Ask HN: How do you bypass, eliminate, reduce friction when starting a project
I feel that I'm out in the yak barn shaving yaks about every time I start these. I have projects "on the books" that I've done the documentation on what I need and after awhile it just gets moved into the too hard pile.
I try to pre-plan and figure out what I need so I don't get to the point that I need a discontinued chip and need to spin up a foundry in my basement. :-) I get parts and file them away by the project. Side effect is I have enough parts to reopen a half dozen Radio Shack stores for staged projects.
I document everything, since task switch could be days / weeks apart. I use a local wiki to make it a little easier since I can cross reference what software/hardware similar projects need. (But one of the projects is updating the wiki software, that will break some of the features I added, so I'll need to fix them as I go)
But there is still so much other software friction going on. My other stumbles come from projects that I'm trying to include that are multi-lingual (lets mash C, Rust, and Python together??). Library gathering has the nuance of make sure everything is on the same level.
None of these are For$ale projects, they are allegedly something that will be fun, but it's like the fun gets sucked out before there is any results.
So hackers, what are your secrets from not getting stuck in the startup friction and staying out of the yak shed? How do you manage having 5-7 inprocess projects and keeping where you are documented. Thanks.