Do the Real Thing

Disagree, strongly.

Gonna learn to lift? Read starting strength. Set yourself up for success by structuring an environment for habit formation. Take the steps to ensure you are getting good nutrition before you start. Don’t just run to the local gym and staring trying to mess around in the squat rack.

Going to build a new app? Get up to date on tech stack choices so you don’t make a dumb decision. Look at prior art and understand the work that has come before you can learn from.

Trying to learn a new skill? Do some research to find out what the best, most high quality resources there are first. Determine what level of mastery you’ll be happy with before diving in unprepared to know how to assess opportunity costs, which can

This doesn’t mean get stuck in analysis paralysis. I feel pretty strongly those who give advice like the advice you mention do not think it’s possible to avoid this obvious trap. It is if you are disciplined and have enough self awareness to know when you have hit the point of diminishing returns. Literally one day of focused thinking can outflank a week or more of pointless execution, even at the beginning. Thinking is underrated, execution without preparation is overrated. The most effective people know how to prepare the mind for execution and not get bogged down doing so. Taking an immediate first step can lead to a bad first impression: a bad lift, a mess of a prototype, or a garbage online course. That can be hard to unwind, since it forms busted initial mental models - and can undermine long term success. There is no hard rule, but aversion to pre-emptive research and analysis is an anti pattern unless you know yourself to be unable to do so.

My takeaway from this is to always try for the thing that is just out of reach.

Doing what's comfortable isn't going to lead to growth. At the same time, as others in this thread say, doing things that are radically out of reach is too likely to fail without a good foundation.

Learning is in the struggle, so find something that is a little harder than you think you can handle.

I’d say they weren’t doing what I suggested (that could still mean my advice is bad).

Before they actually started climbing the rock, they had to do a bunch of other stuff. They at least got their hands on some gear, probably drove and hiked. And this is partly what gets them into trouble.

Fooling yourself into doing something unsafe generally requires a bit of work. I think there’s a safety in doing what’s immediately around you. For most people, it would be really hard to kill themselves rock climbing in the next 15 minutes, but very easy to do it next weekend.

My hope for someone doing what I suggest would be something like “I’m gonna go hang from that pull-up bar / low beam / jungle gym on and off for 5 minutes.” Then they research how to actually start rock climbing safely.

...not if your crappy prgrammer art is the only thing standing in the way of shipping a game that could actually sell.

There are people on both sides and they need opposite advice. The overanalyzers may be overrepresented on HN, and maybe the mindless doers are more in the general population.

Still, just as in startups the minimum viable product is important, you should get to the point of a simple but complete thing.

Think of it as breadth first, not depth first. You want to go through all main stages, see a simplified full experience. Then you can tweak each part. Instead of reading and never finishimg Starting strength and researching the best squat shoes and the other details, just go to the gym, do something lightweight and something you understand, then go home, feel some of the burn. You just had a full gym experience, a simple one, but full.

I recently made a website I've been endlessly procrastinating on. I was paralyzed by choice of static site generator, template style, hosting provider etc etc. Then I decided to go ahead and make a simple version. A single HTML file and a single CSS file, like in the old days, no framework. It is a complete website. It's alive on Github pages. I can now tweak it in many ways. If it becomes tedious, I can switch to a site generator. The main thing is, I've written the bulk of the content now.

Take the time to reflect and research every once in a while, experiment with different setups and ways of doing things, sure. But going ahead and doing an imperfect but functional first version is hugely important. Writers say you should write a draft in a totally non-judgmental way. Consciously accept it will be bad. But often there just isn't a way around it. The road to good involves passing through the not so good, so you see the feedback of reality, you can update your mental model of how this whole thing works so that the next thing you read already has a place to go in your mind. You don't know what you don't know. You know what you need to look up once you have a bit of experience. The thing gets demystified.