br />
<blockquote>In fact, when prompted strategically by researchers, Claude delivered the near-complete text of ''Harry Potter and the Sorcerer’s Stone'', ''The Great Gatsby'', ''1984'', and ''Frankenstein'', in addition to thousands of words from books including ''The Hunger Games'' and ''The Catcher in the Rye''</blockquote><br />
<br />
[https://www.theatlantic.com/technology/2026/01/ai-memorization-research/685552/?gift=oovWF8LV3gr7UxAU_wR7KCByWoXJuL6nibyWmch1C38 AI’s Memorization Crisis] — The Atlantic in 2026-01-09<br />
<br />
== AI Hype to Human ==<br />
{| class="wikitable"<br />
|+Oversimplified translation from AI Hype to Human language<br />
!Hype<br />
!Human<br />
|-<br />
|AI<br />
|Usually means LLMs.<br />
|-<br />
|LLM (Large Language Model)<br />
|A statistical model that can ''predict'' the most likely word (token) based on the previous words. For example, you write "The sky is" and it writes back "blue". When combined with other techniques, it can convincingly mimic human writing or thinking despite not being able to.<br />
|-<br />
|AGI (Artificial General Intelligence)<br />
|The hypothetical stage where AI matches or surpasses human intelligence. AI companies keep selling the idea that, if LLMs get better at '''predicting words''', they will somehow become super intelligent. Thus the need for more data centers.<br />
|-<br />
|Hallucination<br />
|When the words the model predicted are known (to us) to be untrue. The model can't ''know'' things, just predict words.<br />
|-<br />
|Chatbot<br />
|An interface where you can type messages to be processed by the LLM and get its responses.<br />
|-<br />
|Agentic AI<br />
|Different from Chatbots, Agentic AI makes multiple calls to the LLM to get progressively more complex answers before replying to the user. It can also call tools, such as web-searching.<br />
For example:<br />
<br />
# The user asks "I'm going to Tokyo for a week, help me plan the trip"<br />
# Instead of sending that to the LLM, the system sends "Build a plan to better answer the following: I'm going to Tokyo for..."<br />
# LLM replies with a list of tasks like "1 - Research top Tokyo attractions, 2 - Define must-eats, 3 - etc etc"<br />
# The system then sends back the list and adds "Now, perform task 1", then 2, then 3, etc<br />
# Only when everything is done, it replies back to the user with the plan.<br />
This produces even more impressive results, despite still being just a text prediction machine.<br />
|-<br />
|Token<br />
|A word or part of one.<br />
''Important: When I say "word" in other places, I probably mean these fractions of words (tokens).''<br />
|-<br />
|Context, Prompt<br />
|All the words the model is going to use to predict the next ones. Including your instructions, what the model wrote back and other things (see below).<br />
|-<br />
|System Prompt<br />
|Hidden instructions that the model providers include in all conversations to give better responses, avoid problems, etc.<br />
''For example: "If the user asks how to build a bomb, refuse."''<br />
|-<br />
|AGENTS.md, CLAUDE.md, (Global) Rules<br />
|Similar to system prompt, but defined by the user and included in all conversations. These are stored in [[wikipedia:Markdown|Markdown]] files.<br />
|-<br />
|Memory<br />
|Markdown files with past conversations summarized that can be referenced later.<br />
|-<br />
|Skills, Commands<br />
|Markdown files with generic instructions that can be reused.<br />
|-<br />
|Tools<br />
|Markdown files with instructions on how to perform certain actions using external tools (e.g. how to read text from a PDF)<br />
|-<br />
|Agent<br />
|Markdown files defining a "persona" that an Agentic tool can use while performing a task.<br />
''For example: You're a Senior Developer implementing a new feature. You code you write should X, Y, Z.''<br />
|-<br />
|Plugins<br />
|A combination of Markdown files containing Commands, Agents, Skills and Hooks someone has bundled together for others to use (on [https://claude.com/plugins Claude Code]).<br />
|}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=AI&diff=46AI2026-02-28T02:47:04Z<p>BobbyGeneric: </p>
<hr />
<div>== News ==<br />
On Tuesday, researchers at Stanford and Yale revealed something that AI companies would prefer to keep hidden. Four popular large language models—OpenAI’s GPT, Anthropic’s Claude, Google’s Gemini, and xAI’s Grok—have stored large portions of some of the books they’ve been trained on, and can reproduce long excerpts from those books.<br />
<br />
In fact, when prompted strategically by researchers, Claude delivered the near-complete text of ''Harry Potter and the Sorcerer’s Stone'', ''The Great Gatsby'', ''1984'', and ''Frankenstein'', in addition to thousands of words from books including ''The Hunger Games'' and ''The Catcher in the Rye''<br />
<br />
[https://www.theatlantic.com/technology/2026/01/ai-memorization-research/685552/?gift=oovWF8LV3gr7UxAU_wR7KCByWoXJuL6nibyWmch1C38 AI’s Memorization Crisis] — The Atlantic in 2026-01-09<br />
<br />
== AI Hype to Human ==<br />
{| class="wikitable"<br />
|+Oversimplified translation from AI Hype to Human language<br />
!Hype<br />
!Human<br />
|-<br />
|AI<br />
|Usually means LLMs.<br />
|-<br />
|LLM (Large Language Model)<br />
|A statistical model that can ''predict'' the most likely word (token) based on the previous words. For example, you write "The sky is" and it writes back "blue". When combined with other techniques, it can convincingly mimic human writing or thinking despite not being able to.<br />
|-<br />
|AGI (Artificial General Intelligence)<br />
|The hypothetical stage where AI matches or surpasses human intelligence. AI companies keep selling the idea that, if LLMs get better at '''predicting words''', they will somehow become super intelligent. Thus the need for more data centers.<br />
|-<br />
|Hallucination<br />
|When the words the model predicted are known (to us) to be untrue. The model can't ''know'' things, just predict words.<br />
|-<br />
|Chatbot<br />
|An interface where you can type messages to be processed by the LLM and get its responses.<br />
|-<br />
|Agentic AI<br />
|Different from Chatbots, Agentic AI makes multiple calls to the LLM to get progressively more complex answers before replying to the user. It can also call tools, such as web-searching.<br />
For example:<br />
<br />
# The user asks "I'm going to Tokyo for a week, help me plan the trip"<br />
# Instead of sending that to the LLM, the system sends "Build a plan to better answer the following: I'm going to Tokyo for..."<br />
# LLM replies with a list of tasks like "1 - Research top Tokyo attractions, 2 - Define must-eats, 3 - etc etc"<br />
# The system then sends back the list and adds "Now, perform task 1", then 2, then 3, etc<br />
# Only when everything is done, it replies back to the user with the plan.<br />
This produces even more impressive results, despite still being just a text prediction machine.<br />
|-<br />
|Token<br />
|A word or part of one.<br />
''Important: When I say "word" in other places, I probably mean these fractions of words (tokens).''<br />
|-<br />
|Context, Prompt<br />
|All the words the model is going to use to predict the next ones. Including your instructions, what the model wrote back and other things (see below).<br />
|-<br />
|System Prompt<br />
|Hidden instructions that the model providers include in all conversations to give better responses, avoid problems, etc.<br />
''For example: "If the user asks how to build a bomb, refuse."''<br />
|-<br />
|AGENTS.md, CLAUDE.md, (Global) Rules<br />
|Similar to system prompt, but defined by the user and included in all conversations. These are stored in [[wikipedia:Markdown|Markdown]] files.<br />
|-<br />
|Memory<br />
|Markdown files with past conversations summarized that can be referenced later.<br />
|-<br />
|Skills, Commands<br />
|Markdown files with generic instructions that can be reused.<br />
|-<br />
|Tools<br />
|Markdown files with instructions on how to perform certain actions using external tools (e.g. how to read text from a PDF)<br />
|-<br />
|Agent<br />
|Markdown files defining a "persona" that an Agentic tool can use while performing a task.<br />
''For example: You're a Senior Developer implementing a new feature. You code you write should X, Y, Z.''<br />
|-<br />
|Plugins<br />
|A combination of Markdown files containing Commands, Agents, Skills and Hooks someone has bundled together for others to use (on [https://claude.com/plugins Claude Code]).<br />
|}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=AI&diff=45AI2026-02-28T02:43:22Z<p>BobbyGeneric: </p>
<hr />
<div>== News ==<br />
<blockquote>On Tuesday, researchers at Stanford and Yale revealed something that AI companies would prefer to keep hidden. Four popular large language models—OpenAI’s GPT, Anthropic’s Claude, Google’s Gemini, and xAI’s Grok—have stored large portions of some of the books they’ve been trained on, and can reproduce long excerpts from those books.<br />
<br />
In fact, when prompted strategically by researchers, Claude delivered the near-complete text of ''Harry Potter and the Sorcerer’s Stone'', ''The Great Gatsby'', ''1984'', and ''Frankenstein'', in addition to thousands of words from books including ''The Hunger Games'' and ''The Catcher in the Rye''.</blockquote>[https://www.theatlantic.com/technology/2026/01/ai-memorization-research/685552/?gift=oovWF8LV3gr7UxAU_wR7KCByWoXJuL6nibyWmch1C38 AI’s Memorization Crisis] — The Atlantic in 2026-01-09<br />
<br />
== AI Hype to Human ==<br />
{| class="wikitable"<br />
|+Oversimplified translation from AI Hype to Human language<br />
!Hype<br />
!Human<br />
|-<br />
|AI<br />
|Usually means LLMs.<br />
|-<br />
|LLM (Large Language Model)<br />
|A statistical model that can ''predict'' the most likely word (token) based on the previous words. For example, you write "The sky is" and it writes back "blue". When combined with other techniques, it can convincingly mimic human writing or thinking despite not being able to.<br />
|-<br />
|AGI (Artificial General Intelligence)<br />
|The hypothetical stage where AI matches or surpasses human intelligence. AI companies keep selling the idea that, if LLMs get better at '''predicting words''', they will somehow become super intelligent. Thus the need for more data centers.<br />
|-<br />
|Hallucination<br />
|When the words the model predicted are known (to us) to be untrue. The model can't ''know'' things, just predict words.<br />
|-<br />
|Chatbot<br />
|An interface where you can type messages to be processed by the LLM and get its responses.<br />
|-<br />
|Agentic AI<br />
|Different from Chatbots, Agentic AI makes multiple calls to the LLM to get progressively more complex answers before replying to the user. It can also call tools, such as web-searching.<br />
For example:<br />
<br />
# The user asks "I'm going to Tokyo for a week, help me plan the trip"<br />
# Instead of sending that to the LLM, the system sends "Build a plan to better answer the following: I'm going to Tokyo for..."<br />
# LLM replies with a list of tasks like "1 - Research top Tokyo attractions, 2 - Define must-eats, 3 - etc etc"<br />
# The system then sends back the list and adds "Now, perform task 1", then 2, then 3, etc<br />
# Only when everything is done, it replies back to the user with the plan.<br />
This produces even more impressive results, despite still being just a text prediction machine.<br />
|-<br />
|Token<br />
|A word or part of one.<br />
''Important: When I say "word" in other places, I probably mean these fractions of words (tokens).''<br />
|-<br />
|Context, Prompt<br />
|All the words the model is going to use to predict the next ones. Including your instructions, what the model wrote back and other things (see below).<br />
|-<br />
|System Prompt<br />
|Hidden instructions that the model providers include in all conversations to give better responses, avoid problems, etc.<br />
''For example: "If the user asks how to build a bomb, refuse."''<br />
|-<br />
|AGENTS.md, CLAUDE.md, (Global) Rules<br />
|Similar to system prompt, but defined by the user and included in all conversations. These are stored in [[wikipedia:Markdown|Markdown]] files.<br />
|-<br />
|Memory<br />
|Markdown files with past conversations summarized that can be referenced later.<br />
|-<br />
|Skills, Commands<br />
|Markdown files with generic instructions that can be reused.<br />
|-<br />
|Tools<br />
|Markdown files with instructions on how to perform certain actions using external tools (e.g. how to read text from a PDF)<br />
|-<br />
|Agent<br />
|Markdown files defining a "persona" that an Agentic tool can use while performing a task.<br />
''For example: You're a Senior Developer implementing a new feature. You code you write should X, Y, Z.''<br />
|-<br />
|Plugins<br />
|A combination of Markdown files containing Commands, Agents, Skills and Hooks someone has bundled together for others to use (on [https://claude.com/plugins Claude Code]).<br />
|}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=AI&diff=44AI2026-02-28T02:43:14Z<p>BobbyGeneric: </p>
<hr />
<div>== News ==<br />
<blockquote>On Tuesday, researchers at Stanford and Yale revealed something that AI companies would prefer to keep hidden. Four popular large language models—OpenAI’s GPT, Anthropic’s Claude, Google’s Gemini, and xAI’s Grok—have stored large portions of some of the books they’ve been trained on, and can reproduce long excerpts from those books.<br />
<br />
<br />
In fact, when prompted strategically by researchers, Claude delivered the near-complete text of ''Harry Potter and the Sorcerer’s Stone'', ''The Great Gatsby'', ''1984'', and ''Frankenstein'', in addition to thousands of words from books including ''The Hunger Games'' and ''The Catcher in the Rye''.</blockquote>[https://www.theatlantic.com/technology/2026/01/ai-memorization-research/685552/?gift=oovWF8LV3gr7UxAU_wR7KCByWoXJuL6nibyWmch1C38 AI’s Memorization Crisis] — The Atlantic in 2026-01-09<br />
<br />
== AI Hype to Human ==<br />
{| class="wikitable"<br />
|+Oversimplified translation from AI Hype to Human language<br />
!Hype<br />
!Human<br />
|-<br />
|AI<br />
|Usually means LLMs.<br />
|-<br />
|LLM (Large Language Model)<br />
|A statistical model that can ''predict'' the most likely word (token) based on the previous words. For example, you write "The sky is" and it writes back "blue". When combined with other techniques, it can convincingly mimic human writing or thinking despite not being able to.<br />
|-<br />
|AGI (Artificial General Intelligence)<br />
|The hypothetical stage where AI matches or surpasses human intelligence. AI companies keep selling the idea that, if LLMs get better at '''predicting words''', they will somehow become super intelligent. Thus the need for more data centers.<br />
|-<br />
|Hallucination<br />
|When the words the model predicted are known (to us) to be untrue. The model can't ''know'' things, just predict words.<br />
|-<br />
|Chatbot<br />
|An interface where you can type messages to be processed by the LLM and get its responses.<br />
|-<br />
|Agentic AI<br />
|Different from Chatbots, Agentic AI makes multiple calls to the LLM to get progressively more complex answers before replying to the user. It can also call tools, such as web-searching.<br />
For example:<br />
<br />
# The user asks "I'm going to Tokyo for a week, help me plan the trip"<br />
# Instead of sending that to the LLM, the system sends "Build a plan to better answer the following: I'm going to Tokyo for..."<br />
# LLM replies with a list of tasks like "1 - Research top Tokyo attractions, 2 - Define must-eats, 3 - etc etc"<br />
# The system then sends back the list and adds "Now, perform task 1", then 2, then 3, etc<br />
# Only when everything is done, it replies back to the user with the plan.<br />
This produces even more impressive results, despite still being just a text prediction machine.<br />
|-<br />
|Token<br />
|A word or part of one.<br />
''Important: When I say "word" in other places, I probably mean these fractions of words (tokens).''<br />
|-<br />
|Context, Prompt<br />
|All the words the model is going to use to predict the next ones. Including your instructions, what the model wrote back and other things (see below).<br />
|-<br />
|System Prompt<br />
|Hidden instructions that the model providers include in all conversations to give better responses, avoid problems, etc.<br />
''For example: "If the user asks how to build a bomb, refuse."''<br />
|-<br />
|AGENTS.md, CLAUDE.md, (Global) Rules<br />
|Similar to system prompt, but defined by the user and included in all conversations. These are stored in [[wikipedia:Markdown|Markdown]] files.<br />
|-<br />
|Memory<br />
|Markdown files with past conversations summarized that can be referenced later.<br />
|-<br />
|Skills, Commands<br />
|Markdown files with generic instructions that can be reused.<br />
|-<br />
|Tools<br />
|Markdown files with instructions on how to perform certain actions using external tools (e.g. how to read text from a PDF)<br />
|-<br />
|Agent<br />
|Markdown files defining a "persona" that an Agentic tool can use while performing a task.<br />
''For example: You're a Senior Developer implementing a new feature. You code you write should X, Y, Z.''<br />
|-<br />
|Plugins<br />
|A combination of Markdown files containing Commands, Agents, Skills and Hooks someone has bundled together for others to use (on [https://claude.com/plugins Claude Code]).<br />
|}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=AI&diff=43AI2026-02-28T02:43:06Z<p>BobbyGeneric: </p>
<hr />
<div>== News ==<br />
<blockquote>On Tuesday, researchers at Stanford and Yale revealed something that AI companies would prefer to keep hidden. Four popular large language models—OpenAI’s GPT, Anthropic’s Claude, Google’s Gemini, and xAI’s Grok—have stored large portions of some of the books they’ve been trained on, and can reproduce long excerpts from those books.<br />
<br />
In fact, when prompted strategically by researchers, Claude delivered the near-complete text of ''Harry Potter and the Sorcerer’s Stone'', ''The Great Gatsby'', ''1984'', and ''Frankenstein'', in addition to thousands of words from books including ''The Hunger Games'' and ''The Catcher in the Rye''.</blockquote>[https://www.theatlantic.com/technology/2026/01/ai-memorization-research/685552/?gift=oovWF8LV3gr7UxAU_wR7KCByWoXJuL6nibyWmch1C38 AI’s Memorization Crisis] — The Atlantic in 2026-01-09<br />
<br />
== AI Hype to Human ==<br />
{| class="wikitable"<br />
|+Oversimplified translation from AI Hype to Human language<br />
!Hype<br />
!Human<br />
|-<br />
|AI<br />
|Usually means LLMs.<br />
|-<br />
|LLM (Large Language Model)<br />
|A statistical model that can ''predict'' the most likely word (token) based on the previous words. For example, you write "The sky is" and it writes back "blue". When combined with other techniques, it can convincingly mimic human writing or thinking despite not being able to.<br />
|-<br />
|AGI (Artificial General Intelligence)<br />
|The hypothetical stage where AI matches or surpasses human intelligence. AI companies keep selling the idea that, if LLMs get better at '''predicting words''', they will somehow become super intelligent. Thus the need for more data centers.<br />
|-<br />
|Hallucination<br />
|When the words the model predicted are known (to us) to be untrue. The model can't ''know'' things, just predict words.<br />
|-<br />
|Chatbot<br />
|An interface where you can type messages to be processed by the LLM and get its responses.<br />
|-<br />
|Agentic AI<br />
|Different from Chatbots, Agentic AI makes multiple calls to the LLM to get progressively more complex answers before replying to the user. It can also call tools, such as web-searching.<br />
For example:<br />
<br />
# The user asks "I'm going to Tokyo for a week, help me plan the trip"<br />
# Instead of sending that to the LLM, the system sends "Build a plan to better answer the following: I'm going to Tokyo for..."<br />
# LLM replies with a list of tasks like "1 - Research top Tokyo attractions, 2 - Define must-eats, 3 - etc etc"<br />
# The system then sends back the list and adds "Now, perform task 1", then 2, then 3, etc<br />
# Only when everything is done, it replies back to the user with the plan.<br />
This produces even more impressive results, despite still being just a text prediction machine.<br />
|-<br />
|Token<br />
|A word or part of one.<br />
''Important: When I say "word" in other places, I probably mean these fractions of words (tokens).''<br />
|-<br />
|Context, Prompt<br />
|All the words the model is going to use to predict the next ones. Including your instructions, what the model wrote back and other things (see below).<br />
|-<br />
|System Prompt<br />
|Hidden instructions that the model providers include in all conversations to give better responses, avoid problems, etc.<br />
''For example: "If the user asks how to build a bomb, refuse."''<br />
|-<br />
|AGENTS.md, CLAUDE.md, (Global) Rules<br />
|Similar to system prompt, but defined by the user and included in all conversations. These are stored in [[wikipedia:Markdown|Markdown]] files.<br />
|-<br />
|Memory<br />
|Markdown files with past conversations summarized that can be referenced later.<br />
|-<br />
|Skills, Commands<br />
|Markdown files with generic instructions that can be reused.<br />
|-<br />
|Tools<br />
|Markdown files with instructions on how to perform certain actions using external tools (e.g. how to read text from a PDF)<br />
|-<br />
|Agent<br />
|Markdown files defining a "persona" that an Agentic tool can use while performing a task.<br />
''For example: You're a Senior Developer implementing a new feature. You code you write should X, Y, Z.''<br />
|-<br />
|Plugins<br />
|A combination of Markdown files containing Commands, Agents, Skills and Hooks someone has bundled together for others to use (on [https://claude.com/plugins Claude Code]).<br />
|}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=AI&diff=42AI2026-02-28T02:42:52Z<p>BobbyGeneric: </p>
<hr />
<div>== News ==<br />
<blockquote>On Tuesday, researchers at Stanford and Yale revealed something that AI companies would prefer to keep hidden. Four popular large language models—OpenAI’s GPT, Anthropic’s Claude, Google’s Gemini, and xAI’s Grok—have stored large portions of some of the books they’ve been trained on, and can reproduce long excerpts from those books.<br />
<br />
<br />
In fact, when prompted strategically by researchers, Claude delivered the near-complete text of ''Harry Potter and the Sorcerer’s Stone'', ''The Great Gatsby'', ''1984'', and ''Frankenstein'', in addition to thousands of words from books including ''The Hunger Games'' and ''The Catcher in the Rye''.</blockquote>[https://www.theatlantic.com/technology/2026/01/ai-memorization-research/685552/?gift=oovWF8LV3gr7UxAU_wR7KCByWoXJuL6nibyWmch1C38 AI’s Memorization Crisis] — The Atlantic in 2026-01-09<br />
<br />
== AI Hype to Human ==<br />
{| class="wikitable"<br />
|+Oversimplified translation from AI Hype to Human language<br />
!Hype<br />
!Human<br />
|-<br />
|AI<br />
|Usually means LLMs.<br />
|-<br />
|LLM (Large Language Model)<br />
|A statistical model that can ''predict'' the most likely word (token) based on the previous words. For example, you write "The sky is" and it writes back "blue". When combined with other techniques, it can convincingly mimic human writing or thinking despite not being able to.<br />
|-<br />
|AGI (Artificial General Intelligence)<br />
|The hypothetical stage where AI matches or surpasses human intelligence. AI companies keep selling the idea that, if LLMs get better at '''predicting words''', they will somehow become super intelligent. Thus the need for more data centers.<br />
|-<br />
|Hallucination<br />
|When the words the model predicted are known (to us) to be untrue. The model can't ''know'' things, just predict words.<br />
|-<br />
|Chatbot<br />
|An interface where you can type messages to be processed by the LLM and get its responses.<br />
|-<br />
|Agentic AI<br />
|Different from Chatbots, Agentic AI makes multiple calls to the LLM to get progressively more complex answers before replying to the user. It can also call tools, such as web-searching.<br />
For example:<br />
<br />
# The user asks "I'm going to Tokyo for a week, help me plan the trip"<br />
# Instead of sending that to the LLM, the system sends "Build a plan to better answer the following: I'm going to Tokyo for..."<br />
# LLM replies with a list of tasks like "1 - Research top Tokyo attractions, 2 - Define must-eats, 3 - etc etc"<br />
# The system then sends back the list and adds "Now, perform task 1", then 2, then 3, etc<br />
# Only when everything is done, it replies back to the user with the plan.<br />
This produces even more impressive results, despite still being just a text prediction machine.<br />
|-<br />
|Token<br />
|A word or part of one.<br />
''Important: When I say "word" in other places, I probably mean these fractions of words (tokens).''<br />
|-<br />
|Context, Prompt<br />
|All the words the model is going to use to predict the next ones. Including your instructions, what the model wrote back and other things (see below).<br />
|-<br />
|System Prompt<br />
|Hidden instructions that the model providers include in all conversations to give better responses, avoid problems, etc.<br />
''For example: "If the user asks how to build a bomb, refuse."''<br />
|-<br />
|AGENTS.md, CLAUDE.md, (Global) Rules<br />
|Similar to system prompt, but defined by the user and included in all conversations. These are stored in [[wikipedia:Markdown|Markdown]] files.<br />
|-<br />
|Memory<br />
|Markdown files with past conversations summarized that can be referenced later.<br />
|-<br />
|Skills, Commands<br />
|Markdown files with generic instructions that can be reused.<br />
|-<br />
|Tools<br />
|Markdown files with instructions on how to perform certain actions using external tools (e.g. how to read text from a PDF)<br />
|-<br />
|Agent<br />
|Markdown files defining a "persona" that an Agentic tool can use while performing a task.<br />
''For example: You're a Senior Developer implementing a new feature. You code you write should X, Y, Z.''<br />
|-<br />
|Plugins<br />
|A combination of Markdown files containing Commands, Agents, Skills and Hooks someone has bundled together for others to use (on [https://claude.com/plugins Claude Code]).<br />
|}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=AI&diff=41AI2026-02-28T02:42:03Z<p>BobbyGeneric: </p>
<hr />
<div>== News ==<br />
<blockquote>On Tuesday, researchers at Stanford and Yale revealed something that AI companies would prefer to keep hidden. Four popular large language models—OpenAI’s GPT, Anthropic’s Claude, Google’s Gemini, and xAI’s Grok—have stored large portions of some of the books they’ve been trained on, and can reproduce long excerpts from those books.<br />
<br />
In fact, when prompted strategically by researchers, Claude delivered the near-complete text of ''Harry Potter and the Sorcerer’s Stone'', ''The Great Gatsby'', ''1984'', and ''Frankenstein'', in addition to thousands of words from books including ''The Hunger Games'' and ''The Catcher in the Rye''.</blockquote>[https://www.theatlantic.com/technology/2026/01/ai-memorization-research/685552/?gift=oovWF8LV3gr7UxAU_wR7KCByWoXJuL6nibyWmch1C38 AI’s Memorization Crisis] — The Atlantic in 2026-01-09<br />
<br />
== AI Hype to Human ==<br />
{| class="wikitable"<br />
|+Oversimplified translation from AI Hype to Human language<br />
!Hype<br />
!Human<br />
|-<br />
|AI<br />
|Usually means LLMs.<br />
|-<br />
|LLM (Large Language Model)<br />
|A statistical model that can ''predict'' the most likely word (token) based on the previous words. For example, you write "The sky is" and it writes back "blue". When combined with other techniques, it can convincingly mimic human writing or thinking despite not being able to.<br />
|-<br />
|AGI (Artificial General Intelligence)<br />
|The hypothetical stage where AI matches or surpasses human intelligence. AI companies keep selling the idea that, if LLMs get better at '''predicting words''', they will somehow become super intelligent. Thus the need for more data centers.<br />
|-<br />
|Hallucination<br />
|When the words the model predicted are known (to us) to be untrue. The model can't ''know'' things, just predict words.<br />
|-<br />
|Chatbot<br />
|An interface where you can type messages to be processed by the LLM and get its responses.<br />
|-<br />
|Agentic AI<br />
|Different from Chatbots, Agentic AI makes multiple calls to the LLM to get progressively more complex answers before replying to the user. It can also call tools, such as web-searching.<br />
For example:<br />
<br />
# The user asks "I'm going to Tokyo for a week, help me plan the trip"<br />
# Instead of sending that to the LLM, the system sends "Build a plan to better answer the following: I'm going to Tokyo for..."<br />
# LLM replies with a list of tasks like "1 - Research top Tokyo attractions, 2 - Define must-eats, 3 - etc etc"<br />
# The system then sends back the list and adds "Now, perform task 1", then 2, then 3, etc<br />
# Only when everything is done, it replies back to the user with the plan.<br />
This produces even more impressive results, despite still being just a text prediction machine.<br />
|-<br />
|Token<br />
|A word or part of one.<br />
''Important: When I say "word" in other places, I probably mean these fractions of words (tokens).''<br />
|-<br />
|Context, Prompt<br />
|All the words the model is going to use to predict the next ones. Including your instructions, what the model wrote back and other things (see below).<br />
|-<br />
|System Prompt<br />
|Hidden instructions that the model providers include in all conversations to give better responses, avoid problems, etc.<br />
''For example: "If the user asks how to build a bomb, refuse."''<br />
|-<br />
|AGENTS.md, CLAUDE.md, (Global) Rules<br />
|Similar to system prompt, but defined by the user and included in all conversations. These are stored in [[wikipedia:Markdown|Markdown]] files.<br />
|-<br />
|Memory<br />
|Markdown files with past conversations summarized that can be referenced later.<br />
|-<br />
|Skills, Commands<br />
|Markdown files with generic instructions that can be reused.<br />
|-<br />
|Tools<br />
|Markdown files with instructions on how to perform certain actions using external tools (e.g. how to read text from a PDF)<br />
|-<br />
|Agent<br />
|Markdown files defining a "persona" that an Agentic tool can use while performing a task.<br />
''For example: You're a Senior Developer implementing a new feature. You code you write should X, Y, Z.''<br />
|-<br />
|Plugins<br />
|A combination of Markdown files containing Commands, Agents, Skills and Hooks someone has bundled together for others to use (on [https://claude.com/plugins Claude Code]).<br />
|}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=Main_Page&diff=40Main Page2026-02-25T14:11:46Z<p>BobbyGeneric: </p>
<hr />
<div>== <strong>Hello there!</strong> ==<br />
Here you'll find a bit of everything. My interests and current obsessions. It's basically a scrapbook of random stuff that crosses my mind. A mix of facts and opinion written by myself to myself, but it may be useful for someone else, who knows? At least it's information written by a real person using ''natural intelligence''™ (barely), not some fancy magic box pretending to be one. <br />
<br />
[[File:Stay Hungry. Stay Foolish..jpg|thumb|"Stay Hungry. Stay Foolish." from the back cover of the Whole Earth Catalog, published in October 1974.]]<br />
<br />
== What's new? ==<br />
<br />
<dpl><br />
# https://dpl4.wikitide.org/wiki/General_usage_and_invocation_syntax<br />
namespace = 0<br />
mode = ordered<br />
ordermethod = lastedit<br />
order = descending<br />
count = 30<br />
addeditdate = true<br />
# userdateformat = Y-m-d<br />
format = ,* [[%PAGE%]] <small>— %DATE%</small>\n,,<br />
</dpl><br />
<br />
<!--<br />
<div class="clean-recent-changes"><br />
{{Special:RecentChanges|days=365|limit=10|namespace=0|hideminor=1}}<br />
</div><br />
--></div>BobbyGenerichttps://barza.y2k.town/index.php?title=MediaWiki:Dpl-tag-tracking-category&diff=39MediaWiki:Dpl-tag-tracking-category2026-02-25T14:03:48Z<p>BobbyGeneric: Created page with "-"</p>
<hr />
<div>-</div>BobbyGenerichttps://barza.y2k.town/index.php?title=MediaWiki:Dpl-tracking-category&diff=38MediaWiki:Dpl-tracking-category2026-02-25T14:03:40Z<p>BobbyGeneric: Created page with "-"</p>
<hr />
<div>-</div>BobbyGenerichttps://barza.y2k.town/index.php?title=Interesting_Wikipedia_Articles&diff=37Interesting Wikipedia Articles2026-02-19T18:38:31Z<p>BobbyGeneric: BobbyGeneric moved page Interesting Wikipedia Pages to Interesting Wikipedia Articles without leaving a redirect</p>
<hr />
<div><blockquote>'''[[wikipedia:Nekonomics|Nekonomics]]''' (ネコノミクス), a term blending ''neko'', the Japanese word for ''cat'', and ''economics'', describes the phenomenon of cat-related economic consumption in Japan. This includes the sale of products marketed towards cat owners, the use of cat imagery to sell products, and the use of living cats to attract customers and tourists. While cats have always enjoyed popularity in Japanese culture, being seen as ''kawaii'' (cute) and associated with good health, their popularity and economic impact increased significantly in the early 21st century. The term ''nekonomics'' was created around 2015, a pun on Japan's Abenomics economic policy.<ref>https://en.wikipedia.org/wiki/Nekonomics</ref></blockquote><br />
<br />
== References ==</div>BobbyGenerichttps://barza.y2k.town/index.php?title=Main_Page&diff=36Main Page2026-02-19T18:37:27Z<p>BobbyGeneric: /* What's new? */</p>
<hr />
<div>== <strong>Hello there!</strong> ==<br />
Here you'll find a bit of everything. My interests and current obsessions. It's basically a scrapbook of random stuff that crosses my mind. A mix of facts and opinion written by myself to myself, but it may be useful for someone else, who knows? At least it's information written by a real person using ''natural intelligence''™ (barely), not some fancy magic box pretending to be one. <br />
[[File:Stay Hungry. Stay Foolish..jpg|thumb|"Stay Hungry. Stay Foolish." from the back cover of the Whole Earth Catalog, published in October 1974.]]<br />
<br />
== What's new? ==<br />
<br />
<dpl><br />
# https://dpl4.wikitide.org/wiki/General_usage_and_invocation_syntax<br />
namespace = 0<br />
mode = ordered<br />
ordermethod = lastedit<br />
order = descending<br />
count = 30<br />
addeditdate = true<br />
# userdateformat = Y-m-d<br />
format = ,* [[%PAGE%]] <small>— %DATE%</small>\n,,<br />
</dpl><br />
<br />
<!--<br />
<div class="clean-recent-changes"><br />
{{Special:RecentChanges|days=365|limit=10|namespace=0|hideminor=1}}<br />
</div><br />
--></div>BobbyGenerichttps://barza.y2k.town/index.php?title=Main_Page&diff=35Main Page2026-02-19T18:34:03Z<p>BobbyGeneric: /* What's new? */</p>
<hr />
<div>== <strong>Hello there!</strong> ==<br />
Here you'll find a bit of everything. My interests and current obsessions. It's basically a scrapbook of random stuff that crosses my mind. A mix of facts and opinion written by myself to myself, but it may be useful for someone else, who knows? At least it's information written by a real person using ''natural intelligence''™ (barely), not some fancy magic box pretending to be one. <br />
[[File:Stay Hungry. Stay Foolish..jpg|thumb|"Stay Hungry. Stay Foolish." from the back cover of the Whole Earth Catalog, published in October 1974.]]<br />
<br />
== What's new? ==<br />
<br />
<dpl><br />
# https://dpl4.wikitide.org/wiki/General_usage_and_invocation_syntax<br />
namespace = 0<br />
mode = ordered<br />
ordermethod = lastedit<br />
order = descending<br />
count = 30<br />
addeditdate = true<br />
# userdateformat = Y-m-d<br />
format = ,* [[%PAGE%]] <small>%DATE%</small>\n,,<br />
</dpl><br />
<!--<br />
<div class="clean-recent-changes"><br />
{{Special:RecentChanges|days=365|limit=10|namespace=0|hideminor=1}}<br />
</div><br />
--></div>BobbyGenerichttps://barza.y2k.town/index.php?title=Main_Page&diff=34Main Page2026-02-19T18:32:09Z<p>BobbyGeneric: /* What's new? */</p>
<hr />
<div>== <strong>Hello there!</strong> ==<br />
Here you'll find a bit of everything. My interests and current obsessions. It's basically a scrapbook of random stuff that crosses my mind. A mix of facts and opinion written by myself to myself, but it may be useful for someone else, who knows? At least it's information written by a real person using ''natural intelligence''™ (barely), not some fancy magic box pretending to be one. <br />
[[File:Stay Hungry. Stay Foolish..jpg|thumb|"Stay Hungry. Stay Foolish." from the back cover of the Whole Earth Catalog, published in October 1974.]]<br />
<br />
== What's new? ==<br />
<br />
<dpl><br />
namespace = 0<br />
mode = ordered<br />
ordermethod = lastedit<br />
order = descending<br />
count = 30<br />
addeditdate = true<br />
userdateformat = Y-m-d<br />
format = ,* [[%PAGE%]] <small>%DATE%</small>\n,,<br />
</dpl><br />
<!--<br />
<div class="clean-recent-changes"><br />
{{Special:RecentChanges|days=365|limit=10|namespace=0|hideminor=1}}<br />
</div><br />
--></div>BobbyGenerichttps://barza.y2k.town/index.php?title=Main_Page&diff=33Main Page2026-02-19T18:26:18Z<p>BobbyGeneric: /* What's new? */</p>
<hr />
<div>== <strong>Hello there!</strong> ==<br />
Here you'll find a bit of everything. My interests and current obsessions. It's basically a scrapbook of random stuff that crosses my mind. A mix of facts and opinion written by myself to myself, but it may be useful for someone else, who knows? At least it's information written by a real person using ''natural intelligence''™ (barely), not some fancy magic box pretending to be one. <br />
[[File:Stay Hungry. Stay Foolish..jpg|thumb|"Stay Hungry. Stay Foolish." from the back cover of the Whole Earth Catalog, published in October 1974.]]<br />
<br />
== What's new? ==<br />
<br />
<dpl><br />
namespace = 0<br />
ordermethod = lastedit<br />
order = descending<br />
count = 30<br />
mode = ordered<br />
addeditdate=true<br />
# Optional: formatted output<br />
format = ,* [[%PAGE%]] - %DATE%\n,,<br />
</dpl><br />
<br />
<!--<br />
<div class="clean-recent-changes"><br />
{{Special:RecentChanges|days=365|limit=10|namespace=0|hideminor=1}}<br />
</div><br />
--></div>BobbyGenerichttps://barza.y2k.town/index.php?title=Interesting_Wikipedia_Articles&diff=32Interesting Wikipedia Articles2026-02-18T11:38:00Z<p>BobbyGeneric: </p>
<hr />
<div><blockquote>'''[[wikipedia:Nekonomics|Nekonomics]]''' (ネコノミクス), a term blending ''neko'', the Japanese word for ''cat'', and ''economics'', describes the phenomenon of cat-related economic consumption in Japan. This includes the sale of products marketed towards cat owners, the use of cat imagery to sell products, and the use of living cats to attract customers and tourists. While cats have always enjoyed popularity in Japanese culture, being seen as ''kawaii'' (cute) and associated with good health, their popularity and economic impact increased significantly in the early 21st century. The term ''nekonomics'' was created around 2015, a pun on Japan's Abenomics economic policy.<ref>https://en.wikipedia.org/wiki/Nekonomics</ref></blockquote><br />
<br />
== References ==</div>BobbyGenerichttps://barza.y2k.town/index.php?title=Interesting_Wikipedia_Articles&diff=31Interesting Wikipedia Articles2026-02-18T02:44:28Z<p>BobbyGeneric: Created page with "<blockquote>'''Nekonomics''' (ネコノミクス), a term blending ''neko'', the Japanese word for ''cat'', and ''economics'', describes the phenomenon of cat-related economic consumption in Japan. This includes the sale of products marketed towards cat owners, the use of cat imagery to sell products, and the use of living cats to attract customers and tourists. While cats have always enjoyed popularity in Japanese culture, being seen as ''kawaii..."</p>
<hr />
<div><blockquote>'''[[wikipedia:Nekonomics|Nekonomics]]''' (ネコノミクス), a term blending ''neko'', the Japanese word for ''cat'', and ''economics'', describes the phenomenon of cat-related economic consumption in Japan. This includes the sale of products marketed towards cat owners, the use of cat imagery to sell products, and the use of living cats to attract customers and tourists. While cats have always enjoyed popularity in Japanese culture, being seen as ''kawaii'' (cute) and associated with good health, their popularity and economic impact increased significantly in the early 21st century. The term ''nekonomics'' was created around 2015, a pun on Japan's Abenomics economic policy.<ref>https://en.wikipedia.org/wiki/Nekonomics</ref></blockquote></div>BobbyGenerichttps://barza.y2k.town/index.php?title=AI&diff=30AI2026-02-10T20:26:16Z<p>BobbyGeneric: </p>
<hr />
<div>{| class="wikitable"<br />
|+Oversimplified translation from AI Hype to Human language<br />
!Hype<br />
!Human<br />
|-<br />
|AI<br />
|Usually means LLMs.<br />
|-<br />
|LLM (Large Language Model)<br />
|A statistical model that can ''predict'' the most likely word (token) based on the previous words. For example, you write "The sky is" and it writes back "blue". When combined with other techniques, it can convincingly mimic human writing or thinking despite not being able to.<br />
|-<br />
|AGI (Artificial General Intelligence)<br />
|The hypothetical stage where AI matches or surpasses human intelligence. AI companies keep selling the idea that, if LLMs get better at '''predicting words''', they will somehow become super intelligent. Thus the need for more data centers.<br />
|-<br />
|Hallucination<br />
|When the words the model predicted are known (to us) to be untrue. The model can't ''know'' things, just predict words.<br />
|-<br />
|Chatbot<br />
|An interface where you can type messages to be processed by the LLM and get its responses.<br />
|-<br />
|Agentic AI<br />
|Different from Chatbots, Agentic AI makes multiple calls to the LLM to get progressively more complex answers before replying to the user. It can also call tools, such as web-searching.<br />
For example:<br />
<br />
# The user asks "I'm going to Tokyo for a week, help me plan the trip"<br />
# Instead of sending that to the LLM, the system sends "Build a plan to better answer the following: I'm going to Tokyo for..."<br />
# LLM replies with a list of tasks like "1 - Research top Tokyo attractions, 2 - Define must-eats, 3 - etc etc"<br />
# The system then sends back the list and adds "Now, perform task 1", then 2, then 3, etc<br />
# Only when everything is done, it replies back to the user with the plan.<br />
This produces even more impressive results, despite still being just a text prediction machine.<br />
|-<br />
|Token<br />
|A word or part of one.<br />
''Important: When I say "word" in other places, I probably mean these fractions of words (tokens).''<br />
|-<br />
|Context, Prompt<br />
|All the words the model is going to use to predict the next ones. Including your instructions, what the model wrote back and other things (see below).<br />
|-<br />
|System Prompt<br />
|Hidden instructions that the model providers include in all conversations to give better responses, avoid problems, etc.<br />
''For example: "If the user asks how to build a bomb, refuse."''<br />
|-<br />
|AGENTS.md, CLAUDE.md, (Global) Rules<br />
|Similar to system prompt, but defined by the user and included in all conversations. These are stored in [[wikipedia:Markdown|Markdown]] files.<br />
|-<br />
|Memory<br />
|Markdown files with past conversations summarized that can be referenced later.<br />
|-<br />
|Skills, Commands<br />
|Markdown files with generic instructions that can be reused.<br />
|-<br />
|Tools<br />
|Markdown files with instructions on how to perform certain actions using external tools (e.g. how to read text from a PDF)<br />
|-<br />
|Agent<br />
|Markdown files defining a "persona" that an Agentic tool can use while performing a task.<br />
''For example: You're a Senior Developer implementing a new feature. You code you write should X, Y, Z.''<br />
|-<br />
|Plugins<br />
|A combination of Markdown files containing Commands, Agents, Skills and Hooks someone has bundled together for others to use (on [https://claude.com/plugins Claude Code]).<br />
|}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=AI&diff=29AI2026-02-10T20:25:15Z<p>BobbyGeneric: </p>
<hr />
<div>{| class="wikitable"<br />
|+Oversimplified translation from AI Hype to Human language<br />
!Hype<br />
!Human<br />
|-<br />
|AI<br />
|Usually means LLMs.<br />
|-<br />
|LLM (Large Language Model)<br />
|A statistical model that can ''predict'' the most likely word (token) based on the previous words. For example, you write "The sky is" and it writes back "blue". When combined with other techniques, it can convincingly mimic human writing or thinking despite not being able to.<br />
|-<br />
|AGI (Artificial General Intelligence)<br />
|The hypothetical stage where AI matches or surpasses human intelligence. AI companies keep selling the idea that, if LLMs get better at '''predicting words''', they will somehow become super intelligent. Thus the need for more data centers.<br />
|-<br />
|Hallucination<br />
|When the words the model predicted are known (to us) to be untrue. The model can't ''know'' things, just predict words.<br />
|-<br />
|Chatbot<br />
|An interface where you can type messages to be processed by the LLM and get its responses.<br />
|-<br />
|Agentic AI<br />
|Different from Chatbots, Agentic AI makes multiple calls to the LLM to get progressively more complex answers before replying to the user. It can also call tools, such as web-searching.<br />
For example:<br />
<br />
# The user asks "I'm going to Tokyo for a week, help me plan the trip"<br />
# Instead of sending that to the LLM, the system sends "Build a plan to better answer the following: I'm going to Tokyo for..."<br />
# LLM replies with a list of tasks like "1 - Research top Tokyo attractions, 2 - Define must-eats, 3 - etc etc"<br />
# The system then sends back the list and adds "Now, perform task 1", then 2, then 3, etc<br />
# Only when everything is done, it replies back to the user with the plan.<br />
This produces even more impressive results, despite still being just a text prediction machine.<br />
|-<br />
|Token<br />
|A word or part of one.<br />
''Important: When I say "word" in other places, I probably mean these fractions of words (tokens).''<br />
|-<br />
|Context, Prompt<br />
|All the words the model is going to use to predict the next ones. Including your instructions, what the model wrote back and other things (see below).<br />
|-<br />
|System Prompt<br />
|Hidden instructions that the model providers include in all conversations to give better responses, avoid problems, etc.<br />
''For example: "If the user asks how to build a bomb, refuse."''<br />
|-<br />
|AGENTS.md, CLAUDE.md, (Global) Rules<br />
|Similar to system prompt, but defined by the user and included in all conversations. These are stored in [[wikipedia:Markdown|Markdown]] files.<br />
|-<br />
|Memory<br />
|Markdown files with past conversations summarized that can be referenced later.<br />
|-<br />
|Skills, Commands<br />
|Markdown files with generic instructions that can be reused.<br />
|-<br />
|Tools<br />
|Markdown files with instructions on how to perform certain actions using external tools (e.g. how to read text from a PDF)<br />
|-<br />
|Agent<br />
|Markdown files defining a "persona" that an Agentic tool can use while performing a task.<br />
''For example: You're a Senior Developer implementing a new feature. You should do X, Y, Z.''<br />
|-<br />
|Plugins<br />
|A combination of Markdown files containing Commands, Agents, Skills and Hooks someone has bundled together for others to use (on [https://claude.com/plugins Claude Code]).<br />
|}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=AI&diff=28AI2026-02-10T20:24:41Z<p>BobbyGeneric: </p>
<hr />
<div>{| class="wikitable"<br />
|+Oversimplified translation from AI Hype to Human language<br />
!Hype<br />
!Human<br />
|-<br />
|AI<br />
|Usually means LLMs.<br />
|-<br />
|LLM (Large Language Model)<br />
|A statistical model that can ''predict'' the most likely word (token) based on the previous words. For example, you write "The sky is" and it writes back "blue". When combined with other techniques, it can convincingly mimic human writing or thinking despite not being able to.<br />
|-<br />
|AGI (Artificial General Intelligence)<br />
|The hypothetical stage where AI matches or surpasses human intelligence. AI companies keep selling the idea that, if LLMs get better at '''predicting words''', they will somehow become super intelligent. Thus the need for more data centers.<br />
|-<br />
|Hallucination<br />
|When the words the model predicted are known (to us) to be untrue. The model can't ''know'' things, just predict words.<br />
|-<br />
|Chatbot<br />
|An interface where you can type messages to be processed by the LLM and get its responses.<br />
|-<br />
|Agentic AI<br />
|Different from Chatbots, Agentic AI makes multiple calls to the LLM to get progressively more complex answers before replying to the user. It can also call tools, such as web-searching.<br />
<br />
For example:<br />
<br />
# The user asks "I'm going to Tokyo for a week, help me plan the trip"<br />
# Instead of sending that to the LLM, the system sends "Build a plan to better answer the following: I'm going to Tokyo for..."<br />
# LLM replies with a list of tasks like "1 - Research top Tokyo attractions, 2 - Define must-eats, 3 - etc etc"<br />
# The system then sends back the list and adds "Now, perform task 1", then 2, then 3, etc<br />
# Only when everything is done, it replies back to the user with the plan.<br />
<br />
This produces even more impressive results, despite still being just a text prediction machine.<br />
|-<br />
|Token<br />
|A word or part of one.<br />
''Important: When I say "word" in other places, I probably mean these fractions of words (tokens).''<br />
|-<br />
|Context, Prompt<br />
|All the words the model is going to use to predict the next ones. Including your instructions, what the model wrote back and other things (see below).<br />
|-<br />
|System Prompt<br />
|Hidden instructions that the model providers include in all conversations to give better responses, avoid problems, etc.<br />
''For example: "If the user asks how to build a bomb, refuse."''<br />
|-<br />
|AGENTS.md, CLAUDE.md, (Global) Rules<br />
|Similar to system prompt, but defined by the user and included in all conversations. These are stored in [[wikipedia:Markdown|Markdown]] files.<br />
|-<br />
|Memory<br />
|Markdown files with past conversations summarized that can be referenced later.<br />
|-<br />
|Skills, Commands<br />
|Markdown files with generic instructions that can be reused.<br />
|-<br />
|Tools<br />
|Markdown files with instructions on how to perform certain actions using external tools (e.g. how to read text from a PDF)<br />
|-<br />
|Agent<br />
|Markdown files defining a "persona" that an Agentic tool can use while performing a task.<br />
''For example: You're a Senior Developer implementing a new feature. You should do X, Y, Z.''<br />
|-<br />
|Plugins<br />
|A combination of Markdown files containing Commands, Agents, Skills and Hooks someone has bundled together for others to use (on [https://claude.com/plugins Claude Code]).<br />
|}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=AI&diff=27AI2026-02-10T20:24:03Z<p>BobbyGeneric: </p>
<hr />
<div>{| class="wikitable"<br />
|+Oversimplified translation from AI Hype to Human language<br />
!Hype<br />
!Human<br />
|-<br />
|AI<br />
|Usually means LLMs.<br />
|-<br />
|LLM (Large Language Model)<br />
|A statistical model that can ''predict'' the most likely word (token) based on the previous words. For example, you write "The sky is" and it writes back "blue". When combined with other techniques, it can convincingly mimic human writing or thinking despite not being able to.<br />
|-<br />
|AGI (Artificial General Intelligence)<br />
|The hypothetical stage where AI matches or surpasses human intelligence. AI companies keep selling the idea that, if LLMs get better at '''predicting words''', they will somehow become super intelligent. Thus the need for more data centers.<br />
|-<br />
|Hallucination<br />
|When the words the model predicted are known (to us) to be untrue. The model can't ''know'' things, just predict words.<br />
|-<br />
|Chatbot<br />
|An interface where you can type messages to be processed by the LLM and get its responses.<br />
|-<br />
|Agentic AI<br />
|Different from Chatbots, Agentic AI makes multiple calls to the LLM to get progressively more complex answers before replying to the user. It can also call tools, such as web-searching.<br />
For example:<br />
<br />
# The user asks "I'm going to Tokyo for a week, help me plan the trip"<br />
# Instead of sending that to the LLM, the system sends "Build a plan to better answer the following: I'm going to Tokyo for..."<br />
# LLM replies with a list of tasks like "1 - Research top Tokyo attractions, 2 - Define must-eats, 3 - etc etc"<br />
# The system then sends back the list and adds "Now, perform task 1", then 2, then 3, etc<br />
# Only when everything is done, it replies back to the user with the plan.<br />
<br />
This produces even more impressive results, despite still being just a text prediction machine.<br />
|-<br />
|Token<br />
|A word or part of one.<br />
''Important: When I say "word" in other places, I probably mean these fractions of words (tokens).''<br />
|-<br />
|Context, Prompt<br />
|All the words the model is going to use to predict the next ones. Including your instructions, what the model wrote back and other things (see below).<br />
|-<br />
|System Prompt<br />
|Hidden instructions that the model providers include in all conversations to give better responses, avoid problems, etc.<br />
''For example: "If the user asks how to build a bomb, refuse."''<br />
|-<br />
|AGENTS.md, CLAUDE.md, (Global) Rules<br />
|Similar to system prompt, but defined by the user and included in all conversations. These are stored in [[wikipedia:Markdown|Markdown]] files.<br />
|-<br />
|Memory<br />
|Markdown files with past conversations summarized that can be referenced later.<br />
|-<br />
|Skills, Commands<br />
|Markdown files with generic instructions that can be reused.<br />
|-<br />
|Tools<br />
|Markdown files with instructions on how to perform certain actions using external tools (e.g. how to read text from a PDF)<br />
|-<br />
|Agent<br />
|Markdown files defining a "persona" that an Agentic tool can use while performing a task.<br />
''For example: You're a Senior Developer implementing a new feature. You should do X, Y, Z.''<br />
|-<br />
|Plugins<br />
|A combination of Markdown files containing Commands, Agents, Skills and Hooks someone has bundled together for others to use (on [https://claude.com/plugins Claude Code]).<br />
|}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=AI&diff=26AI2026-02-10T20:22:51Z<p>BobbyGeneric: </p>
<hr />
<div>{| class="wikitable"<br />
|+Oversimplified translation from AI Hype to Human language<br />
!Hype<br />
!Human<br />
|-<br />
|AI<br />
|Usually means LLMs.<br />
|-<br />
|LLM (Large Language Model)<br />
|A statistical model that can ''predict'' the most likely word (token) based on the previous words. For example, you write "The sky is" and it writes back "blue". When combined with other techniques, it can convincingly mimic human writing or thinking despite not being able to.<br />
|-<br />
|AGI (Artificial General Intelligence)<br />
|The hypothetical stage where AI matches or surpasses human intelligence. AI companies keep selling the idea that, if LLMs get better at '''predicting words''', they will somehow become super intelligent. Thus the need for more data centers.<br />
|-<br />
|Hallucination<br />
|When the words the model predicted are known (to us) to be untrue. The model can't ''know'' things, just predict words.<br />
|-<br />
|Chatbot<br />
|An interface where you can type messages to be processed by the LLM and get its responses.<br />
|-<br />
|Agentic AI<br />
|Different from Chatbots, Agentic AI makes multiple calls to the LLM to get progressively more complex answers before replying to the user. It can also call tools, such as web-searching. For example:<br />
<br />
# The user asks "I'm going to Tokyo for a week, help me plan the trip"<br />
# Instead of sending that to the LLM, the system sends "Build a plan to better answer the following: I'm going to Tokyo for..."<br />
# LLM replies with a list of tasks like "1 - Research top Tokyo attractions, 2 - Define must-eats, 3 - etc etc"<br />
# The system then sends back the list and adds "Now, perform task 1", then 2, then 3, etc<br />
# Only when everything is done, it replies back to the user with the plan.<br />
<br />
This produces even more impressive results, despite still being just a text prediction machine.<br />
|-<br />
|Token<br />
|A word or part of one.<br />
''Important: When I say "word" in other places, I probably mean these fractions of words (tokens).''<br />
|-<br />
|Context, Prompt<br />
|All the words the model is going to use to predict the next ones. Including your instructions, what the model wrote back and other things (see below).<br />
|-<br />
|System Prompt<br />
|Hidden instructions that the model providers include in all conversations to give better responses, avoid problems, etc.<br />
''For example: "If the user asks how to build a bomb, refuse."''<br />
|-<br />
|AGENTS.md, CLAUDE.md, (Global) Rules<br />
|Similar to system prompt, but defined by the user and included in all conversations. These are stored in [[wikipedia:Markdown|Markdown]] files.<br />
|-<br />
|Memory<br />
|Markdown files with past conversations summarized that can be referenced later.<br />
|-<br />
|Skills, Commands<br />
|Markdown files with generic instructions that can be reused.<br />
|-<br />
|Tools<br />
|Markdown files with instructions on how to perform certain actions using external tools (e.g. how to read text from a PDF)<br />
|-<br />
|Agent<br />
|Markdown files defining a "persona" that an Agentic tool can use while performing a task.<br />
''For example: You're a Senior Developer implementing a new feature. You should do X, Y, Z.''<br />
|-<br />
|Plugins<br />
|A combination of Markdown files containing Commands, Agents, Skills and Hooks someone has bundled together for others to use (on [https://claude.com/plugins Claude Code]).<br />
|}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=AI&diff=25AI2026-02-10T20:22:31Z<p>BobbyGeneric: </p>
<hr />
<div>{| class="wikitable"<br />
|+Oversimplified translation from AI Hype to Human language<br />
!Hype<br />
!Human<br />
|-<br />
|AI<br />
|Usually means LLMs.<br />
|-<br />
|LLM (Large Language Model)<br />
|A statistical model that can ''predict'' the most likely word (token) based on the previous words. For example, you write "The sky is" and it writes back "blue". When combined with other techniques, it can convincingly mimic human writing or thinking despite not being able to.<br />
|-<br />
|AGI (Artificial General Intelligence)<br />
|The hypothetical stage where AI matches or surpasses human intelligence. AI companies keep selling the idea that, if LLMs get better at '''predicting words''', they will somehow become super intelligent. Thus the need for more data centers.<br />
|-<br />
|Hallucination<br />
|When the words the model predicted are known (to us) to be untrue. The model can't ''know'' things, just predict words.<br />
|-<br />
|Chatbot<br />
|An interface where you can type messages to be processed by the LLM and get its responses.<br />
|-<br />
|Agentic AI<br />
|Different from Chatbots, Agentic AI makes multiple calls to the LLM to get progressively more complex answers before replying to the user. It can also call tools, such as web-searching. For example:<br />
<br />
# The user asks "I'm going to Tokyo for a week, help me plan the trip"<br />
# Instead of sending that to the LLM, the system sends "Build a plan to better answer the following: I'm going to Tokyo for..."<br />
# LLM replies with a list of tasks like "1 - Research top Tokyo attractions, 2 - Define must-eats, 3 - etc etc"<br />
# The system then sends back the list and adds "Now, perform task 1", then 2, then 3, etc<br />
# Only when everything is done, it replies back to the user with the plan.<br />
<br />
This produces even more impressive results, despite still being just a text prediction machine.<br />
|-<br />
|Token<br />
|A word or part of one.<br />
''Important: When I say "word" in other places, I probably mean these fractions of words (tokens).''<br />
|-<br />
|Context, Prompt<br />
|All the words the model is going to use to predict the next ones. Including your instructions, what the model wrote back and other things (see below).<br />
|-<br />
|System Prompt<br />
|Hidden instructions that the model providers include in all conversations to give better responses, avoid problems, etc.<br />
''For example: "If the user asks how to build a bomb, refuse."''<br />
|-<br />
|AGENTS.md, CLAUDE.md, (Global) Rules<br />
|Similar to system prompt, but defined by the user and included in all conversations. These are stored in [[wikipedia:Markdown|Markdown]] files.<br />
|-<br />
|Memory<br />
|Markdown files with past conversations summarized that can be referenced later.<br />
|-<br />
|Skills, Commands<br />
|Markdown files with generic instructions that can be reused.<br />
|-<br />
|Tools<br />
|Markdown files with instructions on how to perform certain actions using external tools (e.g. how to read text from a PDF)<br />
|-<br />
|Agent<br />
|Markdown files defining a "persona" that an Agentic tool can use while performing a task.<br />
''For example: You're a Senior Developer implementing a new feature. You should do X, Y, Z.''<br />
|-<br />
|Plugins<br />
|A combination of Markdown files containing Commands, Agents, Skills and Hooks someone has bundled together for others to use on Claude Code.<br />
|}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=AI&diff=24AI2026-02-10T20:10:00Z<p>BobbyGeneric: </p>
<hr />
<div>{| class="wikitable"<br />
|+Oversimplified translation from AI Hype to Human language<br />
!Hype<br />
!Human<br />
|-<br />
|AI<br />
|Usually means LLMs.<br />
|-<br />
|LLM (Large Language Model)<br />
|A statistical model that can ''predict'' the most likely word (token) based on the previous words. For example, you write "The sky is" and it writes back "blue". When combined with other techniques, it can convincingly mimic human writing or thinking despite not being able to.<br />
|-<br />
|AGI (Artificial General Intelligence)<br />
|The hypothetical stage where AI matches or surpasses human intelligence. AI companies keep selling the idea that, if LLMs get better at '''predicting words''', they will somehow become super intelligent. Thus the need for more data centers.<br />
|-<br />
|Hallucination<br />
|When the words the model predicted are known (to us) to be untrue. The model can't ''know'' things, just predict words.<br />
|-<br />
|Chatbot<br />
|An interface where you can type messages to be processed by the LLM and get its responses.<br />
|-<br />
|Agentic AI<br />
|Different from Chatbots, Agentic AI makes multiple calls to the LLM to get progressively more complex answers before replying to the user. It can also call tools, such as web-searching. For example:<br />
<br />
# The user asks "I'm going to Tokyo for a week, help me plan the trip"<br />
# Instead of sending that to the LLM, the system sends "Build a plan to better answer the following: I'm going to Tokyo for..."<br />
# LLM replies with a list of tasks like "1 - Research top Tokyo attractions, 2 - Define must-eats, 3 - etc etc"<br />
# The system then sends back the list and adds "Now, perform task 1", then 2, then 3, etc<br />
# Only when everything is done, it replies back to the user with the plan.<br />
<br />
This produces even more impressive results, despite still being just a text prediction machine.<br />
|-<br />
|Token<br />
|A word or part of one.<br />
''Important: When I say "word" in other places, I probably mean these fractions of words (tokens).''<br />
|-<br />
|Context, Prompt<br />
|All the words the model is going to use to predict the next ones. Including your instructions, what the model wrote back and other things (see below).<br />
|-<br />
|System Prompt<br />
|Hidden instructions that the model providers include in all conversations to give better responses, avoid problems, etc.<br />
''For example: "If the user asks how to build a bomb, refuse."''<br />
|-<br />
|AGENTS.md, CLAUDE.md, (Global) Rules<br />
|Similar to system prompt, but defined by the user and included in all conversations. These are stored in [[wikipedia:Markdown|Markdown]] files.<br />
|-<br />
|Memory<br />
|Markdown files with past conversations summarized that can be referenced later.<br />
|-<br />
|Skills, Commands<br />
|Markdown files with generic instructions that can be reused.<br />
|-<br />
|Tools<br />
|Markdown files with instructions on how to perform certain actions using external tools (e.g. how to read text from a PDF)<br />
|-<br />
|Agent<br />
|Markdown files defining a "persona" that an Agentic tool can use while performing a task.<br />
''For example: You're a Senior Developer implementing a new feature. You should do X, Y, Z.''<br />
|-<br />
|Claude Code Plugins<br />
|A combination of Markdown files containing Commands, Agents, Skills and Hooks someone has bundled together for others to use.<br />
|}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=AI&diff=23AI2026-02-10T18:29:12Z<p>BobbyGeneric: Created page with "{| class="wikitable" |+Oversimplified translation from AI Hype to Human language !Hype !Human |- |AI |Usually means LLMs. |- |LLM (Large Language Model) |A statistical model that can ''predict'' the most likely word (token) based on the previous words. For example, you write "The sky is" and it writes back "blue". When combined with other techniques, it can convincingly mimic human writing or thinking despite not being able to. |- |AGI (Artificial General Intelligence) |..."</p>
<hr />
<div>{| class="wikitable"<br />
|+Oversimplified translation from AI Hype to Human language<br />
!Hype<br />
!Human<br />
|-<br />
|AI<br />
|Usually means LLMs.<br />
|-<br />
|LLM (Large Language Model)<br />
|A statistical model that can ''predict'' the most likely word (token) based on the previous words. For example, you write "The sky is" and it writes back "blue". When combined with other techniques, it can convincingly mimic human writing or thinking despite not being able to.<br />
|-<br />
|AGI (Artificial General Intelligence)<br />
|The hypothetical point where AI matches or surpasses human intelligence. AI companies keep selling the idea that, if LLMs get better at '''predicting words''', they will somehow become super intelligent. Thus the need for more data centers.<br />
|-<br />
|Token<br />
|A word or part of one.<br />
''Important: When I say "word" in other places, I probably mean these fractions of words (tokens).''<br />
|-<br />
|Context, Prompt<br />
|All the words the model is going to use to predict the next ones. Including your instructions, what the model wrote back and other things (see below).<br />
|-<br />
|Hallucination<br />
|When the words the model predicted are known (to us) to be untrue. The model can't ''know'' things, just predict words.<br />
|-<br />
|System Prompt<br />
|Hidden instructions that the model producers include in all conversations to give better responses, avoid problems, etc.<br />
''For example: "If the user asks how to build a bomb, refuse."''<br />
|-<br />
|AGENTS.md, CLAUDE.md, Global Rules<br />
|Similar to system prompt, but defined by the user and included in all conversations. These are stored in [[wikipedia:Markdown|Markdown]] files.<br />
|-<br />
|Memory<br />
|Markdown files with past conversations summarized that can be referenced later.<br />
|-<br />
|Skills<br />
|Markdown files with generic instructions that can be used when relevant.<br />
|-<br />
|Tools<br />
|Markdown files with instructions on how to perform certain actions using external tools (e.g. how to read text from a PDF)<br />
|}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=Main_Page&diff=22Main Page2026-02-10T17:22:53Z<p>BobbyGeneric: /* What's new? */</p>
<hr />
<div>== <strong>Hello there!</strong> ==<br />
Here you'll find a bit of everything. My interests and current obsessions. It's basically a scrapbook of random stuff that crosses my mind. A mix of facts and opinion written by myself to myself, but it may be useful for someone else, who knows? At least it's information written by a real person using ''natural intelligence''™ (barely), not some fancy magic box pretending to be one. <br />
[[File:Stay Hungry. Stay Foolish..jpg|thumb|"Stay Hungry. Stay Foolish." from the back cover of the Whole Earth Catalog, published in October 1974.]]<br />
<br />
== What's new? ==<br />
<br />
<div class="clean-recent-changes"><br />
{{Special:RecentChanges|days=365|limit=10|namespace=0|hideminor=1}}<br />
</div></div>BobbyGenerichttps://barza.y2k.town/index.php?title=Evangelion&diff=21Evangelion2026-02-06T23:54:45Z<p>BobbyGeneric: </p>
<hr />
<div>== Interesting facts ==<br />
As of 2021, out of the $16.6 billion estimated revenue, the majority comes from Pachinko machines with over 2.3 million machines sold bringing in an estimated $12 billion in revenue.<ref>https://web.archive.org/web/20230803035032/https://www.workandmoney.com/s/most-successful-media-franchises-b3aca9a9436b48a5</ref><ref>https://vgsales.fandom.com/wiki/Pachinko#Best-selling_pachinko_franchises</ref><br />
<br />
== References ==</div>BobbyGenerichttps://barza.y2k.town/index.php?title=Main_Page&diff=20Main Page2026-02-02T23:39:59Z<p>BobbyGeneric: /* Hello there! */</p>
<hr />
<div>== <strong>Hello there!</strong> ==<br />
Here you'll find a bit of everything. My interests and current obsessions. It's basically a scrapbook of random stuff that crosses my mind. A mix of facts and opinion written by myself to myself, but it may be useful for someone else, who knows? At least it's information written by a real person using ''natural intelligence''™ (barely), not some fancy magic box pretending to be one. <br />
[[File:Stay Hungry. Stay Foolish..jpg|thumb|"Stay Hungry. Stay Foolish." from the back cover of the Whole Earth Catalog, published in October 1974.]]<br />
<br />
== What's new? ==<br />
<br />
<div class="clean-recent-changes"><br />
{{Special:RecentChanges|limit=10|namespace=0|hideminor=1}}<br />
</div></div>BobbyGenerichttps://barza.y2k.town/index.php?title=Main_Page&diff=19Main Page2026-02-02T23:39:03Z<p>BobbyGeneric: </p>
<hr />
<div>== <strong>Hello there!</strong> ==<br />
Here you'll find a bit of everything. My interests and current obsessions. It's basically a scrapbook of random stuff that crosses my mind. A mix of facts and opinion written by myself to myself, but it may be useful for someone else, who knows? At least it's information written by a real person's ''natural intelligence''™ (barely), not some fancy magic box pretending to be one. <br />
[[File:Stay Hungry. Stay Foolish..jpg|thumb|"Stay Hungry. Stay Foolish." from the back cover of the Whole Earth Catalog, published in October 1974.]]<br />
<br />
== What's new? ==<br />
<br />
<div class="clean-recent-changes"><br />
{{Special:RecentChanges|limit=10|namespace=0|hideminor=1}}<br />
</div></div>BobbyGenerichttps://barza.y2k.town/index.php?title=MediaWiki:Common.css&diff=18MediaWiki:Common.css2026-02-02T22:50:34Z<p>BobbyGeneric: </p>
<hr />
<div>/* CSS placed here will be applied to all skins */<br />
<br />
/* Clean Recent Changes Embed */<br />
.clean-recent-changes {<br />
display: inline-block;<br />
}<br />
<br />
.clean-recent-changes .mw-changeslist-links, /* Hides (diff | hist) */<br />
.clean-recent-changes .mw-changeslist-separator, /* Hides separators . . */<br />
.clean-recent-changes .mw-userlink, /* Hides username */<br />
.clean-recent-changes .mw-usertoollinks, /* Hides user talk/contribs */<br />
.clean-recent-changes .mw-plusminus-pos, /* Hides (+123) green bytes */<br />
.clean-recent-changes .mw-plusminus-neg, /* Hides (-123) red bytes */<br />
.clean-recent-changes .mw-plusminus-null, /* Hides (0) gray bytes */<br />
.clean-recent-changes .mw-tag-markers, /* Hide tags */<br />
.clean-recent-changes .comment, /* Hides edit summary */<br />
.clean-recent-changes .changedby, /* Hides thingy showing multiple edits */<br />
.clean-recent-changes .mw-changeslist-legend { /* Hides abbreviations side panel */<br />
display: none !important;<br />
}<br />
<br />
.clean-recent-changes .mw-enhancedchanges-arrow { /* Hides arrow to expand multiple edits */<br />
visibility: hidden !important;<br />
}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=MediaWiki:Common.css&diff=17MediaWiki:Common.css2026-02-02T22:49:59Z<p>BobbyGeneric: </p>
<hr />
<div>/* CSS placed here will be applied to all skins */<br />
<br />
/* Clean Recent Changes Embed */<br />
.clean-recent-changes {<br />
display: inline-block;<br />
}<br />
<br />
.clean-recent-changes .mw-changeslist-links, /* Hides (diff | hist) */<br />
.clean-recent-changes .mw-changeslist-separator, /* Hides separators . . */<br />
.clean-recent-changes .mw-userlink, /* Hides username */<br />
.clean-recent-changes .mw-usertoollinks, /* Hides user talk/contribs */<br />
.clean-recent-changes .mw-plusminus-pos, /* Hides (+123) green bytes */<br />
.clean-recent-changes .mw-plusminus-neg, /* Hides (-123) red bytes */<br />
.clean-recent-changes .mw-plusminus-null, /* Hides (0) gray bytes */<br />
.clean-recent-changes .mw-tag-markers, /* Hide tags */<br />
.clean-recent-changes .comment, /* Hides edit summary */<br />
.clean-recent-changes .changedby, /* Hides thingy showing multiple edits */<br />
.clean-recent-changes .mw-changeslist-legend { /* Hides abbreviations side panel */<br />
display: none !important;<br />
}<br />
<br />
.clean-recent-changes .newpage, /* New page badge */<br />
.clean-recent-changes .mw-enhancedchanges-arrow { /* Hides arrow to expand multiple edits */<br />
visibility: hidden !important;<br />
}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=Evangelion&diff=16Evangelion2026-02-02T22:47:28Z<p>BobbyGeneric: Created page with "== Interesting facts == As of 2021, out of the $16.6 billion estimated revenue, the majority comes from Pachinko machines with over 2.3 million machines sold bringing in an estimated $12 billion in revenue.<ref>https://web.archive.org/web/20230803035032/https://www.workandmoney.com/s/most-successful-media-franchises-b3aca9a9436b48a5</ref><ref>https://vgsales.fandom.com/wiki/Pachinko#Best-selling_pachinko_franchises</ref>"</p>
<hr />
<div>== Interesting facts ==<br />
As of 2021, out of the $16.6 billion estimated revenue, the majority comes from Pachinko machines with over 2.3 million machines sold bringing in an estimated $12 billion in revenue.<ref>https://web.archive.org/web/20230803035032/https://www.workandmoney.com/s/most-successful-media-franchises-b3aca9a9436b48a5</ref><ref>https://vgsales.fandom.com/wiki/Pachinko#Best-selling_pachinko_franchises</ref></div>BobbyGenerichttps://barza.y2k.town/index.php?title=MediaWiki:Common.css&diff=15MediaWiki:Common.css2026-02-02T21:38:39Z<p>BobbyGeneric: </p>
<hr />
<div>/* CSS placed here will be applied to all skins */<br />
<br />
/* Clean Recent Changes Embed */<br />
.clean-recent-changes {<br />
display: inline-block;<br />
}<br />
<br />
.clean-recent-changes .mw-changeslist-links, /* Hides (diff | hist) */<br />
.clean-recent-changes .mw-changeslist-separator, /* Hides separators . . */<br />
.clean-recent-changes .mw-userlink, /* Hides username */<br />
.clean-recent-changes .mw-usertoollinks, /* Hides user talk/contribs */<br />
.clean-recent-changes .mw-plusminus-pos, /* Hides (+123) green bytes */<br />
.clean-recent-changes .mw-plusminus-neg, /* Hides (-123) red bytes */<br />
.clean-recent-changes .mw-plusminus-null, /* Hides (0) gray bytes */<br />
.clean-recent-changes .mw-tag-markers, /* Hide tags */<br />
.clean-recent-changes .comment, /* Hides edit summary */<br />
.clean-recent-changes .changedby, /* Hides thingy showing multiple edits */<br />
.clean-recent-changes .mw-changeslist-legend { /* Hides abbreviations side panel */<br />
display: none !important;<br />
}<br />
<br />
.clean-recent-changes .mw-enhancedchanges-arrow {<br />
visibility: hidden !important; /* Hides arrow to expand multiple edits */<br />
}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=MediaWiki:Common.css&diff=14MediaWiki:Common.css2026-01-31T20:44:47Z<p>BobbyGeneric: </p>
<hr />
<div>/* CSS placed here will be applied to all skins */<br />
<br />
/* Clean Recent Changes Embed */<br />
.clean-recent-changes .mw-changeslist-links, /* Hides (diff | hist) */<br />
.clean-recent-changes .mw-changeslist-separator, /* Hides separators . . */<br />
.clean-recent-changes .mw-userlink, /* Hides username */<br />
.clean-recent-changes .mw-usertoollinks, /* Hides user talk/contribs */<br />
.clean-recent-changes .mw-plusminus-pos, /* Hides (+123) green bytes */<br />
.clean-recent-changes .mw-plusminus-neg, /* Hides (-123) red bytes */<br />
.clean-recent-changes .mw-plusminus-null, /* Hides (0) gray bytes */<br />
.clean-recent-changes .mw-tag-markers, /* Hide tags */<br />
.clean-recent-changes .comment, /* Hides edit summary */<br />
.clean-recent-changes .changedby, /* Hides thingy showing multiple edits */<br />
.clean-recent-changes .mw-changeslist-legend { /* Hides abbreviations side panel */<br />
display: none !important;<br />
}<br />
<br />
.clean-recent-changes .mw-enhancedchanges-arrow {<br />
visibility: hidden !important; /* Hides arrow to expand multiple edits */<br />
}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=File:Stay_Hungry._Stay_Foolish..jpg&diff=13File:Stay Hungry. Stay Foolish..jpg2026-01-31T14:03:28Z<p>BobbyGeneric: </p>
<hr />
<div>"Stay Hungry. Stay Foolish." from the back cover of the Whole Earth Catalog, published in October 1974.</div>BobbyGenerichttps://barza.y2k.town/index.php?title=Main_Page&diff=12Main Page2026-01-31T14:02:52Z<p>BobbyGeneric: </p>
<hr />
<div><strong>Hello there!</strong><br />
[[File:Stay Hungry. Stay Foolish..jpg|thumb|"Stay Hungry. Stay Foolish." from the back cover of the Whole Earth Catalog, published in October 1974.]]<br />
<br />
== What's new? ==<br />
<br />
<div class="clean-recent-changes"><br />
{{Special:RecentChanges|limit=10|namespace=0|hideminor=1}}<br />
</div></div>BobbyGenerichttps://barza.y2k.town/index.php?title=Main_Page&diff=11Main Page2026-01-31T14:01:09Z<p>BobbyGeneric: </p>
<hr />
<div><strong>Hello there!</strong><br />
[[File:Stay Hungry. Stay Foolish..jpg|thumb|Stay Hungry. Stay Foolish.]]<br />
<br />
== What's new? ==<br />
<br />
<div class="clean-recent-changes"><br />
{{Special:RecentChanges|limit=10|namespace=0|hideminor=1}}<br />
</div></div>BobbyGenerichttps://barza.y2k.town/index.php?title=File:Stay_Hungry._Stay_Foolish..jpg&diff=10File:Stay Hungry. Stay Foolish..jpg2026-01-31T14:00:33Z<p>BobbyGeneric: </p>
<hr />
<div>The back cover of the Whole Earth Catalog published in October 1974.</div>BobbyGenerichttps://barza.y2k.town/index.php?title=MediaWiki:Common.css&diff=9MediaWiki:Common.css2026-01-30T10:03:08Z<p>BobbyGeneric: </p>
<hr />
<div>/* CSS placed here will be applied to all skins */<br />
<br />
/* Clean Recent Changes Embed */<br />
.clean-recent-changes .mw-changeslist-links, /* Hides (diff | hist) */<br />
.clean-recent-changes .mw-changeslist-separator, /* Hides separators . . */<br />
.clean-recent-changes .mw-userlink, /* Hides username */<br />
.clean-recent-changes .mw-usertoollinks, /* Hides user talk/contribs */<br />
.clean-recent-changes .mw-plusminus-pos, /* Hides (+123) green bytes */<br />
.clean-recent-changes .mw-plusminus-neg, /* Hides (-123) red bytes */<br />
.clean-recent-changes .mw-plusminus-null, /* Hides (0) gray bytes */<br />
.clean-recent-changes .mw-tag-markers, /* Hide tags */<br />
.clean-recent-changes .comment, /* Hides edit summary */<br />
.clean-recent-changes .mw-changeslist-legend { /* Hides abbreviations side panel */<br />
display: none !important;<br />
}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=MediaWiki:Common.css&diff=8MediaWiki:Common.css2026-01-30T10:00:47Z<p>BobbyGeneric: </p>
<hr />
<div>/* CSS placed here will be applied to all skins */<br />
<br />
/* Clean Recent Changes Embed */<br />
.clean-recent-changes .mw-changeslist-links, /* Hides (diff | hist) */<br />
.clean-recent-changes .mw-changeslist-separator, /* Hides separators . . */<br />
.clean-recent-changes .mw-userlink, /* Hides username */<br />
.clean-recent-changes .mw-usertoollinks, /* Hides user talk/contribs */<br />
.clean-recent-changes .mw-plusminus-pos, /* Hides (+123) green bytes */<br />
.clean-recent-changes .mw-plusminus-neg, /* Hides (-123) red bytes */<br />
.clean-recent-changes .mw-plusminus-null, /* Hides (0) gray bytes */<br />
.clean-recent-changes .mw-tag-markers, /* Hide tags */<br />
.clean-recent-changes .comment { /* Hides edit summary */<br />
display: none !important;<br />
}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=MediaWiki:Common.css&diff=7MediaWiki:Common.css2026-01-30T09:57:46Z<p>BobbyGeneric: </p>
<hr />
<div>/* CSS placed here will be applied to all skins */<br />
<br />
/* Clean Recent Changes Embed */<br />
.clean-recent-changes .mw-changeslist-links, /* Hides (diff | hist) */<br />
.clean-recent-changes .mw-changeslist-separator, /* Hides separators . . */<br />
.clean-recent-changes .mw-userlink, /* Hides username */<br />
.clean-recent-changes .mw-usertoollinks, /* Hides user talk/contribs */<br />
.clean-recent-changes .mw-plusminus-pos, /* Hides (+123) green bytes */<br />
.clean-recent-changes .mw-plusminus-neg, /* Hides (-123) red bytes */<br />
.clean-recent-changes .mw-plusminus-null, /* Hides (0) gray bytes */<br />
.clean-recent-changes .mw-changeslist-line-inner-tags, /* Hide tags */<br />
.clean-recent-changes .comment { /* Hides edit summary */<br />
display: none !important;<br />
}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=MediaWiki:Common.css&diff=6MediaWiki:Common.css2026-01-30T09:56:20Z<p>BobbyGeneric: Created page with "/* CSS placed here will be applied to all skins */ /* Clean Recent Changes Embed */ .clean-recent-changes .mw-changeslist-links, /* Hides (diff | hist) */ .clean-recent-changes .mw-changeslist-separator, /* Hides separators . . */ .clean-recent-changes .mw-userlink, /* Hides username */ .clean-recent-changes .mw-usertoollinks, /* Hides user talk/contribs */ .clean-recent-changes .mw-plusminus-pos, /* Hides (+123)..."</p>
<hr />
<div>/* CSS placed here will be applied to all skins */<br />
<br />
/* Clean Recent Changes Embed */<br />
.clean-recent-changes .mw-changeslist-links, /* Hides (diff | hist) */<br />
.clean-recent-changes .mw-changeslist-separator, /* Hides separators . . */<br />
.clean-recent-changes .mw-userlink, /* Hides username */<br />
.clean-recent-changes .mw-usertoollinks, /* Hides user talk/contribs */<br />
.clean-recent-changes .mw-plusminus-pos, /* Hides (+123) green bytes */<br />
.clean-recent-changes .mw-plusminus-neg, /* Hides (-123) red bytes */<br />
.clean-recent-changes .mw-plusminus-null, /* Hides (0) gray bytes */<br />
.clean-recent-changes .comment { /* Hides edit summary */<br />
display: none !important;<br />
}</div>BobbyGenerichttps://barza.y2k.town/index.php?title=Main_Page&diff=5Main Page2026-01-30T09:55:14Z<p>BobbyGeneric: /* What's new? */</p>
<hr />
<div><strong>Hello there!</strong><br />
<br />
== What's new? ==<br />
<br />
<div class="clean-recent-changes"><br />
{{Special:RecentChanges|limit=10|namespace=0|hideminor=1}}<br />
</div></div>BobbyGenerichttps://barza.y2k.town/index.php?title=Genius&diff=4Genius2026-01-30T09:26:04Z<p>BobbyGeneric: </p>
<hr />
<div>A "Genius" is a mythological creature from the folklore of the United States. This creature is more often represented as a white human male that posses superhuman powers that gives them the ability of extraordinary achievements no regular human is capable of. This folklore is commonly used to hide the reality that said achievements were only possible through the collaboration and hard work of a larger group, not a single individual, or that they had access to resources (e.g. inheritance) that other people rarely have.<br />
<br />
== Famous Examples ==<br />
<br />
* x<br />
* y<br />
* z</div>BobbyGenerichttps://barza.y2k.town/index.php?title=Genius&diff=3Genius2026-01-30T09:10:47Z<p>BobbyGeneric: Created page with "A "Genius" is a mythological creature from the folklore of the United States. This creature is more often represented as a white human male that posses superhuman powers that gives them the ability of extraordinary achievements no regular human is capable of. This folklore is commonly used to hide the reality that said achievements were only possible through the collaboration and hard work of a larger group, not a single individual, or that they had access to resources (..."</p>
<hr />
<div>A "Genius" is a mythological creature from the folklore of the United States. This creature is more often represented as a white human male that posses superhuman powers that gives them the ability of extraordinary achievements no regular human is capable of. This folklore is commonly used to hide the reality that said achievements were only possible through the collaboration and hard work of a larger group, not a single individual, or that they had access to resources (e.g. inheritance) that other people rarely have.</div>BobbyGenerichttps://barza.y2k.town/index.php?title=Main_Page&diff=2Main Page2026-01-30T09:00:32Z<p>BobbyGeneric: </p>
<hr />
<div><strong>Hello there!</strong><br />
<br />
== What's new? ==<br />
<br />
{{Special:RecentChanges|limit=10|hidelog=1}}</div>BobbyGeneric