TL;DR we should bring back dotplans using Markdown and HTTP(S)

What's a dotplan?

.plan files are a holdover from a period of time when everyone who was on a networked computing environment could reasonably use a command line. They used to be a way of finding information about a specific user (like whether they were online, what their office hours were, and in the case of remote workers, where in the world they actually were). All a user had to do was run finger <user>@<host> and they would be given this information.

A famous use of the .plan file: John Carmack would use them to document what he was doing. In the days of Quake and Doom, there would even be websites which were a frontend to finger which would archive what he was doing, and this practice spread out of Id Software and into the wider Dallas game design bubble.

I think many people are sick of the always-on nature of modern communications.

Why can't we bring .plan files back?

What has changed since those days

1. Most networked devices are not a UNIX and do not have the finger daemon running, and even if they did, most routers sinkhole it from the outside world in the case of residential connections, and most sysadmins now frown upon running any unneeded services on developer machines.

2. Hosting a website is very easy. .plan files came around because it was easier to write to a text file for an already-running daemon to pick up than it was to stand up a web server with which to write text to. It's the opposite now. For example, this site is currently hosted on GitHub Pages, which is very easy for anyone with a GitHub account and a modicum of Git experience.

3. We have X the everything app now. Even John Carmack gave up his blog in favour of Xitter. The nice thing about this distinction is that you can keep your .plan file somewhere which requires a little technical skill to get at, so you can write about technical topics and anything else with much greater candour.

With this in mind, let's update the concept of a .plan file to the modern day.

Proposing: neoplan

I propose neoplan: a HTTP(s) dotplan file. Back to the future with neoplan. An anti-Facebook or anti-Xitter for people with technical chops. No images, no comments, no dogpiles, no algorithms, just text.

I used something similar in a previous job: we were all on servers running Apache on a VPN, and before Slack and Teams, we would put our office hours on index.html, along with what we were doing that day and where in the world we were working from.

Hosting

Stick it in plain text on <host>/.plan. Serve it over HTTP(s) in plain text. Write it in Markdown (but this is entirely up to you, even write it in Org Mode if you want). Update it in the text editor of your choice. It's literally a text file. You could even set your web server up to periodically copy notes from a Dropbox or something like that. It's up to you, but you will be constrained by it being literally plain text and that is absolutely fine.

Fetching

It should be accessible over HTTP(S) with a single curl request. The following should work:

curl <host>/.plan | $EDITOR

Alias this command to neoplan or something like that. It's just a curl oneliner.

Goals

I don't want this to become gentrified like many other technical works. I don't want a GUI for browsing .plan files. It should be down to the developer to make something that works for them in this respect.

I want this to be for technical people only. The only way it can stay this way is for it to be both complex for outsiders and ridiculously simple for insiders. Tough enough to be significant but simple enough to protect. I want the specification for this to fit in one sentence with a client in one line of Bash so that it becomes part of software oral tradition.

It should be a safe space away from algorithms, indexers, and AI-generated slop. While anything accessible over HTTP will be absorbed by the borg at some point, I want this to be something that developers actually care about enough to not want to ruin it for everyone.