Ask PG: If Viaweb hadn't worked out, what would you have likely done next? | Hacker News Reader

Ask PG: If Viaweb hadn't worked out, what would you have likely done next?

187 pointsSnail_Commando43 comments

I don't think I had a backup plan at the time. I suppose I would have gone back to consulting and writing books. Maybe after a while I would have tried to start another company. That would have been hard, though, because if Viaweb hadn't worked out it would have been hard to talk Robert Morris into trying again.

I was pretty excited about Web apps, which were a new thing then. (We planned eventually to make a whole suite of them; Viaweb Store was just the first.) So I probably would have worked on those in some way. Maybe I would have written some sort of general platform for building them, and/or written a book about that topic.

What type of consulting work did you do before Viaweb?

I used to consult for a company called Interleaf, which was one of the few places that used Lisp at the time.

I used interleaf publisher for my part time job in university. I remember the fonts looked nice on our 20 inch displays.

Assuming the market hadn't changed, and you still had a big market for something like Viaweb, would you still use the same technology stack? Would you still use CLisp, or Clozure/SBCL/ABCL?

Are you asking what I'd use if I were building Viaweb today? The answer to that is probably Arc, though that says more about me than about currently available technologies.

Would you recommend Arc to modern startups in general?

No, I don't think so, not in its current state.

Are there any Lisps you would recommend (other than Clojure)? (Or languages?)

Clojure is probably the best bet. I don't know much about it but lots of people seem to use it, and any decent Lisp that lots of people use is probably a good bet.

Why not? Have you used a language other than Arc to write a web app recently?

If writing web apps in other languages isn't still painful where have other languages/frameworks caught up?

It's still missing some things that most people take for granted.

I've kept an eye on Arc for years, with my 100-year language glasses on. How do you feel about Arc as the 100-year language as described in your essay? It's been 9 years since the essay; any plans for a followup, or do you feel that it's held up over time?

I don't get to work on it much now. I hope there is little in it that wouldn't be in the 100 year language. But there is also little in it.

You mean Go, right?

Why would have been difficult to convince Robert to try again?

Because he didn't even want to do it the first time.

Sorry for going off in a tangent but PG, would you consider doing an AMA on reddit. I've seen two posts in last week with the sole purpose of asking you a question. Seems like the community wants an interaction with you. If you've already done it, I couldn't find it.

Seems like this site is a pretty good way of interacting with him.

There is a meaningful subset of HN'ers who have fairly direct access to PG and the other YC partners - the current class and alumni of Y-Combinator. For the rest of us, PG's essays cover a lot of his background as does his book Hackers and Painters.

The format of Ask PG seems to work pretty well. A mix of technical questions regarding YC applications, changes to HN, and interesting business questions tend to get deeper answers than is common within the limits of AMA.

Maybe I would have written some sort of general platform for building them

There's still time for that. The current solutions leave much to be desired (although they're slowly moving in the right direction).

Doubtfully a popular opinion, but I'll throw it out there anyway. In finance it's common to assume the winners (aka. those that float to the top of hedge funds, investment banks, etc) are worth observing as a result of their apparent success, however it's considered prudent to ignore the winners and instead observe what the losers have done, as few intelligent people in that sector believe a consistent winner is little more than someone who has had a good run of luck.

Following from that, if pg wasn't a "winner" there is very little to differentiate him from what 1,000,000 others have tried and failed. Circumstance and context are a frighteningly arbitrary decider, and therefore I don't think this question is particularly meaningful.

[edit: this is not to suggest that dumb luck alone is enough to create a success, it's just that intelligence and tact alone are often not enough to create success either]

I do believe there's a pg essay on this very topic.

Do you have a link?

http://www.paulgraham.com/really.html #16

It's not on this topic but it definitely hits on it

Thanks!

http://www.paulgraham.com/really.html

  With two such random linkages in the path between startups and money, 
  it shouldn't be surprising that luck is a big factor in deals. 
  And yet a lot of founders are surprised by it.

http://paulgraham.com/wealth.html

"The other catch is that the payoff is only on average proportionate to your productivity. There is, as I said before, a large random multiplier in the success of any company. So in practice the deal is not that you're 30 times as productive and get paid 30 times as much. It is that you're 30 times as productive, and get paid between zero and a thousand times as much. If the mean is 30x, the median is probably zero. Most startups tank, and not just the dogfood portals we all heard about during the Internet Bubble. It's common for a startup to be developing a genuinely good product, take slightly too long to do it, run out of money, and have to shut down."

Y-combinator has had a long run of successes, far beyond what could reasonably be attributable to luck.

YC is an investor, putting money into lots of startups. As a result, it benefits from the law of large numbers (https://en.wikipedia.org/wiki/Law_of_large_numbers), so luck isn't as big a factor as it is for individual startup founders.

Regardless of the applicability to pg, I think this is a smart conversation. When evaluating hypotheses, we have a cognitive bias in favor of past success, when past failure might be the more informative signal. Planet Money did an interesting podcast with the author of this book:

http://www.amazon.com/Adapt-Success-Always-Starts-Failure/dp...

discussing just such a hypothesis:

http://www.npr.org/blogs/money/2012/09/04/160555540/episode-...

Just ordered the book, thanks!

The reason it is considered prudent (by some) in finance to not invest in the latest success story is that you want to buy low and sell high. Once a fund is a success, it is hard to do that.

How would you envision Viaweb not working out? Would it be a technical problem where Graham et al weren't able to produce viaweb, a business problem where they weren't able to sell or make money off of it, or a market problem where there simply wasn't anyone interested in online stores, or do you simply mean if Viaweb was something else that would fail?

To me it feels like any of those problems would require the world to be quite different from how it is.

Are you suggesting that the success of Viaweb was inevitable?

There are a million possible failure cases for any startup, including a large number that are for all practical purposes out of your control - a car accident could disable a key participant, a better competitor could suddenly pounce on the scene, fraudsters could decide to make you a target, etc. With all due respect to PG, you don't have to change the world much to see Viaweb fail. Even with hindsight.

All good points. Additionally the timing worked as well and allowed them "not to die".

http://www.paulgraham.com/die.html

"If you can just avoid dying, you get rich. That sounds like a joke, but it's actually a pretty good description of what happens in a typical startup. It certainly describes what happened in Viaweb. We avoided dying till we got rich."

That quote brings to mind the following from the Hitchhiker's Guide:

"There is an art, it says, or rather, a knack to flying. The knack lies in learning how to throw yourself at the ground and miss."

I think with respect to startups the quote is more than just funny, it captures the bravery and apparent foolishness but great rewards of the endeavor.

> The knack lies in learning how to throw yourself at the ground and miss

It's also a perfect description of how orbital mechanics work - the Moon is actually falling down towards us, it's just moving so fast in the other direction that it constantly misses the Earth :).

Having read pg's book (ANSI Common Lisp) and all of his articles; I liked this question and the answers pg has given.

I agree with forgotusername's opinion about the luck plays a good part in life, but as long as the Viaweb's concerned, the luck had its minimal share because I still remember in one of pg's articles he was talking about how frequently they have written their code base in order to serve a user experience better than the competition. That's hardly got anything to do with luck.

And if you've ever read all his articles, you might have spotted sections talking about when and how the luck has been playing really impressive role; and there you got it: In the case of Bill Gate's Microsoft. How it suddenly became a multi billion dollar company almost overnight is plain luck.

Also when you consider for how much Viaweb was sold, which was about $50M, and compare it to other sales most of them as high as ~$1B, again there is no luck to be talked about.

Related question: what happened to the code for it? PG promoted it as an example of what you can do when you're free to do pure Lisp, but as best I can tell Yahoo has since gutted most or all of the Lisp that it ran on.

Edit: essay I had in mind: http://www.paulgraham.com/avg.html

I think a salient elaboration might be specifying what is meant by "not worked out".

Do you mean not getting acquired, or never making any money or getting any traction?

If the former, he (I would surmise) would've been a lifestyler.

Reply on news.ycombinator.com