Notices by Russ Garrett (russss@chaos.social)

@whitequark that one's written in Latin but then there was a period where they were in French https://www.legislation.gov.uk/aep/Edw1/3/5/section/wrapper1

@whitequark this is the oldest one still in force! https://www.legislation.gov.uk/aep/Hen3cc1415/52/1/contents

The UK now has the world's first ETCS-enabled steam locomotive, because we're a very silly country.

https://www.theguardian.com/uk-news/2025/apr/19/nobody-has-done-this-before-britains-beloved-steam-trains-trial-pioneering-technology

@harry_wood I don't think it's the big AI companies which are scraping excessively, it's random people who probably got a LLM to write their scraper bots... To make it more confusing, in some cases they're cloning the user-agents of the major AI bots.

@harry_wood oh it's probably LLMs or something LLM-adjacent. But I doubt it's the big AI players who are responsible for the excessive scraping.

This is not a narrowly-construed piece of legal text. This is some boilerplate that a lawyer hammered out with very little oversight and nobody thought to question it.

And that makes me more nervous about Mozilla than anything else they've done in the last few years.

How is Mozilla "operating Firefox" on my machine??

@amapanda it's also telling that people suddenly care about this and were apparently fine about Google both-sidesing borders of Crimea or Kashmir or Arunachal Pradesh.

Maps are always political and multinationals will happily do whatever horrible diplomacy is necessary to allow them to make money.

@ryanc @ljrk yeah their nudity policy is extremely vague: https://support.google.com/youtube/answer/2802002?sjid=12852059540298367797-EU#age_restricted&zippy=%2Cage-restricted-content%2Ceducational-content

It is of course available to everyone on media.ccc.de, and that's the version we embed on the EMF website.

And yeah it still has that "topical context" note...

@grifferz it wasn't quite an open S3 bucket this time:

https://chaos.social/@russss/113729717983729088

@grifferz the talk was really good but it's in German - here's the English simultaneous translation:

https://media.ccc.de/v/38c3-wir-wissen-wo-dein-auto-steht-volksdaten-von-volkswagen#l=eng

@gsuberland @ryanc yeah it's still there. We have found a Google employee to file an internal issue though

@ryanc Doesn't seem to be anything we can do as a channel owner. I assume it's based on a rather broad keyword match in the transcript.

Not sure if it still happens but there used to be one which came up if you even mentioned covid once...

@cstross I've spent the last week down this rabbit hole and it's not been pleasant:

https://russ.garrett.co.uk/2024/12/17/online-safety-act-guide/

I don't think it's impossible for small sites to comply, but the guidance is terrible and it seems like it's always going to leave you with some risk.

New version of my visualisation of world power networks mapped on OpenStreetMap, with data from today.

I've added power plants to this iteration - although these are covered by power lines in denser areas, I think they give a bit more detail in darker areas.

Rendered from 38.1 million power lines and 85,000 power plants in 86 seconds using datashader.

#energy #gis #openstreetmap

If you run a small online community in the UK (or with significant UK users), I spent most of yesterday reading hundreds of pages of Online Safety Act guidance and summarised it:

https://russ.garrett.co.uk/2024/12/17/online-safety-act-guide/

There's definitely stuff you have to do if you want to comply with this law...

@siege I only vaguely remember that night but I do remember getting home at 6am wearing lipstick

@siege on Dalston Lane near the bridge over the railway?

@siege now I'm wondering if I may have visited this secret bar!

I still find it slightly astonishing that we sell £150k worth of tickets in 40 seconds using some software we wrote in our spare time.

World electricity networks mapped on @openstreetmap: here's my approximately-yearly update of this visualisation.

#osm #energy #maps