ic_fly2

Generally not a fan of geohot, but this is pretty possible, especially once self driving tech becomes available through suppliers like mobile eye / Bosch/ continental etc. (it’s possible it will stay a Tesla / Waymo duopoly but I doubt it)

No matter the merit of the claim, the appeal to the fascists in the US government invalidates any legitimacy.

And that is an achievement given how moronic the current Italian government is.

That has two causes, dependency on natural gas, which would be worse without renewables and taxes, which is unrelated to renewables and related to general policy goals of reducing energy demand.

Also even in a European context UK power prices aren’t as high as many of its peers.

My favourite is always:

breakpoint( and then some nonsense arguments.

Apparently a good chunk of the code that these LLMs are trained on is python, yet setting a debugging breakpoint still causes difficulties.

The first part is wrong, it’s a question of org size. A lot of large orgs hand roll a lot of these things, they call it developer excellence.

And your last paragraph hits the nail on the head, people are afraid to run their own software.

Really?

I grant you pipelines are the best bit about ado, but the fact that you can’t test them is a pain.

And the webhooks and templating are pretty messy and unpleasant quickly.

We’re changing from ADO to GitHub (had to be an MS product for corporate) and the infra people are looking forward to GHA as they prefer their maintenance to ADO pipelines.

IBM might not have a data strategy or AI plan but he isn’t wrong on the inability to generate a profit.

A bit of napkin math: NVIDIA claims 0.4J per token for their latest generation 1GW plant with 80% utilisation can therefore produce 6.29 10^16 tokens a year.

There are ~10^14 tokens on the internet. ~10^19 tokens have been spoken by humans… so far.

I’m in Europe too.

We pay well. Very well in fact. We’re a small company though.

I have a harder time hiring here than at my previous position with a much larger company, even though my current employer is superior to my old one in every way except for brand recognition.

And from others in this space (who have at times tried to recruit me) this is not a problem unique to my company.

I don’t know, I’m hiring and the candidates so far have not been impressive.

Maybe that is restricted to my area / region but I got one am confused.

If in the end we can just have .spy on some files that have performance critical functions in them and the rest is just normal python, this could be down right amazing.

We recently swapped out mypyc optimised module for a rust implementation to get a 2-6x speed up, and not having to do that would be great.

If you can’t work out the email, you won’t get past the interview, sorry.

Waiting for the lack of returns on LLM investments to come and bite back.

Together with the debt payments needed then, this will do wonders for the stock. I’m sure.

I’m pretty sure it’s smartphones and aging society.

I’ve been gathering data on the natural experiment that occurred due to differences in proliferation of smartphones across countries. The sex surveys aren’t consistent but that is a very strong factor. Look at the hockey stick curve in the paper here.

The rest of the decline is (in my evaluation) best explained with the increase in average age of the 18-64 year old demographic.

The thing that is often missed is that with high taxes and resulting high labour cost, you get high productivity. Simply because low productivity is not profitable.

This results in bad service, high quality goods and strong utilisation of capital goods.

But as others have noted the wealth disparity is increasing thanks to new policies and low interest rates leading to asset inflation.

The Yaris is more expensive than 10k.

Simple as that. Allow people to shift value in larger units without AML and the crooks will use that route.

The AML form will not be the most unpleasant part of buying a Yaris.

As someone who sees the outcome of people losing everything to sophisticated scammers/ fraudsters and thieves and how little authorities are able to do, nah, the overreach is not in sight.

There are more criminals than abusive IRS agents. And usually when people tell me stories like that, there is more to it..

Meh, even with basic async I’ve been able to overload azure’s premium ampq offering memory capacity.

But yes managing db connections is a pain. But I don’t think it’s any better in Java (my only other reference at this scale)

It’s a good article on speed.

But honestly the thing that makes any of my programs slow is network calls. And there a nice async setup goes a long way. And then k8 for the scaling.

I’ve used the reverse. When making a new module I’ve let the llm make up an api and I’ve used the names suggested as inspiration of what might come natural to others, to make the use more intuitive.

Not all IPOs pop

pydantic basemodel has made dataclasses redundant.

Very neat looking tool.

Do you expose an api to set and get network information like valve placement, demand at nodes or pump schedules?

In my old research group we ran a forked versions of epanet to do some of these things and there was a previous effort called oompnet that tried to bring oo into working with epanet.

If researchers can use epanet-js to give their researched algos and methods for wdn control or management, the combination could actually give Bentley a run for their money.

You might want to present this at ewri and ewra or ccwi, there are usually quite a few people working with epanet there.

This is such a weird cannon on sparrows approach.

The local builds are fast, why would you rebuild docker for small changes?

Also why is a personal page so much rust and so many dependencies. For a larger project with more complex stuff you’d have a test suite that takes time too. Run both in parallel in your CI and call it a day.

More money made with batteries means more batteries installed. How is that a bad thing?

I have programmed batteries / algos to do this in some European markets. This is being done right now.

The yield you could make from batteries in the UK dropped from double digits to 2% in 8months once some hedge funds figured out how to build and bid (or commission companies like my employer) batteries in the UK short term reserve market.

There are a few firms in northern Jutland and London specialised in this sort of trading.

What I don’t understand in such reports is why there is no mention at all how this is done in other countries. Do they all still use floppy disks? Did they do an upgrade? How did it go? Surely this would be valuable information.

The LLM generated unit tests pass. Obviously!

Azure API for LLM changes depending on what datacenter you are calling. It is bonkers. In fact it is so bad that at work we are hosting our own LLMs on azure GPU machines rather than use their API. (Which means we only have small models at much higher cost…)

I bet the next AI powered IDE will be called sailboat or some other preppy water themed sport.

Good luck down the stream.

In the past when I needed this I just made a function that processed the f string, often enough a simple lambda function would do. This looks like additional complexity for not a lot of gain.