This seems insightful. I was going to comment elsewhere that some people seem to find structured systems really bothersome, whereas I personally just do not. But I am an INTJ, if anything. The idea of not taking a systems approach feels incredibly foreign, to the extent I can’t even really imagine where one would start.
How do you approach improving your personal outcomes related to ADHD without developing a system?
The short answer is that I try to maximize the strengths and minimize the struggles of ADHD.
Some practical examples:
- I have a job without any long-term projects (only long-term goals), so I never get bored from my tasks.
- I keep my evening schedules clear so I have room to be spontaneous in my hobbies after work. This is prabably made easier by being an introvert as I don't have the need to socialize often.
- I nurture an attitude toward my hobbies which is that the process in itself is always the goal. That it is okay to not finish a project as long as I keep learning new things.
- I depricate any stressfull commitment which is neither important or meaningfull long-term.
Also medication helps to overcome the menial hurdels.
As an ADHD person, this app looks like a repackaging (with nice design) of all the stuff I’ve built up over years - habit tracking, daily/weekly/yearly reflection, detailed task management, etc.
This isn’t for me (because I’ve already built a system that works), but this looks like something that would be very useful. For the target user who does feel stuck and hasn’t successfully built their system, this looks like a phenomenal product.
I appreciate the emphasis on self-reflection and perhaps the implied focus on continuous improvement.
Over the last few years I implemented a weekly self-review + planning practice (think solo agile retrospective), and my life has been on a steady trajectory of improvement since.
Edit: commenting on the product concept, not the company, pricing, or concerning tracking practices.
Glad to hear you've built a system that works for you! We've also heard from a lot of our beta users that they've tried to cobble together something similar, and a lot of their feedback and ideas is what we used to build this initial version (in collaboration with our Research Lab to integrate the latest methods too). Many of them weren't able to push their self-built systems over the finish line or maintain it, due to ADHD challenges though. Our goal is to build a flexible enough system that it can be adapted for various learning styles (in practice we're still far off from where we want to be) and continue building agents on top of it that make science-backed exercises and methods more accessible. A lot of the best practices are currently gated behind long textbooks and scattered PDF worksheets so I'm really excited about making this more accessible. For example, this week we're working on an "energy accounting" agent that's widely used (in varying formats) across ADHD practitioners that many ADHDers know they want to do theoretically but haven't found the way to follow through on it.
I love the weekly self review and planning practice you mention; I do a similar one with myself and my co-founder each week and have started moving that process into Indy recently!
Here is my template from obsidian that I use for my weekly reflection - customize reflection based on your values and priorities. I have goals to improve work-life balance, social connections (social isolation was a factor driving poor outcomes, and through deliberate consistent effort I have solved this problem).
Hyper scheduling: https://dev.to/maxpatiiuk/series/32301 (I stumbled upon this and implemented a form of it, although mostly I just like the colors in my calendar)
- [ ] Review year compass
- [ ] Review journal entries from the week
- [ ] Review last week's reflection
## Quick summary
> *Headline for the week*:
## Basic planning
- [ ] Set up outline of the week in Outlook
- [ ] Plan a fun weekend activity:
- [ ] Plan to visit one new restaurant:
- [ ] Plan one meet up or social activity:
- meditation (I combine with a fancy LED face mask to help reinforce the habit via my desire to combat wrinkles and acne - the cryoglow is better at acne than wrinkles so far)
- exercise (you can add notes)
- evening leisure time (if I don’t have dedicated leisure time, I end up revenge bedtime procrastinating/doomscrolling)
- stretching (there are two simple band stretching exercises that solved what I thought would be life-long neck and shoulder pain)
When I really struggle with productivity, I find the pomodoro system is a good bootstrap, and TickTick makes it easy to start. I like seeing the pomos on the built-in calendar.
I have a ton of respect for your approach. That said, as someone without ADHD, it seems somewhat odd that an inability to kick off executive function would be well addressed by adding an additional activity that requires executive function. Like, if I had to plan my day out with this document before doing things, I think I'd grow to dread the process, and be even more stymied - i.e. if it was hard to go clean the kitchen, why wouldn't it be hard to go write my dayplan?
Yet, I do hear this sort of thing works for people. I'd love to know more about what you experience and why this helps.
I have ADHD & use a very similar kind of daily notes template. I think the reason that it works for me at least is that it shortcuts some of the "not knowing where to start" executive function issues at the start of the day. I only need to remember one thing: open a new daily notes template. The checklist itself chains other good habits off the back of that, reminding me to e.g. make sure I've taken my meds, make coffee, put music on, think about what I've done recently & if there's anything to pick up again or dive back into & what I'm going to do for the day etc.
Yes, it can feel like a chore at times (especially forcing myself to do more in-depth periodic reflections) but without a system like this, I quickly find myself completely rudderless & my mental health & productivity both collapse pretty quickly.
You'll actually hear quite frequently that people diagnosed relatively late in life went under the radar for so long exactly because we developed our own massively over compensating systems like this & can appear (externally at least) to almost have the exact opposite of "executive function issues"!
TickTick gives me reminders to do some of these things - daily reflection is a “habit” and weekly reflection is a recurring task. That helps me not forget. There’s still value in doing the daily plan regardless of whether I do it right after I wake up (things are going well) or if I do it many hours later when I realize my day has not gone well and I want to get back on track.
The calendars and checklists really help with not forgetting things, and getting back on task once distracted. I can have 100 adhd moments; the system can’t prevent that, but it can help me find my way back to shore when I’m lost at sea.
I don’t really struggle with kitchen cleaning, but sometimes I do set out to clean the kitchen and end up folding laundry or scrolling instead. I can do that but if I haven’t checked off “clean the kitchen”, I know to come back to it.
The weekly reflection is a chore but I set aside time for it, and I keep doing it because it works. I can spend an hour doing deep reflection, or I can rush through it, there is value either way. It is really just a check list, and check lists are very ADHD friendly.
Separately, I find that physical and mental health improve performance regardless - so a system that improves these factors _is_ a system that helps with ADHD. Getting to bed on time, and building the system that produces that outcome, is an ADHD-friendly system. I struggle with this, but I try to get better over time by experimenting and adjusting.
The reality is that the system will not solve ADHD problems, the system is just a tool - you still have to do the work. Same for the app that is being shared.
> Look at my drafts that were started within the last three months and then check that I didn’t publish them on simonwillison.net using a search against content on that site and then suggest the ones that are most close to being ready
This is a very detailed, particular prompt. The type of prompt a programmer would think of as they were trying to break down a task into something that can be implemented. It is so programmer-brained that I come away not convinced that a typical user would be able to write it.
This isn’t an AI skepticism post - the fact that it handles the prompt well is very impressive. But I’m skeptical that the target user is thinking clearly enough to prompt this well.
Since LLMs were introduced, I've been of the belief that this technology actually makes writing a *more* important skill to develop than less. So far that belief has held. No matter how advanced the model gets, you'll get better results if you can clarify your thoughts well in written language.
There may be a future AI-based system that can retain so much context it can kind of just "get what you mean" when you say off-the-cuff things, but I believe that a user that can think, speak, and write clearly will still have a skill advantage over one that does not.
FWIW, I've heard many people say that with voice dictation they ramble to LLMs and by speaking more words can convey their meaning well, even if their writing quality is low. I don't do this regularly, but when I have tried it, it seemed to work just as well as my purposefully-written prompts. I can imagine a non-technical person rambling enough that the AI gets what they mean.
Thats a fair counterpoint, and it has helped translate my random thoughts into more coherent text. I also haven't taken advantage of dictation much at all either, so maybe I'll give it a try. I still think the baseline skill that writing gives you translates to an LLM-use skill, which is thinking clearly and knowing how to structure your thoughts. Maybe folks can get that skill in other ways (oration, art, etc.). I don't need to give it essays, but I do need to give it clear instructions. Every time it spins off and does something I don't want, its because I didn't clarify my thoughts correctly.
Setting up SpeechNote with Kokoro is one of the the best things I've ever done.
I can speak faster than I type, and the flow state is much smoother when you can just dump a stream of consciousness into the context window in a matter of seconds. And the quality of the model is insane for something that runs locally, on reasonable hardware no less.
Swearing at an LLM is also much more fun when done verbally.
The prompt the user enters is actually not the prompt. Most agents will have an additional background step to use the user's prompt to generate the actual, detailed instructions, which is then used as the actual prompt for code generation. That's how the ability to build a website from "create a website that looks like twitter" is achieved.
My 85 year-old father could probably resolve 90% of his personal technology problems using an LLM. But for the same reason every phone call on these subjects ends with me saying "can it wait until I come over for lunch next week to take a look?", an LLM isn't a viable solution when he can't adequately describe the problem and its context.
Yeah, we've already seen that over the past few decades. It's both a limitation and a benefit, but until recently it was the only thing we had (well that, and just hiring another person to act as an LLM for us). LLMs are an upgrade.
> No matter how advanced the model gets, you'll get better results if you can clarify your thoughts well in written language.
This definitely agrees with my experience. But a corollary is that written human language is very cumbersome to encode some complex concepts. More and more I give up on LLM-assisted programming because it is easier to express my desires in code than using English to describe what forms I want to see in the produced code. Perhaps once LLMs get something akin to judgement and wisdom I can express my desires in the terms I can use with other experienced humans and take for granted certain obvious quality aspects I want in the results.
> So far that belief has held. No matter how advanced the model gets, you'll get better results if you can clarify your thoughts well in written language.
I've heard it well described as a k-type curve. Individuals that already know things will use this tool to learn and do many more things. Individuals that don't know a whole lot aren't going to learn or do a whole lot with this tool.
I agree 100% - it's a very programmer-coded prompt. It was pretty much the first thing I thought to try.
I expect we'll see an enormous quantity of "cool prompts to try in Cowork" content show up over the next few months, which makes sense - regular non-programmers will benefit enormously from cookbooks and prompting guides and other tools to help them figure out what they can ask this thing.
It takes a certain amount of expertise to use LLMs effectively. And I know that some people claim otherwise but they simply aren't worth listening to.
Just because Claude Cowork is for "other" kinds of work, not just software engineering, doesn't in any way change that. It's not like other kinds of knowledge work aren't being done by intelligent professionals who invest time into learning how to use complicated software and systems
That is to say, I don't know who the "target user" of this is, but it is a $100/month subscription, so it's presumably someone who is a pretty serious AI user.
The notion that people who aren't developers couldn't figure out how to use this tool well, or be trained to, is a little too negative. Programmers aren't special snowflakes. Everyone with a brain is capable of describing a problem and breaking the solution into steps.
The popularity of LLMs proves this. That's how most people use them - building up a detailed prompt in steps, and learning how to put more detail in to get the result you want.
This is why I think (at least given the current state of AI code generators) that senior engineers will benefit more from AI than less experienced engineers. I don't know exactly what the chart of experience (on the x-axis) and amount of productivity gain from AI (on the y-axis) will look like, but I'm pretty sure it will be roughly (given suitable error bars around the input) a monotonically increasing function.
One part I like about LLMs is that they can smooth over the rough edges in programming. Lots of people can build pretty complicated spreadsheets, can break down a problem into clear discrete tasks, or can at least look at a set of steps and validate that solves the issue they have & more easily updated it. Those people don’t necessarily know json isn’t a person, how to install python or how to iterate over these things. I cant give directions in Spanish but its not because I don’t know how to get to the library its just I can’t translate precisely.
Also you may only need someone to write the meta prompt that then spits out this kind of thing given some problem “I want to find the easiest blog posts to finish in my drafts but some are already published” then a more detailed prompt out of it, read it and set things going.
Of all the lessons society failed to learn from 2020, the importance of clean air is perhaps the most disappointing.
I wear N95 masks on local trains, long distance trains, planes, buses (most of the time), and movie theaters when busy. The few times I haven’t, in particular local train (unfortunately, lots of unhealthy people) and Amtrak, I have gotten sick - strep and covid. Being sick is in some ways not a big deal, but is serious enough I will keep wearing the masks forever. I get sick far less often than my non mask wearing peers.
The tragedy is that the people operating these services - and schools and hospitals - should have installed filters and UV lamps to make this less unnecessary. At least planes have air circulation, the Amtrak trains are an absolute disaster.
Flu is spreading like wildfire right now. With the advent of these and other technologies, that is essentially an opt-in choice society is making. Totally unnecessary. You don’t have to stop many flu cases before a lamp pays for itself.
> I wear N95 masks on local trains, long distance trains, planes, buses (most of the time), and movie theaters when busy. The few times I haven’t, in particular local train (unfortunately, lots of unhealthy people) and Amtrak, I have gotten sick - strep and covid. Being sick is in some ways not a big deal, but is serious enough I will keep wearing the masks forever. I get sick far less often than my non mask wearing peers.
I used to get sick when I went into the city. Once I started commuting, it happened a lot less as I built up immunity. I'm not saying everyone should lick doorknobs for maximum health.
Anyways, I got to a lot of shows as well, and wearing masks is joyless. I wear them when I'm feeling sick, to reduce the range of transmission (although I just try and stay home)
That said, for institutions where there are sick and weak people, it's almost criminal that they aren't investing even more heavily in testing these sort of technologies.
> Once I started commuting, it happened a lot less as I built up immunity.
To give another anecdotal evidence: before COVID I used to catch 3 to 4 colds per year. Winter was basically a nightmare season where I was always living in fear of when I would get sick next. After COVID I started wearing an N95 in populated places. As a result, I went to 0 cold per years instead of getting sick even more often because of the additional virus in town. Now I feel I can live normally in winter without always worrying of getting sick and I always feel healthy.
I tried switching to a regular surgical mask (and in general being less careful) to try to find a good middle ground between cost, appearance and protection; while I did not catch colds I did get COVID at the same time as unmasked people around me, so I'm back to N95s (this was likely following something like hours and hours of continuous exposure so makes sense a leaky surgical mask did not prevent it). (my bout of covid was quite mild fortunately; but first time being mild does not mean future ones will be or won't lead to long-term symptoms).
Some people told me that wearing masks will "weaken" my immune system, I still need to see that; after two/three years I just feel healthy and this is refreshing after decades of getting sick all the time.
Plus, instead of the advice of "cook your own food to eat well, sleep well, do sports" that probably requires something like 28 hours per day with a standard-issue job while likely not being as effective as respiratory protection, putting a mask on takes only 30 seconds per day. That's probably good general advice anyway, but not the shortest path to solving the "getting sick often" problem.
I got similar results when i changed 3 things: making sure d levels were around 60, adding glycine and nac supplements and adding a high quality fish oil supplement. My 3 young children have been their usual trainwreck of sickness this winter and inatead of getting sick with them 80 % of the time i have had zero sickness.
i am not here to advertise any of this stuff so i am not going to link or even get the units or amounts right unless i remembered them off the top of my head, just here to point out there are two ways to not get sick: eliminate exposure or increase resistance. For some of us limiting exposure is a painful experience (i cant comfortably wear masks i feel like i am dieing slowly from oxygen deprivation the whole time and the extra moisture gives me a rash eventually)and there does appear to be options on the other path to limit our own sickness.
> I wear them when I'm feeling sick, to reduce the range of transmission (although I just try and stay home)
I really, really wish more people would do this. If you are feeling sick and need to go out into public, put on a mask. This is doubly true especially if you need to take a plane flight. I understand that you probably can't reschedule your plane flight for a lot of reasons, but, for the love of God, if you're hacking up a lung on an airplane put on a damn mask.
Does everyone know how often their peers fall sick? I’ve had maybe two illnesses in the last five years and I don’t do all this. In fact, I was masks off as soon as we were allowed to, traveled often via plane and train. I also was at many EDM shows as soon as they opened up.
I used to wear masks whenever I fell sick but it’s been a while. After the pandemic, though, everyone started using these as political symbols and I want to opt out of that personally. A thing I’d been doing for decades is suddenly some kind of statement. I’ve no problem looking out for others but if there’s enough schoolmarming over this, I might just not.
I think people are neurotic about this stuff but probably the reality is that others fall sick often and need to take these precautions and I don’t and therefore see no need to.
At the population level, if it’s worth it, then that’s fine but I think it’s not a given that these things are worth it.
I used to get sick a lot more often, pretty consistently 3-4 times per year. Since I’ve started masking on trains (specifically because others who are sick and coughing aren’t wearing a mask) and other things (elderberry, zinc, NAC at signs of cold) I’ve gone down to 1-2 times per year. And that is mostly from hooking up with people who had COVID and didn’t realize it.
It’s hard to say exactly what led to the reduction in illnesses, but the N95 masks have the strongest evidence base of the interventions I’ve tried.
That paper is about mandatory masking and social distancing at the population level. It does not speak to the question of whether it's "worth it" to wear a mask on the train if you're the only one who is doing it.
We also have found out that Covid decays less quickly in air with high CO2. So highly polluted areas and poorly ventilated ones compound the problem by trapping CO2 in and allowing the virus to survive longer in the air. That is very likely the case for other viruses as well as it impacts the aerosols.
This is a story about the risks of AI-induced brainrot. You get so used to having the computer just do your work, that the second you need to engage your noggin you’re lost at sea. Or at least just frustrated.
Reading and understanding the docs and reference material has always been part of the work.
Aside from the commentary it read like an advertisement for how great the swift/macos translation APIs are. PEBCAK
Gotta say as a Swift dev I agree—followed the link the to Translate docs and was pleasantly surprised to see a discussion section clearly explaining the usage, which is not always the case for Apple APIs! But this wasn’t really just an article about the API. It was about the complexity of trying to build on the stack of Swift/SPM/ParseableCommand/Foundation/Concurrency/Translation without having a good grasp of any of them. I was frustrated reading it, but I think it does point to the underlying knowledge that’s needed to be proficient at something like this. None of it is a particular indictment of Swift as an ecosystem (though there are lots of valid criticisms)-it’s just the nature of development and something that’s massively eroded by relying too much on these ghosts
The problem is there are a wide class of problems that you want solved but putting on the work will prevent you from actually doing the task because the cost isn't worth the reward. Because it's for a low impact tool. Or you can't imagine yourself dealing with this API again within a year or two by which time it will probably be completely different with v2 of the API.
So, you reach for AI and it works really well. So you start reaching for that more and more...
1. You can run decent local AI now - see /r/LocalLlama. You pay the electricity cost and hardware capex (which isn't that expensive for smaller models).
2. Chinese APIs like Moonshot and DeepSeek have extremely cheap pricing, with optional subscriptions that will grant you a fixed number of requests of any context size for under $10 a month. Claude Code is the bourgeois option, GLM-4.7 does quite well on vibe coding and is extremely cheap.
I remember reading and hearing similar rants from programmers 15 years ago, long before LLMs. The author kept going and figured it out, and probably got some pride and enjoyment from finishing the project in spite of the frustrating moments. That’s what learning to code has always been like.
How did these come to be associated with command-k and not command-p, which is the more common association at least for developer tools (chrome, vs code, sublime, figma, obsidian)? Command P makes sense since it is a Palette. Maybe it is just a weird quirk that all developer-facing tools use one binding while tools for the hoi polloi use the other.
Particularly offensive to try to name the whole concept after Slack’s default keybinding.
Half of the things you mentioned only came out after 2014. My Chrome wants to print with Comamnd-P.
But I'm not disagreeing in principle, I personally use so many of these fuzzy search things in so many apps, all with different keybinds (which is maddening in another way) - and maybe that's just the objectively most well-known one?
You’re right command-p prints in the common view, but opens command palette in devtools. That actually catches me out pretty frequently if I forget I moved focus.
Safari exclusivity is the only reason we aren’t living in a 100% “this site built for chrome” world. I think folks must forget the IE days and how bad that was.
There is zero percent chance developers are wasting a second making sure their sites actually work cross platform if not for iOS (and iOS more moneyed user base).
We were in a “built for Netscape” world right before IE had its brief window of innovation in versions 4-5. The fact that people were building to IE though was only painful for a few specific reasons:
1. the versions of IE targeted were exclusive to Windows (Mac IE was way different, so it wasn’t that useful for when the site had targeted Windows IE)
2. IE stopped all development of useful UI or web standards features, meaning if you needed the compatibility you were stuck with a stagnant browser
3. Due to #2, of course web devs hands were tied when it comes to adopting things like HTML5, <video> tags etc. Users would have needed to switch between the two constantly — Firefox for cool new sites and IE for their bank, school, government, whatever.
I would posit that none of the above seems true about Chromium. They do continue developing it, they add new web standards the most aggressively of anyone, and it’s available on basically every platform except the one Apple bans it from.
Mind you I don’t really want Google to own it, because they are way too damn big even without Chrome… but honestly it’s no IE situation.
It’s an impoverished view of what an OS is. A great OS is also providing APIs that enable a cohesive desktop, integrations that enable apps, and other core services.
For example, iOS has a built-in API for managing calendar events. That means my third-party to-do list can show a calendar view with everything from my iOS-managed calendar. Similarly, iOS has an API that enables third-parties to build apps that interact with Apple Music - so you don’t need to use the default client. Another example is focus modes/contextual computing - the OS enables configuring various settings in an automatable way. Another service is the health-tracking database - all my health apps share a common view of data, so my nutrition app can see my weigh ins, calories burned, and glucose levels - all coming from different devices.
On macOS, it goes beyond being just a launcher by providing rich file system services - for example, features to automate working on directories as contents change. It provides integration points to enable providing actions/services that work on text fields. It lets you configure dictionaries that support right-click/force press to look up. It implements a rich eMacs-inspired shortcut system in text fields.
Windows used to have APIs and integrations to support showing calendar events in the clock area. It was going to have a unified tab system that let you have windows with tabs provided by other apps (Sets).
In other words, an OS isn’t just a launcher. It’s a system.
> Too many bugs. Too many changes. Too little control. Windows 11's reputation might be at its lowest it's ever been as 2025 comes to a close.
That, sadly, also applies almost perfectly to macOS. And yet, as bad as macOS has become, it is still a distant third in the race to become the worst desktop OS.
Very difficult spot to be in when the entire industry is racing to make desktops awful.
Linux starts to look very interesting, but is held back by the lack of good efficient high quality mobile hardware - the only such devices, Macs and snapdragon x elite devices, have poor Linux support.
Do you have a trackpad? If you re-enter focus by right clicking (very easy to do with gestures), focus stops working until you leave and re enter by left clicking. Unfortunately not a new issue.
How do you approach improving your personal outcomes related to ADHD without developing a system?
reply