-
Hacky workaround while I decide what to do about #Mastodon: Python script takes various Mastodon URIs, converts them to RSS URLs, and subscribes to them
All that I had to do on top of that is add the script as a system “Service” (in the menus, via Automator) and now I can subscribe to Mastodon as RSS from any application. Once I am happy with it, I’ll publish it somewhere.
-
Asking #ChatGPT: Please cite a few things which are legal online but illegal offline. #OnlineSafetyBill
This is an issue which has peppered debate on the Online Safety Bill, for example: The noble Baroness is right: what is illegal offline should be illegal online, and it is very clear that the social media companies should remove that content. Where there is harmful but legal content, they need to have very clear
-
Asking #ChatGPT: Please write a 600 word essay on the decline of politeness in society that will result from younger generations not saying “please” or “thank you” to generative AI bots, causing those phrases to drop out of fashion in daily use.
Response from ChatGPT The decline of politeness in society is a growing concern, particularly as younger generations increasingly interact with generative AI bots. These bots are designed to make our lives easier and more efficient, but as they become more advanced and widespread, there is a danger that certain social norms and customs will begin
-
Asking #ChatGPT: Could you please write 500 words in the style of a Daily Telegraph op-ed about whether the #OnlineSafetyBill will meet its objectives?
The following response was generated by ChatGPT… As the Online Safety Bill makes its way through Parliament, many are questioning whether it will be able to achieve its stated objectives. On the surface, the bill appears to be a well-intentioned effort to protect children from harmful content online. But upon closer examination, it becomes clear
-
EU Post-Pegasus Report, drawing on EDRi paper (?) calls for regulation of vulnerability disclosure
I don’t know who needs to hear this, but this (in the current report draft) is a very, very bad idea. https://www.europarl.europa.eu/doceo/document/PEGA-PR-738492_EN.pdf > Sophie in ‘t VeldVulnerabilities 151. Without vulnerabilities in software, it would be impossible to install and deploy spyware. Therefore, in order to regulate the use of spyware, the discovery, sharing and exploitation of
-
Notes for interview with @TonyaJoRiley of @CyberScoopNews regarding #EndToEndEncryption of Direct Messenger on Twitter post-Musk
So, I got interviewed for this; as usual I took a big set of questions via email, and then some back-and-forth over Signal for clarifications and followup: Interview Notes So, to answer the usual last question first, as I am sure you already know otherwise you would not be contacting me: I’m Alec Muffett, I
-
Suddenly #Mastodon: history teaches what will happen when Twitter-expatriates discover “federated social networks” are a volatile blend of unsafe spaces & stifled speech with blunt, imprecise instruments to address their woes @jonronson @jamesrbuk @bradtem
Elon took over Twitter, both the community and the products are tearing themselves apart, and a bunch of people who have probably never used a federated social network before are beginning to experience the joy of having nobody they can usefully complain to. But those of us who are old enough and nerdy enough, have
-
Slides, Scripts and PDFs for the Encryption Panel session at #UKIGF #UKIGF22
I am chairing the session on encryption at UKIGF on the 1st of November, 2022. For reasons of size and timing the slides did not make it into the conference’s monolithic megadeck, so I attach them below, along with the draft speaker notes, in a variety of formats; hopefully attendees can use them to follow
-
More evidence of how the #DigitalMarketsAct demand for #Interoperability will turn #EndToEndEncryption Messenger Apps into festering pools of Spam
Previously Previously Previously Via a discussion on HackerNews, I came across this 2014 email to the ModernCryptography list, outlining spam prevention from the content-centric approach that GMail takes, via Mike Hearn — a former GMail engineer. It’s a great read, and I particularly noted this bit, with my emphasis: Systems like WhatsApp don’t seem to
-
“XCheck at Meta: Why it exists and how it works” by Krishna Sundarram
What a timely posting; I particularly like the war stories of how trolls and other abusers circumvent and then leverage “approval” processes, in order to create strongly authenticated abusive accounts which can do significant damage because they require escalation to have them finally removed.
-
“An adversary can imperceptibly perturb an image that would normally be placed in this bank, and all live instances of the benign image will then be either incorrectly removed or flagged”
Adversarial collision attacks on image hashing functions https://aisecure-workshop.github.io/amlcvpr2021/cr/9.pdf
-
One aspect that I really appreciate about getting older is the increasing number of past events that you thought were “missed opportunities” being later revealed as “dodged bullets”
Yes, of course there are a few regrets, but actually over-all the trend is downwards.