petersumskas

Ai generated text being too concise is not a problem I’m encountering. What we’ve begun to term “Claudish” tends to more closely approximate those “become an expert in X using this one weird trick“ articles that go on and on and never quite seen to get to the point.

Usually I’m getting better results by telling it to be more concise.

The writing in that blog has the distinct whiff of Claudish to me!

Yes! I wish I had more than one upvote to give!

Because the limiting factor for what you can deliver has never been, and still isn’t, developer productivity.

Talking point: I think it is communication, and AI (notably information exchange in “Claudish”) is not as effective there.

In Australia there is a scale model of the Solar System where the sun is the main dome of Siding Springs observatory and the planets are on billboards at scale size and distance. Pluto is several hundred kilometres away (and the size of a tennis ball on the billboard)

This discussion has made me wonder what it would look like to have a different scale where the distances are based on the “speed limit” (which is mostly 100km/hr)…so that earth, for example, would be 8min from the Sun. Neptune or Pluto would be 4 - 7 hours away.

It would still fit in Australia but be significantly larger.

Current me if I’m wrong but I think that was a calculation based on a specific velocity at some fraction of C.

Whereas a constant 1g acceleration would far exceed that fraction and thus shorten the time significantly.

Chris Newman fan?

You should have lead with “What is it now?”

Discreamination FTFY

There are only two hard problems in computer programming:

1. Naming things 2. Cache invalidation 3. off-by-one errors

> Markdown files are slightly annoying to read/preview

Maybe I’ve missed the intentions of markdown, but the ability to easily read the plain text version has always been the killer feature.

Rendering as html is a nice bonus.

I understand there are plenty of useful things to say “but what about…” to, like inline images, and I use them. But they still detract from what differentiated markdown in the first place.

The more of that you add, the more it could have been any document format.

Keep going! The idea has merit. But it is US centric.

I tried the example with my local post code and of course it didn’t work as I’m not in the US.

> You can now spin up five agents to implement five different versions of the thing you're building and simply pick the best one.

Or you end up with five different mediocre solutions where the best parts are randomly distributed amongst all five.

The described SDLC is a recipe for rigorously and predictably building the wrong thing.

Does anyone actually work like this? Have they ever?

At the least it misses all the feedback loops between the stages. Even the actual waterfall model isn’t as linear as the one given as an example.

AI is like electricity. It may be the “product” you think you are selling. But it isn’t the product people are buying.

People are buying what these things facilitate (lights, tvs, air conditioning, etc in the case of electricity)

I think the AI folk have generally done a terrible job of connecting the dots to show people what they are actually getting.

It’s worth noting that some things mentioned above already existed before electricity. So people needed to be shown that electric light is in almost all ways better, cheaper, more convenient than existing alternatives.

Yes! I have a similar reaction.

I often think of Jack Vance’s “Eyes of the Overworld”. They hid the reality of a decaying world behind a fabulous facade.