Well, to really know how things would pan out, you have to indeed do the thing.

And it happens that people will procrastinate with preparation. But as someone who does not and is fast at doing things I can tell you that if I want something done really fast preparation will be a good chunk of a projects time. If I want to toy around and the project doesn't matter that much, just doing the thing may do the trick.

Now the advice on that site aims to get people started with the actual work part and tries to do so by telling you preparation doesn't matter. That is not true. But it is true that people prepare forever because they are afraid to cross the threshold where they start translating grand ideas into concrete, protentially flawed reality. That means actually good advice would deal with the question where this fear comes from, how to address it and how to notice you have been preparing too long.

If you want to build a house, drawing the plans and finding the right location where to put said house is part of building a house. Or you could just mash bricks on top of each other in your backyard and then realize you forgot the foundation.

Yes. Also, attempting to do the thing and failing is indeed doing the thing. (although you may only know in retrospect)

That's right, you're not producing paperclips unless one minute you have metal wire and the next minute you have a paperclip.

Oh you are absolutely doing the thing in this situation, as in you're in the process of doing it, somewhere between 1% and 99% of the way there depending on the context. You haven't done the thing until you actually fully do the thing, it is true. But most of the work could be in this preparatory meta work, and of course it means nothing without the final all important step of doing the thing, but there you have it, doing the thing before having fully done the thing.

Au contraire! Almost all the time you spend doing the thing is actually doing the things required before the final completion of the thing. I've taken to calling this "the work before the work" and it's at least 90% of the work.

Example: I get asked at the start of a project to provide ModbusTCP comms mapping for a new control panel, so that the client can start integrating it into their SCADA system. It's just a spreadsheet, maybe 100 rows, how hard could it be? They need it right now, why am I telling them it'll take 6 months?

Typing the addresses and descriptions into the spreadsheet is 'the work', and it only takes an afternoon, but it can't start until we do the work before the work:

- To document the ModbusTCP mapping I need to the PLC program

- To finish the PLC program I need the electrical drawings

- To finish the electrical drawings, the electrical engineer needs the device list, datasheets for all the devices, and the functional spec

- To finish the device list and functional spec, we need to agree with the client exactly what we're building and what it's meant to do

None of these things are 'the work' but all of them are 'the work before the work' and usually nobody wants to do, to wait for, or pay for this work.

You can’t know in the moment whether the thing you are doing is doing the thing. In retrospect alone does it become clear.

Business school is a great way to postpone doing the thing...

You have no idea what my thing is and whether or not I'm doing it via this post on HN.