Transcript: Thinking a little bit about input types for burrito and kind of pondering what a minimum sort of ingredient for the filling is. So as we've been testing, voice note is one of these, image is another. Image comes with the GPS tag which is sort of a combination of ingredients. All of these come with time stamps as metadata. So it's kind of crucial for making any sense of anything. But that does make me think now, what happens when you combine more of these ingredients at entry time. So in some sense this is a video I suppose, but like I often want to give my photo a specific caption of some sort. And I also want to see the AI-generated caption. So I'm curious if there's a any way of combining the two in a useful way. My voice note is about what I'm seeing, and these two things are connected to a GPS. So I guess there's questions here of implicit or explicit. Do I need to tell the app that these things are connected, or is it just enough to send the minutes back-to-back and have the query layer figure out that these are actually talking about the same moment in time. And it really does start to like raise a question of when I'm looking at something. It's one of the inputs to my thoughts about what I'm saying, but it's not the only one. I mean my mood and my tasks earlier that day and who I'm around, all of these matter too. So if we're going down the logic of I need to explicitly bundle these separate media artifacts together, then how far do you go? A little point, you know, would make more sense to just scan my brain and everything around it. So which seems easier and more in line with the underlying structure of the data. It's just to say if these things went in at similar times and similar pinpoints to GPS locations, then they don't need to be explicitly related. So that's my thinking there. I guess these are some sort of lab notes.
Voice notes for creating burritos can vary in length; they can be long if needed, but sometimes a short description suffices. Despite not always understanding the thought fully, there's an instinct to describe it with high fidelity to AI. Short descriptions can be beneficial as they can connect to other ideas, implying a hypothesis that the connection between ideas can be explored through these descriptions.
The text discusses the concept of using screen recording to capture and organize thoughts, particularly when mapping them out with supportive graphics or diagrams, enhancing the process with features like rich audio and linking possibilities. The author suggests that a system similar to rewind.ai's capture format could be utilized, allowing for full-text search and leveraging metadata from shared Figma files to extract links and possibly map these as concept maps. This method aims to enhance the searchability, filtering, and querying of content, integrating into a platform the author refers to as "burrito dot place." The author contemplates the addition of robust social context to screen recordings, considering them as potential raw input for content understanding, akin to the role of audio, and builds upon themes previously explored in R-Log.
A shared 'brain' is being discussed as a platform for asynchronous voice note conversations where metadata could enhance understanding and visualization of conversational threads. The speaker suggests a focus on DEMO rather than DEC as a fork in the road, believing it better suits the work they've been doing with building prototypes. A group experiment is proposed with four members to delve into how these voice notes can overlap and interconnect, with the idea of marking chapters within responses to clarify dialogue. The concept also touches on the nuances of information retrieval, preferring vector databases over direct text searches, hinting at a similarity to the speaker's initial voice note exchanges with Savannah after meeting on a dating app. Voice communication offers significant advantages as a medium, and there's an idea presented here that its power should extend beyond just live conversations. Current messaging apps are filled with voice notes that are often difficult to search, filter, or respond to, though iMessage now has transcripts, which are generally reliable and useful once you've listened to the original voice note. The ability to refer back to transcribed voice notes can aid in crafting thoughtful responses and engaging in more meaningful discussions. The sender of the message suggests that by embracing this approach to communication, we could enhance our conversations and is curious to see how it will develop.
The speaker is contemplating how to ensure a substrate recognizes the relationship between two related but unlinked entries. They consider whether to trust the system's ability to connect them or address the issue using the Cray layer. The role of metadata is questioned; whether it could enhance the process or complicate it. Ultimately, the speaker is weighing the benefits of a simpler approach against a more complex but precise one.
85.88% similar
The speaker is considering the research question of how to achieve distributed compute, particularly the need for parallelism in executing pipelines and AI agents. They question the potential for building a Directed Acyclic Graph (DAG) that allows for agents to dynamically contribute to it and execute in parallel, emphasizing the need for pipeline development to accommodate this level of complexity. The discussion also touches on the scalability and parallel execution potential of the mixture of experts model, such as GPT-4, and the potential for hierarchical or vector space implementation. The speaker is keen on exploring the level of parallelism achievable through mixture of experts but acknowledges the limited understanding of its full capabilities at this point. They also express curiosity about fine-tuning experts for personal data. The speaker is discussing the data they are generating and the value of the training data for their system, particularly emphasizing the importance of transforming the data to suit their context and actions. They mention meditating and recording their thoughts, which they intend to transform into a bullet point list using an AI model after running it through a pipeline. The individual also discusses making their data publicly accessible and considering using GPT (possibly GPT-3) to post summaries of their thoughts on Twitter. They also ponder the potential of using machine learning models to create a personal Google-like system for individual data. The text discusses using data chunking as a method for generating backlinks and implementing PageRank in an agent system. It mentions steep space models and the continuous updating of internal state during training. It also compares the level of context in transformer models and discusses the idea of transformer as a compression of knowledge in a language. The speaker expresses interest in understanding the concept of decay in relation to memory and its impact on the storage and retrieval of information. They draw parallels between the processing of information in their mind and the functioning of a transformer model, with the long-term memory being likened to a transformer and short-term memory to online processing. They speculate on the potential of augmenting the transformer model with synthetic training data to improve long-term context retention and recall. Additionally, they mention a desire to leverage a state space model to compile a list of movies recommended by friends and contemplate the symbiotic relationship between technology and human sensory inputs in the future. In this passage, the speaker reflects on the relationship between humans and computers, suggesting that a form of symbiosis already exists between the two. They acknowledge the reliance on technology and the interconnectedness of biological and computational intelligence, viewing them as mutually beneficial and likening the relationship to symbiosis in nature. They express a preference for living at the juxtaposition of humans and computers, while acknowledging the potential challenges and the need to address potential risks. Additionally, they mention that their thoughts on this topic have been influenced by their experiences with psychedelics. The speaker discusses the potential increase in computing power over the next five years, mentioning the impact of Moore's Law and advancements in lithography and semiconductors. They refer to the semiconductor roadmap up to 2034, highlighting the shift towards smaller measurements, such as angstroms, for increased transistor density. They emphasize that the nanometer measurements are based on nomenclature rather than actual transistor size, and the challenges in increasing density due to size limitations and cost constraints. The conversation touches on different companies' approaches to transistor density and the role of ASML in pushing lithography boundaries, before concluding with a reference to the high cost and potential decline in revenue for semiconductor production. The speaker discusses the importance of semiconductor manufacturing in the U.S. and China's significant focus in this area. They mention watching videos and reading sub stacks related to semiconductor technology, specifically referencing industry analysts and experts in the field. The speaker expresses enthusiasm for staying updated on developments and offers to share information with the listener. The conversation concludes with a friendly farewell and the possibility of future discussions.
85.83% similar
The author contemplates the process of converting an audio note into a transcript, then summarizing it on their "burrito" page. They express a desire to adjust the summarization voice to better represent themselves on the page. Recognizing that this feature may not have widespread appeal, the author nonetheless sees value in providing users with controls to personalize their "burrito." The concept of allowing users to fine-tune their experience is seen as an intriguing possibility.
85.13% similar
The speaker is reflecting on their experience with making audio burrito posts, noting that it often requires multiple attempts to get into the correct mindset—similar to drafting written posts. They're grappling with the challenge of monologuing without a clear understanding of the audience, as they are aware that at least John and CJ will hear it, but uncertainty about the wider audience affects their ability to communicate effectively. This creates a 'contextual membrane shakiness' as the speaker finds the lack of audience boundaries difficult to navigate, which they recognize may vary among different people. The speaker concludes by deciding to end the current note and start a new one.
84.94% similar
Considering the potential of using Brian Eno's diary as inspirational material, the speaker is pondering on how to reflect and experiment through a year's time. They're impressed by the quick and easy transition from voice note to summary and transcript, finding these tools useful for reflecting on the day. To enhance this reflective process, the speaker contemplates setting up a service to receive a text message summary at day’s end. They also consider the feasibility of sending voice memos directly as an input surface and the possibility of extending this service to friends, acknowledging that it aligns with current developmental efforts.
84.83% similar
The speaker mentions using AI to convert voice notes into tweets or Instagram captions, expressing an interest in conveying their thoughts effectively and using it as a springboard for more thoughtful content. They highlight the potential for a "spiffy remark" on Twitter without needing to use their exact words, indicating a desire for increased flexibility and creativity in their social media posts. The overall focus is on exploring new ways to express their ideas and promote long-form, thoughtful content.