I think of English as being actually not a single language but rather dialects with many different pronunciations which share most of their written form. I also think this means any centralised effort to change how it is written will fall flat.

We can’t agree in my house how to pronounce “bath” so how will the entire English speaking world agree on the spelling of every such word with consistent meaning but differing pronunciation…

That one stood out for me too. I think there’s a lot of pronunciation nuance that would be lost with SoundSpel.

I’d distrust any top-down effort to change a language anyway. It belongs to the users and they’ll adapt it to their needs the way they see fit.

What are your thoughts on humans learning from LLM output ?

I’ve been encouraging the new developers I work with to ensure they read the docs and learn the language to ensure the LLM doesn’t become a crutch, but rather a bicycle.

But it occurred to me recently that I probably learned most of what I know from examples of code written by others.

I’m certain my less experienced colleagues are doing the same, but from Claude rather than Stack Overflow…

Not Simon, but here is my take.

Code are practical realizations of concepts that exists outside of code. Let's take concurrency as an example. It's something that is common across many domains where 2 (or more) independent actors suddenly need to share a single resource that can't be used by both at the same time. In many cases, it can be resolved (or not) in an ad-hoc manner. And sometimes there's a basic agreement that takes place.

But for computers, you need the solution to be precise, eliminating all error cases. So we go one to define primitives and how these interacts with each other and the sequence of the actions. These are still not code. Once we are precise enough, we translate it to code.

So if you're learning from code, you are tracing back to the original concept. And we can do so because we are good at grasping patterns.

But generated code are often a factor of several patterns at once, and some are not even relevant. Just that in the training phases, the algorithm detected similarity (and we know things can be similar but are actually very different).

So I'm wary of learning from LLMs, because it's a land of mirages.

I've been programming since 2009 and lately I've been also learning a ton from LLM output. When they review my code or architecture ideas, they sometimes suggest approaches I outright didn't know because my day-to-day "rut" has been different so far.

LLMs are like a map of everything, even it's a medieval map with distorted country sizes and dragons in every sea. Still, it can be used to get a general idea of where to go.

I think an important skill to develop as a software developer (or any other profession) is learning to learn effectively.

In software these days that means learning from many different sources: documentation and tutorials and LLM-generated examples and YouTube talks and conference sessions and pairing with coworkers and more.

Sticking to a single source will greatly reduce the effectiveness of how you learn your craft. New developers need to understand that.

what’s the vector for microplastics in shower water causing you harm ? swallowing some of it or through the skin or something else ?

Microplastics was only one of my concerns. Their “why” page does a good job explaining the benefits. https://weddellwater.com/pages/why-duo

Make has its place as a build tool for large C codebases.

People sometimes treat it as a generic “project specific job runner”, which it’s not a good fit for. Even simple conditionals are difficult.

I’ve seen several well-intentioned attempts at wrapping Terraform with it, for example, which have ended terribly.

It’s not a generic job runner. It’s a generic way to transform linear shell scripts into declarative dependencies. It’s a general tool for the shell.

> Make has its place as a build tool for large C codebases.

This is no longer true imo.

More robust and well-defined build systems have been created in the last 2 decades. Time to update.

Is there a good generic job runner?

Edit: Sorry, it looks like I totally misunderstood what you meant by "job runner".

Sure, a bash script.

People keep writing and using other alternatives (like just), which provide a very slight improvement on pure shell at the cost of installing yet another tool everywhere.

I stick with bash, write every task as a separate function, and multiplex between them with a case statement (which supports globs et al. and is very readable).

Years ago, I discovered git-rev-parse's option parsing, and it completely removed any excuse I had not to write my own personal bash scripts to a professional standard.

Now when I need a tool, I can knock it out in bash with proper option parsing, usage, etc.

bash is awful on a lot of fronts, but if you're writing code that's primarily calling a bunch of tools and mucking with their output, it's still the best thing out there I've found just due to piping syntax.

Taskfile and Justfile are pretty solid.

Let’s assume starship works out and they come up with a nifty wide-opening payload door solution, one of the advantages will be payload volume as well as mass - the JWST’s main mirror would have fit inside without being folded (although the heat shield would not have).

Was this the BT demo exchange somewhere in Essex ? Bishop Stortford maybe ?

I did that tour around the same time and it was fascinating ! right in the middle of the 21CN (ethernet core network) transition.

No, it was just a normal BT exchange in Suffolk

A local fibre co here in York dug a cable to all the way my parent’s farm house, about 250m outside the nearest village (and probably even further into the village to reach the PCP), then left the cable coiled up on the outside of their property and haven’t done anything further with it in over a year.

I’ve heard other similar stories from friends in the city too.

It’s almost like there’s money for the infrastructure but not for the staff required to actually run it as a service…

Where as in London, where there is at least one house every 8 meters, they were in and out of my street in two days and now we have constant door to door sales people offering deals.

Smells like they "installed" fibre in York to meet a contract/regulation but they really focus on Urban density. Makes sense for them but not for the rest of us.

Sometimes I think everyone travelling to an office was a good thing because it reinforced the idea that your home and work lives are separate and it’s healthy to draw a line between the two.

P.S. I’m a permanent homeworker, I like it. I like the fact I don’t have to commute. But man, loads of people I work with seem to conflate work with childcare and I dunno if that’s right for anyone involved.

Or maybe our continued focus on excluding our families from the way we spend the majority of our day is toxic in a way that we can’t even see because we’re so deep in it

Sitting at your desk trying to write code whilst a toddler is screaming for your attention nearby ?

I’m honestly not sure what you’re getting at, but this isn’t the middle ages where the kids learn how to weave baskets by watching their mother do it and that’s going to be their job too.

The thing is, nuclear was never on such a steep learning curve as solar and batteries are today.

It’ll never be too cheap to meter, but electricity will get much cheaper over the coming decades, and so will synthetic hydrocarbons on the back of it.