This was the top comment on the launch post for a file sharing app in 2007. It reads, "For a linux user, you can already build such a system for yourself quite trivially by ...", and "it does not seem very viral".

The app: Dropbox.

Of course, the commenter was technically right – but the value proposition was Dropbox's secret sauce on Windows and the arrival of AWS S3 only a few months earlier. And Dropbox eventually did have difficulty making money and shifting to enterprise sales, but by all measures is was a home run business.

Looking back, I cringe at some of the things I said on the internet. Here's my own faux pas on Hacker News.

At the time, $3000 was an exorbitant amount of money for me (still is, honestly). The author and CEO, Vlad, wrote about how he had purchased an expensive domain name (webflow.com) in college while he was $10,000 in debt before he had an idea of what he wanted to build.

Webflow is currently valued at $2.1B and a well-known brand.

Another thing I've been wrong about on Hacker News (in a comment since deleted) is about Plaid, a company that was initially screen scraping banking data.

In college and tight for money, I wanted to analyze my finances in real time. Services like Mint existed but were slow and didn't give me raw access to my data. My bank didn't offer an official API either. As an engineer, I thought screen scrapers like Plaid and Yodlee were terrible ideas. Screen scraping breaks every time a website updates its interface, and having a service use your credentials in plaintext is always a security issue.

So with a little ingenuity, I reverse engineered the APIs that the mobile banking apps used to pull my data programmatically. I thought this was the "right" approach. Later on, the courts would say that screen scraping == legal, but unauthorized use of APIs == illegal. Plaid bootstrapped its business off of screen scraping and later cut deals with the banks for direct access (like Mint before it).

Luckily, I learned my lesson early on in my career. Be optimistic. Technology isn't everything. Solving problems is more important than being right. And most importantly, don't be a jerk!