robfitz

An extremely eye-opening comment, thank you. I haven't played with the image generators for ages, and hadn't realized where the workflows had gotten to.

Very interesting to see differences between the "mature" AI coding workflow vs. the "mature" image workflow. Context and design docs vs. pipelines and modules...

I've also got a toe inside the publishing industry (which is ridicilously, hilariously tech-impaired), and this has certainly gotten me noodling over what the workflow there ought to be...

It's all about that value-per-word ;)

There are plenty of river ports/quays with power and water hookups. If you want to houseboat it, you really only need to move when you want to dump the septic tank (and in some areas, not even that).

The "good" spots are highly desirable and usually privately owned (or on a very long-term lease), and you often need to buy a boat to get the mooring spot beneath it. (A buddy of mine in London ended up with 3 boats while upgrading his mooring spot.)

Another options is to vagabond it, where you move your boat to a new area every week or two -- this is usually unserviced, so it's more like camping, but lets you use unofficial spots and is free.

(The last option is to find goofy loopholes. E.g., in some areas of the Thames, you can create a nesting habitat for an endangered bird on your boat, and then once the bird settles in, they can't force you to move, since doing so would destroy an important habitat. Sounds stupid but a surprising number of permanent city boats are there on some sort of loophole.)

Any chance you've written more about the system design and incentives side of it somewhere (whether about medium in particular, a previous biz, or just the mechanisms in general)? If so, I'd absolutely love to dig in... It's got big overlap w/ some stuff I'm trying to figure out for building useful communities, and it's not so common to find folks with a deep view on it ;)

I got into the fifth batch (s07) and remember my other startup friends staging an intervention to dissuade me from accepting because "the valuation is really bad."

Interesting to see how long even the industry insiders failed to take YC seriously. And then once it was working, they flipped immediately to complaining that YC was too powerful and too influential (unbeatable network effect, seed/A valuation inflation, and so on).

I also remember the constant naysaying about scalability and batch size. Our batch was ~19 companies. People kept naysaying, "Well this model is fine for now, but it will never work past 20 teams."

PG would always reply with something like, "Yeah, they said that when we had fewer than 10 teams also. We aren't thinking too far ahead; each batch, we just find the next bottleneck and solve it, and we'll see how far that gets us." Which evidently got them pretty far. A lovely example of doing things that don't scale.

Aw, you're a star, Andrew. I really appreciate (and admire) the sentiment and thoughtfulness. (Although I still think that I dropped a bit of an unsalvageable mess on your lap!)

As a side note since you're here, I'm super psyched that you've taken the time to share what you've learned in your new book -- it's the next one up on my list for serious study, and I couldn't be more excited about it. Based on the reviews so far, it looks like you've written something really special and that I'm in for a real treat. Can't wait :)

The mundane answer is that there's some stuff that I worry about optimizing and other stuff that I don't, and this little personal site is the latter ;).

To a lesser extent, I also didn't exactly expect this to hit frontpage, so I'd sort of assumed that the only people who would see it were people who already knew me.

I'm still not sure about the answer to that, but this has at least prompted me to give it a proper think and decide whether I want to try to appeal to "the world" or just stay focused on my own little orbit. No idea where I'll end up, but it's an interesting question, and I appreciate you giving me the nudge to take it seriously.

Exactly. Really opens up some beautiful coastal areas to explore where the tides are big.

It's also quite nice to know that if you happen to run into an unmarked sandbank, you'll only have to sit there, and not sit there sideways ;)

Yeah, I looked at trailer sailers (although I wanted a cabin), but it wasn't really viable for the types of travel I wanted to do. Now that my life and location are slightly more predictable, a 7m with a small cabin that could go on a trailer would be ideal. (Although I'm sure I won't end up making that sensible choice.)

The user is always right ;). I'll keep working on it. Thanks again.

I don't have a proper answer (because it's a good and difficult question), but I sort of think of it as a context-dependant thing.

E.g., During a customer interview where I'm trying to understand what they've done (and predict what they'll do), I have vanishingly low confidence in their self-predicted behavior. So I prefer to try to look at what they've actually done in the past, talk to them to understand why they did it that way instead of some other way, tease out the invisible data (like what they tried or researched but didn't continue with), and then come to my own conclusion about how that's all likely to extrapolate.

But in terms of understanding and ascribing purpose to my experiences? Or understanding and empathizing with a friend or stranger? Or sending a message out into the world in a way that can stick and spread? In that case, it's kind of a subjective thing to start with, and it's all about the stories. (To clarify: not intentionally fanciful or fictional stories. Just "stories" in the sense of how we tie all these things that happen into something coherent and ordered.)

That was kind of a non-answer, and I'm sorry I can't be more helpful. I only understand a very, very small number of things. (Three, I think, although I'm working hard on understanding the fourth.) Apart from those few narrow domains, I'm just fumbling through it like everybody else.

An important moment and transformation, well-described. Thanks for sharing it :)