81–90 of 239 posts
The only software worth writing is tools for agents that contain something hard for them to vibe code in a couple of sessions. Making someone’s agents 20% better, cheaper or faster will be a measurable and easy sales goal.
> Big companies can copy your product in no time Your advantage in this case, now or 10 years ago, is that this is simply not true. If your business is "a flashlight app", yeah, eventually they'll copy it (as it happened). However they'll take an unusual long time to do that simple thing (as it also happened). Why? Because everything at big companies is a political game, full of internal conflicts, multiple prioritie…
The real moat isn’t just code it’s speed, focus, user trust, and the ability to actually ship. Those are things bloated orgs struggle with, with or without AI. If you’re solving a real problem and building a real brand, you’re already ahead.
Earlier quoted context omitted.
Facebook cannot make products, its that simple.
Oculus?
Lots of talented people, and decent hardware, once the stupid politics was overcome.
But.
In all the time I worked there, we never managed to get "join my friends" feature get made. Trying to join your friends in a game was so fucking hard. You'd think being a social company, who's whole point is connecting people, this would be the first feature.
But no.
Threads is the only product they have launched in the last 5 years that anything close to successful. Even then it only launched because management ignored it, let them get an MVP out the door before swelling the size of the team from ~8(?) to >>2k The time to fun was also waaay too high. it got better towards the end, but in 2020-2023 it would take ages to log in, update, load, get kicked out, reboot and then join.
Earlier quoted context omitted.
Of course profit is a priority how else are you going to pay the bills/employees etc?
“The”, not “a”. There is a chasm of difference between “I care about this problem and want to solve it, but I also need to think how to make it sustainable” and “I don’t really care what I’m tackling, as long as I make bank”. Both exist. The second one is ever more the norm.
Earlier quoted context omitted.
I can imagine a future where it will be possible for a family to host their own essential services ... There got to be something between homelabs and cloud services bc the gap is too big.
With IPv6 (and/or NAT-forwarding) it was already possible to host stuff. However, E-mail's horrible protocols and spam-blocking security monopolies mean you're stuck with one of the big cloud providers, even if you could automate/solve e-mail server complications.
Earlier quoted context omitted.
Have you actually had it do anything substantial and then tried to work with the code it produces afterwards? It may "work" but it's a horrific mess. Good luck bug fixing that.
Yes, Opus 4.5 is not business as normal. If you haven't tried it, you really need to, to update your opinion.
> Big companies can copy your product in no time Your advantage in this case, now or 10 years ago, is that this is simply not true. If your business is "a flashlight app", yeah, eventually they'll copy it (as it happened). However they'll take an unusual long time to do that simple thing (as it also happened). Why? Because everything at big companies is a political game, full of internal conflicts, multiple prioritie…
Don’t worry big companies still can’t copy anything quickly, even with AI. Why? Because before they can ship a single feature, they’ll need to schedule 42 alignment meetings, debate AI-generated slide decks, and log their “strategic pivots” into an AI-curated Jira board. The real moat isn’t just code it’s speed, focus, user trust, and the ability to actually ship. Those are things bloated orgs struggle with, with or…
Yet, Ford didn't copy Tesla, why not?
Checkout https://en.wikipedia.org/wiki/The_Innovator%27s_Dilemma but overall the idea is that large corporations have totally different goals and incentive than startups. Also check https://steveblank.com and his startup owner manual trying to clarify that a startup is NOT a small company, rather it's searching for a product/market fit.
So is it worth pursuing a (software) startup?
Honestly if you want to make money, no. It's much easier to work for a large corporation (big economical benefits) or a public institution (work safety). You might not get ridiculously rich either way but you will be safer than most.
If you want to make a ridiculous amount of money though, well you might have to make sacrifices you are underestimating. I don't want to promote the grind culture but checkout Silicon Valley, the TV show, in particular when they discuss VC and promises done. Basically if you ask for a very high valuation you have to deliver on those claims and obviously they investors are going to keep you accountable on that. You might have to work more than you wanted (which might make the whole process a nightmare more than a joy) but more dangerously you might lose the very raison d'etre of the company you founded. If it's "just" about making money, you won't care, if you have a genuine mission, those compromises are going to eat you alive.
So... is it "worth it" very much depend on who you are, who you want to be and how saturated the market is.
I wouldn't worry about the big corporations, I would worry about what you can actually deliver and how not being able to do so will change you along the way.
It’s still hard to create something successful. Remember that Mark Zuckerberg has had “AGI” for over a decade in the form of tens of thousands of human software engineers and Facebook still has barely been able to create a new successful product on their own without acquiring it from a smaller company.
> Mark Zuckerberg has had “AGI” for over a decade in the form of tens of thousands of human software engineers This makes me think that companies like Open AI, Anthropic etc are simply new types of body shoppers. body shoppers don’t have their own products they just supply talent to help other companies make products and services.