Via X/Vikhyat

The diagram uses the Cynefin Framework

One of the better blog posts I have read in a while. Going into depth about the motivation, and how to practice the “trade” etc.

Being an artist always seemed like “not-a-job”, and appeared to me like something fueled by motivations different than those animate people to hold a job or non-pecuniary reasons. This blog does a good job of uncovering that layer of ashes.

Another thought I have been mulling over the past few days is — You must develop your own unique view of the world. Find the “gold” seam, a unique perspective that gives you a leverage over one more affairs of the world. Art, original art is definitely a unique perspective of the world.

---

web archive link

AWS’s Marc Brooker offers a sober view of the evolving split in the road for software development career.

 The slow end of programming as an economic discipline, as weaving, ploughing, and coopering went before. It is reasonable and rational to feel a sense of loss, and a sense of uncertainty. With the loss of the craft comes the loss of the economic moment where that craft was valued beyond nearly any other.

Craft was valued beyond nearly any other — At least since the turn of this century, software jobs were the veritable gold mines of making over the top money, even if you were slightly ambitious, and worked your way into high paying jobs. That is already gone.

But new ideas, real transformative new ideas, remain hard to come by. And, as the lever gets longer, more and more valuable.

We haven’t seen this yet through the haze, and the dense fog of the present, but we need to be patient, and keep looking for that lever.

Nothing wrong in refusing the touch “AI” because there will be still work for those want to do it the “old way”.

The stubborn who stick to the old ways, and hustle to squeeze out the remaining economic value. That value will remain, because the world always changes slower than we would like.

Many fields only change through the death of the current crop of “experts”, and software is no different.

But one this is for certain — Software’s first act is over.

Congratulations… your days of whining are over.

In this room, we deliver success, we don’t whine.

Look, I’m not confused, I know you walk through fields of shit every day.

Your job is to find the rose petals.

Don’t come whining that you don’t have the resources you need.

We’ve done our homework.

We’ve evaluated the portfolio, considered the opportunities and allocated our available resources to those opportunities.

That is what you have to work with.

Your job is to manufacture success with the resources you’ve been allocated.

And yes – you have a hard job.

You only have 2 controls:

1. The clarity, culture, and energy you give your teams; and
2. Resource allocation.

And I want to be clear with you.

If you are in this room, you need to deliver outsized success.

To do that, you will need to allocate resources ahead of conventional wisdom​.

Conventional wisdom will generate conventional success and that won’t allow you to stay in this room.

You need to have courage and be bold.

And when you do that, you may fail.

BUT.

If you fail, I will back you if, and only if, you are “intellectually honest”.

Intellectually honest means: 2) You always have a plausible theory of success. 3) You allocate your resources in accordance to that theory 4) You monitor your theory 5) When you find it is no longer plausible, you make changes to get a new plausible theory of success.

If you are doing these things, I will back you even if you have a failure… As long as you don’t make it a habit.

(Satya Nadella’s address to his senior exec team, as shared by Jeffrey Snover)

Similar notes from Peter Cundill from his two books — “There is always something to do” and “Routines and Orgies” via The Ichanist

With the popularity of exe.dev, sprites (from fly.io), and shellbox, one wonders if there is a way to give sandboxed environment on existing, local boxes without having to pay for a cloud hosted service?

Michael Stapelberg lays out one such possible solution using NixOS and microvm.nix.

Another interesting addition to this space is nanoclaw that uses Apple Container to provision Linux containers on MacOS.

Between the two, the microvm.nix is appears to be handy for provisioning reproducible, complex, and repeatable containers (think if you are trying to provision hundreds of openclaws.. say as a service provider), while the nanoclaw, is good for your own personal (1) setup. Apple Container also feels very handy for provisioning throw away containers. The example on the README shows a good example.

Ajeya Cotra goes into the “Stable Marriage Problem” algorithm to explain why Asking for what you want opens up you to opportunities that may be outside your usual band of possibilities.

… I think the core dynamic in the proof of asker-optimality and askee-pessimality does apply to real life. If you only ever pick from offers you get, you never try anything unless someone out there already knew you and liked you enough that they took the trouble of coming to you. If you ask for stuff, you get to pick from among the entire universe of potential options theoretically available to you — and who knows, it might work out.

In my own personal experience, I can count many, many things I have gotten by asking, even (especially) things in retrospect, I had no business of having. However, it is always good to remind ourself that it is a muscle that atrophies if you don’t exercise frequently. So, go ahead, and ask away!

This is a great phrase — “Reliable Signals of Honest Intent”, one that you should be reminding yourself every time you are making something for “others” to see/consume/approve/buy/appreciate.

In the current atmosphere, where everyone is drowning in an attempt to keep up with rapid, exponential changes in AI.

Why are we all doing that? Is it just to have fun? Is it to signal that we are current? If so, to whom? and whatever you are doing - doing side projects, writing LinkedIn listicles, doing yet another MOOC, getting another certificate, prowling grad school websites to see if you should get another MS or an MBA, or start a side hustle that might get acqui-hired.. The list of things we conjure to signal are endless, and they all take a lot of energy.

I think it is important to stop, pause and consider — who are we signaling to? what is the signal we are sending? and more importantly — *is what I am doing A Reliable Signal of Honest Intent?

Time is finite, things to learn/do are, well, infinite. An ounce of clarity, an hour of planning, a spoon of taste, an earful of listening carefully to what people(users/buyers/investors/what-have-you) are saying might help calm the nerves, make targeted bets, that have higher odds of working out.

More than all that, a reliable signal is something that takes something from you. It is not cost-free. If you haven’t put your skin in the game, sweat in the brow, and metaphorical blood in the soil, your effort may be hardly better than someone else’s AI generated slop at worst, or someone’s couple hours of noodling.

The bar has definitely jumped a couple of stories high. It is disorienting to have to adjust to the new expectations. But, this is how it has been for many professions.

Going back to the original subject of Sutherland’s  “Alchemy: The Surprising Power of Ideas That Don’t Make Sense” — largely about marketing and consumer psychology, it is a dark art for people who are not in that profession. But, almost everything that is made, sold, and consumed in the capitalistic system is made possible through what he calls the “Alchemy” of synthesizing ideas that appeal to humans.

We, the technologists, have among our midst, many such mavens that understood human psychology very well (e.g: Steve Jobs). But, it is no longer sufficient that one in 10,000 is good that stuff. Perhaps, we all have to get 10x better at persuasion, and it is hard to do that without producing signals that are very hard to ignore.

Produce more reliable signals of honest intent.

The biggest takeaway from reading “Poor Charlie’s Almanack” is that the late great investor Charlie Munger preached the importance of having, acquiring, and applying mental models.

Lot of success comes down to identifying how to increase our odds in a crowded marketplace where many are vying for the same pie. This post by Sean Muirhead provides good insight in a terrible job market (2023-ongoing).

The biggest takeaway for me is that you need to be intentional, and think hard about what is that I want? Even if it is - I just need a job, I don’t care which (many people say this, but there is always a “but” when you listen), sending resumes to blackholes and keeping fingers crossed has not worked at all for much of the people on the market.

Sean’s mental model of targeted bet is applicable for much more than just job hunt, and it definitely gave me a pause and prompted me to reconsider my approach to how I am going about my current job search.

After coming back to US from India after in mid 2024, I was in the boat of “I need a job right away”, and I was lucky enough to find a job that is very close to where I lived, in a respected organization, doing something that is currently hot (Generative AI), as well tap into my experience. I am currently on the lookout, attempting to leverage my experience into a robust role where I can make a bigger impact.

I will be making targeted bets.

I thought I would resist the temptation to make the first post on this blog about the tools used to write this blog, but ironically, this news pushed me to finally stop polishing the apple and get the first post out.

I am generally (still) averse to Javascript, and Typescript programming; however, the framework at the top of the exceptions list is — Astro. I use astro to write this blog, and publish it on Cloudflare. I am very happy with this development because Astro finds reliable funding, and dedicated staff to work on Astro. You can manifest things into existence and you should try!

I have more to say about how I setup this blog, and how I went about coding this up, but that’s for another day, but for now I’m letting the cat out of the bag — the whole site was coded from the get-to using Claude Code, OpenAI Codex, and a tool that I only heard earlier in this week - Impeccable Style.