"God Rewards Fools"

post by Czynski (JacobKopczynski) · 2020-04-30T22:51:41.974Z · LW · GW · 6 comments

Contents

6 comments

In an otherwise gripping but misleading book about the history of cryptography called The Code Book, by Simon Singh, there is a quote from one of the two inventors of Diffie-Hellman key exchange which has stuck with me two decades later. (I've been unable to determine whether it was Diffie or Hellman; sources conflict and I no longer own the book.)

…the way to get to the top of the heap in terms of developing original research is to be a fool, because only fools keep trying.  You have idea number 1, you get excited, and it flops.  The you have idea number 2, you get excited, and it flops.  Then you have idea number 99, you get excited, and it flops.  Only a fool would be excited by the 100th idea, but it might take 100 ideas before one really pays off.  Unless you’re foolish enough to be continually excited, you won’t have the motivation, you won’t have the energy to carry it through.  God rewards fools.

I am a habitual naysayer, like many people in this community. Recently, I've been thinking about this in the context of holiday and ritual design, which is my primary preoccupation in terms of contributing to the rationalist project. As I've written about recently, this is an extremely high-risk, high-reward enterprise; the upside potential is substantial but the downside potential is enormous. Which means that many ideas are needed before you get something that's +EV to actually try and implement, and many of those attempts will be needed before you get something that's +EV to try and sustain. I'd guess that 1000 ideas : 100 fleshed-out ideas : 10 meatspace attempts : 1 sustainable result is about right.

This has a couple important consequences:

Ray Arnold and others possess the incredibly vital capacity of actually doing things rather than waiting for something perfect. However, this is in accordance with the old Facebook motto: "Move fast and break things." They move fast, but they will, accordingly, break things. Which for Facebook meant "we will sometimes ship code that breaks the site" and here means "we will sometimes establish holidays and rituals which are not actually good for our community". Years of iteration have improved several MVPs which were probably -EV initially into fleshed-out versions which are solidly +EV. But they're still probably not good enough. I'd guess, based purely on a gut attempt to gauge what my aliefs are about it, that Solstice is good enough and everything else isn't, though I think everything that's still being tried is +EV overall, i.e. better than nothing.

It does take enough ideas that only a fool would be excited, but you have to keep going before you get to one that really pays off.  "Unless you’re foolish enough to be continually excited, you won’t have the motivation, you won’t have the energy to carry it through." The filtering provided by naysayers is vital, because most of the ideas actually will flop, and when you're playing with alief inception, a flop is very dangerous. But we do need to keep trying, because the upside is, correspondingly, very large.

I suspect this also applies to pretty much any other endeavor in creating novel ideas, but I have no direct experience so I can't be at all sure.

6 comments

Comments sorted by top scores.

comment by quanticle · 2020-05-01T14:40:09.191Z · LW(p) · GW(p)

I'm not sure I agree. I don't think that one of the problems with the rationality community today is that it has insufficient holidays. I think the problem with the rationality community today is that it has insufficient accomplishments that can justify holidays.

If one looks at the actual mythology of holidays in major religions, they're not invented, they're earned. Christmas was earned by the 3 magi making the journey from the East to visit the infant Jesus. Easter was earned by the son of God redeeming his earthly body for the sins of mankind. Passover was earned by Moses choosing to confront Pharaoh and telling him to release the Israelites. Diwali was earned by Rama making the journey to Sri Lanka, defeating Ravana, and securing his bride, Laxmi.

Secular holidays, too, have to be earned. Memorial Day was earned by the sacrifices of the Civil War. July 4th was earned by winning a war of independence against Great Britain. Labor Day was earned by the struggle that working classes endured to gain the right to bargain collectively. Thanksgiving was earned by the starvation and cold that the Puritan Pilgrims suffered in their first winter in the New World.

To put it another way, holidays are justified as commemorations and celebrations. So, before we ask ourselves what holidays we ought to create, we ought to ask ourselves, what have we done, as a community, that is worth creating a holiday around?

Replies from: JacobKopczynski, JacobKopczynski, Bucky
comment by Czynski (JacobKopczynski) · 2020-05-06T16:58:53.863Z · LW(p) · GW(p)
Memorial Day was earned by the sacrifices of the Civil War. July 4th was earned by winning a war of independence against Great Britain. Labor Day was earned by the struggle that working classes endured to gain the right to bargain collectively. Thanksgiving was earned by the starvation and cold that the Puritan Pilgrims suffered in their first winter in the New World.

Also, half of these are serious stretches. Labor Day was 'earned' only by the need to supplant May Day to prevent expression of Communist sympathies/solidarity. Memorial Day celebrations, such as they are, never include mention of the Civil War in my experience. Even Thanksgiving's mythology is much more about commemorating the spirit of cooperation which helped the Pilgrims get through the winter than about the difficulty of the situation.

comment by Czynski (JacobKopczynski) · 2020-05-02T18:57:49.348Z · LW(p) · GW(p)

I don't think that's a fair description of most holidays, and it's definitely not a requirement for new ones.

The biggest source of inspiration I have used in my own designs is the Neopagan/Wiccan holidays; Beltane, Imbolc, Luggasnadh , etc. Whatever their proponents may claim, these are de novo holidays they assembled from whole cloth with only scraps of historical practice and misinterpreted or outright fabricated mythology to back them up. And, despite being entirely novel, they're sticky and genuinely emotionally impactful. In the whole eight points of the wheel of the year, only Yule (winter solstice) can really be construed as "earned". (Via the standard "we have brought back the Sun" winter solstice narrative. YMMV on how much this actually applies to celebrations of Yuletide.)

So given that, I don't think holidays need to commemorate things. And actually I think they generally shouldn't, because a commemoration is backward-looking. We are an extremely forward-looking movement/community, and our holidays should reflect that. Which isn't to say we should ignore the past - Solstice cares a lot about the past, and how we've dug ourselves out of a hole nature made for us, and this is good - but we should focus on the future and the present, doing collective values affirmation and trying to establish a tradition that can sustain us across a couple generations. Values I've considered include 'be ready for tail risks' ("Prepper's Night"), 'community connection' ("Day of Warmth"), 'humanity's potential is enormous' ("High Summer" or "Awe Summer"), and 'appreciate skill'/'share skill' ("Day of Achievement").

Replies from: gworley
comment by Gordon Seidoh Worley (gworley) · 2020-05-04T21:24:35.905Z · LW(p) · GW(p)

On the other hand, the holidays quanticle points out are "more successful" than the ones you bring up, in that they are holidays that have existed for longer and are celebrated by more people, so there is likely something going on where there is something strongly gained from making a holiday more earned.

Replies from: JacobKopczynski
comment by Czynski (JacobKopczynski) · 2020-05-05T22:34:26.281Z · LW(p) · GW(p)

That doesn't follow. Traditions are self-perpetuating; the more entrenched a holiday is, the more momentum it has to keep being entrenched. Therefore, an old holiday tells us merely that it had something major going for it at some point in the past.

comment by Bucky · 2020-05-06T08:23:06.143Z · LW(p) · GW(p)

Halloween as a counterexample? (Or possibly the exception which proves the rule?)