Most startup theory is ex-post, therefore bs • Hacker News Board

Re: Most startup theory is ex-post, therefore bs

#121

Post by _ptgt » Tue, Aug 20, 2019, 9:09 PM UTC

tlb wrote: post #2

Not all ex-post theories are BS. For example, I slam my hand in the drawer and theorize, "Slamming hands in drawers hurts; one should not do it." That theory is pretty solid. It doesn't cover all the edge cases, but following it is better than not following it. There are lots of startup theories that are equally solid, despite being ex-post. I know a startup that didn't pay payroll tax for a couple years, and it ende…

The difference is the prior.

There's strong prior reason to believe not paying payroll tax is not going to end well. The theoretical basis is simple, clear, well-reasoned, and supported by strong evidence in adjacent realms (i.e. the IRS takes taxes taxes pretty seriously).

Whereas the prior on the validity of something like "early-stage founders should live with their customers" is much weaker.

Even though the advice for each presented example is based on sample size n=1, the resulting belief can and should be quite different depending on your prior.

The real meta-advice is to be able to understand and evaluate your prior for any given piece of advice you receive. This mostly comes down to thinking rationally, being knowledgeable, and critically considering each piece of advice for whether it's worth heeding or not.

Which is not so far from the author's message I think ("figure everything out yourself", "I don't mean to say don't take any advice").

(Obligatory shout-out to Bayes Theorem)

Re: Most startup theory is ex-post, therefore bs

#122

Post by downandout » Tue, Aug 20, 2019, 9:12 PM UTC

Ma8ee wrote: post #50

I can't help thinking of this wonderful talk about how to win the lottery: https://youtu.be/l_F9jxsfGCw

That's actually a pretty great video. The tldw version: advice on how to buy more lottery tickets is useful; advice on how to win the lottery is not.

Re: Most startup theory is ex-post, therefore bs

#123

Post by owens99 » Tue, Aug 20, 2019, 9:15 PM UTC

zaind wrote: post #115

varjag wrote: post #96

Earlier quoted context omitted.

This extends to methodologies. At this point am convinced that a competent team will complete the same project under formally waterfall, XP or Agile governance with only minor schedule differences.

I'm a huge fan of Alistair Cockburn's thinking on this point. He argues that people - and their characteristics - are a far greater driver of project success than methodology. The argument is not a formal one, but I think still holds a lot of weight: https://ameyakarve.wordpress.com/2012/07/22/characterizing-p...

I agree with this for the most part, though there are many false positives of “great teams” on paper who screw it all up purely because of a backwards methodology.

Methodology and people are not mutually exclusive. And it’s a lot harder to swap in and out people (though it may be necessary).

Re: Most startup theory is ex-post, therefore bs

#124

Post by travbrack » Tue, Aug 20, 2019, 9:18 PM UTC

Kind of ironic that this is advice for startups.

Re: Most startup theory is ex-post, therefore bs

#125

Post by rmason » Tue, Aug 20, 2019, 9:25 PM UTC

vikramkr wrote: post #109

marcus_holmes wrote: post #106

Earlier quoted context omitted.

no, they failed because they failed. If she'd won some of her gambles she'd be a hero. My go-to ex-post example is FedEx, whose founder took their last month's payroll to Vegas, and won big enough to keep the business going. I was taught about that in business school (as an example of entrepreneurial spirit). Founders who do bad things and lose are convicted of fraud (and gambling with your employees' payroll is defi…

Tell that to the ceo of uber, who built a multi billion dollar company (even after the recent underperformance) but also did some bad things, and then got kicked out of his own company for doing those bad things. Success isn't a guarantee of your sins being forgiven at all

How many people do you know that can spin up an $76 billion dollar business from scratch? Damned few. His board let him make mistakes while providing little or no guidance.

I see Uber's problems more as a failure of the board to govern than solely the fault of the founder.

Re: Most startup theory is ex-post, therefore bs

#126

Post by rmason » Tue, Aug 20, 2019, 9:30 PM UTC

I did two startups in the nineties. There was little or no information out there and the Internet quite frankly was a brand new world. Or so we thought at the time ;Now there's too much information out there. Lots of it is total malarkey to be polite. If I had the choice I'd much rather have too much information than no information. I honestly believe that I would have made fewer mistakes.

Re: Most startup theory is ex-post, therefore bs

#127

Post by newswasboring » Tue, Aug 20, 2019, 9:43 PM UTC

ionforce wrote: post #86

kd5bjo wrote: post #70

Earlier quoted context omitted.

I’ve read a lot of time management/productivity books, and there seem to be only two things all of the authors agree on: write stuff down, and continuously tweak your system. The latter is usually mentioned offhand in the preface, as the author wants to deliver a fait accompli instead of just some tips and tricks.

Can you elaborate on "write stuff down"?

I keep an Excel control table. Every task assigned goes on that table with a project, type (report, code, algorithm design or integration work) and module (which part of the project needs an update). There is also a remark column which I one day intend to process with nlp. I also like to generate a pivot table at the end of each week to know which type or modules of work I'm neglecting. I've also got some conditional formatting to mark rows as red when I blow the deadline (red) or when I need to finish something by today (yellow). Has worked well for me when I was an engineer and I'm evaluating this work flow for an integrator (it's an official post for someone who makes sure that several teams deliver compatible outputs on time).

Re: Most startup theory is ex-post, therefore bs

#128

Post by hinkley » Tue, Aug 20, 2019, 9:46 PM UTC

monocasa wrote: post #23

hinkley wrote: post #8

I have had a couple of managers over the years who had dreams of writing a book, and with the exception of one, they were all mostly wrong about the secrets of their success. They didn't have any awareness of how their team and especially their leads supported them. Responses varied from humor to hurt feelings. Ask the people who work for them why the team was successful. They'll probably have very different opinions…

Would you mind expanding on their respective theses? I can see a lot of value in looking at these in a anti-pattern kind of way.

You know, it's funny for all of things I can enumerate on 'how not to do something', there are some pretty big areas where I either don't keep score or tear up the scorecard the moment I don't work there anymore. In this case probably the latter.

The guy I said did have it figured out? I'd like to retract that. He wanted to think of the process of moving stories from the backlog to done as a physics problem. As managers go he was fairly emotionally sophisticated, but that was his big blindspot.

Some of the others had elements of 'mushroom management' (keep them in the dark and feed them bullshit), and a lot of quick fix-feeling techniques. Those can be helpful in small doses, as long as someone else is beating a drum for technical excellence (not perfectionism, but get stuff done, keep it done, and keep getting stuff done).

Re: Most startup theory is ex-post, therefore bs

#129

Post by 8note » Tue, Aug 20, 2019, 9:46 PM UTC

sha_r_roh wrote: post #9

tlb wrote: post #2

Not all ex-post theories are BS. For example, I slam my hand in the drawer and theorize, "Slamming hands in drawers hurts; one should not do it." That theory is pretty solid. It doesn't cover all the edge cases, but following it is better than not following it. There are lots of startup theories that are equally solid, despite being ex-post. I know a startup that didn't pay payroll tax for a couple years, and it ende…

> "Slamming hands in drawers hurts; one should not do it." This is a fact and don't think this is what the author is trying to say. I guess he's more focused on the kind of theories which people throw around as some sort of self-evident truth, when in fact it's a personal truth which they learned through trial and error. But I guess there's a point in learning from some of these "self-evident" truths (this is more ra…

slamming your hands in drawers is also a personal experiment. I might be wearing a good set of gloves, and it won't hurt

Re: Most startup theory is ex-post, therefore bs

#130

Post by roynasser » Tue, Aug 20, 2019, 9:49 PM UTC

Advice and startup theory should be taken (as any advice) from anecdotal situations... In that sense, most theory will prbably not be BS...

Does it make sense to sleep with your customers if you are building a car? probably not... it may make sense if you are building something hospitality related (in case of airbnb they could have been thought of as "trying the service themselves")... Should all startup founders sleep with their customers? obviously not, but try their service? (probably!)

I think this post lacks sufficient depth to be relevant