3 ms·

I find these 'Valley Koans' popularised in the last long while to be a little troublesome. Things like 0->1->n Do things that don't scale Andreesen Horowitz bi

by jariel 7y ago

I find these 'Valley Koans' popularised in the last long while to be a little troublesome. Things like 0->1->n Do things that don't scale Andreesen Horowitz bit on 'The Good Ideas That Seem Bad' Paul G's essays. They all have nice little nuggets of wisdom, but what the almost always lack are good examples and concrete evidence. Funny that many HN comments are better researched than a lot of these 'Deep Thoughts'. Since these guys have access to startup data like nobody else, golly, wouldn't it be nice to see examples and break-downs? i.e. here's 1000 YC startups, and how they categorise in terms of 'hard/simple' problems. Here's their trajectory. Here's some mitigating factors etc.. The reason I write this is because of my flinch on this note from Sam. He's worth listening to, absolutely, but that doesn't mean we can't disagree. The problem with his thesis is that YC makes all their money from very regular businesses. There is no 'magic' to Stripe. It's a really great product that did mostly what previous products did, with a much cleaner API that devs loved. This is not koolaid/AI kind of stuff. For pragmatically minded people, yes, it does seem like it could 'really help change the world', but really ... it's not sexy. AirBnB because it's more consumer, yes, I can see some koolaid about 'sharing' etc. but again I don't think it's in the category of 'tough problems'. DoorDash, Instacart, Dropbox, Faire, Brex, Flexport, Checkr, Mixpanel. So many of the big YC companies are solving regular kinds of business problems, in big market places. The 'heady' startup would be 'Cruise' but that was an odd man out for a few reasons. All the things it would seem to me that Engineers would really like to solve are not quite on that top list. The 'coolest tech' YC startups aren't really the big hits.

keenmaster 7y ago
The biggest "deep problem"-type leaps in the past few decades have been due to innovations in technological infrastructure: the advent of the internet, GPS, and smartphone for example. All three innovations benefited from the positive externalities of government funding, as well as routine R&D and large corporations. The entities that collectively solved those hard problems only captured a fraction of the profits/benefits of their innovations. The quickest route to success for a "hard problem" startup is to see the obvious Next Big Thing that is still in early adoption phases, and find the best way to translate it to the broader market. I know, that's not solving a hard problem in the same sense as the OP, but I think it's an underrated strategy with a quicker feedback loop and higher likelihood of success. What's the Next Big Thing? One of them is AR combined with AI, cloud-computing, and 5G for real-time rendering of 3D objects. That will enable so many incredible things. Imagine the most profitable possibilities, for software, peripherals, physical objects that play a part in the new physical-digital realm, etc...and start hacking now. By the time AR devices are mainstream, you'll be ready to profit immensely while everyone else is scrambling to learn the fundamentals of AR. VR is fairly mediocre right now. Of course, you know it will get better. However, do you think you correctly predict the magnitude and speed of that change? It's fairly likely that VR will be very, very good in the next ten years. If you're ready for that scenario, you can strike gold. Self-driving cars are another Next Big Thing. Right now all the attention is on Velodyne, Waymo, etc...and those are promising companies. However, there's plenty of room for startups in that space. Anticipate how self-driving cars will change the world, and profit off of that change. That can make you rich. I'd rather ride the tsunami of new technologies than try to capture the energy from late-stage ambient tidal waves (such as making another camera app, or a niche ERP tool for enterprise). On the other side of the spectrum, it's noble to try to invent the Next Big Thing yourself, but the odds are against you even if you're extremely smart. You might be just as likely to stumble into the Next Big Thing doing routine work at a big company.
- JabavuAdams 7y ago
  Second mouse gets the cheese.

by jariel 7y ago

keenmaster 7y ago

The biggest "deep problem"-type leaps in the past few decades have been due to innovations in technological infrastructure: the advent of the internet, GPS, and smartphone for example. All three innovations benefited from the positive externalities of government funding, as well as routine R&D and large corporations. The entities that collectively solved those hard problems only captured a fraction of the profits/benefits of their innovations. The quickest route to success for a "hard problem" startup is to see the obvious Next Big Thing that is still in early adoption phases, and find the best way to translate it to the broader market. I know, that's not solving a hard problem in the same sense as the OP, but I think it's an underrated strategy with a quicker feedback loop and higher likelihood of success. What's the Next Big Thing? One of them is AR combined with AI, cloud-computing, and 5G for real-time rendering of 3D objects. That will enable so many incredible things. Imagine the most profitable possibilities, for software, peripherals, physical objects that play a part in the new physical-digital realm, etc...and start hacking now. By the time AR devices are mainstream, you'll be ready to profit immensely while everyone else is scrambling to learn the fundamentals of AR. VR is fairly mediocre right now. Of course, you know it will get better. However, do you think you correctly predict the magnitude and speed of that change? It's fairly likely that VR will be very, very good in the next ten years. If you're ready for that scenario, you can strike gold. Self-driving cars are another Next Big Thing. Right now all the attention is on Velodyne, Waymo, etc...and those are promising companies. However, there's plenty of room for startups in that space. Anticipate how self-driving cars will change the world, and profit off of that change. That can make you rich. I'd rather ride the tsunami of new technologies than try to capture the energy from late-stage ambient tidal waves (such as making another camera app, or a niche ERP tool for enterprise). On the other side of the spectrum, it's noble to try to invent the Next Big Thing yourself, but the odds are against you even if you're extremely smart. You might be just as likely to stumble into the Next Big Thing doing routine work at a big company.

JabavuAdams 7y ago
Second mouse gets the cheese.