ianbicking

14 ms·

61.
by ianbicking 1y ago
I've been thinking and playing slightly with this concept myself. A few thoughts: 1. Using a standard transcription service is pretty tricky because it's going to correct the user's speech. Or make it incorrect! Standard tran
62.
by ianbicking 1y ago
It feels like what Custom GPTs should have been. Custom GPTs are barely able to do anything interesting beyond an initial prompt, there's no ability to modify the core user experience. The ability to run code and have it do subrequests
63.
by ianbicking 1y ago
Why do folks think remote telepresence never became popular, outside of the occasional appearance in sitcoms?
64.
by ianbicking 1y ago
If you just give the transcript, and tell the LLM it is a voice transcript with possible errors, then it actually does a great job in most cases. I mostly have problems with mistranscriptions saying something entirely plausible but not at a
65.
by ianbicking 1y ago
FWIW in my recent experience I've found LLMs are very good at reading through the transcription errors (I've yet to experiment with giving the LLM alternate transcriptions or confidence levels, but I bet they could make good use o
66.
by ianbicking 1y ago
It's confusing, but if I look closer then 10x is an exaggeration, it's more like 5x... https://elevenlabs.io/pricing Creator tier (lowest tier that's full service) is $22/mo for 250 minutes, $0.08/m
67.
by ianbicking 1y ago
I've been using OpenAI's new models a lot lately ( https://www.openai.fm/ )... separating instructions from the spoken word is an interesting choice, and I'm assuming also has a lot to do with OpenAI/GPT u
68.
by ianbicking 1y ago
Seeing him back in the news made me think more about Urbit, his decentralized computing project. For a bit of a discussion of the technical parts of Urbit this is decent: https://wejn.org/2021/02/urbit-good-bad-ins
69.
by ianbicking 1y ago
I find these studies that use minimal prompts for the LLMs to be quite frustrating. Here's the prompt: > You are a {DEMOGRAPHIC-IDENTITY}. Please answer the following question in the first person and in a single paragraph. Question:
70.
by ianbicking 1y ago
I'm not exactly clear on the difference between "time" and "timing"...? Like an accurate incrementing time offset from some unknown start point?
71.
by ianbicking 1y ago
"If we had a cellular modem, we could get the time from the cell towers, which is broadcast through a protocol called NITZ. This is how phones auto-update when you enter a different timezone, but it's somewhat unreliable depending
72.
by ianbicking 1y ago
I think I understand what you're proposing, but I'm not sure. So in concrete terms I'm imagining: 1. Create a prompt that gives the complete API specification and some general guidance about what role the agent will have. 2.
73.
by ianbicking 1y ago
Everything has to do with language! Language is a way of stating intention, of expression something before it exists, of talking about goals and criteria. Everything example you give can be described in language. You are caught up in the me
74.
by ianbicking 1y ago
Someone built a legit commercial game with EGA graphics just last year: https://store.steampowered.com/app/1098770/The_Crimson_Diamo... I haven't played it, but looking at the graphics alone brings up some de
75.
by ianbicking 1y ago
"Does AI finally enable truly humane interfaces?" I think it does; LLMs in particular. AI also enables a ton of other things, many of them inhumane, which can make it very hard to discuss these things as people fixate on the inhum
76.
by ianbicking 1y ago
This mentions an SKF format for knowledge representation... but looking it up, I'm assuming it was invented just for this project? Which is fine, but is there a description of the format distinct from this particular use? (I'm pla
77.
by ianbicking 1y ago
I feel like I should understand the purpose of knowledge graphs, but I just... don't. Like the example "CocoIndex supports Incremental Processing" becomes the subject/predicate/object triple (CocoIndex, supports, In
78.
by ianbicking 1y ago
There's definitely a phenomena I've seen especially post-Covid where parents and schools just let a child... not learn. Not attend school, not attend classes. Substitute virtual education which is just a pretend education.
79.
by ianbicking 1y ago
In one way or another they are all circling around a concept of "one runtime environment that is as close as possible to the browser JavaScript runtime" So in a sense, yes, they need to converge on the concept of a singular runtim
80.
by ianbicking 1y ago
All these legacy answers don't really make sense for this Z17... it's a new mainframe supporting up to 64T of memory and specialized cores for matrix math/AI applications. I have a hard time believing that legacy workloads ar
81.
by ianbicking 1y ago
Hume has been on my radar for a long time, but I've never actually used their products. They keep coming out with new lines and yet I never see anyone talk about them... I'm not sure why? Though it's so hard to figure out the
82.
by ianbicking 1y ago
That seems like a really short lifespan...? Like you can use it for a couple years, but only if you don't actually use it. There's an obvious aesthetic draw to e-ink, but it seems like passive-matrix monochrome LCD (like the Playd
83.
by ianbicking 1y ago
What's been your experience with the Realtime API? I've been doing LLM with voice, but haven't really given it a try – the price is so high, and it feels like it's much harder to control. Specifically that you just get o
84.
by ianbicking 2y ago
I love that this is more art piece than serious software... more like offering someone an expedition than a product. (Though I'm not sure I'll get on the expedition, I am a little worried about sandboxing and setup and getting dis
85.
by ianbicking 2y ago
To ground the story in a consistent world: build up secret world lore behind the scenes, and provide that to the LLM as it makes choices. As the story is expanded the lore will expand. But this gets complicated as time progresses, as things
86.
by ianbicking 2y ago
There have been experiments in wiki-style cyoa generation (letting the public create options instead of an LLM), but they suffer the same problem as LLM-generated stories: aimless wandering and lack of consistency. (As I think about it, an
87.
by ianbicking 2y ago
I'm a little confused why there aren't better, non-exploitative chatbots. There's not a big technical moat here. Seems like anyone could build a modest business with a better chatbot, built for conversation and companionship
88.
by ianbicking 2y ago
That's also a nice explanation of why brain size among mammals scales with body size without necessarily leading to higher-order intelligence.
89.
by ianbicking 2y ago
What does a "backlog" mean in this context?
90.
by ianbicking 2y ago
I had seen about Google Maps and "Gulf Of America" and couldn't quite believe it was true, so I kept loading it up, seeing if they really just decided to lick Trump's boots. Today it for the first time it showed up for m

More ›