_doctor_love

Btw, I'm going to try an experiment and respond to your GP comment with a non-moderation reply, just because I find the topic interesting, and also in the hope of communicating what I said above: the thoughtful component of the conversation is fine. (No obligation to reply of course.)

I had logged off to chill out a bit so missed that you wrote this. Sorry to admit my ignorance but I don't know what this means? I appreciate your engaging with me fwiw.

I'm sucked back into work for most of the rest of the day so if you don't see a reply from me that's why, I'm not off stewing in a huff.

Clearly we have different definitions of what constitutes an attack, that said, I know you're fighting an eternal uphill battle on HN so I'll rest my case.

I will say that I find some of this kind of moderation very confusing. I feel that I often see folks going way harder than anything I've said today and there's no moderation hand-slaps visible. Possibly the sensitivity is greater here since it's a Launch HN thread, so I appreciate that. But please take my feedback for what it's worth and reflect on it a bit.

Sorry I don't follow what argument you're making?

Funny enough, calculators went through this exact same thing when they came out. "If the calculator can do math for the students, will they still learn?"

Please see my other reply to you on this.

You bet. Like I wrote to 'dang I'm really not trying to hate on what you're doing, just offering my very candid take on products like this. I care very much about education and I'm happy to see someone trying to do something about it. In that sense I applaud your efforts and hope that Bloomy leads to some good outcomes for somebody. I simply feel that more AI is not the answer.

your previous comment was a personal attack as well:

Brother, I meant what I wrote.

That said, I'll cool off for a bit. I'm not here to rain on anyone's parade but as you said, products like this stir extremely strong feelings on my end. I find the framing offensive, exploitative, and dehumanizing.

I would want to solve the problem way upstream. Obviously I want children to receive education.

I would also say that looking at the problem like it's teacher headcount that's the issue is assuming a specific education model. If there is only 1 teacher and 300 students, you can't meaningfully do a lecture-style one-way transmission of information. That's outdated anyway and not what the students need.

[flagged]

[flagged]

That's how it works though. The moment we have "AI" and see something working, it immediately ceases to be magic because "it's just a program after all." Aligning on a true definition of Artificial Intelligence is a very vexing problem.

What evidence or product behavior would you need to trust something like this with a student?

I would need the product to not exist. I don't want this anywhere near a child. I want humans instructing humans, teaching empathy, connection, and - most importantly - learning to learn.

Anything short of that is just building Brave New World. Electroshock the kids if they reach for forbidden knowledge.

I'm sorry to be so negative since I'm sure you worked hard on this but I find all products in this category to be 100% reprehensible.

"Computer-assisted learning. What an insult, to have the computer teach the human."

- Russ Ackoff

Late to the party but wanted to say thank you and contribute an idea - let the user community of this tool henceforth be known as juggalos.

Stuff like this is what makes me not worry about AI overtaking humanity. The human brain + mind is so sophisticated, we have no idea what untapped potential still remains.

And anecdotally, this headline feels like it's confirming something already well-known. I once did a presentation for a software team visiting the US from South Korea and their translator was real-time translating my words into Korean for them. The translator had an earpiece in and so did the clients. Once I adjusted to it, it became very natural to present and interact this way (but it was certainly weird at first).

I like the podcast features still for getting an overview or orientation on a new topic where I know I need a big context dump before I can begin to do anything under my own speed.

I have tried the "walk and talk" pattern with AI, and it's okay, but I find that it can be kind of janky if the network connection isn't very good. And I find it very frustrating to use still in terms of how the interrupt feature works. For example, if I'm out hiking and there's even a little bit of wind, the voice mode of ChatGPT and Gemini both will think that I'm done talking and begin responding.

In fairness to the AI vendors, I imagine that this is a very difficult thing to get right. Humans have a very natural sense of flow in conversation, and knowing when a speaker is done talking and it's appropriate to begin responding.

They exist but their defining characteristic seems to be that they are not well-known and generally much less wealthy than celebrity CEOs.

Similar to the music world, the better you are, usually the more obscure you are as well. (e.g., Allan Holdsworth is a name known to most pros but the average Jack or Jill have no idea who he is or why he's considered important.)

What do you do for work if it's not writing code? I'm always fascinated by the non-paid-professional-programmers who frequent HN.

I am sad to say it but I am not sure it is possible in the short term. It seems that there is a taste for conflict in the air and we are cruising towards another world-level spasm of violence before we can return to some semblance of sanity.

This is partially an on-ramp for young people. No experience? Not sure where to start? Buy this gadget! Hook it up to your machine then take lessons on how to use it. All in the OpenAI ecosystem, of course.

Best outcome for OpenAI is that this becomes a status symbol / cool shiny thing that "leet" devs have.

For someone with a lot of experience already, this looks semi-retarded. For a newbie / newcomer it looks like someone finally thought of them.

That's the key, a physical object is much more easily made part of a devs identity.

Projecting a hacker image has become very fashion based. e.g., some devs love their clickity-clackity keyboards with LEDs, others have those all-blacks ones with nothing on the keys.

Hey, as long as the masses are looking to the left and the right for their enemies, rather than up, that's what's desired.

Reminds me of an observation made when they removed Maus from a library in Tennessee I think it was: "these people want a kinder gentler Holocaust."

It's totally idiotic the state we have reached, with an absolutely equal number of stupid people on the left and the right.

Wasn't it Jeff Skilling who once waxed poetical about hypothetical future value?

I think your attitude is pretty lazy. It's a low-effort answer to say "skill issue" and then walk away feeling superior. Much harder to lean in to the other person, understand them, then help guide them onto the path. When you do this, you cause them to accelerate tremendously.

IMHO as you get more senior in software, after a while the only interesting metric becomes: are you raising the level of the people around you?

EDIT: looking at your Github seems like you are still in university. So I'll say "age and experience issue" on your end ;)

The key is to remove degrees of freedom from the agent. IMHO this is the really hard work of agentic engineering. You need to strongly constrain what the agent can even do in the first place to force it into a certain area.

That runs the danger of putting the model out of distribution, of course, so it takes some experimenting.

I'm working on a Ruby project where - on purpose - I'm disallowing any frameworks other than what's standard Ruby. So no Rails or Sinatra, just the inbuilt Webrick web server. Initially it was hard to keep the model on the rails but as I've learned more it gets easier. That said, it's clear to me that the model pulls very strongly in familiar directions in terms of how it writes code, i.e. a lot of what gets written still ends up looking "Railsy" even though my learning project doesn't use it.

"I just want a deluge. Just one little deluge. Is that too much to ask?"

Yes, we all remember the Vasa, and understand how it applies to any untested overengineered idea.

We clearly do not given that this story repeats over and over again.

The deeper issue was that no-one had the ability to push back on the king's demands. When the king rules by divine right, pushing back on the stakeholder is pushing back on God. Back in the day that was a great way to expedite the process of your head being separated from your torso.

Seriously credit to you on that ASCII art. I grew up sailing and you captured it beautifully. Well done.

Agree with the observation others have made. The only true solve if a specific model version is critical to your application or workflow, you need to host the model yourself so you have control over it. You don't want to be stuck getting rug-pulled by a model provider.

And as another commenter pointed out - in particular for Google of all companies - expect that the rug pull can and will happen. They're not known for keep anything around for very long.