[User23]

It's been done. In one of his books Feynman describes how the scientists on the Manhattan project would "estimate" their results as what they had accomplished in the prior quarter. It's a bit messy to get it started, but once you've got a cadence you can always maintain that three months of buffer.

Edit: Now I'm thinking about what kind of tooling I'd want to make the process seamless.

[phyzome]

I've had a researcher describe to me how a lot of times they have to submit a grant proposal for what they're already doing, funded by a previous grant. If approved, it will actually be used for the next chunk of research...

(I don't recall why, exactly. Something about how there's a chicken and egg problem, where they wouldn't have enough results to support the grant proposal if they hadn't already been doing the research...)

[atleta]

It had been done in sentence as quoted from the post as well ("hey, I did some research on X"). GGP suggested waiting for some time purely for psychological purposes. But yeah, I don't think it matters a lot after a while either.

If the management doesn't like it, they'll figure it out that if you keep coming up with these then you must be spending time on them - continuously. :)

[cutemonster]

> what kind of tooling

A private Git repo where all the work happens, and a public repo where everything is 3 months old, but history rewritten so the dates are up to date?

And a cron job and scripts to automate it all

[rubyist5eva]

In agile/scrum isn’t this just measuring velocity?

[killingtime74]

Use a second ledger, like the mafia