cateye

Turkish TV series have been operating on this exact "second screen" logic for decades. They are massive global exports specifically because their 140-minute episodes rely heavily on meaningful stares, flashbacks, and circular dialogue.

They are designed for people ironing or cooking; you can leave the room for twenty minutes and miss absolutely nothing.

Personally, I can't bear them, the constant spoon-feeding is torture if you are actually paying attention, but Netflix is effectively just adopting this proven, low-attention retention strategy.

Unclear if this is actually intended or just a temporary malfunction.

It feels like they might be pushing users to the Mac app, but there’s no official communication about it yet.

That’s why I posted it here on HN, hoping someone knows more.

Isn’t giving LLMs “IQ scores” a category error?

Human IQ is norm-referenced psychometrics under embodied noise. Calling both “IQ” isn’t harmless, it invites bad policy and building decisions on a false equivalence. Don’t promote it.

we are inspired by the recent advancements in reinforcement learning (e.g., o1)

It is interesting to see what the future will bring when models incorporate chain of thought approaches and whether o1 will get outperformed by open source models.

Me too. I use now https://fireworks.ai/ for vision models to test.

It is strange that this model is not available on Together.ai to try it out after reading the blog artcile.

Some currencies use more than 2 decimal places. For instance, the currencies of Algeria, Bahrain, Iraq, Jordan, Kuwait, Libya, and Tunisia are subdivided into 3 decimals.

There is a logical error in this statement:

"governments and corporations silencing dissidents has done far far more harm to humanity than people losing their accounts"

People can not loose their accounts, because they are governed which makes silencing possible.

Odds Ratio Preference Optimization (ORPO): https://arxiv.org/abs/2403.07691

Last week, chatgpt literally generated a pptx file that I have downloaded, with the content that I asked for.

This article and many people attempting to analyze this topic (and many others) are missing a key element: "incentives"!

Many developers are more passionate about technology itself – coding, solving complex technical problems, and creating efficient solutions. Their formal rewards (like promotions, bonuses, and recognition) are often tied to technical achievements. If a developer spends 40 hours on customer research, this effort might not be as formally recognized or rewarded as much as a technical achievement like developing a new authentication solution that benefits the entire team.

Shifting this dynamic is not just about encouraging developers to spend more time on product management. It requires a systemic change in how organizations value and reward.

Changing the incentive structure to be more user-centric and less technically focused would significantly impact middle and upper management roles. A shift would challenge the very foundations of their roles and responsibilities, potentially leading to a loss of control and influence.

It's likely that such changes would face subtle yet strong resistance from middle and upper management. These individuals often have significant political power within an organization and a vested interest.

I have mixed thoughts about this approach.

Unification should mean creating a single interface that seamlessly interacts with multiple underlying services without needing custom adaptations for each case.

The issue with allowing for customization is that it leads to fragmentation. This defeats the purpose of having a unified API in the first place.

I'm not sure whether the convenience of unifying leftover common functionalities like authentication outweighs the complexity and reliance introduced by adding another layer to the tech stack.

The expected pricing is very strange:

Regular Twitch Neurons (RTN) - running wherever there's capacity at $0.01 / 1k neurons

Fast Twitch Neurons (FTN) - running at nearest user location at $0.125 / 1k neurons

Neurons are a way to measure AI output that always scales down to zero. To give you a sense of what you can accomplish with a thousand neurons, you can: generate 130 LLM responses, 830 image classifications, or 1,250 embeddings.

Who came up with this? This is ridiculous. I understand the underlying issues but would still prefer a metric like seconds of utilization multiplied by the size of worker.

Besides this, the expected pricing doesn't talk about the expected pricing but just the pricing model. Have the feeling that this is not going to be competitive to platforms like Vast.ai

I have not read it very carefully, but couldn't spot price information.

Without any indication, the whole question whether someone is interested becomes a little meaningless.

It makes more sense to embed stable diffusion capabilities into well-established image editors such as Gimp, Photoshop, Krita, or Figma, which come with layered, non-destructive functionalities, rather than attempting the opposite approach.

https://github.com/Interpause/auto-sd-paint-ext https://github.com/thndrbrrr/gimp-stable-boy https://www.magicbrushai.com/

The conventional wisdom around burnout often emphasizes taking a break or pursuing hobbies, and while I'm not against that approach, I'd like to propose a potentially more contentious angle. Consider that burnout might not stem from the sheer volume of work but rather the cognitive load it imposes.

Many of us in the tech industry feel overwhelmed, not necessarily by the actual tasks, but by a lack of clear understanding of what needs to be done or what you have accomplished.

To tackle this, ask yourself precise questions: What are the specific tasks at hand? What have you truly accomplished? What's causing the stress? Identifying and rigorously addressing these underlying stressors might lead to a more fundamental solution to burnout.

Always open to conversations; you never know what can come out of it. Feel free to reach out to me at info@boostpixels.com.

Introducing a free (for a limited time) tool to generate SVG logos & icons using text prompts like 'Minimalist mountain logo' or 'Retro game icon.' Simply input your idea, and the tool will craft a crisp, scalable design that's easily editable in Figma. The creative flexibility of the output, running on Stable Diffusion 1.5 (no SDXL yet), is impressive.

I'm considering enhancements such as quantized color tracing and gradient tracing improvements (to reduce speckles), but I'd love to gauge interest first from the HN community.

The generation process is blazing fast at around 3 seconds per image. Your feedback and insights would be invaluable as I continue to develop this tool.

Greg Rutkowski's art style is similar to Gustave Moreau's, Rutkowski shouldn't be permitted to create his artwork.

One way is trying to sneak in a specific structure/pattern that is difficult for a human to notice when reading, like using a particular sentence length, paragraph length, or punctuation pattern. Or use certain words in the text that may not be frequently used by humans etc.

Watermarking needs to be subtle enough to be unnoticeable to opposing parties, yet distinctive enough to be detectable.

So, this is an arms race especially because detecting it and altering it based on the watermark is also fun :)

"The only constant in life is change." Indeed, the rate of technological change seems ever-accelerating, leading us through cycles of adaptation and learning.

While this constant shift may sometimes be seen as tedious or even chaotic, it's also a reflection of our growth as an industry. Each new tool or practice arises from the shortcomings of the previous ones. Serverless wasn't born in a vacuum, it was born out of the limitations of servers. Microservices weren't born out of nothing, they came from the monolithic application era's shortcomings.

The key lies in discerning between meaningful progress and trend-induced churn. Not every new tool or practice is necessarily an improvement, and even those that are may not be the right fit for every organization or use case.

On a personal level, how can one keep up? Find your anchors in the fundamentals. Concepts like abstraction, encapsulation, concurrency, state management, and others are applicable across a multitude of tech stacks and paradigms.

As an avid Runpod user, I have come across some information that is not entirely accurate.

- Although the number of models is limited, the platform has a community feature where users can fork models.

I'm not entirely sure of the intended meaning, but contradicts with being able to bring any container.

However, if it pertains to the selection of pre-made template containers, it is true that the options may be limited. Nonetheless, there aren't a significant number of commonly-used open-source models available either...

Or it is referring to the API endpoints. That is indeed correct but it is confusing what this review is exactly about.

- Post-deployment, it can be confusing to understand how the platform works, which may result in users receiving a bill if they are not careful.

Although documentation is sometimes lacking, it is very clear what it costs and there is not such thing as a bill...

- There is no bot or instant support mechanism available.

They have a live chat available, and their response time was good. They are also very active on their Discord channel, providing speedy support to users.

I think there are 2 problems at interplay here. First, there needs to be nomenclature. This is a weird semantic problem. Music is quite abstract in that sense. The schema is just a small sub problem of this.

Second, how to systematically tag the music based on the defined nomenclature? Without an automated system, it will be subjective and error prone. This problem could be maybe tackled on long term with machine learning.

http://musicontology.com/

https://donotpay.com/ should be involved in this case :)

I get what you are trying to explain but this analogy is completely wrong. It would mean that the two subjects are complementary and are often used together which is clearly not the case.

While a JavaScript framework like React can be used to build a company website or CMS, it is often not a good idea to use React to do this.

Off course this is highly context specific and depends on the particular requirements of the "company website". Without a strict definition of what it is, it will be always a discussion about an ambiguous and moving target.

The choice also depends on the technical knowledge and experience. For someone with years of experience with Javascript and really an expert on it, it might be still a smart choice. Or maybe we are not talking about 1 developer but a team of 10 developers etc.

I was just trying to suggest that this is not often the case.

For most use cases, I wouldn't advice actually a JavaScript framework.

The author is seriously suggesting Next.js to build a company website.

In a lot of "website" scenarios, it makes much more sense to choose a CMS like WordPress, Drupal, Django-cms etc.

In case of a custom application development, frameworks like Ruby on Rails, Phoenix/Elixir or Django is usually a smarter choice.

Next.js is not necessarily a good fit for every problem, and it may be more suited to a specific subset of problems than is often assumed.

I just tried it out. This prompt (without any additional description) didn't had a satisfying expected output with Stable Diffusion 1.5.

With some other keywords, it generated some cool looking images but not any where there was a clear transparent horse and visible legs of an astronaut or something.

It is generally just very hard to compose a prompt where multiple subjects interact in a specific instructed way.

This is really cool.

I have quite some wifi issues at home but when I check things like ping everything seems normal.

I'm going to try this out when I have issues in a video meeting and see if there is a correlation.

This explains a lot and a cap sounds great!

Is 9% fee per successful referral not crazy expensive? Or am I missing something here?