Simone SIf you're using Proton and/or Threema or other similar software, you might want to keep an eye on this:
@extenebris
- 9 Followers
- 50 Following
- 57 Posts
biased opinions on subjects that scarcely matter
chesheerThe more I think about it, the more it seems to me that in this day and age with all the modern threats having a text editor that is capable to not only connect to the Internet, but also install some code packages from repositories (and probably do dependency resolving) is a recipe to catastrophe. Sooner or later.
It's probably one thing when you use a curated list of half a dozen addons that you can even personally peruse (or even contribute to). It's a whole other thing when you use some huge "distro" with probably hundreds of packages that also receive constant updates you cannot possibly control.
It's mostly about #Emacs, of course, but #vim is fully capable of it too. I won't even mention the likes of #VSCode.
We had a fair share of supply chain attacks in the recent years (npm, pip, even xz in some way). No reason to think no one's gonna use this channel of attack.
Maybe it's just my fibs. But there is some uneasy feeling about the fact that you edit, perhaps, extremely private, personal or sensitive texts while your editor runs some background code doing who knows what. It's one thing to trust people who wrote vim or Emacs and a whole other thing to trust a hundred other unknown parties at the same time.
It's probably one thing when you use a curated list of half a dozen addons that you can even personally peruse (or even contribute to). It's a whole other thing when you use some huge "distro" with probably hundreds of packages that also receive constant updates you cannot possibly control.
It's mostly about #Emacs, of course, but #vim is fully capable of it too. I won't even mention the likes of #VSCode.
We had a fair share of supply chain attacks in the recent years (npm, pip, even xz in some way). No reason to think no one's gonna use this channel of attack.
Maybe it's just my fibs. But there is some uneasy feeling about the fact that you edit, perhaps, extremely private, personal or sensitive texts while your editor runs some background code doing who knows what. It's one thing to trust people who wrote vim or Emacs and a whole other thing to trust a hundred other unknown parties at the same time.
unfa🇺🇦If you're using an android phone you need this:
https://keyboard.futo.org/
https://www.youtube.com/watch?v=cFP5bp3JvaU
I have been on the lookout for a sensible Gboard replacement that wasn't making my (voice) typing experience painful, and so far only FUTO Keyboard managed to provide that.
It has really good offline voice typing as well, which is something I use a lot.
I can not recommend this enough!
#FUTO #Android #Privacy #SpeechToText #VoiceTyping #Swype #Gboard #Heliboard #Florisboard
people use ai to write exactly what I don't want to read:
- ads
- corporate emails
- fanfics
- books that are collections of ”tropes”
- mediocre longreads
- motivational posts (”daily grind” 💀)
- clichéed affirmations (”you are enough”)
- lists (jk, I love lists)
this is not a llm-hate post, I use llms daily; it says more about the so called "content" and the ultimate soullessness thereof
David Fi&er"Emacs is a Gnostic cult. And you know what? That’s fine. In fact, it’s great. It makes you happy, what else is needed? You are allowed to use weird, obscure, inconvenient, obsolescent, undead things if it makes you happy." -- You can choose tools that make you happy
https://borretti.me/article/you-can-choose-tools-that-make-you-happy
This is cool, I didn't know about FSRS. Maybe that's why Anki kind of never grew on me 🤔 going to try that out
Simon WillisonI got access to Gemini Diffusion, Google's first diffusion LLM, and the thing is absurdly fast - it ran at 857 tokens/second and built me a prototype chat interface in just a couple of seconds, video here: https://simonwillison.net/2025/May/21/gemini-diffusion/
mobidic@tealeg Same source - another view: https://www.technologyreview.com/2025/05/20/1116274/opinion-ai-energy-use-data-centers-electricity/
By the way - you can read the ALT-Text for more ( ;
TutaDeGoogling is possible, and it doesn't need to be difficult. đź‘Ź
Take a look at our in-depth guide of Google alternatives to learn how you can take back your privacy in 2025. ❤️🔒
👉 https://tuta.com/blog/how-to-leave-google-gmail
Have you already DeGoogled? If so, let us know your favorite Google-free apps.
Terence TaoI was planning to write here a short text about how, when considering whether to use AI for a task, that one should take into account not only the difficulty/complexity of the task, but also the acceptable failure rate; for instance, using an AI to suggest a recipe for dinner has an acceptable failure rate when just cooking for oneself, but would be inadvisable for a head chef preparing a state banquet for a high-profile diplomatic function, even if the two tasks are essentially of comparable difficulty (per person served, at least). But I realized that since this writing task was itself simple and with a high acceptable failure rate, it made sense to just to let an AI summarize this point directly in a table form, as enclosed below; it contains minor imperfections, but certainly suffices for the task at hand. [My prompt for this can be found at https://chatgpt.com/share/67e813bc-590c-800e-90a3-79d4115a5053 ]
