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

LLMs being LLMs.

Reading this makes me a feel a little more secure in my job.

They have a long way to go.


Certainly!


it would be great to have a raw json file or something of this list so we can just start blocking the requests instead of using robots.txt. do they really respect it? probably not.


Most of these bots already publish their IP ranges, so it would be just a simple firewall rule.

For chatgpt user:

https://platform.openai.com/docs/plugins/bot

And for gtpbot:

https://platform.openai.com/docs/gptbot https://openai.com/gptbot.json


Lets make ourselves redundant.

At least I won't have to relearn JS every 2 weeks.


Don't worry, in 5 years it will just be another layer in the JS dev stack, so then not only the version of npm, dependencies, docker, and 3 API keys have to match but you'll also need the correct commit of the 20GB LLM so the build doesn't fail. All for the sake of simplicity and increased productivity, of course.


We’ll have dependency files for LLMs in our mixture of experts config and we’ll get paged at 2am to update a version of an LLM because there’s a new social engineering CVE making it vulnerable to disclosing secrets.


let chatgpt fix our dependency issues


required dependency: leftpad-ai-gpt4


These little demos of toy UIs are cute but humans can still hold the context of a whole codebase in their heads.

We're not quite at the level (yet) of feeding a whole codebase to an LLM and making it add features or make changes while understanding the big picture of the problem being solved, being consistent with the design principles and coding style of the overall existing codebase. And I'm not even talking about creating complex UIs where performance matters

Another issue I've ran into a lot is staleness of the knowledge the LLM was trained on, a lot of libraries and frameworks get really frequent, quite often breaking changes, and LLMs have a cutoff date.

Try using ChatGPT for something like Godot's GDScript, it will always try to use old Godot v3 style scripting because that's what it's been taught, and the whole documentation for Godot v4 is not something small enough to just fit into context

Maybe this would be a better fit for some agent type workflow where it can decide what to lookup from the documentation and then retrieve it, but it also needs to know and decide what to look up and how. There is still a lot to figure out


> We're not quite at the level (yet) of feeding a whole codebase to an LLM and making it add features or make changes while understanding the big picture of the problem being solved

We are one step away from that. All we need is a more advanced form of fine tuning.


lol, this magical thinking; also not AI, it's ML


ML is a kind of AI


nope


Sorry to do this to you, but you're actually incorrect.


sorry to dither your ego, but it's not; prove it, but i guess lying to yourself is easier than accepting you're wrong


If there is one good thing going to come out of EVEN the fairly clumsy LLMs, it is that we probably can forget about this js frontend framework crap and can focus on work instead of figuring out what the taste-du-jour is. It won't matter; the LLM will translate it.


Probably not. Instead we’ll be stuck using the old frameworks driver because LLMs won’t be able to learn anything new without years of stack overflow examples.


Already I bet stack overflow usage has gone down because LLMs can help fixing bugs.

I think the documentation, open source code, and code examples are more important training sets.

And in. my experience the code is often maybe 95% correct, so there will be a greater premium on expert developers who can spot and fix bugs (with the aid of LLMs, since stack overflow will no longer have any answers since no one will use it having moved to LLMs)


Stackoverflow has less that 50% of the traffic now than it used to a few years ago.


I 10000% agree.


More like - let's give ourselves better tools. The people who hire us won't be able to build this stuff themselves no matter how good an AI you give them. The only developers who will lose their job are ones who fail to embrace AI.


You don't have to now too


Much like cockroaches JS will survive a nuclear explosion


China: Look at our internet, users only do what we want.

UK: Hold my beer, I'm going to do this for the whole internet.

---

All the companies that said they would stop operating in the UK if this bill was passed, need to now do so. There are provision in place that say companies won't be forced to do so unless technologically possible. But I think we all know how that plays out.


ffs, who is thinking of this shit. personally i don't own any and now i certainly won't. Smart homes are a great example of a good idea with a shitty implementation.


I've been looking for something like this solution for a while. Never thought of using ublock filters. Cheers


If you went back to the office, then would you rather an open planned office or your own cubicle? If its the latter, your own cubicle, then why the hell are you going back to the office?

Open planned offices are a pain the arse for distractions.

I for one couldn't imagine going back, far more productive WFH.


Its 2023, how have we come so far with this nonsense. I would have thought Firefox would be king of the browsers by now. But nope, people love their data being sucked up and used by the big G.


Mozilla focused on the wrong things for a few years


It doesn't really matter what Mozilla did. They never had the budget to compete with Google on browser advertising anyway.

If you look at the market share progression of the old opera and mozilla vs. internet explorer, and you compare it to the stellar rise of Chrome when it came out, it's really obvious that Chrome's success is due mostly to advertising.


Blows my mind that people think that there’s some magical way that Firefox could have competed. Chrome had ads on google.com, like the home page. No one except Google gets those ads. 90-second prime time tv spots. And they also scattered nudges throughout their entire website. Google ran a relentless ad campaign to get people to switch, and no browser on Earth was able to compete with that.


I agree Firefox never had a chance, but there is losing a battle and there is diving head first onto the floor. Mozilla played the few cards it had not that well.


Mozilla dug their own grave, to be fair.


Firefox is just less usable. It's slower and a lot of config options are missing. Stuff that is common in enterprises and has buttons in Chrome (e.g. "add the intranet search to your browser!") is hard in Firefox (you have to write and serve a specific XML file somewhere).

Also it has a big incredible redesign every 2 weeks, as far as I can tell from HN, while Chrome's UI is mostly boring and stable.



These are not solutions, these are workarounds. Adding a bookmark with a keyword, because the internal UI isn't good enough? Come on.


Yeah. For me, the main feature that has me use chrome instead is that it can remember my credit card details. I don't understand how it can be that firefox doesn't do this, it's so annoying.


Firefox actually also does it, maybe limited based on location.

For instance here they announce it for UK, France, Germany (in 2022) and say it was available in the US/Canada since 2018: https://blog.mozilla.org/en/products/firefox/firefox-news/fi...


That's good news I guess. Any idea about why it's geographically limited? I'm not in one of those countries, unfortunately.


Firefox has a "saved credit cards" feature. I'm not using it, so maybe I'm unaware of its limitations?


Firefox does do this.


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

Search: