4 ms·
If you compare the pg essays from 2000-2010 to 2010-now, there's a very noticeable shift in tone on this exact topic, even moreso if you zoom out of pg and cons
by waprin 4y ago
If you compare the pg essays from 2000-2010 to 2010-now, there's a very noticeable shift in tone on this exact topic, even moreso if you zoom out of pg and consider YC messaging then and now.
Tons of old pg essays sound straight from something like Indie Hackers or Microconf with their "avoid raising VC money" angle. He even literally predicts the death of VC in web SaaS, saying "investors aren't worth the trouble": http://www.paulgraham.com/divergence.html http://www.paulgraham.com/divergence.html .
His YC cofounder Jessica Livingston wrote a book "Founders at Work" and about half the founders in the book are bootstrappers (DHH, Joel Spolosky, Craigslist). The other half are mostly founders recounting horror stories of interacting with VCs.
Compare that to modern YC where there are videos where they say _everyone_ should consider applying to YC with the _only_ exception being people that want to bootstrap. Modern Startup School says: "Without startup funding the vast majority of startups will die." (https://www.ycombinator.com/library/4A-a-guide-to-seed-fundraising https://www.ycombinator.com/library/4A-a-guide-to-seed-fundr...).
There's something very important to note which is that YC did not change the deal from 125k for 7% to 500k for 7%. It's still 125k for 7% plus 375k worth of equity on your next raise. Which only makes sense if there is a next raise. So obviously they discourage bootstrapping since their whole model has been built around you raising at least one more round after YC.
Seems a pretty simple case of "follow the incentives." YC basically became more of a traditional VC over time. And I think pg was still closer to a founder in the early 2000s, from a founder perspective, it's more of a set of tradeoffs whether you should bootstrap or seek VC. But from a VC's perspective, obviously they want you to seek VC since they can't get involved if you bootstrap and bootstrapped startups won't get the outsized returns they need.
My opinion is, pg is a smart guy, there's still a ton of wisdom to learn from in his essays, but as with every other person on the planet, consider their motivation and incentives for telling you what they're telling you.
- dasil003 4y agoAlso worth noting the 14 year tech bull run and ever increasing amounts of capital looking for a return. In 2000-2010 it was in many ways easier to bootstrap as everything was smaller, complexity was less with simpler browsers and no smart phones, UX expectations were less, and salaries for software engineers were more in line with average professional salaries. Not that I disagree with your assessment on pg's incentives and YC's change in position over time, but the environment for software startups is also materially different now that affect you whether you take VC money or not.
- waprin 4y agoI'd argue the exact opposite. Key costs like hosting are way cheaper, back then you probably had to rack server in a data center, now you can deploy a free/cheap PaaS with a free/cheap database while you find PMF. Marketing is way cheaper, social media is a grind but it's a game you can play to acquire users/customers for _free_. There's more stuff going on online, more people, more businesses, more everything. Think about the entire "creator economy", other bootstrappers, Shopify sites, etc. Yeah some stuff has gotten harder, more platforms, higher UX standards. But on the whole I'd rather bootstrap in 2023 then 2010. Check in with me in a year though.
- logifail 4y ago> Key costs like hosting are way cheaper, back then you probably had to rack server in a data center, now you can deploy a free/cheap PaaS with a free/cheap database while you find PMF. Back in the day I have delivered several different cheap 1U rackmount servers to data centres. Once took one - using public transport - to whatever suitable spot in the Docklands in East London. Travelling on the DLR with a server under one arm was a a memorable experience. Long story short: you can do a lot from a $1000 server, if you put your mind to it. Yet these days it seems PaaS is better, (over-)paying GCP/AWS/Azure, while you grope around to find a product that will sell for $$$ before your cloud credits run out?
- bluedino 4y agoBack in 2009 I had a dedicated server for $89/month It wasn't the greatest thing in the world, but it was dual-core, 2GB RAM, 160GB HD. I ran a small forum, email server, and a couple buddies used it for shells and running whatever PHP apps they wanted. An $80 Linode these days only has 16GB and 320GB.
- kjksf 4y agoA better comparison is something like Hetzner dedicated server. Here's what you can get for $89+: https://www.hetzner.com/sb?price_from=89 https://www.hetzner.com/sb?price_from=89 For 92 euro: 128 GB RAM, 3 TB SSD, 8 core i9-9900K, unlimited bandwidth. You get 50x value for the 2009 price.
- FlyingSnake 4y agoI wish this comment was on the top level comment in this thread. The divergence of pg from his original message is really sad. The more you gaze into the abyss, the abyss gazes back into you.
- dopeboy 4y agoThis is an excellent point. I'm with you - I still trust his advice if not for his judgement for the fact that the sample population of startups he's worked with is so high. But this bias is there. I'd wager if YC started HN today, it wouldn't be called HN. That name is uniquely related to the times YC & PG came into.
- serf 4y agoI think you're right, but I think it's unsurprising. It's a bit like trusting an insurance broker's opinion on your own personal insurance necessity -- it'd be unsurprising to me for a broker to tell me that I need the most comprehensive (expensive) plan that they sell.