gofreddygo

The unreasonable difficulty of predicting (actual) future effects based on past effects with no regards to their cause.

in a decade, we're all going to say how developers job is so different and yet still the same as it was 10 years ago.

hobby projects I still code by hand

So true. I was too blinded to not notice the joy of my hobby getting sucked away by not doing it by hand.

"How can we dig a well here" and "should we dig a well here" are neither mutually exclusive nor sufficient in their own regard.

Cant answer should we without having done a lot of it before. Cant learn unless you actually do it and see the consequences through.

For the producer, a nominal (at cost) fee is worse than an exorbitant one. better to have a few high paying customers first then drop the price inviting more than the other way around of charging nothing first then going up

screens distract everyone. I avoid them. in meetings, I use a piece of paper or a whiteboard. Everyone can see what is being written on it.

on that note, the small whiteboard right next to my desk (at home) was a great idea too

my good notes have their origins in thoughts I had while in the shower or while driving or watching a movie or a conversation with someone close to me. Ideas evade me when I'm at my desk in front of my notebook with my awesome pen. This was a huge problem I never noticed that my notebook is never there for me when i need it the most.

the second problem, because its my original thought, I'm naturally attached and biased to my precious thought in that moment. The same thought could feel dumb or obvious or wrong next week. But right then it feels like I've stumbled upon something so insightful. So I need to write it down to revisit.

I've now distributed manila folders with 5-10 blank college ruled sheets of paper across different rooms in my home. I'm never too far from a blank piece of paper.

in the moment, I write it down on a sheet. Fold it and put it in my pocket. Its with me the entire day to review, add on or scratch off. It ends either in the bin or the binder at my desk. As the binder gets thick, i trash or consolidate to keep it thin. The really insightful ideas as far too few for a $2 binder.

Works way better than any other form of capturing insights. As i look through, my binder has so many backs of mailer's i've received and scribbled on. cant figure why.

Oh that is such a bad heuristic ! The commits and history of how a PR was put together is no indicator of the quality of the PR or the thought process that led to it. Thats the equivalent of rejecting a (handwritten) essay for having too many corrections. ridiculous.

The code is all that should matter. Maybe comments for being nice to others and my future self. thats it.

Ahh.. gotcha. My bad.

That is diametrically opposite to how I read non-fiction. Its designed on how I approach a new codebase effectively.

first is skimming and context. table of contents, the preclude, intro etc. Then author (interviews, actual work done, other books, friends, enemies) to build a picture of who it is that's talking and what the topic is. 70% books fail this filter.

then the book itself, i usually pick 2 (sometimes 3) books on the topic fed by the research I just did. and i'd build a picture of the main points. skimming through the pages to see what the structure is and what stands out for all books.

now comes the actual reading. i now start reading in chunks. that make me question things, clear out my understanding and the picture of events, groups and people involved in the subject. this is the most time consuming and usually where i lose interest. i make notes in a few index cards or binder leaves or a text file for when I want to revisit.

this format of consumption has proved vastly superior to reading books start to end. I can usually see where the author is biased, pushing agenda or trying to please or oppose some one. and this works amazingly well for old texts (art in 1500s, finance in 1800s, history, programming languages, etc)

yes and that led me to a more fundamental question : is it even possible for an easy way to do fine grained adjustments. The finer granularity brings complexity. that makes it unreadable (especially for the untrained) and hence set aside and forgotten. The other way is to narrow down and focus on a well defined subset of problems.

either way the number of people willing or compelled to learn it will be tiny. and it hence becomes a niche, perfect for a long-term side project but with no real return.

my conclusion was to stop looking and use D3 or custom code ( good looking charts for humans).

this project is trying to do the same, balance verbosity in text with granularity in the charts with a narrowish usecase : agent consumption.

Graphviz and mermaid are a shitshow.

anything more than their own handpicked examples and you're better off using d3 or yfiles. layering, clustering, boundaries, rearranging are all basic needs for text to diagrams. None support it.

Both suck at being any good for rendering diagrams from readable structured text. There is a gap to be addressed.

Like area 51.

recently read a post on a moms' forum about her having a hard time getting her kid to practice violin and the skill plateauing.

the breakthrough was a trip to a relative's remote house with no screens / internet. Kid found violin better than other things to do. More Hours led to improvement. That fed on itself and kid spent more time with the violin and improved much faster.

The busier our lifestyle, the more cheap, meaningless pursuits we engage ourselves in, it all chips away at out natural ability to focus on one thing for long term and we're left with a recursive livelock of infinite choices laden with hidden opportunity costs.

how can humans from 1500 with everything stacked against them have better art sense and science insight than humans in 2026 ?

How to ask for help is an under-discussed important topic.

Women are way better at asking for help and getting it as compared to men.

there's some very well-known psychology tricks that you can use to get better outcomes. Cialdini's book and Munger's speech on psychological bias are first class.

you want to be opportunistic but not too attached to your ask.

ask for a lot and then back off works tremendously better even with kids

perception (looks, reputation, stereotypes) have an outsized impact on outcomes. look at any successful sales department for proof of this. perception can be faked unfortunately.

put hard limits (of time) on your ask. make it clear that its a one time ask.

The most counterintuitive aspect (covered in the article) is that you'll get a yes more often if it's easy for them to say no.

The programming language is the shared vocabulary

Everything above are made up leaky abstractions with a handful of exceptions

Its valid but useless.

Think someone says i'm thirsty all the time because there's little clean water available, what's available is expensive but it could be better if we did so-and-so.

and someone replies I'm not thirsty.

This is true in broader contexts too. Bunch of experts can't agree on something fundamental which is hard to prove/ disprove, and they have strong opinions on the topic.

AI is much worse.

Its not _just_ that. Orgs aren't remotely sensible at measuring anything that isn't counted in dollars.

employees who are on the ai bandwagon are there for the free management attention.

Management is cooked because the damn market is hard, money is tight and they can't afford to fight the top down love and $$$ thrown at AI.

If you zoom out, all the real money spent on energy to keep AI alive isn't going to be held in nvidia stock for too long. it will burst, but its stupid to time it.

we are consultants

This is the key insight. Design patterns were developed by a set of consultants. Promoted by other consultants. Consultants have perverse incentives, like bankers.

Realizing this made me critical of the design pattern kool aid. I've come to terms that these are going to be around longer than I'm going to be employed writing code. i keep the criticism to myself and avoid them when i dont see fit. Works ok.

As Hoare said:

    There are two ways of constructing a software design: One way is to make it so simple that there are obviously no deficiencies and the other way is to make it so complicated that there are no obvious deficiencies.
  The first method is far more difficult.

There are two ways of constructing a software design: One way is to make it so simple that there are obviously no deficiencies and the other way is to make it so complicated that there are no obvious deficiencies. The first method is far more difficult.

That is so real. Brilliant !

For months, Employees had the option to choose claude code or copilot. Now they dont.

Underlying model choice still has no restrictions. Opus 4.6 is by far the most popular. there's still big $$$ bills going anthropic's way.

This got me thinking, so what happens in two years?

every tom, dick and harry who can type english has the tools to attack any software that isn't patched.

tools that were accessible to specialized groups, now made available to anybody with a grudge and a few dollars for tokens.

and what does anthropic and openai do? They form an inner ring to make the latest models available first to Enterprises. Enterprises will cough up the prices that anthropic and openai set, they have no choice here. e

Eventually everybody pays. This does not sound good

Software today is worse in every possible way. Subscription solicitation is one such dimension.

BBEdit, Sublime et al. are beacons of what software quality, distribution and pricing ought to be.

three quarters of the saas industry is built around such made-up needs. Not much to be proud of there with a handful of exceptions.

as for price, it feels 100% fair to bleed your enterprise customers to subsidize individual customers.

I keep them embedded in the codebase or an artifact right next to the source.

And the key thing is that i dont need too many details at all. A few cues and its all back in my head.

I very much dislike this behavior, personally dont do this and want it stopped.

people that I like to be nice to have done this to me. I dont have a good response that wont trigger them.

They chose to do this cz its easy. I now have to choose : read that garbage and make sense and ask questions or reject it asking for your opinions not claude's.

Former leads to more walls of text. Latter makes me come across as not-nice and get broad stroke painted as an AI hater (which I'm not). Not many where I work will voice any of their discontent about anything AI. Its the hotness.

The irony of that article having an AI smell is not lost on me

yes I don't like the term investor anymore, because there aren't anymore. in my mind I just switch the word to a compulsive gambler or at best a speculator. and everything else is just bots.

exactly what else do you expect them to do?

They can see peers cheating the system using AI to get ahead, future job prospects, directly affecting time to pay off student loans are being crushed by the AI narrative which is a reminder of how the tuition money is never coming back

and then to have someone come in on commencement day and sing praises of AI just totally shows how tone deaf, blind, and off track the college system really is

related [1] Glendale Community College's screws up names as students walk up to the stage on graduation day. Blamed on AI

[1] https://www.azfamily.com/2026/05/19/ai-system-fails-during-g...

It helps to think of investors in tiers. The lower tiers mimic higher ups. Each tier has two orders of magnitude deeper pockets than the lower tier.

At the very top are the big investment banks and fund houses, berkshire. Second are smaller institutions and third retail/individual.

The top two layers demand a steady return, never losing money on average in any 36 month window. Otherwise it triggers a selloff top down to cover for it.

The bottom follows the top so the selloff or buy just gets mimicked, with the top tier never losing (the bottom layers make sure of it by following blindly)

With wild indicators already set a massive selloff should have already been in motion, but its not. The top tier is getting more greedy.

No one is betting on AI long term. Everyone's in for the ride. As always the bottom will feed the top.

ash trays give them a safe place to put it out

ha. i always thought they were remnants from old airplane plans that were too much effort to update to remove them. thanks for that