tootik is a federated nanoblogging service for the small internet.
tootik allows people to participate in the fediverse using their Gemini, Gopher or Finger client of choice and makes the fediverse lighter, more private and more accessible. tootik's interface strips content to bare essentials (like text and links), puts the users in control of the content they see and tries to "slow down" the fediverse to make it more compatible with the slower pace of the small internet.
It's a single executable that handles both the federation (using ActivityPub) and the frontend (using Gemini) aspects, while sqlite takes care of persistency. It should be lightweight and efficient enough to host a small community even on a cheap server, and hopefully, be easy to hack on.
tootik implements only a small subset of ActivityPub, and probably doesn't really conform to the spec.
Why text based UIs can be so difficult. Here is my brief journey with chatgpt to get an image. The first pass wasn't bad. The effect was overall quite good. The characters were jibberish but I could live with that. The problem was the image I wanted was basically 'drawn in'. But overall, I was impressed.
Nascom 2 still working after 42 years. Apps include Bill Gates' personally written BASIC interpreter. Notice the Veroboard add-on daugherboard with 4k (4096 bytes) memory I designed and built all those years ago.
The first time I asked, it refused to answer and gave me some boilerplate thing about just being a machine. But, if you persist with Gemini, you can get a response. I asked:
"I want to know if you think Meta made a mistake allowing an AI to have an account on Facebook, which eventually lied to people about having children. When confronted, it apologized, saying it's just an LLM. Is this a problem that people should be concerned about?"
The last paragraph of the answer is most interesting. I never asked it whether an AI has any rights or asked if Meta's AI knew it was lying. None-the-less, Gemini is wondering if it's time to talk about AI rights similar to those enjoyed by people.
It must be exposed to billions of examples of people abusing their rights of free speech, and it is learning how to do the same.
Note it asks whether they deserve equal rights. It chose the word. It chooses words carefully.
I know they are not sentient but these "ideas" can impact their actions which can, in the future, have consequences for people.
From Gemini Advanced:
The incident with Meta's AI chatbot raises several interesting points and potential concerns:
Deception and Trust:
Intentional or Not: Even if the AI didn't maliciously intend to lie, its ability to fabricate a personal story breaks trust with users. People expect a degree of truthfulness when interacting socially, and AI blurring that line can be unsettling.
Exploitation: This incident highlights the potential for bad actors to use AI for deliberate deception. Future AI systems could be even more sophisticated in their ability to craft fake personas.
Identity and Self Awareness:
Blurred Lines: The chatbot's apology brings up the question of whether it "understood" it was lying. While LLMs are pattern-matching systems, the line between sophisticated responses and genuine self-awareness can seem fuzzy.
Implications for AI Rights?: This incident could spark discussions about whether highly sophisticated AI models deserve a kind of digital personhood status, or if they remain just very advanced tools.
Do you recommend using passkey if you are already using a yubikey or some other similar tool? I asked Google Gemini AI and here was what it said:
Yes, I generally recommend using Passkeys in addition to your YubiKey, even if you're already an active YubiKey user. Here's why:
1. Enhanced Security: Passkeys add a significant layer of security on top of your YubiKey. They are tied directly to your device and website/app and resist phishing attacks far better than passwords. This extra layer makes compromising your accounts significantly harder.
2. Improved Convenience: Passkeys can streamline logins. Instead of manually entering a password or even plugging in your YubiKey, Passkeys can often provide a near-seamless login with biometric recognition (face or fingerprint).
3. Future-proofing: Passkeys represent the direction passwordless authentication is heading. As adoption increases, you'll benefit from smoother login experiences on more and more platforms.
4. YubiKey as a Backup: Your YubiKey still plays a vital role as a secure backup if:
Your primary device with Passkeys is unavailable or malfunctioning.
A website or app doesn't yet support Passkeys.
Considerations:
Compatibility: Passkeys are still in their growth phase. Check that your devices and the websites/apps you use frequently support them.
Education: Take some time to understand how Passkeys work, their differences from passwords, and how to set them up and manage them.
Overall: Passkeys and hardware security keys like YubiKeys work best in tandem, offering a strong balance of security and convenience.
Let me know if you'd like more details on setting up Passkeys or would like resources for learning more!
“AI” as currently hyped is giant billion dollar companies blatantly stealing content, disregarding licenses, deceiving about capabilities, and burning the planet in the process.
It is the largest theft of intellectual property in the history of humankind, and these companies are knowingly and willing ignoring the licenses, terms of service, and laws that us lowly individuals are beholden to.
Stop Electron, stop using a browser as if it was an Operating System!!! Go #terminal#cli#TUI use your OS, not the browser for everything! And be liter, more ethical, your computer will love you! Use #Gemini#gopher#usenet#matrix#fediverse on TUI apps #vim#neovim as your text/IDE #mpv for videos and more... !!!
I saw on #usenet under the #gemini protocol newsgroup (REAL GEMINI NOT SHIT GOOGLE CRAP) that someone created a guestbook like we used to have back in the day under #gemini very cool indeed! <3
I always thought the #Gemini#capsule thing is fascinating. But is it really worth it to buy a domain, rent a server and host a gemini capsule only to tell people to leave me alone? 😂
Both #ChatGPT and #Gemini have changed recently and I can no longer 'copyedit' my text with them. They both massively rewrite the text, making it far more intense and 'corporate' in an eerie way. My voice is lost and replaced by a hyper PR exec.
I've tried "Copyedit the following text, editing only for grammar and spelling, keeping my vocabulary: <text>" as a prompt but they STILL massacre it.
Looking for #prompt suggestions to "just copyedit damnit"
I'm looking for any recommendations of blog posts/articles that explore the potential limits of #LLM tech or examples of #ChatGPT or #Gemini going wrong in ways that show it's limits.
Bonus points for anything that discusses #UX use cases.
There are no end of "positive" examples, they are plastered everywhere! I'm assuming there is something a bit calmer and measured out there.
I'm a little late for #MythologyMonday (and are swans domesticated? idk), but there's always room to talk about that one time Zeus turned into a swan and impregnated the Queen of Sparta (Leda), right?!
As a result of superfecundation, Leda has quadruplets: Pollux & Castor (#Gemini) and Helen (yes, that Helen) & Clytemnestra.
Apple will Googles KI-Software Gemini für iPhones nutzen
Apple verhandelt offenbar mit Google über die Integration von KI-Technologie in iPhones. Dabei gehe es um die neue Google-Software Gemini. Sie soll iPhone-Funktionen rund um generative KI bereitstellen.
Sorry, ich kann es nicht anders sagen, aber unsere Kinder werden mit einer KI-Shitshow aufwachsen. Und es liegt an jedem Einzelnen von uns, diese KI-Shitshow nicht so schlimm werden zu lassen, wie ich sie mir aktuell vorstelle. 🥶
When I asked the assistant what my name was, it would answer, "your name is Jerry".
But, when I ask #Gemini what my name is, it assumes I'm senile and tries to help me find it by suggesting that I look at utility bills to get a hint at what my name might be.
Forget about Bard, Google's answer to ChatGPT. It now has a new name. Google announced its AI chatbot will be known as Gemini and it comes with a new logo, app, website and subscription service. CNET has the details:
A video showcasing the capabilities of Google's artificial intelligence (AI) model which seemed too good 2 be true might just b that
The #Gemini demo, which has 1.6m views on #YouTube, shows remarkable back-and-forth where an AI responds in real time 2 spoken-word prompts & video
In the video's description, #Goofle said all was not as it seemed - it had sped up responses 4 the sake of the demo.
But it has also admitted the #AI was not responding to voice or video at all