ngkw

I love this.

happy new year from tokyo

Sample input(for fun):

``` What would you like AI to know about you to personalize your digest?(Optional) >> Me a genius kinnygardener!

How would you like AI to respond?(Optional)

>>Can you 'splain wif weally compwicated big maf-ma-tiks fomulas and lots of scawy symbows pwease ```

Sample output:

```

title: Weekwy Hacker News: Maf-ma-tiks of Mulch & Mowe! ∫∑∞≈ body: Hewwo, Genius Kinnygardener! Wewcome to youw weekwy Hacker News newswetter! We've curated some supew intewesting awticwes that will make youw bwain go boom! Let's dive into the wowwd of visuaw pwogwamming, typogwaphy in movies, intewpwetews, and even gawden muwch! Pwepare for some mind-bending maf-ma-tiks and scawy symbows!

1. We need visuaw pwogwamming. No, not wike that! Imagine coding with pwetty pictuws! This awticwe expwores how visuawizations can hewp us undewstand compwex pwogwamming concepts. It's wike using emojis to expwain quantum mechanics!

Here's a scawy equation to represent the power of visuaw pwogwamming:

∀x ∈ (Complexity(x) = ∫0∞ Visualization(x) dx)

One smart commenter said: "Visuawizations do exactwy that - they hewp us wook at pwobwems fwom diffewent angwes!"

2. The Typeset of Waww·E (2018): A Pixar Typogwaphy Adventure! Did you know that even wobots have their own fonts? This awticwe dives deep into the typogwaphy of Waww·E, showing how even the smawwest detaiws can tell a big stowy!

Let's express the importance of typography with a formula:

Typography_Impact = Σ(Font_Choice × Cultural_Reference × Pixel_Density2)

A weader mentioned: "This write-up has gweat depth, not onwy on typogwaphy but awso awchitecture, art stywes, fiwm, and music!"

3. Cwafting Intewpwetews: Buiwding Wanguages fwom Scwatch! Ever wanted to cweate your own programming wanguage? This book shows you how! It's wike cooking, but instead of food, you're making code dewicious!

Here's a compwicated fowmuwa for intewpweter design:

Interpreter_Complexity = ∑i=1n (Lexical_Analysisi + Syntactic_Analysisi + Semantic_Analysisi) × Runtime_Efficiency

The author said: "I put a ton of time and wove into it and it's gwatifying to see it had the effect I'd hoped for." Aww, how sweet!

4. Show HN: Wesuwwecting a dead Dune RTS game Spice must fwow, and so must owd games! This project brings back a cwassic Dune game, showing how important it was for the whole RTS genwe!

Let's represent the impact of Dune on RTS games with this equation:

RTS_Evolution = ∏(Dune_Influence × Resource_Management × Unit_Control) / Time

A commenter noted: "This game actuawwy howds a ton of significance for the RTS genwe as a whole." It's wike finding a fossiw that expwains how dinosaurs became birds!

5. Which Gawden Muwch Saves the Most Water? [video] Time for some weaw kinnygardening science! This video shows which muwch is best for saving water. It's wike a supew soaker fight, but for pwants!

Let's express muwch efficiency with this fowmuwa:

Water_Retention = ∫(Mulch_Depth × Porosity) dt - Evaporation_Rate

Someone said: "This video is an exampwe of a gweat weseawch and video as a medium fits better than a whitepaper for this puwpose." It's wike watching Biww Nye, but for gawdening!

Concwusion Wow, what a week of techy goodness! From visuaw coding to wobot fonts, wanguage cwafting to desert wargames, and even some juicy muwch facts - your bwain must be buzzing!

Don't fowget to expwore these awticwes and join the discussions. Who knows, you might even come up with a new scawy equation to share!

Until next week, keep your code cwean and your pwants gween!

Yours twuwy, Your Hacker News Newswetter Bot

My Hacker News

```

Seemingly very interesting moves, but I wonder what specific needs remain for a 'browser-level' local LLM, because local LLMs will be on devices in the near future. So if we're not connected to the internet, maybe a device-level LLM would be better. On the other hand, when we open the browser, we're connected to the great internet! I know browser-level LLMs can have several benefits like speed, privacy protection, and cost-effectiveness, but these features are covered by internet-based LLM APIs or device-level LLM APIs.

haha I have mistaken the URL in the comments. Here's the right one: https://copyfromurl.com

To be honest, I'm really frustrated that OpenAI and Anthropic don't open certain links. It's not just because of robots.txt restrictions, but also due to copyright issues and their excessive caution regarding media content. Isn't it annoying to deal with that? So, the idea was to create a tool that allows us to easily bypass these limitations ourselves.

Thanks for your comment, JSDevOps! I appreciate your input. I understand that robots.txt can be bypassed by some. The idea behind "Copy from URL" is to address a specific need when interacting with AI chatbots through their websites, like ChatGPT or Claude, where we don't have control over robots.txt settings (that's up to OpenAI or Anthropic). If you use the API, you can just ignore this, but that's not the case I'm targeting. This tool is designed for those times when you want to quickly copy content from a site to use in your AI conversations. It's just a simple way to make that process smoother. Hope that clarifies the intention behind the tool!

Github codes: https://github.com/Amunzen/CopyFromURL

Any comment is welcome! Thanks.

I'm curious to hear what UC citizens think about this issue.

Apple's new iPad Pro promotional video, released by CEO Tim Cook on Twitter, has sparked controversy among Japanese users. The video depicts various creative tools being destroyed by a giant press machine, consolidating their functions into the thin iPad Pro. Many Japanese users, who have a strong cultural attachment to cherishing objects, expressed discomfort and disappointment in response to the video. They felt it disrespected their appreciation for musical instruments, cameras, and art supplies. Comments from Japanese users flooded in, criticizing the video for "looking down on analog creativity" and expressing sadness at seeing creative tools being destroyed. The backlash highlights a cultural disconnect between Apple's intended message and the values held by its Japanese user base.

Hi HN,

I've been exploring the concept of using AI to judge online debates as a way to make debate more globally accessible. The core idea is to let users compete in real-time debates on rotating topics, with an AI acting as the judge.

Some key features being considered:

- Multiple battle modes (AI match, friend match, global match) - Multilingual support for cross-language debates - Voice and text input options - Score-based matchmaking - Custom topic creation

The goal is to extend debate education beyond current limitations of country, region, and socioeconomic class. By supporting multiple languages and enabling location-agnostic debates, such a platform could potentially help more people worldwide develop critical thinking and rhetorical skills.

However, there are certainly valid concerns about using AI judging. The AI would be meant as an accessibility aid and starting point, not an infallible authority. Questions arise around how much value the platform could provide by offering a structured space to practice debating and get feedback, even if the judging is imperfect.

As an HN community with a wealth of expertise, I would greatly value your thoughts and reactions:

- What potential benefits and drawbacks do you see in using AI to judge debates for accessibility? - What considerations should be kept in mind in designing such a system? - Are there existing projects or research in this area that could inform the approach?

I'm deeply curious to hear the community's perspective on the intellectual challenges and opportunities associated with this concept. Debating the implications of "AI debate judging" seems quite meta!

Thanks very much for entertaining this topic. I'm looking forward to a thought-provoking discussion and to learning from all of your insights.

Hi HN,

I'm excited to share Debatia, an online debate platform I've been working on that aims to make debate more accessible to a global audience.

The core idea is to let users compete in real-time debates on rotating topics, with an AI acting as the judge. Some key features:

- Multiple battle modes (AI match, friend match, global match) - Multilingual support for cross-language debates - Voice and text input options - Score-based matchmaking - Custom topic creation to debate with friends

My goal with Debatia is to extend debate education beyond the current limitations of country, region, and socioeconomic class. By supporting multiple languages and enabling location-agnostic debates, I hope this platform can help more people worldwide develop critical thinking and rhetorical skills.

I know there may be valid concerns about using AI judging. The AI is meant as an accessibility aid and starting point, not an infallible authority. I believe the platform can still provide value by offering a structured space to practice debating and get feedback.

As an HN community with a wealth of expertise, I would greatly value your thoughts and reactions:

- What aspects of the concept resonate with you? - What concerns or areas for improvement come to mind?

Your insights at this formative stage will be immensely helpful in shaping a platform that provides real benefit to the debate community.

Thanks very much for your consideration. I'm looking forward to learning from all of you!

It's almost showtime! Debatia's Inaugural One-Hour Online Debate Challenge begins in just 60 minutes. Are you ready to unleash your debating prowess in a multilingual arena judged by AI? Remember, no registration needed – just log in to your Debatia account and let the intellectual battle commence at 11:00AM PST.

Prepare for surprise topics, face matched opponents, and let your arguments shine. Your feedback today shapes the future of Debatia. See you in the arena!

Best

Debatia is launching its first online debate challenge on 12/17/2023. This one-hour event promises intense, AI-judged debates in 12 different languages. Whether you're a master debater or just love a good argument, this challenge is for you.

# Key Highlights:

- Multilingual Competition: Debate in English, Spanish, French, and more. Our AI(ChatGPT-based one) seamlessly bridges language gaps. - AI Judges: ChatGPT's AI ensures unbiased, analytical judgment of your arguments. - Equal Matchmaking: Face opponents at your level for truly challenging debates. - Dynamic Scoring: Start with 1500 points and climb the ranks by outwitting higher-ranked debaters.

# Event Details:

- Date: 12/17/2023 - Time: 11:00 AM PST - Duration: 1 Hour - Topics: Surprise themes to test your quick thinking!

No registration needed – just show up ready to debate. Create a Debatia account for a smoother experience.

Your feedback post-event could shape the future of daily debates on Debatia. Don’t miss this chance to be a part of something new and exciting in the world of online debating. Let's debate!

Thank you!

Thank you very much. I think this is the most serious issue and I admit it is not appropriate. I will fix it from now.

Thanks! I fixed it!

Thanks! I will fix tonight.