3 ms·
I've been at the start-up game non-stop since the mid 1990s. You know what's really amazing today? What would have cost me $30,000 per year in infrastructure in
by adventured 8y ago
I've been at the start-up game non-stop since the mid 1990s. You know what's really amazing today? What would have cost me $30,000 per year in infrastructure in 1999 or 2003, plus the obnoxious management of it all, is now $2k or $3k at DigitalOcean (and others) and rather easy to script and manage. After you adjust for the better hardware vs increased total Internet users scale and often increased resource demands, I also think that value proposition heavily tilts toward a greater than 10x gain versus eg 15 years ago.
Most everyone here will fully understand that effect. I'm skeptical the NYTimes et al. understands what it has done to the start-up vs capital equation however.
When Excite got started, you know what they needed? A $10,000, 10gb drive. Vinod Khosla had to provide it just so they could properly test out their search engine. Adjusting to today's scale, the equivalent might be three dozen virtual servers for tens of hours per month at a cost of $100.
I don't need venture capital to scale to a large size. It's beautiful, the VC aspect has almost entirely become optional outside of a few niche scenarios. That has provided the large capital raising start-ups with a lot of sustained bargaining power that they have never had before. If nothing else, you can simply wait longer to deal with VCs, boosting your position. That's not going away with the next crash (it will deflate some of course), that cost edge is here to stay. The leverage point is: can you actually build the thing, can you execute, do you have the ability / persistence / etc to get it done.
It really wasn't that long ago that infrastructure was the biggest cost in being able to prove out an Internet business and scale. Just to put on a good test of a product meant for scale you had to sink large sums of money in upfront. Today the people - their time - have become the biggest cost (unless you're in SF etc., then it's that and real-estate). If I want to start something today, my time is the (personal) red ink risk, rather than very large amounts of capital plowed into infrastructure.