Sven Slootweg (soft-deprecated) @joepie91@pixie.town

@arubis @eniko Sorry, I should've probably been a bit less concise/flippant there. Specifically the problem I see time and time again is that the people claiming it "makes them more productive" are developers who hold no responsibility for how well the end product works, and therefore are not factoring the externalized costs (and the costs of fixing the problems caused by it) into their assessment.

And on the other side of the wall that their code is being thrown over, there's the team leads, senior devs, etc. who are expected to fix (and in some way 'pay for') the results of insecure, unreliable, unmaintainable etc. code, who seem to pretty much universally feel that LLMs are *harming* the overall productivity of the team. Meanwhile there seems to be no objectively measurable evidence of LLMs increasing productivity, just a lot of marketing puff pieces.

So as far as I can tell these claims of 'productivity' are purely marketing, aimed at those who can afford to ignore the externalized costs.

@hazelnot @schratze @gsuberland@chaos.social Yep, this is why I use a physical planner too

@arubis @eniko But is it though? Because the last bits of research I've seen on this don't really support that notion, to say the least, and it seems to be mostly just Microsoft (and developers with no responsibility for the end product...) claiming that it is.

@eniko This framing reminds me a lot of https://xkcd.com/1425/, except it being weaponized to sell people bullshit hype... 😐

@hazelnot A variant of this, that works especially for lower 'chore loads', is to not keep a 'todo' list but only a 'done' list, where you track per day what you have done that day, but there's no queue of stuff to do. This mainly helps to fight the sense of "I've done nothing today" (which very often is your brain playing tricks on you and not actually true) by keeping an explicit record of achievements.

This doesn't work as well when many external obligations are placed on you, though, as it will not help you ensure you're not missing anything.

@hazelnot The important part here is that the end of a day's todo list is the point where I no longer have an obligation (to myself *or* others) to do any further chores; I have done my part for today, there is another day tomorrow. In other words, I know where the end of the work queue is, so it doesn't feel perpetual.

@hazelnot @schratze @gsuberland@chaos.social I currently do this in my... planner? ("agenda" in Dutch), where for every day that I'm not intentionally keeping free of chores, I write down something like 4-5 chores that I still need to do.

Then each day I look at the entry for that day, and therefore the todo list for that day, and do those chores in one go (where possible). Once I've done them, I am done with chores for the day, and the rest of the day is free time (modulo work and such). I tick off each task as I go.

If I *don't* manage to do all of them in one day, or I run out of spoons before finishing all of them, then I just take the ones that haven't been ticked off yet, and find another spot in my planner where there's still space, and re-schedule them for that day. That's not a failure to me, just a change of plans.

Normally it's more like maybe 1 or 2 chores a day, the 4-5 right now is because I have a crapton of things to prepare for my transplantation, and so my days are mostly chores for a while. The important part is that the chore selection for a day is a genuine estimate as to what is possible with your executive function, so *not* just cramming in as much as possible. If the
answer is one thing a day, then the answer is one thing a day.

Sometimes I'm feeling particularly motivated and pre-empt some chores from a later day but that's very rare.

For large chores that span more than one day, I break them up into smaller chores that can be completed in like 1-2 hours each, and then schedule those parts individually (on different days, usually).

@hazelnot @schratze @gsuberland@chaos.social For this sort of thing I've found it easier to just read it immediately when I encounter it, instead of filing it away - often the "having to remember to look at this" ends up costing more spoons than just assessing it immediately (and only marking it for later processing if it looks helpful). May not work like that for everyone though.