Never been one to let the carrier drop

OKay, so, it has become very popular to cowrite code with a LLM. And I cannot argue that it is a tremendous force multiplier, however, I do have some suggestions

#1) Painful as it might be, *read every diff*. Do not ever commit code you haven’t read.
a) While you’re reading, see if you can shorten the comments. LLMs overcomment – the reason this matters is if you are going to use
LLMs to maintain the code over long terms, you do *not* want to be burning tokens with excessive comments

b) Check the logic and approach. LLMs make mistakes.

#2) If the LLM says something improbable, *definitely* push back. *Expect* the LLM to draw incorrect conclusions from individual failures – one example, recently Claude concluded that the shopify metaobject filter didn’t work because you can only create a metaobject filter for a metaobject
which exists. It didn’t know that, got errors, and concluded the errors were fatal in scope to the operation. It was all set to create a implementation with 30 webhook registrations where only 3 were needed. I repeatedly pushed back, and led it down the correct road, until it understood the right way to do things.

#3) Point out when the LLM says something blatantly wrong. among other things, there isn’t any real reasoning going on – remember, LLMs have no aboutness, they’re in no way anchored in reality – they can’t tell a cat from a cow, they just are a sliding set of activations of weights moving through a transformer, also thought of as a feed-forward ANN with a attention head. By pointing out a blatently wrong statement, you may well nudge them into the correct alignment.

#4) *TEST* everything. One *good* thing about LLMs is they are wonderful at writing unit tests. Use this ability – but also end to end test the system under development unless you can tell by reading the code that it’s going to work – and maybe even then. Certainly don’t release anything to production that hasn’t been tested.

#5) Sometimes, write it yourself. You need to keep the ability to write code, and it will atrophy if you don’t use it. You want to occasionally exercise those pathways – neural skills are use it or lose it.

So, I was going through my dad’s office and I found a front panel from a early ARPAnet router.

This led me eventaully to a memory of “Take your kid to work day” in 1988 or 1989. My dad took me to the Pentagon and set me down in front of a early Sun Sparcstation – I’m fairly sure a Sparc 1, in fact, from looking at photos and from the timeline. I also remember I was super fascinated by the NEXTcube that had just come out and it’s ability to emulate a modem in software, and DARPA had a few of those somewhere.

Anyway, my dad wandered off for a meeting, and I discovered the video teleconferencing software on the sparc and tried to open several video conferencing links. I am pretty sure I temporarily broke the ARPAnet, which it turns out was still made of 256kbit links.

Shortly thereafter, my dad decided to decomission the ARPAnet. I had never considered there as being any possible connection between those events until today, when I realised that it must have to him looked as a demonstration of how outdated the ARPAnet was. NSFnet was already held together with T1 lines, which were becoming the industry standard, and would not have broken quite so easily.

So, one thing you can safely say about DOGE, the cryptocurrency. It is headed for zero. It has a *unlimited supply*, with a fairly high mint rate. It was originally intended as a joke.

One thing you can say about DOGE, the government agency, is it is also a terrible idea. Musk is the poster child for Dunning-Kruger – we’re talking about the guy who thought you could have a 600 mile long evacuated above ground tunnel, among other things. He also doesn’t understand that the more people you make homeless and hungry, the less wealth we as a species have. DOGE is also likely to destroy many useful functions that the government performs.. protecting our bank accounts, protecting us against bad loans, protecting us from e coli, protecting us from predatory corporations who otherwise would destroy all the land and poison all the water to make a few bucks.

Musk himself has says there will be some hardship. I vote all Musk’s money be taken away and he be forced to live with the hardships he is causing.

#1: Property values – from my point of view, the sane person wants their property values to stay the same, or to go down. If you want your property values to go up, A: you hate your neighborhood and are planning on leaving B: you don’t want other people to be able to afford to live there C: possibly you’re doing two things which are a bad idea – 1) gauging your value as a person on your net worth and 2) including the value of your home in your assessment of your net worth – not only might you lose it at any time to the capricious whims of earth, but this encourages all kinds of unhealthy thinking surrounding the family unit – healthy families have more than a few people and the home may be owned amongst everyone or titled in one person’s name for convenience sake.

In any case, lowering property values is a good thing. Which is good, because I want to live in a neighborhood that looks lived in, with science projects and houses of all different colors and sports equipment and dogs and kids and general mayhem.

#2: The cheapest option is often *not* *the best. For example in southern california the argument against desal plants is often the cost of water. However, what SoCal is doing now will sooner or later lead to a catastrophe – indeed one could argue it already has (more on this in another post). SoCal should be self sufficient, water wise, and that is better than having the cheapest water.

#3: Ideally we would be getting rid of money in favor of a better system. Money, as I’ve discussed elsewhere, encourages bad decisions, some so bad that they may lead to the destruction of our species.

#4: We have undoubtedly made Earth too difficult. In general we should be thinking in every system how can we make this easier – how can we make less things illegal, how can we accept more people for who they are, how can we deprioritize GDP and prioritize health, happiness, and sanity.

I have to say, contempt of court should not be a crime, because most courts in the USA deserve contempt. (Of course, anyone who thinks they are worthy to judge others is starting out in a extremely questionable place, and when we look at the US where murder is encouraged if done by a cop or soldier, where ethics are for sale to the highest bidder).. In any case “Your Dishonor” seems like a honest way to refer to a judge.

Please make a whole series of movies about the space dog from Over The Moon.. thank you 😉

One thing that I do worry about, post the previous post, is what absolutely wrong beliefs *my* mind is defending. No real way to know, either.

I didn’t really feel like this accurately captured me.. I feel like the benevolence should be higher. I maybe should have thought a little more about the data I was feeding it.