Hacker Newsnew | past | comments | ask | show | jobs | submit | sghiassy's commentslogin

In Portland Oregon back in 2000s we had - Incredible Universe - CompUSA - Fry’s (later on)

I’d beg my dad to drive me to them on a Friday night. Great times!


When I would go there for many years I wondered what the hell that giant globe was for!

Agree

But would you share your salary here right now?

If not, is it purely cultural? Or are there other motivations?


I think it's purely cultural, and totally to our (meaning workers') detriment. See my above comment about working in Germany.

LLMs can understand ASCII diagrams

LLMs nowadays can understand png diagrams too.

But not all CLIs handle pngs as easily as a text diagram. Syncing images is also multiple times slower.

png diagrams should never be the default.


They can’t update it though. In docs it makes sense to use that as a basis and have the Llm update it when needed

Mermaid diagrams are even better because you don't waste characters on the visual representation but rather the relationships between them. It's the difference between

    graph TD
            User -->|Enters Credentials| Frontend[React App]
            Frontend -->|POST /auth| API[NodeJS Service]
            API -->|Query| DB[(PostgreSQL)]
            API --x|Invalid| Frontend
            DB -->|User Object| API
            API -->|JWT| Frontend
and

    +-------+           +-------------+           +---------+
    |  User |           | React App   |           | NodeJS  |
    +-------+           +-------------+           +---------+
        |                      |                       |
        |  Enters Creds        |       POST /auth      |
        |--------------------->|---------------------->|
        |                      |                       |
        |      Invalid         |    <-- [X] Error -----|
        |<---------------------|                       |
        |                      |       Query DB        |
        |                      |---------------------->| [ DB ]
Plus while an LLM can understand relationships via pure ASCII or an image, it's just easier to give it the relationship data directly.

But the point is to have something easy to read both for humans and LLM, no?

It’s harder to read mermaid in a terminal or a markdown file…


Mermaid diagrams automatically render on Markdown and IDE chat windows as in VSCode or Cursor. So you get the best of both worlds, a graph you can look at a ND manipulate with the mouse but also in a format LLMs can read.

Ah thanks I didn’t know that…

more tokens, less reliable, dont work in all CLI agent harnesses

I agree with your point, that you need to find something that works for you.

I have ADHD and use the start-from-zero or as I would call it Inbox-Zero method personally


FWIW, I do aim for inbox-zero for email, and similar for chat apps (Slack/Teams). Otherwise it piles up and gets overwhelming. I'm referring more to things like - "only the exact thing you're currently working on open" part. I agree systems are needed. For me it's Obsidian for notes, inbox zero, and OneTab extension to allow me to remove tabs without fear of "losing" them completely. I've learned that it's also a trap to over-complicate my system, even something like Todoist which is fairly minimal was semi-problematic, although I may come back to it - just using manual TODO checklists in Obsidian with a small table that pulls them all into a single dashboard file for reference.

I always wonder what type of moat systems / business like these have

edit: referring to Anthropic and the like


Subsidized plans that are only for their Agent (Claude Code). Fine tuning their models to work best with their agent. But it's not much of a moat once every leading model is great at tool calling.


Capital, both social and economic.

Also data, see https://news.ycombinator.com/item?id=46637328


The only moat in all of this is capital.


Is it a moat if new start ups avoid competing in the space because there is inherently no moat?


Its open source. Where does it say he wants to monetise it?


None, basically.


I do think Claude Code as a tool gave Anthropic some advantages over others. They have plan mode, todolist, askUserQuestion tools, hooks, etc., which greatly extend Opus's capabilities. Agree that others (Codex, Cursor) also quickly copy these features, but this is the nature of the race, and Anthropic has to keep innovating to maintain its edge over others


The biggest advantage by far is the data they collect along the way. Data that can be bucketed to real devs and signals extracted from this can be top tier. All that data + signals + whatever else they cook can be re-added in the training corpus and the models re-trained / version++ on the new set. Rinse and repeat.

(this is also why all the labs, including some chinese ones, are subsidising / metoo-ing coding agents)


(I work at Cursor) We have all these! Plan mode with a GUI + ability to edit plans inline. Todos. A tool for asking the user questions, which will be automatically called or you can manually ask for it. Hooks. And you can use Opus or any other models with these.


N00b Question - how do you measure performance for AI agents like the way they did in this article? Are there frameworks to support this type of work?


That’s not a zero-sum game.

Pivoting to zero-planning, would also have a basket of flaws.


Is planning, like deciding how to position your troops in battle, doing the thing?


Planning is doing the planning thing, but it is not doing the battle thing.


And running the marathon is just running the marathon? I disagree. Big part of running the marathon is in the preparation. Weeks after weeks of training and not skipping a single session. The marathon itself is the tip of the iceberg; important but not the whole "thing".


There are some things that you just can't do without preparation. But never mistake the preparation for doing the thing. You can be "getting in shape for a marathon" forever without ever running a marathon.


It depends on your thing. If the marathon was just the motivation, your thing is running... if the marathon was the bucketlist item, it is the thing.


No matter how much preparation and training one does, if they haven’t run the marathon, they haven’t run the marathon.


Right, but conversely you can’t run the marathon unless you train. You can’t skip straight to doing the thing.


But both are doing the winning thing, which is more valuable than just the battle thing. Unless you do it just for fun and don't mind the result.


I don't know anything about planning and battle.

But as a metaphor for other creative pursuits, my experience is that most of the time when people are "planning" or working on other things that they like to believe will help them do the thing... they are really just avoiding doing the thing.

People spend years doing "world-building" and writing character backgrounds and never write the damn book. Aspiring musicians spend thousands collecting instruments and never make a song.

As you say, if it's just for fun, that's all fine. But if the satisfaction you want comes from the result of the thing, you have to do the thing.


The YouTube app on iOS is superior to the Android app for one


A YouTube client that can't AdBlock and SponsorBlock automatically is strictly worse.


Is there clients that block? I just use Brave browser


NewPipe, but it is an Android exclusive.


PipePipe and Revanced both do both perfectly.


This used to be true, but really is not anymore.


Also, I wasn't aiming at the official Youtube app, but at PipePipe etc. The great alternative Youtube clients Android has.


LLMs help me read code 10x faster - I’ll take the win and say thanks


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: