Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Codex for almost everything

openai.com

Post reply on HN
Print view View original

401–410 of 600 posts

Re: Codex for almost everything

#401

Post by byzantinegene »

solenoid0937 wrote: post #346
luddit3 wrote: post #230

Earlier quoted context omitted.

So Anthropic degraded their product. OAI updated their product to meet for exceeded Anthropic old product. This is normal behavior and not a cause for such a hyperbolic response.

There is good competition and bad competition. Pricing your product unsustainably vs a competitor to gain market share is regarded as "bad competition" and has historically been seen as anticompetitive. It does not benefit the consumer in the long run, because the goal is to use your increased funding or cash reserve to wipe your competition out of the market, decreasing competition in the long term. Then, once your…

you're right but for now it doesn't matter if both competitors are running on infinite vc money, we as consumers benefit from it. it only matters if they cause negative externalities in the meantime

Re: Codex for almost everything

#402

Post by yoyohello13 »

rommelsLegacy wrote: post #353

I am quite worried that people are continuing to use OpenAIs offerings just because it works. Everyone here seem to gloss over the fact that this is a project funded by Peter Thiel. Thousands of morslity posts, complaints about ICE, Tump etcand yet you all choose to use a tool created and funded by the same person enabling this dictatorial machine. I am speechless everytime I see posts like this and the comments foll…

If you’re expecting morality from the HN crowd they will disappoint you every time. Most of the people here wish they could be as ruthless and successful as someone like Sam Altman.

Re: Codex for almost everything

#403

Post by badrequest »

someotherperson wrote: post #224
solenoid0937 wrote: post #218

Codex is HN's darling now because Anthropic lowered rate limits for individuals due to compute constraints. OAI has so few enterprise users they can afford to subsidize compute for this group a lot more than Anthropic. Eventually once they have more users they'll do the same thing as Anthropic, of course. It's all a transparent PR play and it's kind of absurd to see the X/HN crowd fall for it hook, line, and sinker.

Competition is bad? Who cares - let the big players subsidize and compete between each other. That's what we want. We want strong models at a low price, and we'll hype up whoever is doing it. Simultaneously, we also hype up the open models that are catching up. That are significantly more discounted, that also put pressure on the big players and keep them in check. People aren't falling for PR; people are encouraging…

It's hilarious how much this post reads as drafted by an LLM. The emdash, "it's not X, it's Y" framing, incredible.

Re: Codex for almost everything

#404

Post by sofixa »

someotherperson wrote: post #281
watwut wrote: post #258

Earlier quoted context omitted.

Big players subsidizing is what kills medium and small players which then kills competition. What follows is monopoly. Big players operating at loss to distort the market is not a good thing overall.

The medium and small players are literally just distilling the larger models. It's not the smaller players spending billions on training data.

No, the medium and small players are the Mistals, DeepSeek and H Company of the world, with their own models using quirky optimisation techniques to be able to compete.

Re: Codex for almost everything

#405

Post by obrajesse »

moomin wrote: post #316

Wait, did they just send out a press release boasting that they’re bundling Jesse Vincent’s Superpowers?!

They did! I didn't actually think we were going to make it into one of the launch videos for this. That was a very pleasant surprise.

And they've been lovely to work with as we got this put together.

Re: Codex for almost everything

#406

Post by briHass »

vunderba wrote: post #397
culopatin wrote: post #387

Earlier quoted context omitted.

Do you think this will continue growing if we stop struggling and posting our findings on forums?

Yeah, I think that's a legitimate concern. It's hard to know, even with sufficient training data, how far these systems can actually generalize their problem-solving abilities when they become data starved in the future either because of scarcity or that any potential new training data is contaminated by LLM radiation. Too bad we don’t have a portal gun to access an infinite number of parallel universes where large l…

I'm more optimistic about LLMs tracking down and fixing issues in software, even without SO/forum posts, at least for OSS. I've seen enough unique insights from agents on tricky problems to know it wasn't extrapolating from a helpful comment somewhere.

It hit me that as it's deciphering some verbose log file, it has also read through all the source code that wrote that log, and likely all of the discussions/commits that went into building that (broken) feature.

Re: Codex for almost everything

#407

Post by sph »

badrequest wrote: post #403
someotherperson wrote: post #224

Earlier quoted context omitted.

Competition is bad? Who cares - let the big players subsidize and compete between each other. That's what we want. We want strong models at a low price, and we'll hype up whoever is doing it. Simultaneously, we also hype up the open models that are catching up. That are significantly more discounted, that also put pressure on the big players and keep them in check. People aren't falling for PR; people are encouraging…

It's hilarious how much this post reads as drafted by an LLM. The emdash, "it's not X, it's Y" framing, incredible.

Dogfooding by the slop factory. The artificial centipede.

Re: Codex for almost everything

#409

Post by 4b11b4 »

ogig wrote: post #187
jampekka wrote: post #143

Lots of scepticism here, but I think this may really take off. After 25 years of heavy CLI use, lately I've found myself using codex (in terminal) for terminal tasks I've previously done using CLI commands. If someone manages to make a robust GUI version of this for normies, people will lap it up. People don't want to juggle applications, we want computers to do what we want/need them to do.

I agree. As a long time linux user, coding assistants as interface to the OS has been a delight to discover. The cryptic totality of commands, parameters, config files, logs has been simplified into natural language: "Claude, I want to test monokai color scheme on my sway environment" and possibly hours of tweaking done in seconds. My setup has never been so customized, because there is no friction now. I love it and…

Never been a better time to Emacs

Re: Codex for almost everything

#410

Post by rachel_rig »

KronisLV wrote: post #306
the__alchemist wrote: post #298

Earlier quoted context omitted.

Call it fall for it, but here are my two experiences, with both applications open. ($20/month plan for both) - Claude: Good for ~20 minutes of work once every 4 hours - Codex: Good for however long I want to use it. Claude nerfed their product so that it's not usable, so I use something else.

I'm on the 100 USD plan with Anthropic, I hit the 5 hour limits about 75% of the time during working hours, but almost never the weekly ones - by the time they're reset I've usually used up between 50% - 75% of the quota. There are periods of more intense usage ofc, but this is the approx. situation I'm in (also it doesn't work on tasks while I'm asleep, because I occasionally like having a look at WIP stuff and inte…

[dead]
Post reply on HN