inmysocks (version: awoo) @inmysocks@awoo.space

I still maintain that it is objectively bad design to have a programming language where the default behaviour is to run commands synchronously and then when you have an asynchronous function instead of marking that it does something differently, you make the version that behaves the way everything else is the language does.

*glares at javascript and node*

It turns out that my simulations were running about 5x longer than necessary.

It wasn't wrong, but the system reached steady state after about 50k steps and I was running each one for 250k steps.

So instead of only doing one more batch I get to do more batches and hope this gives good data.

it turns out that my table of ~50 million rows had no index. So that may be part of it.

it is really really annoying but otherwise you get 1000 emails spread out over a week each asking for something tiny that should have been included in the original contract.

Before having to work directly with clients I always thought it was weird that people charged for things like quotes and consultations to determine what a job would be.

Now I fully understand. People have absolutely no respect for your time or expertise and the only way to get them to be at all specific or have their shit together is to change them for every tiny thing.

You want a tiny update? You pay for my time telling me you want the update, and my time making the update

I don't think I will ever understand why some SQL things take as long as they do

Client < we want to make an rss scraper so we don't have to pay for access to a news ticker api for our historical data so we can get better results by only looking at high quality sources for our market sentiment data!
Me > That isn't how this works, the noise means we only see trends when you have huge amounts of data.
< So can you make the rss scraper? we only need like 10 sources! I want the better results!
> .........
> That isn't how math works.
< But high quality sources! Better results!!

Julia can do 10000 recursive function calls just fine, but it looks like 100000 is too many.

So I guess I can't do this simulation using a pure functional style and tons of recursion.
I am a bit disappointed.

Windows file systems break my git repos.

Forget viruses and climate change, windows file system pathing problems are going to cause the end of humanity.

There are always more details to work out when trying to implement something that actually respects privacy.

If you try and use that as an excuse to not try and make your application respect privacy, fuck you.

The word 'server' is used for too many things in computers.

Because there is a server where my server is running my server.

the number of big projects that still rely on python 2.7 is a bit disturbing.

And once again, a feature request for one of the basic features that have been available since before the first release.

The person who keeps saying that localStorage is equivalent to saving things on the local file system just got really excited to find out that browser add-ons may let you store persistent data in a browser that may not get cleared without warning the way localStorage can.

But we have browser plugins to just save the stuff to the local file system, if you are installing something anyway saving it in the browser is a step backwards in terms of functionality.

This dude wouldn't bug me so much if he wasn't so fucking loudly telling everyone about how all the saving problems are solved by just using progressive web apps because localStorage.

I must not forget about list comprehensions when coding in python