Replying to
@futurebird@sauropods.win I hadn't thought of the ant-surveillance angle! And this is my company logo 😁
Remote
paulwalk
@paulwalk@mastodon.social
mastodon 4.8.0-nightly.2026-10-06Technical consultant and developer, working with Web technologies to support open access to research.
Lives in Frome, UK.
0 Followers
0 Following
19 Posts
Joined August 27, 2018
Personal website:
Business website:
Football Club:
Portsmouth
RE: https://social.openrightsgroup.org/@jim/116753611803911869
This is really interesting - not just the immediate issue, but also the general point about how the U.K. government claims it is designing/evaluating policy interventions like this.
Quoting
The social media policy ratchet:
How #onlinesafety measures are costing users but leaving #socialmedia companies just fine thank you.
https://www.openrightsgroup.org/blog/the-social-media-policy-ratchet/
Open quoted postChapter and verse on the Mythos bullshit:
"The flagship demonstration of “unprecedented cyber capability” is in fact a model that weaponized two bugs that a different Anthropic model had already found, in software Mozilla had already patched, in a harness with the actual defenses turned off, where the “triage” step it performed is also performed by its predecessor."
https://www.flyingpenguin.com/the-boy-that-cried-mythos-verification-is-collapsing-trust-in-anthropic/
Replying to
@carusb@mastodon.scot yes - the author there covered the claims about the testing of Firefox in that article too
Replying to
@carusb@mastodon.scot It's pretty much debunked in the article I shared. But I appreciate that that was an enormously detailed piece!
@mia@hcommons.social just saw your announcement on MCG - congratulations! 🥳 😁. 2026 is working out for you 👍
Looking for advice: I want to experiment with some data repository functionality and to do this I need some public-domain (e.g. CC0) & heterogeneous datasets to ingest for testing. Mainly I'm looking for things like:
- datasets - various file formats & data structures
- text (may try Proj. Gutenburg for this?)
- documents/papers
- images
I don't need huge quantities or data volumes - just enough to play with some processes.
Can anyone suggest some useful collections I can "borrow"?
The use of Signposting in COAR Notify (written by @hochstenbach@openbiblio.social).
https://coar-notify.net/guide/signposting/
I'm very pleased to be able to provide this resource: I believe Signposting and COAR Notify complement each other very well!
Replying to
@wragge @cameronneylon
There were various services related to tagging which were used by researchers - I remember “del.icio.us” and I think there were others
Replying to
@mia Good suggestion - thanks! I guess i should revisit BL Labs too - maybe some dig hums datasets there I can use?
Candidate for goal of the tournament 😮 Senegal Belgium 2-0 #worldcup
Replying to
@hochstenbach@openbiblio.social that's an interesting datapoint. I wonder if that is the common case for other repos.