Interesting data. My first reaction was that lisp and scheme are still underutilized and underappreciated.
But then, the metrics he's using wouldn't count downloads of the PLT books, or Shriram Krishnamurthi's Programming Languages: Application and Interpretation, and it's not clear SICP or EOPL would be counted as Scheme books in their reckoning.
And that's assuming you accept popularity as a useful metric.
It seemed as though pg was talking about the unions of the 1950's. Those unions weren't heroic. The heroism was in the union movement of the late 1800s/early 1900s. Companies/capitalists of that gilded age were often enough a law unto themselves, and workers organizing to have their legal rights respected were in very real danger.
It's a good thing that union organizers are in no real danger anymore, and companies allow a fair negotiation process between workers and the corporate offices who determines their pay.
Sure, in a field where there is a high demand for workers, unions are unnecessary, because any deficiencies in pay will be punished. In manufacturing and retail, however, where the demand is not as high, there has to be someone out there establishing a minimum standard for their employees to receive in compensation for their work. Since minimum wage in the U.S. is not enough to live on, the only other groups with the possibility to enforce such standards are unions.
I think that the recent trend (as in the last couple decades) of discounting the importance of unions is in large part due to comparing dissimilar organizations. Manufacturing and retail is a completely different beast from the service industry, and to lump them all into one "unions fight for overcompensation" group is shameful.
I don't know why anyone other than the pre-acquisition investors would be aggravated by it. Is anyone posting here willing and able to quantify what he contributed and really judge whether he was "not worthy" of his share of the equity? If you haven't been paying attention to pg's essays, "time spent" is not equal to "value produced," so a vesting schedule does not fairness make.
It does sound as though he was not well-suited to corporate life. Where's the crime in that?
This is a good question. The other investors may already be in control of the company and just attempt replacing the founders with professional management. They may liquidate, since sunk costs are sunk.
According to Paul's essays, the software of Viaweb was powerful because it was written in a high-level (but bottom-up) way. It seems like besides this being an edge over your competitors, it would also make your investors leery of getting rid of you. Who else would want to buy it if the Viaweb team wasn't there to walk them through it? And wouldn't that consideration also be true for lenders? I recall reading something about Trump's period of declining fortunes years ago - to whom, exactly, were the banks going to sell his gaudy yacht and other outrageously expensive and virtually unsellable properties? They were better off pretending he could still be successful for a while, and they got lucky.
I can't tell why point 22 shouldn't be titled "Sometimes you need capital other than sweat equity". It's not clear to me why this couldn't come in the form of debt. The only type of debt I recall Paul mentioning is the convertible kind.
I mean, other than the fact you'd have to talk someone into giving you that loan at a reasonable interest rate. Talking investors into giving you the money as a bet for an astronomical interest rate would be easier for companies with no collateral. Still, if you have a revenue stream and can show how a capital expenditure would increase it, it would at least be worth considering.
It's not just investors. This is a good practice anytime you're trying to sell a big ticket item to a potential customer. It gives them permission to believe the price is worth it.
It's been a few years since I was a sysadmin, but I believe either his DNS system or qmail system will show it. When I first read them, I thought they were very annoying, because it's not one monolithic program, and there are a lot of one-line shell scripts that do sequential operations using exec.
The way he describes it, he's getting rid of a lot of parsing by using the directory tree instead of text files following. But this use of directories can also be good for some of the more degenerate uses of databases.
Mostly I recall the "aha" moment of "So, this is the Unix philosophy in action." The design is very elegant.
Let's say you have developed your own programming language/infrastructure, such as Arc or GOAL or whatever. You want to use your tools for a new venture because they give you a lever your competition doesn't have. On the other hand, if the venture doesn't work out, you don't want your language/tools to be part of the liquidation, so you are not deprived of them in the future.
This is impolitic, but if you were starting a new venture and Arc were ready to be used in production, would you want to put it into the new company and, if not, how would you handle investors? This is a little more than hypothetical since you are using Arc in Ycombinator business, but it seems like a different situation than when forming a company with investors you barely know.
But then, the metrics he's using wouldn't count downloads of the PLT books, or Shriram Krishnamurthi's Programming Languages: Application and Interpretation, and it's not clear SICP or EOPL would be counted as Scheme books in their reckoning. And that's assuming you accept popularity as a useful metric.