I do use worktrees occasionally (especially during times where I'll have a very sticky problem that I make the LLM run in a loop on until it satisfies acceptance criteria, and want to isolate the potential fallout of Claudes Gone Wild), and I run Claude and Codex side by side, but I rarely have them work on truly-different tasks simultaneously.
The main reason is because if there's a significant bug or large optimization going on, that shit needs to be done, tested and merged before building more stuff on top, otherwise you run the risk of wasted time, tokens and effort having a bunch of parallel work running that may not end up compatible at the end.
Lately I've had a lot more success having Claude generate a plan, send the plan to Codex for co-validation/amendments, have Claude implement the plan, then have Codex PR review the commit (and likely make some edits of its own), then I test out the code/changes.
Meanwhile, my actual management of what I'm asking them to do is just a text file in Notepad where I'll write like BUG: xyz thing does abc or IDEA: let's change this to that as I'm testing in-app, with the actual code opened in Notepad++ tabs (lol feel free to roast me, I'm in front of 2 screens, one Windows (primary), one Mac (to the right), sharing keyboard and mouse -- LLMs are 99% on the Mac, planning/testing/verification/manual coding/graphic design on Windows, committing and pushing to a repo both machines have checked out)
I haven't yet found a scenario where many Claudes and many Codexes running simultaneously on 35 concurrent features makes any sense, but I'd definitely encourage people to try multi-model cooperation since they all seem to have different sensibilities. I haven't made much use of Gemini in this context though because two's company, three's a crowd. YMMV.
I’m on the $100 Claude and $20 GPT plans. I almost never run out of weekly usage on Claude and occasionally blow my Codex allowance, but OpenAI lets you buy credits ad-hoc (either $40 per 1000, or just turn on and monitor the “top-off” and set it to buy like $5 at a time. The one or two times I’ve run out, by the time the week resets, I’ve only spent an extra 5 or 10 bucks.
Using somebody's stuff is different than hot-linking directly to a hosted version of it, even just from the perspective that dude could delete it at any time and break the whole app.
Are you aware that common libraries like Bootstrap, FontAwesome, and HTMX walk developers through linking to their CDNs directly? In fact, FontAwesome recommends it for CDN performance.
I think you're dangerously mistaken if you believe that it "literally never" happens. It literally does happen all the damned time. And, for your own safety and others', you should assume that when you use any app for which you don't have the source code.
Linking to a CDN is for development only. Once the app is build you build your dependencies into the app. You don't fetch them at runtime and run them. Not only for security, but also for performance.
There's also a difference between using a CDN for, say, React and a random github project hosted by some dude.
Assuming everything physical gets tokenized (as occasionally gets predicted), people could soon literally lose their house on a bet! Maybe even a bet placed by their swarm of agents. The future's so bright, I gotta wear shades!
Worse: It won't even be your own bet that does the damage.
It'll be bets collected (or "lost" bets which are actually payments) by other parties that have an interest in an outcome which involves whether you lose your house.
Hence why we drastically reduced the number of wagers/addictive risk taking because of its cost on society.
The libertarians here will say "oh yay, we've won". Then a few years later they'll cry about your mom losing her house to a gambling scam with no recourse. Then they'll cry again when the voters finally rid themselves of the gambling scourge years later.
I don’t even know what to say here -- you’re entitled to your opinion obviously, and I disagree with it deeply, and the spirit of HN is to avoid personal attacks and reply with curiosity, but you kinda laid it out very plainly above. Where’s your imagination gone? Your connection to child-like wonder? Empathy for your fellow man?
Project Hail Mary isn’t Arrival, it’s ET mixed with Castaway. It’s about friendship and loneliness and the fragility of the human experience and the triumph of the human spirit!
Normally I’d just say “you didn’t get it, it wasn’t for you” but given the insufferable and total dismissal above, I’d wager it actually IS for you LOL but you chose not to receive the message.
Anyways, everybody’s a critic these days, I get that. I’d just encourage people to soften a bit and appreciate things for what they are (not what we want them to be)
>> It’s about friendship and loneliness and the fragility of the human experience and the triumph of the human spirit!
So is every Disney movie and that is what this but with the crappy Amazon Studios take on it.
>> Anyways, everybody’s a critic these days,
Do you believe a movie can objectively be considered good or bad? If you do you then believe some are better critics than others, the same some way some are better Coders than others or better Basketball players than others?
You're asking the wrong person lol. I can give you a list of "objectively bad" movies that I think are incredible for a variety of defensible reasons.
Just off the top of my head as I briefly scan shit sitting on the shelves of my office:
- Joe Dirt
- Death Wish 3
- Thrashin
- Hackers
- Mortal Kombat
- Uncle Buck
- The Incredible Burt Wonderstone
- Tapeheads
- Prayer of the Rollerboys
- Weekend at Bernie's
Not exactly Fellini, and some are barely even Andy Sidaris if we're being honest, but every movie in that list is amazing for different reasons. An objective critique of any of them (especially in context with "film", as a shapeless, vague concept) misses the point and the spirit of each and every one. But I am an uncultured heathen, so ...
Devil's advocate (because honestly I do agree with you, but..) -- help/encouragement often ends up turning into far more time and effort than it sounds like up front.
~18 months ago a friend of mine had a very viable, good idea for a physical product, but very fuzzy on the details of where to begin. My skillset backfilled everything he was missing to go from idea to reality to in-market.
I began at arm's length with just advice and validation, then slowly got involved with CAD and prototyping to make sure it kept moving forward, then infrastructure/admin, graphic design, digital marketing and support, etc, while he worked on manufacturing, physical marketing, networking, fulfillment, sales, etc.
Long story short, because I both deeply believe in the vision and know that teamwork makes the dream work, I am fully, completely, inextricably involved LOL -- and I don't have a single complaint about it either, but man, watch out, because if you don't believe in the vision but do have skills/expertise they're lacking, and opt out, friends and family will be the quickest and most aggrieved people you'll ever meet that think you're gatekeeping them from success.
It's always said that a lot of success and opportunities are attributed to being in the right place at the right time (aka "luck"), but in a lot of cases, those folks had the tenacity to be in the right place ALL the time; when opportunities arise, they typically go to whoever's present and available.
Chatting with professors after class or attending office hours might be a grift, but it's not necessarily unfair. Specific circumstances aside, anybody can do it to get some leverage.
I might be the odd-man-out here, but I almost exclusively sit at a desk with a full-size mechanical keyboard and mouse, 2 monitors, and a tower in the closet nearby LOL -- I can't get any real work done hunched over a laptop! Not that I'm incapable of adapting, I just dislike the cramped ergonomics of screen and keyboard, and really need a mouse for precision in Photoshop etc. It always blew my mind when I worked in an office how coworkers were cool with coding on a 13 or 14 inch screen, rarely docked the laptop, sitting around on couches or whatever instead.
OpenClaw guy (he's Austrian, it's relevant) much prefers Codex over Claude and articulated it as being due to Claude's output feeling very "American" and Codex's output feeling very "German", and I personally really agree with the sentiment.
As an American, Claude feels much more natural to me, with the same overly-optimistic "move fast, break things" ethos that permeates our culture. It takes bigger swings (and misses) at harder-to-quantify concepts than Codex, cuts corners (not intentionally, but it feels like a human who's just moving too fast to see the forest for the trees in the moment), etc. Codex on the other hand feels more grounded, more prone to trying to aggregate blind spots, edge cases, and cover the request more thoroughly than Claude. It's far more pedantic and efficient, almost humorless. The dude also claimed that most of the Codex team is European while Claude team is American, and suggested that as an influence on why this might be.
Anyways, I've found that if I force Claude and Codex to talk to each other, I can get way better results and consistency by using Claude to generate fairly good plans from my detailed requests that it passes to Codex for review and amendment, Claude incorporates the feedback and implements the code, then Codex reviews the commit and patches anything Claude misses. Best of both worlds. YMMV
Oh, interesting perspective. I'm Italian, but from an Alpine valley not far from Austria, so I don't know what I should prefer. :D
But joking aside, putting it like that I'd think I'd prefer the German/Codex way of doing things, yet I'm in camp Claude. But I've always worked better with teammates that balance my fastidiousness, so maybe that's my answer.
Childcare and education requires a specific tolerance, mindset and passion to be effective though. I'd be curious how many previously-PMs or HR drones or email jockeys would be adequate (let alone thrive) in an environment where there are next-to-nonexistent budgets, and you're servicing literal babies and tiny children lol
On second thought, client service folks might do extremely well here!
The main reason is because if there's a significant bug or large optimization going on, that shit needs to be done, tested and merged before building more stuff on top, otherwise you run the risk of wasted time, tokens and effort having a bunch of parallel work running that may not end up compatible at the end.
Lately I've had a lot more success having Claude generate a plan, send the plan to Codex for co-validation/amendments, have Claude implement the plan, then have Codex PR review the commit (and likely make some edits of its own), then I test out the code/changes.
Meanwhile, my actual management of what I'm asking them to do is just a text file in Notepad where I'll write like BUG: xyz thing does abc or IDEA: let's change this to that as I'm testing in-app, with the actual code opened in Notepad++ tabs (lol feel free to roast me, I'm in front of 2 screens, one Windows (primary), one Mac (to the right), sharing keyboard and mouse -- LLMs are 99% on the Mac, planning/testing/verification/manual coding/graphic design on Windows, committing and pushing to a repo both machines have checked out)
I haven't yet found a scenario where many Claudes and many Codexes running simultaneously on 35 concurrent features makes any sense, but I'd definitely encourage people to try multi-model cooperation since they all seem to have different sensibilities. I haven't made much use of Gemini in this context though because two's company, three's a crowd. YMMV.
reply