HN user

keizo

273 karma

I make stuff. Mostly canoes. Co-founder Kamanu Composites. Design & manufacturing in Hawaii. Weekend warrior programmer http://grugnotes.com

http://www.kamanucomposites.com

Posts9
Comments143
View on HN

interesting read albeit over my head, but i spent half of yesterday comparing Gemini Live (websockets) vs gpt-realtime-2 and while gpt is super good, seemingly more robust. Gemini connects faster.

Interesting! Like less overall engagement? Was it fully mixed or in distinct columns. I feel like it makes a difference.

I would at least interact with the new content if it was on the page. Vs almost never now. I assume people that are active on the new page must be 1% of users or mostly those directly involved with the story.

Anyway, thanks for the response and keep the place sane.

This. Lately for some harder problems, I'll open two sessions. One writes a draft spec to a file. Then in the 2nd, i ask it to analyze, critique, etc. Often feed that response back to the first. A few ping pongs later, get a pretty polished plan. Open a new session to execute.

gemini flash and groq are pretty fast, and that part is streamable. curiosity got the best of me so i had claude code write a quick test. given this test is simply is 20 requests, with 1 second delay between requests ran once. so take with a grain of salt, but interesting still. Extra half second in a search is super noticeable so google looking like a reasonable improvement.

  OpenAI Statistics:

  - Average: 0.360 seconds
  - Median: 0.292 seconds
  - Min: 0.211 seconds
  - Max: 0.779 seconds
  - Std Dev: 0.172 seconds

  Google Gemini Statistics:

  - Average: 0.304 seconds
  - Median: 0.273 seconds
  - Min: 0.250 seconds
  - Max: 0.445 seconds
  - Std Dev: 0.066 seconds

  The key insights from these numbers:
  - Google has much lower standard deviation (0.066 vs 0.172), meaning more consistent/predictable performance
  - Google's worst-case (max) is much better than OpenAI's (0.445s vs 0.779s)
  - OpenAI had a slightly better best-case (min) performance (0.211s vs 0.250s)
  - Google's performance is more tightly clustered around its average, while OpenAI has more variability

i like this. complexity bad, delete it! Most pkm, tools for thought, second brain apps confuse me. I drank the coolade with roam research but it drove me kinda nuts I've spent almost 3 years making my own tool. I mostly use it as a paste-bin, todos, lists -- and for the only thing i would never delete, voice notes on funny sayings or interactions with my 3 y/o daughter. my project is over at https://grugnotes.com if anyone else fits the anti note app vibe i'm kinda leaning into.

I've been working on a roam research replacement for myself for almost 3 years now. And initially, my main thought with self-organizing notes was to tag things automatically. And use ai for retrieval. Generally I avoid ai from 'touching' your notes. But 'breaking apart' a single note with ai is something i think about doing. But haven't done yet. Either way, i always want human in the loop. Given this is hn i'm sure you're working on your own solution, but feel free to check out my project! :) https://grugnotes.com

GCP Outage 1 year ago

Yup, intermittent db connection issues and cloud storage problems.

The Fall of Roam 1 year ago

Roam got me so frustrated I spent 3 years and likely many more making an html/htmx ai first replacement. It’s still pretty raw, but it does everything I want and more. https://grugnotes.com

didn't know that about roam research. I was a user, but also that app convinced me that front-end went in the wrong direction for a decade...

Rocicorp Zero Sync, instantdb, linear app like trend is great -- sync will be big. I hope a lot of the spa slop gets fixed!

I really hope someone displaces quickbooks -- I was hoping someone would build one on top of Rocicorp Zero Sync!

This looks like a great start. More pricing transparency would be good. A lot of smb, mine included, might not fit in starter, but also can't/won't spend more than we're spending with qb.

I implemented differential sync (https://neil.fraser.name/writing/sync/) mostly because I couldn’t understand anything else and seemed simplest in my grugnotes.com app -- and while the app is pretty janky and not fully real-time, it does get your example merge right regardless of who comes back online first. In the case the deletion comes online first, the 'colour' version is thrown out and not saved in edit history. I’m sure there’s a lot more wrong with it, have no idea what would happen with more than two users, but for my case I’m happy with it. :)

Awesome, thanks for going back to try, and great feedback!

1. It's been fun to implement. Always room for improvement, formatting results is up there. But surprisingly good with simple keywords and rag. There is some fun NLP in there, like filtering by date or navigating to that day if you just type a date. And doing rag only for questions.

2. Yep, high priority -- for me it's the small things like bringing focus to the text area by default, etc.

3. Absolutely. I'm into density and packing as much onto the screen as possible, so I ultimately have to click around as little as possible. But you're right; it's confusing. I occupy this weird space, being super nerdy but also wanting ridiculous simplicity, though not in a super consumer way. Wanting to stick to plain text, for example, markdown, etc, but also targeting smb and tradespeople is an odd path. My dad, of all people, had the same feedback. :)

4. Search/ask is for looking in your notes. Find/create is more like: I know this prompt/entity exists or add it. It's mostly a direct search for the 'prompt' names or entities and a shortcut to adding blocks to your day. For example, if you're having a meeting with a person or company and don't want to clutter your daily notes. Or even my changelog, I hit ctrl-k 2x, type chan, hit tab to get 'changelog', enter, and type away. It's also the only current way to find out what prompts are in your account. Still, very worth exploring if there's a clean way to get rid of find/create. I suppose there could just be an 'add block' button, and then you can pick your prompt on creation. Simpler UI, but I'm also not convinced it's entirely better. Probably worth trying, I'm sure there'll be a time I can circle back to iterate on it. Development is mostly dictated by what's most annoying at this point.

And regarding the intermittent errors during this hn post, I think it was my db max connection django setting causing issues. Reset back to the default of 0, and it's been good so far.