jonfw

I love ASCII diagrams! The fact that I can write a diagram that looks equally wonderful in my terminal via cat as it does rendered on my website is incredible.

A good monospaced font and they can look really sharp!

I will definitely give this tool a shot.

I will also shout out monodraw as a really nice little application for building generic ASCII diagrams- https://monodraw.helftone.com/

Like all other job functions tangential to development- it can be difficult to organize the labor needed to accomplish this within a single team, and it can be difficult to align incentives when the labor is spread across multiple teams.

This gets more and more difficult with modern development practices. Development benefits greatly from fast release cycles and quick iteration- the other job functions do not! QA is certainly included there.

I think that inherent conflict is what is causing developers to increasingly managing their own operations, technical writing, testing, etc.

Coding agents are excellent at solving merge and rebase conflicts according to natural language instruction.

In my experience (this was about 5 years ago mind you) it was no more complex than an arch installation, but with a smaller community and less documentation.

Thank you for the kind words and feedback! Maybe I could have been more clear on the implementation details.

You were on the money with “UI just a reflection of the folders/files on disk that the AI updates.

There is no real UI integration with the AI, other than the UI embedding the chat interface of a coding agent (using an embedded terminal)

The UI just uses a filesystem watcher to keep track of updates to the relevant directories or files.

In my experience, people building APIs w/ python are almost always using frameworks, while people building APIs w/ golang are almost always using stdlib

what if I payed a content marketing expert to craft my blog post and title in such a way that drew attention? Would that be paying for

Many discord servers don't even have voice channels and are just collections of text chat channels.

How do you modify the url exactly?

In theory, sure. In practice? Chinese companies ignore your patent, you waste money suing, it takes a long time.

If you win? Good luck collecting damages from China, and have fun suing the next brand that starts selling the same machine in different plastic

It sounded like he was trying to one shot things when he mentioned he would ask it to fix problems with no luck. It's an approach I've tried before with similar results, so I was sharing an alternative that worked for me. Apologies if it came across as dismissive

For those surveys to mean anything (for or against AI), we'd have to have an effective measure of developer productivity

If you have not started working on a new codebase while adopting AI, it may be harder to realize the gains.

I switched jobs somewhat recently. At my previous job, where I was on the codebase for years, I knew where the changes should be and what they should look like. So I tried to jump directly to implementation with the AI because I didn't need much help planning and the AI got confused and did an awful job.

In a new codebase, where I had no idea how things are structured, I started the process by using AI to understand where the relevant code is, the call hierarchies and side effects, etc.

I have found by using the AI to conduct the initial investigation, it was then very easy to get the AI to generate an effective spec, and then it was relatively easy to get the AI to generate the code to that spec. That flow works much better than trying to one shot implementation

"Company that has been left in the dust claims it was a good business decision"

They have 30 days to disclose trades. Even with that massive limitation, the NANC fund that tracks pelosi does well- https://www.nasdaq.com/market-activity/etf/nanc/advanced-cha...

Tools are more reproducible than prompts w/ instructions to hit apis. They are helpful for agentic workflows that you intend to run multiple times or without supervision.

They aren't worth bothering with for one off tasks or supervised workflows.

The major advantage is that a tool can provide a more opinionated interface to the API then your openAPI definition.If the API is generic, then it may have more verbose output or more complex input then is ideal for the use case. Tools are a good place to bake any opinion in that might make it easier to use for the LLM

There is a Goldilocks effect. Dependency just came out a few minutes ago? There is no time for the community to catch the vulnerability, no real coverage from dependency scans, and it's a risk. Dependency came out a few months ago? It likely has a large number of known vulns

Distribution channels to orgs or countries that don't buy from nvidia. Ability to cut discounts w/o discounting the Nvidia brand

> ChatGPT going to stop in the middle of it and go "Hey, you know what sounds good right now? A nice bowl of soup..."

Google/facebook do that today, because the content they're showing is created pre-ad, and the ads have to be injected after the fact.

With AI- the content is being generated in the same place that the ads are being injected, which allows us to be much more subtle about it.

How much do you think a car company would pay for to put special training weight on their marketing materials? I would guess big money

Investors are non fungible. Partnering w/ your supplier can provide benefits such as preferential treatment and early access to hardware that are non-monetary but enormously valuable.

> The problem is that no 3rd party seems willing to just invest cash

It seems to me that these AI companies are seeing no shortage of cash investments either.

Access to infrastructure is a core part of the AI business.

If you are a startup, and you want to buy AI infrastructure, you would typically sell equity to a VC and use that money to buy infrastructure.

If the AI infrastructure vendor would offer to buy that equity, rather than a VC, that’s preferable because you may get ongoing special treatment from the vendor once they have a stake. So it’ a great deal for the startup.

For the infra vendor- you can likely get a good deal on the investment, you get some exposure to higher upside, and it may be useful to manage your cash flow.

Seems like strategic high value moves to me. It’s no wonder the market likes it

These CLIs are just thin request clients a la curl. The code execution time is peanuts in comparison to the request latency

Are you implying that 1B investment gives Nvidia control over Nokia procurement?

I don't think anybody accepts school shootings, and anybody accusing half of the population of "accepting" this obvious problem is likely making a bad faith argument attempting to paint their political opposition in a bad light.

MCP provides a convenient packaging for tools, and generally workflows, for LLM clients.

You can debate all day whether bringing your own tools is a good thing vs giving the LLM a generic shell tool and an API doc and letting it run curls. I like tools because it brings reproducibility.

MCP is really just a json RPC spec. json RPC can take place over a variety of transports and under a variety of auth mechanisms- MCP doesn't need to spec a transport or auth mechanism.

I totally agree with everybody that most MCP clients are half assed and remote MCP is not well supported, but that's a business problem

Every LLM tool today either runs locally (cursor, zed, IDEs, etc.) so can run MCP servers as local processses w/ no auth, or is run by an LLM provider where interoperability is not a business priority. So the remote MCP story has not been fleshed out

> the freedom to be able to choose to engorge yourself on twinkies is maybe a good thing

In isolation, freedom is wonderful. In practice, maybe not? If a member of the star trek crew abused twinkies to the extent that they were lethargic or unhealthy, it would negatively impact the mission.

There is some compromise between freedoms and social resposibility

This is a very valuable use case for me personally. I frequently have the problem that I change things on the filesystem in a separate tab and the agent context gets out of sync. It fails on subsequent edits, often tries to reverse the changes that I made, and many times I have to copy/paste the command I ran and it's output back into the agent window.

Your complaint is likely a product design decision rather than a engineering capacity prioritization one. As you've noted the fix is pretty trivial. I imagine that some designer or product person is intentionally holding this back for one reason or another

It's less about the punctuation used, and more about the necessity of the punctuation used.

In the sentence you provided, you make a series of points, link them together, and provide examples. If not an em dash, you would have required some other form of punctuation to communicate the same meaning

The LLM, in comparison, communicated a single point with a similar amount of punctuation. If not an em dash- it could have used no punctuation at all.

Low income work just doesn't make financial sense for many americans. https://www.cato.org/commentary/welfare-better-deal-work#