kamaal

>>I would consider that overly wasteful.

ChatGPT tells me thats very likely Puya PY32F002A and that can be as cheap as $0.0682 a piece if you order 5000 and upwards, which should be fairly the case for a mass market vape.

Hearing the term microcontroller makes you feel like a lot of compute capacity is being wasted. This is what you get in that microcontroller-

Puya PY32F002A: Cortex-M0+, up to 24 MHz, up to 20 KB Flash, 3 KB SRAM, ADC, timers, I²C/SPI/USART; available in QFN20/TSSOP20.

Its basically a small battery, a display and a microcontroller to show you battery charge and vape fluid levels. Mostly ADC work and display.

>>I expect to see many stories from parents, non-technical colleagues, and students who irreparably ruined their computer.

I do believe the approach Apple is taking is the right way when it comes to user facing AI.

You need to reduce AI to being an appliance that does one or at most a few things perfectly right without many controls with unexpected consequences.

Real fun is robots. Not sure no one is hurrying up on that end.

>>Edit: most comments are focused on pointing out that version control & file system snapshot exists: that's wonderful, but Claude Cowork does not use it.

Also in my experience this creates all kinds of other issues. Like going back up a tree creates all kinds of confusions and keeps the system inconsistent with regards to whatever else it is you are doing.

You are right in your analysis that many people are going to end up with totally broken systems

Around a decade back, I and a bunch of colleagues explored these theories and despite knowing they were all bunkum, the sheer entertainment value they served was gold.

Think of it like Marvel universe stuff.

We'd go on long walks and let our 'what if' imagination run wild.

It works because if you want some information on React or say Python, or say Prolog. Whatever information ChatGPT generates is quickly verifiable, as you have to write code to test it.

Even better many times, it shows me new insights into doing things.

I haven't bought a book in a while, but Im reading a lot, like really a lot.

Its almost like he is asking users to take a stand between using LLMs and using Tailwind.

LLMs, or Tailwind. Pick one!

>>Adam released an amazing book about UI/UX called Refactoring UI[0] and it really helped me become better and understand subtleties of design.

In the age of AI, if you have Table of Contents. ChatGPT can write the book for you.

Only books I buy these days are in fiction genre. Everything else is derived from facts that already exist some where and AI can derive and write the whole book.

>>Which is why much better advice is to try to get promoted by switching companies and jumping a level in the process.

This is why most companies don't offer a promotion as a part of hiring process, and are mostly hiring at currently levels, but at a pay raise.

In some cases where a promotion is available, they often pay below your current compensation, which defeats the whole point of the process.

>>Taking on extra responsibility is all well and good until someone figures out that they can just get you to do more work for the same amount of money.

Wait a minute. Why are you accepting more work, responsibility without increase in compensation? Promotion de facto means getting paid more.

Otherwise its just some one updating a row in the employee database with fancy text. How does it matter what designation you are called with?

I had a similar situation few years before COVID where a company offered a fancy designation albeit for 50% lesser the pay. All said and done, when I did all the calculations, even with me rapidly changing companies with newly acquired designation, and building from there. It would take more than a decade to merely arrive to the salary I was then. And that would still mean more than a decade of wasted raises, bonuses and RSU vesting at the then current job. By the time that fork got profitable, I'd be due for retirement.

Promotion == Pay/Compensation Raise.

That's not the definition of stagnant I would use. Its mostly on the lines of people learning the language, starting new projects, discussions etc.

There are more discussions on Perl being dead, than Ruby being alive.

>>Python is more likely to make the average programmer happy.

Its a weird place to be. I was making ChatGPT write lots of Python code to do some analysis on the Stock market, and it was crazy how much code I was able to write in a day. I'm talking like a million+ lines of code. In a day.

To that end, it also means the cost of Python code today is $0 given how much can be generated so quickly.

Its a useful language, but pretty much anything you do with it today doesn't have as much value.

>>WHERE ARE THE NEW RUBY USERS!

Ruby adoption has been low to non existent for as long as I remember. Lets say, 15+ years now. Python kind of took over scripting space. That also means Perl ceded its space to Python as well. But man Perl does one thing really well, and that is acting as a glue language for anything Unix. So it will always have one good use and it does that really well.

Ruby revival was a thing during the time of Rails, but that went away with React + Node taking over the frontend world almost entirely.

>>TIOBE has many issues, but ruby was doing better in the past there, so something changed.

Tiobe indeed has its issues. But their results do not surprise me. Perl is in the top 10. Python is no. 1. Out side of these things you are going to write SQL for database. And mostly Java for apps, and C for embedded systems. C++ for performant applications. And JS for anything on browser.

Ruby just doesn't have a space and a sufficient following in that space.

There is also that problem of not having a Killer app.

Index doesn't go up. Individual stocks do. I mean a small section of the stocks within the index contribute to most returns. This is of course different in a bull market where lots of them go up at once. But even in choppy and bear markets, its a small section of stocks that do most of the upward movement.

You need to study index carefully, like pick each stock and study it. You will see Buffet isn't exactly investing outside the index, his investments come from inside the index.

You are not looking for needle in a haystack here. More like a thick loaf from a pound of bread. If you squint enough, you find them.

I think people who are looking at long term wealth building are better off looking at John Bogle's work. The key is this, you absolutely must stick to investing in the index as a whole, but the issue there is fees you pay is a percentage of your investment, even if small, over years compounds against you. That might look small, but its a double digit percentage if done over decades. Hence picking a fund that charges you the lowest is the key.

Buffet does this better. He basically takes the float from insurance business(Geico) and invests in the index his company manages. This basically sets you up for a long compounding cycle. Translation- You are basically investing in the US economy.

Apart from this there are some base businesses in an economy that's under long term multi year climb. Now anybody who is staring at stock charts(weekly) will tell you eventually you do learn to 'Pyramid upwards' on stocks that are clear winners. Lots of literature has been written on this.

All said and done, you can be rest assured there only at best 10 such stocks in any economy. Once you find them, you do periodic audit if the companies are holding up good and just keep putting in money as time passes.

I don't remember where- but I remember reading, Fidelity found out the most profitable accounts belonged to people who were dead. That says two things. 1) You have to hold stocks for long periods of time, like really long periods of time(Index, and some of its constituents that are carrying most of the freight) . Most people won't make it. 2) Live long enough.

Quite literally once you gain the skills to pick stock. Next single biggest skill you can master is to be healthy and live long. These things take lots of time to work.

>>none of your logic makes any sense to me.

Compounding is a function of time. More time you are alive, more you see in that time. This follows from definition of compounding itself.

Whole point of being rich is to have freedom to do whatever you want. Including fancy dress like poor and eat like them for personal marketing at times.

Pretty sure Mr Buffet mostly eats salads.

>>Here are some things Buffet has done that I admire (notice that phrasing):

Perhaps the best thing Buffet has managed to do is to live long. Most of compounding magic begins at ages 60-65, a time where most investors start to die out.

Second best thing he did was to start/acquire a insurance firm. The 'float' helps them to run a kind of in house index fund on other peoples money, without having to pay TERs/Fees. Thats basically no effort bogle style compounding. Even if you end with a situation where you have to return ALL the premiums collected, you still get to keep the returns.

Other wise everything else is just fairly normal, if you are sitting in front of charts for long, its impossible to miss something that's going up on a weekly timeframe for long periods of time. You just pyramid upwards and wait, patiently.

The real issue is waiting doesn't work too well for most people as you start to die out after 60.

>>there is less overhead.

There have been methods to reduce overhead available over the history of our industry. Unfortunately almost all the times it involves using productive tools that would in some way reduce the head counts required to do large projects.

The way this works is you eventually have to work with languages like Lisp, Perl, Prolog, and then some one comes up with a theory that programming must be optimised for the mostly beginners and power tooling must be avoided. Now you are forced to use verbose languages, writing, maintaining and troubleshooting take a lot of people.

The thing is this time around, we have a way to make code by asking an AI tool questions. So you get the same effect but now with languages like JS and Python.

>>Feynman was a uniquely gifted teacher that made things intuitive and simple.

I think explainers like Neil deGrasse Tyson have a job harder than people imagine. Historically the problem with science education has been, that, as the conceptual universe gets bigger and complicated there's a tendency to assume the common person is too stupid and beneath the subject to understand it.

To simplify and demystify science to a point to get people interested in it as a intuitive iterative process helps a lot in increasing participation of the general crowd.

>>As others have joked it's a "write only"language.

The pet project you write to replace that Perl one liner is also write only once project. The only difference is you waste years of your life doing what we do in minutes.

There is nothing to boast about wasting your life doing things that don't even have to be done.

>>Not sure your time-saving thing really works, especially if 99% of problems you have are not solved by a perl oneliner.

That's because if you don't have a paradigm of thought, you are not likely to reach for a tool that works there.

Most of the times, I have seen Java programmers watch in total awe that their pet weekly project, that took 40+ hours to accomplish, was basically a vim macro that they could have done in like 3 - 5 mins.

That was my whole point I made, when I said, when a Java dev says they have like 10 years of experience, thats barely a Junior perl dev.

I'm guessing most Python/Java programmers would find it impossible to work in project which doesn't involve a database, XML, or json. Those languages are not designed to work with a data or a compute paradigm outside of these formats.

>>I think the reason python won was that it was easy to learn and read and was batteries included.

This is also a big reason why AI assisted programming will wholesale replace Python programmers.

If your are optimising for people who wish to remain at beginner levels all life, and that replaced people using power tools to solve harder and bigger problems. It shouldn't be surprising that now some automation will replace you.

Every programmer worth their salt must learn to use power tools, and also train to work beyond the beginner-intermediate levels, that has now become the permanent situation in our line of work.

Nobody should be spending a day writing Python to do work that can be done using a vim macro in like 5 mins. Or spend a year doing in Python what can be done in a week in Perl.

>> $\=$/;for(@ARGV){open F,$_;while(<F>){print if/$_/}}

There are lots Perl one-liner patterns, once you learn to write them. You are basically saving yourself writing many dozen man-hours of Python/Java coding work.

I remember telling my manager during the Perl days. A 10 year Java experience guy looks great, but thats basically a junior Perl dev with 6 months experience.

There is nothing to feel great about wasting effort and more importantly time.

Given the install base of Linux, yes. It did win the OS wars.

>>So was 2010 or 2011 the Year of the Linux Desktop?

Desktops are the of the game of the past. Mobile devices is what you need to look at.

Windows doesn't even have a presence there.

So yes, Linux won!

This is a correct view point to have.

As you get to the downswing of your career, you should have already worked and made enough mistakes to have most of your experience. You must cruise on that experience when you are older.

When you are old, that is not the time to work and make mistakes.

That is, most of the heavy work must be done as early as you can. Eat that frog.

>>Being focused on the goal you defined 30 years ago is almost certainly wrong for most people.

I think there are certain things that are not likely to change, and must be aimed for. For starters, being healthy, proactively working towards a retirement nest egg, so that you don't end up homeless and starving in case things go south too fast.

There are many such things I hold as things I would want several years from now. Good health, free time and enough money to not need a job to just put food on the table, and a roof above my head.

>>I've noticed that car alarms that go off for no good reason seem to be back.

So in the 2000s, this was fairly common in India. Then one day a installation mechanic told me the sensor had various calibration settings. You could get the trigger to be as sensitive as you wanted.

At times a rodent or a crow could trigger the alarm.

I haven't used Microsoft office products in more than a decade. And Windows in more than 15 yrs.

Google offers so many things free out of the box. And for serious spreadsheet sort of work, I use numpy.

Google pretty much won the Office software war.

>>That is to reduce their populations through sanctions, blockades, starvation and outright war.

Speaking as Indian, Our population growth happened when we were poor, and masses were uneducated/illiterate.

I have one more theory that- Population growth happens when poor people have access to lots of carbohydrates. Plus having lots of children is somewhat akin to having meat robot automatons whom you can send for physical work and make money.

If you want to use sanctions to do reduce population it doesn't work in the modern era. Poor people eat well and make babies.

>>If top is bunch of sloppy clowns, whole company is going to be the same.

This happens so often at big companies as well. Management is always assumed to be correct, and the pay grade argument always kicks in(They are paid more because you are beneath them). And this starts to show up everywhere. You can't take any initiative without sanction from the top, and they are often clueless as to what the ask is. Most of the times its rejected just to assert authority, and not on the grounds of merit.

Top bosses are also very envious and proactively trying to kill rising talent out of fear- people better than them, will replace them. To that end no good thing ever happens, if you push too hard you will be eliminated in interest of self preservation.

So by and large no good thing is ever suggested, or tried or happens. Eventually until whole business(es) die out. This happens in every company, no matter what companies claim about hiring, retaining and promoting talent. This is just how every place works.

I was casually chatting with my uncle who is a doctor, he says something along the lines that if a chemical can kill a rat or a mosquito, to assume it won't do any damage to humans is kind of hilarious.

Of course humans who inhale this thing in small quantities won't die, but you can be sure they will kill some tissues that they go into. Now comes another problem of regular exposure, and these chemicals having an entry, but no exit path. That just means there are tissues, that are likely dying out every time there is a exposure.

Again none of this might kill you at the first exposure, but if there are enough dead tissues, there sure is likely to be things like Parkinson's or may be even diabetes.

Im guessing combined with this, if you already some bad genetics it could cause issues like these.