Searching txt.sour.is

Twts matching #Gemini
Sort by: Newest, Oldest, Most Relevant

I used Gemini (the Google AI) twice at work today, asking about Google Workspace configuration and Google Cloud CLI usage (because we use those a lot). You’d think that it’d be well-suited for those topics. It answered very confidently, yet completely wrong. Just wrong. Made-up CLI arguments, whatever. It took me a while to notice, though, because it’s so convincing and, well, you implicitly and subconsciously trust the results of the Google AI when asking about Google topics, don’t you?

Will it get better over time? Maybe. But what I really want is this:

  • Good, well-structured, easy-to-read, proper documentation. Google isn’t doing too bad in this regard, actually, it’s just that they have so much stuff that it’s hard to find what you’re looking for. Hence …
  • … I want a good search function. Just give me a good fuzzy search for your docs. That’s it.

I just don’t have the time or energy to constantly second-guess this stuff. Give me something reliable. Something that is designed to do the right thing, not toy around with probabilities. ā€œAI for everythingā€ is just the wrong approach.

⤋ Read More
In-reply-to » Just a small update, on my birthday (on the 5th), I accidentally deleted the main page, of my website, so I'm using that as an opportunity, to try something new, at https://thecanine.smol.pub or gemini://thecanine.smol.pub - depending on your preferred protocol.

@bender@twtxt.net to work through both https and gemini, the site is not written in HTML, but in Gemtext, automatically converted to HTML, when needed. Gemtext is nicely explained for example here: https://garden.bouncepaw.com/hypha/gemtext . In short, it is so limited, no line can be more than one thing, so no links in a list are possible, othar than doing it through something like this primitive workaround.

⤋ Read More

Just a small update, on my birthday (on the 5th), I accidentally deleted the main page, of my website, so I’m using that as an opportunity, to try something new, at https://thecanine.smol.pub or gemini://thecanine.smol.pub - depending on your preferred protocol.

Any feedback is welcome!

⤋ Read More
In-reply-to » @bender Thanks for this illustration, it completely ā€œmisunderstoodā€ everything I wrote and confidently spat out garbage. šŸ‘Œ

… and now I just read @bender@twtxt.net’s other post that said the Gemini text was a shortened version, so I might have criticized things that weren’t true for the full version. Okay, sorry, I’m out. (And I won’t play that game, either. Don’t send me another AI output, possibly tweaked to address my criticism. That is besides the point and not worth my time.)

⤋ Read More
In-reply-to » @bender Thanks for this illustration, it completely ā€œmisunderstoodā€ everything I wrote and confidently spat out garbage. šŸ‘Œ

@prologic@twtxt.net Let’s go through it one by one. Here’s a wall of text that took me over 1.5 hours to write.

The criticism of AI as untrustworthy is a problem of misapplication, not capability.

This section says AI should not be treated as an authority. This is actually just what I said, except the AI phrased/framed it like it was a counter-argument.

The AI also said that users must develop ā€œAI literacyā€, again phrasing/framing it like a counter-argument. Well, that is also just what I said. I said you should treat AI output like a random blog and you should verify the sources, yadda yadda. That is ā€œAI literacyā€, isn’t it?

My text went one step further, though: I said that when you take this requirement of ā€œAI literacyā€ into account, you basically end up with a fancy search engine, with extra overhead that costs time. The AI missed/ignored this in its reply.

Okay, so, the AI also said that you should use AI tools just for drafting and brainstorming. Granted, a very rough draft of something will probably be doable. But then you have to diligently verify every little detail of this draft – okay, fine, a draft is a draft, it’s fine if it contains errors. The thing is, though, that you really must do this verification. And I claim that many people will not do it, because AI outputs look sooooo convincing, they don’t feel like a draft that needs editing.

Can you, as an expert, still use an AI draft as a basis/foundation? Yeah, probably. But here’s the kicker: You did not create that draft. You were not involved in the ā€œthought processā€ behind it. When you, a human being, make a draft, you often think something like: ā€œOkay, I want to draw a picture of a landscape and there’s going to be a little house, but for now, I’ll just put in a rough sketch of the house and add the details later.ā€ You are aware of what you left out. When the AI did the draft, you are not aware of what’s missing – even more so when every AI output already looks like a final product. For me, personally, this makes it much harder and slower to verify such a draft, and I mentioned this in my text.

Skill Erosion vs. Skill Evolution

You, @prologic@twtxt.net, also mentioned this in your car tyre example.

In my text, I gave two analogies: The gym analogy and the Google Translate analogy. Your car tyre example falls in the same category, but Gemini’s calculator example is different (and, again, gaslight-y, see below).

What I meant in my text: A person wants to be a programmer. To me, a programmer is a person who writes code, understands code, maintains code, writes documentation, and so on. In your example, a person who changes a car tyre would be a mechanic. Now, if you use AI to write the code and documentation for you, are you still a programmer? If you have no understanding of said code, are you a programmer? A person who does not know how to change a car tyre, is that still a mechanic?

No, you’re something else. You should not be hired as a programmer or a mechanic.

Yes, that is ā€œskill evolutionā€ – which is pretty much my point! But the AI framed it like a counter-argument. It didn’t understand my text.

(But what if that’s our future? What if all programming will look like that in some years? I claim: It’s not possible. If you don’t know how to program, then you don’t know how to read/understand code written by an AI. You are something else, but you’re not a programmer. It might be valid to be something else – but that wasn’t my point, my point was that you’re not a bloody programmer.)

Gemini’s calculator example is garbage, I think. Crunching numbers and doing mathematics (i.e., ā€œcomplex problem-solvingā€) are two different things. Just because you now have a calculator, doesn’t mean it’ll free you up to do mathematical proofs or whatever.

What would have worked is this: Let’s say you’re an accountant and you sum up spendings. Without a calculator, this takes a lot of time and is error prone. But when you have one, you can work faster. But once again, there’s a little gaslight-y detail: A calculator is correct. Yes, it could have ā€œbugsā€ (hello Intel FDIV), but its design actually properly calculates numbers. AI, on the other hand, does not understand a thing (our current AI, that is), it’s just a statistical model. So, this modified example (ā€œaccountant with a calculatorā€) would actually have to be phrased like this: Suppose there’s an accountant and you give her a magic box that spits out the correct result in, what, I don’t know, 70-90% of the time. The accountant couldn’t rely on this box now, could she? She’d either have to double-check everything or accept possibly wrong results. And that is how I feel like when I work with AI tools.

Gemini has no idea that its calculator example doesn’t make sense. It just spits out some generic ā€œargumentā€ that it picked up on some website.

3. The Technical and Legal Perspective (Scraping and Copyright)

The AI makes two points here. The first one, I might actually agree with (ā€œbad bot behavior is not the fault of AI itselfā€).

The second point is, once again, gaslighting, because it is phrased/framed like a counter-argument. It implies that I said something which I didn’t. Like the AI, I said that you would have to adjust the copyright law! At the same time, the AI answer didn’t even question whether it’s okay to break the current law or not. It just said ā€œlol yeah, change the lawsā€. (I wonder in what way the laws would have to be changed in the AI’s ā€œopinionā€, because some of these changes could kill some business opportunities – or the laws would have to have special AI clauses that only benefit the AI techbros. But I digress, that wasn’t part of Gemini’s answer.)

tl;dr

Except for one point, I don’t accept any of Gemini’s ā€œcriticismā€. It didn’t pick up on lots of details, ignored arguments, and I can just instinctively tell that this thing does not understand anything it wrote (which is correct, it’s just a statistical model).

And it framed everything like a counter-argument, while actually repeating what I said. That’s gaslighting: When Alice says ā€œthe sky is blueā€ and Bob replies with ā€œwhy do you say the sky is purple?!ā€

But it sure looks convincing, doesn’t it?

Never again

This took so much of my time. I won’t do this again. šŸ˜‚

⤋ Read More
In-reply-to » You do raise very good points though, but I don't think any of this is particularly new because there are many other examples of technology and evolution of change over time where people have forgotten certain skills like for example, changing a car tyre

@prologic@twtxt.net when I first ā€œfedā€ the text to Gemini, I asked for a three paragraphs summary. It provided it. Then I asked to ā€œelaborate on three areas: user experience, moral/political impact, and technical/legal concernsā€. The reply to that is too long for a twtxt.

I then asked to counter the OP opinions—as in ā€œhow would you counter the author’s opinion?ā€. The reply was very long, but started like this:

ā€œThat’s an excellent question, as the post lays out some very strong, well-reasoned criticisms. Countering these points requires acknowledging the valid concerns while presenting a perspective focused on mitigation, responsible integration, and the unique benefits of AI.ā€

What followed was extensive, so I asked for a summary, which didn’t do justice to the wall of text that preceded it.

⤋ Read More
In-reply-to » @bender Thanks for this illustration, it completely ā€œmisunderstoodā€ everything I wrote and confidently spat out garbage. šŸ‘Œ

@movq@www.uninformativ.de I am genuinely curious as to why you think Geminis summarization and the categorization of your gopher post was and is as you say misunderstood?

I asked this very genuinely because before reading @bender@twtxt.net’s comments and Gemini summarization I actually went and unplugged your post into flood gaps go for proxy, and then listen to the text intently with my own human ears šŸ‘‚

⤋ Read More
In-reply-to » @bender Thanks for this illustration, it completely ā€œmisunderstoodā€ everything I wrote and confidently spat out garbage. šŸ‘Œ

@movq@www.uninformativ.de this I find more worrisome, and saw no mention of it on your text: Right-Wing Chatbots Turbocharge America’s Political and Cultural Wars (gift article).

Enoch, one of the newer chatbots powered by artificial intelligence, promises ā€œto ā€˜mind wipe’ the pro-pharma biasā€ from its answers. Another, Arya, produces content based on instructions that tell it to be an ā€œunapologetic right-wing nationalist Christian A.I. model.ā€

⤋ Read More
In-reply-to » For the innocent bystanders (because I know that I won’t change @bender’s opinion):

@movq@www.uninformativ.de Gemini liked your opinion very much. Here is how it countered:

1. The User Perspective (Untrustworthiness)

The criticism of AI as untrustworthy is a problem of misapplication, not capability.

  • AI as a Force Multiplier: AI should be treated as a high-speed drafting and brainstorming tool, not an authority. For experts, it offers an immense speed gain, shifting the work from slow manual creation to fast critical editing and verification.
  • The Rise of AI Literacy: Users must develop a new skill—AI literacy—to critically evaluate and verify AI’s probabilistic output. This skill, along with improving citation features in AI tools, mitigates the ā€œgaslightingā€ effect.
2. The Moral/Political Perspective (Skill Erosion)

The fear of skill loss is based on a misunderstanding of how technology changes the nature of work; it’s skill evolution, not erosion.

  • Shifting Focus to High-Level Skills: Just as the calculator shifted focus from manual math to complex problem-solving, AI shifts the focus from writing boilerplate code to architectural design and prompt engineering. It handles repetitive tasks, freeing humans for creative and complex challenges.
  • Accessibility and Empowerment: AI serves as a powerful democratizing tool, offering personalized tutoring and automation to people who lack deep expertise. While dependency is a risk, this accessibility empowers a wider segment of the population previously limited by skill barriers.
3. The Technical and Legal Perspective (Scraping and Copyright)

The legal and technical flaws are issues of governance and ethical practice, not reasons to reject the core technology.

  • Need for Better Bot Governance: Destructive scraping is a failure of ethical web behavior and can be solved with better bot identification, rate limits, and protocols (like enhanced robots.txt). The solution is to demand digital citizenship from AI companies, not to stop AI development.

⤋ Read More

So, in addition to HTTPS and Gemini, my twtxt should now also be available over Gopher (gopher://hashnix.club:70/0/~dce/twtxt.txt). Not sure who, if anyone, would need this; but since my tilde provides Gopher hosting, I’d may as well mirror my twtxt there as well.

⤋ Read More
In-reply-to » Bloody AI clowns:

Here’s an interesting thought/angle on this topic:

gemini://gemini.conman.org/boston/2025/08/21.1

A further check showed that all the network blocks are owned by one organization—Tencent [4]. I’m seriously thinking that the CCP (Chinese Communist Party) encourage this with maybe the hope of externalizing the cost of the Great Firewall [5] to the rest of the world.

⤋ Read More

When I chose the MIT license for all of my software, I thought:

ā€œShould I use GPL, which I don’t really understand? Is that worth it? Yeah, there is a theoretical possibility that some company might use my code in their proprietary product … and then what? Should I sue them to enforce the GPL? I’m not going to do that anyway, so I’ll just use the MIT license.ā€

And now we have those LLM scrapers and now it’s suddenly a reality that these companies (ab)use my code. I can see it in my logs. I didn’t expect that back then.

GPL wouldn’t help, either, of course. (Regardless, I now think that GPL would have been the better choice anyway.)

I’m honestly considering taking my code and website offline. Maybe make it accessible through some obscure protocol like Gopher or Gemini, but no more HTTP.

(Yes, Anubis might help. Temporarily.)

I’m just tired.

⤋ Read More

Computers in school (updated)

Introduction

A much shorter version of this post was initially published on
2022-05-23 (Pungenday, the 70 day of Discord in the YOLD 3188) in my
gemlog at:

gemini://gem.hack.org/log/computers-in-school.gmi

The text has been edited after speaking with some old school mates and
trying to remember more. I also added a few photos.

The beginning

When I started upper secondary school as a sixteen year-old in 1988 my
school had what I think were IBM PC/XT computers, one classroom of
… ⌘ Read more

⤋ Read More

Computers in school (updated)

Introduction

A much shorter version of this post was initially published on
2022-05-23 (Pungenday, the 70 day of Discord in the YOLD 3188) in my
gemlog at:

gemini://gem.hack.org/log/computers-in-school.gmi

The text has been edited after speaking with some old school mates and
trying to remember more. I also added a few photos.

The beginning

When I started upper secondary school as a sixteen year-old in 1988 my
school had wha … ⌘ Read more

⤋ Read More

Computers in school

Introduction

A version of this post was initially published on 2022-05-23
(Pungenday, the 70 day of Discord in the YOLD 3188) in my gemlog at:

gemini://gem.hack.org/log/computers-in-school.gmi

The text has been edited after speaking with some old school mates and
trying to remember more. I also added a few photos.

The beginning

When I started upper secondary school as a sixteen year-old in 1988 my
school had what I think were IBM PC/XT computers, one classroom of
16(?) computers with co … ⌘ Read more

⤋ Read More

Grok AI Gains Vision and Voice Features in iOS App
xAI has launched a new Voice Mode for its Grok chatbot, introducing a feature called Grok Vision that lets users interact with the world through their smartphone camera. Much like ChatGPT and Google Gemini, Grok can now interpret what your phone sees and respond in real time.

Image

Using Grok Vision, iPhone users can point their camera at an object and ask ā€œWhat am I looking a … ⌘ Read more

⤋ Read More

Google Gemini Advanced & 2TB Storage Free for Students
Google is offering their Gemini Advanced AI model for free to students, along with 2TB of free storage. This is a limited time offer where students must sign up by June 30, 2025. Students will need a valid .edu email address to be able to signup for the deal. Google says you can use Gemini … Read More ⌘ Read more

⤋ Read More
In-reply-to » Wait! What's going on?! 🧐

@prologic@twtxt.net Gemini has an answer for you:

This is a conversation thread from a twtxt network, detailing a user’s (movq) frustration with the Mastodon ā€œexport dataā€ feature and their consideration of self-hosting a fediverse alternative. Here’s a summary:

  • movq’s initial issue:
    • movq is concerned about the volatility of their data on their current Mastodon instance due to a broken ā€œexport dataā€ feature.
    • They contacted the admins, but the issue remains unresolved.
    • This led them to contemplate self-hosting.
  • Alternative fediverse software suggestions:
    • kat suggests gotosocial as a lightweight alternative to Mastodon.
    • movq agrees, and also mentions snac as a potential option.
  • movq’s change of heart:
    • movq ultimately decides that self-hosting any fediverse software, besides twtxt, is too much effort.
  • Resolution and compromise:
    • The Mastodon admins attribute the export failure to the size of movq’s account.
    • movq decides to set their Mastodon account to auto-delete posts after approximately 180 days to manage data size.
    • Movq also mentions that they use auto-expiring links on twtxt to reduce data storage.

⤋ Read More
In-reply-to » My twtxt feed is now also available at gemini://roccodrom.de/twtxt.txt

well, I assume by syntax you mean Gemtext (which I like a lot, my personal blog is built on top of it), so I think it might work for twtxt clients…

I knew of twtxt in Gemini Antenna, so at least the 2017 spec might work on that protocol. I think the main issue with extensions is that they weren’t designed with many URLs and protocols in mind.

Also I have to admit that the Gemini community significantly reduced in the last few years. I don’t know how worth it is to add support for Gemini now.

⤋ Read More
In-reply-to » My twtxt feed is now also available at gemini://roccodrom.de/twtxt.txt

Timeline and twtxt-php, don’t support Gemini, only HTTP/S, as a design choice (although originally it was intended to work on Gemtext, it was a niche inside a niche, so it was discarded very soon).

At the moment of building the engine there weren’t many Gemini URLs supporting twtxt 1.1 (with twtxt.dev extensions).
Also User-Agent won’t work there, and many Gemini URLs are a mirror of the HTTP one, so I think is not strictly necessary.

my 2c

⤋ Read More
In-reply-to » (I keep thinking that going back go Gopher or Gemini might be a good idea at this point. They don’t care about that, probably. 🫣)

well, Gemini clients like Lagrange allow to show inline images when you click on an image link. Text based clients, like Amfora, usually allow to watch the image in another ā€˜window’.

For example here: gemini://text.eapl.mx/en-making-a-tic-tac-toe-variant and there https://text.eapl.mx/en-making-a-tic-tac-toe-variant

I agree that some topics require images to make it easier to explain.

⤋ Read More
In-reply-to » reviewing logs this morning and found i have been spammed hard by bots not respecting the robots.txt file. only noticed it because the OpenAI bot was hitting me with a lot of nonsensical requests. here is the list from last month:

(I keep thinking that going back go Gopher or Gemini might be a good idea at this point. They don’t care about that, probably. 🫣)

⤋ Read More

Orbbec Unveils Gemini 215 Stereo 3D Camera and Pulsar SL450 at CES 2025
At CES 2025, Orbbec introduced the Gemini 215 Stereo 3D Camera and the Pulsar SL450 dToF Single-Line LiDAR. The Gemini 215, designed for high-precision scanning, features depth measurement accuracy of less than 0.5 millimeters, multi-camera synchronization, and a lightweight build, making it suitable for various short-range 3D scanning applications. The Gemini 215 is engineered for […] ⌘ Read more

⤋ Read More

Living together: Reflections on collective living
[A version of this post was initially published on 2022-05-30 (Setting
Orange, the 4 day of Confusion in the YOLD 3188) in my gemlog at:

gemini://gem.hack.org/log/collectives.gmi

It’s been slightly edited and a few photos added.

Posted on the blog on 2024-12-15 02:33 +0100. Later updated with how
Area 41 ended.]

I dreamt about Dial House last night. I’ve never been there, but it
was like I belonged, like I was meant to be th … ⌘ Read more

⤋ Read More

Living together: Reflections on collective living
[A version of this post was initially published on 2022-05-30 (Setting
Orange, the 4 day of Confusion in the YOLD 3188) in my gemlog at:

gemini://gem.hack.org/log/collectives.gmi

It’s been slightly edited and a few photos added.]

I dreamt about Dial House last night. I’ve never been there, but it
was like I belonged, like I was meant to be there.

Last I heard anything about Dial House they were trying to gather
enough money … ⌘ Read more

⤋ Read More

ā€˜A cheap magic trick’: Amazon, Google, Meta accused of dodging Senate questions
ChatGPT, Google’s Gemini and Meta’s Llama should be deemed ā€˜high risk’, a Senate committee has found, with executives in the firing line for their appearances. ⌘ Read more

⤋ Read More
In-reply-to » Righto, @eapl.me, ta for the writeup. Here we go. :-)

@eapl.me@eapl.me here are my replies (somewhat similar to Lyse’s and James’)

  1. Metadata in twts: Key=value is too complicated for non-hackers and hard to write by hand. So if there is a need then we should just use #NSFS or the alt-text file in markdown image syntax ![NSFW](url.to/image.jpg) if something is NSFW

  2. IDs besides datetime. When you edit a twt then you should preserve the datetime if location-based addressing should have any advantages over content-based addressing. If you change the timestamp the its a new post. Just like any other blog cms.

  3. Caching, Yes all good ideas, but that is more a task for the clients not the serving of the twtxt.txt files.

  4. Discovery: User-agent for discovery can become better. I’m working on a wrapper script in PHP, so you don’t need to go to Apaches log-files to see who fetches your feed. But for other Gemini and gopher you need to relay on something else. That could be using my webmentions for twtxt suggestion, or simply defining an email metadata field for letting a person know you follow their feed. Interesting read about why WebMetions might be a bad idea. Twtxt being much simple that a full featured IndieWeb sites, then a lot of the concerns does not apply here. But that’s the issue with any open inbox. This is hard to solve without some form of (centralized or community) spam moderation.

  5. Support more protocols besides http/s. Yes why not, if we can make clients that merge or diffident between the same feed server by multiples URLs

  6. Languages: If the need is big then make a separate feed. I don’t mind seeing stuff in other langues as it is low. You got translating tool if you need to know whats going on. And again when there is a need for easier switching between posting to several feeds, then it’s about building clients with a UI that makes it easy. No something that should takes up space in the format/protocol.

  7. Emojis: I’m not sure what this is about. Do you want to use emojis as avatar in CLI clients or it just about rendering emojis?

⤋ Read More
In-reply-to » So I am really curious, now that I am building upon @sorenpeter's Timeline app, how other users write/add their twtxt, and how you follow conversations. Comment svp!

@codebuzz@www.codebuzz.nl I have some shell scripts that handle some of the log formatting details, but I mostly write my mesages by hand. Lately I’ve been browsing twtxt.net since they aggregate most of the known network. I have a couple of demo aggregators sitting around, but I’m in the middle of some infra rebuilds so a lot of my services are offline rn. They’re both built on a simple social graph analysis that extracts urls for your direct follows the follows listed on each of those feeds (friend-of-a-friend replication). certain formatting operations are awkward with my setup, so I may write an app of some kind in the future. likely gemini-based, but I have a number of projects ahead of that one in the queue.

⤋ Read More

Bringing developer choice to Copilot with Anthropic’s Claude 3.5 Sonnet, Google’s Gemini 1.5 Pro, and OpenAI’s o1-preview
At GitHub Universe, we announced Anthropic’s Claude 3.5 Sonnet, Google’s Gemini 1.5 Pro, and OpenAI’s o1-preview and o1-mini are coming to GitHub Copilot—bringing a new level of choice to every developer.

The post [Bringing developer choice to Copilot with Anthropic’s Claude 3.5 Sonnet, Google’s Gemini 1.5 Pro, and OpenAI’s o1-preview](https://github.blog/news-in … ⌘ Read more

⤋ Read More

Orbbec Introduces Perceptor Dev Kit for Advanced AMR Development with NVIDIA Isaac
Orbbec unveiled the Orbbec Perceptor Developer Kit at ROSCon 2024 in Odense, Denmark, offering a comprehensive, out-of-the-box solution for autonomous mobile robot development. Developed in collaboration with NVIDIA, the OPDK is designed to streamline application development for dynamic environments, such as warehouses and factories. The OPDK integrates four Gemini 335L Depth+RGB … ⌘ Read more

⤋ Read More

Simplified twtxt - I want to suggest some dogmas or commandments for twtxt, from where we can work our way back to how to implement different feature like replies/treads:

  1. It’s a text file, so you must be able to write it by hand (ie. no app logic) and read by eye. If you edit a post you change the content not the timestamp. Otherwise it will be considered a new post.

  2. The order of lines in a twtxt.txt must not hold any significant. The file is a container and each line an atomic piece of information. You should be able to run sort on a twtxt.txt and it should still work.

  3. Transport protocol should not matter, as long as the file served is the same. Http and https are preferred, so it is suggested that feed served via Gopher or Gemini also provide http(s).

  4. Do we need more commandments?

⤋ Read More
In-reply-to » should i delete gemini support from twet? iirc in twtxt v2 it starts prohibited. And all of my fields are https

@doesnm@doesnm.p.psf.lt No.

iirc in twtxt v2 it starts prohibited

This is not true. There are no issues supporting fetching feeds via Gemini/Gopher. This is totally fine. What will likely happen is ā€œrecommendationsā€ and ā€œdrawbacks of using Gemini/Gopherā€

⤋ Read More
In-reply-to » @bender Re that broken thread (#bqor23a). Its the same one. My pod doesn't have the Root Twt: https://twtxt.net/twt/bqor23a => 404 Not Found.

Oh, and I think I said this before, but just in case, fuck Gemini. Hell, fuck Gopher too. Bring on telnet, and UCCP. 😈

⤋ Read More

如何用 Go 來構建 LLM 應用
éšØē€å¤§åž‹čŖžčØ€ęØ”åž‹ļ¼ˆLLMļ¼‰åŠå…¶ē›øé—œå·„å…·ļ¼ˆå¦‚åµŒå…„ęØ”åž‹ļ¼‰åœØéŽåŽ»äø€å¹“äø­čƒ½åŠ›é”Æč‘—ęå‡ļ¼Œč¶Šä¾†č¶Šå¤šēš„é–‹ē™¼č€…č€ƒę…®å°‡ LLM é›†ęˆåˆ°ä»–å€‘ēš„ę‡‰ē”ØēØ‹åŗäø­ć€‚ē”±ę–¼ LLM é€šåøøéœ€č¦å°ˆē”Øē”¬ä»¶å’Œå¤§é‡čØˆē®—č³‡ęŗļ¼Œå®ƒå€‘é€šåøøä½œēˆ²ē¶²ēµ”ęœå‹™ę‰“åŒ…ļ¼Œęä¾› API ä¾›čØŖå•ć€‚é€™å°±ę˜Æ OpenAI 和 Google Gemini ē­‰é ˜å…ˆ LLM ēš„ API å·„ä½œę–¹å¼ļ¼›å³ä½æę˜Æåƒ Ollama é€™ęØ£ēš„č‡ŖčØ—ē®” LLM å·„å…·ļ¼Œä¹Ÿå°‡ LLM å°č£åœØ RES ⌘ Read more

⤋ Read More
In-reply-to » @bender It's just a simple twtxt2html and scp ... it goes like:

@quark@ferengi.one Mine is a little overkill šŸ˜‚ but I need to do something for practice:

#!/bin/bash
set -e
trap 'echo "!! Something went wrong...!!"' ERR

#============= Variables ==========#

# Source files
LOCAL_DIR=$HOME/twtxt

TWTXT=$LOCAL_DIR/twtxt.txt
HTML=$LOCAL_DIR/log.html
TEMPLATE=$LOCAL_DIR/template.tmpl

# Destination
REMOTE_HOST=remotHostName     # Host already setup in ~/.ssh/config

WEB_DIR="path/to/html/content"
GOPHER_DIR="path/to/phlog/content"
GEMINI_DIR="path/to/gemini-capsule/content"

DIST_DIRS=("$WEB_DIR" "$GOPHER_DIR" "$GEMINI_DIR")


#============ Functions ===========#

# Building log.html:

build_page() {
	twtxt2html -T $TEMPLATE $TWTXT > $HTML
}

# Bulk Copy files to their destinations:

copy_files() {
	for DIR in "${DIST_DIRS[@]}"; do
    # Copy both `txt` and `html` files to the Web server and only `txt`
    # to gemini and gopher server content folders
		if [ "$DIR" == "$WEB_DIR" ]; then
			scp -C "$TWTXT" "$HTML" "$REMOTE_HOST:$DIR/"
		else
			scp -C "$TWTXT" "$REMOTE_HOST:$DIR/"
		fi
	done
}

#========== Call to functions ===========$

build_page && copy_files

⤋ Read More

ę§‹å»ŗē”±å¤§åž‹čŖžčØ€ęØ”åž‹ļ¼ˆLLMļ¼‰é©…å‹•ēš„ Go ę‡‰ē”ØēØ‹åŗ
éšØē€éŽåŽ»äø€å¹“å¤§åž‹čŖžčØ€ęØ”åž‹ļ¼ˆLLMļ¼‰åŠå…¶ē›øé—œå·„å…·ļ¼ˆå¦‚åµŒå…„ęØ”åž‹ļ¼‰ēš„čƒ½åŠ›é”Æč‘—å¢žé•·ļ¼Œč¶Šä¾†č¶Šå¤šēš„é–‹ē™¼č€…é–‹å§‹č€ƒę…®å°‡ LLM é›†ęˆåˆ°ä»–å€‘ēš„ę‡‰ē”ØēØ‹åŗäø­ć€‚ē”± ę–¼ LLM é€šåøøéœ€č¦å°ˆē”Øē”¬ä»¶å’Œå¤§é‡ēš„čØˆē®—č³‡ęŗļ¼Œå®ƒå€‘ęœ€åøøč¦‹ēš„å½¢å¼ę˜Æä½œēˆ²ęä¾› API čØŖå•ēš„ē¶²ēµ”ęœå‹™ć€‚é€™å°±ę˜Æé ˜å…ˆēš„ LLM 如 OpenAI ꈖ Google Gemini ēš„ API ēš„å·„ä½œę–¹å¼ļ¼›å³ä½æę˜Æé‹č”Œä½ č‡Ŗå·±ēš„ LLM å·„å…·ļ¼Œå¦‚ Ollama[1]ļ¼Œä¹Ÿęœƒå°‡ ⌘ Read more

⤋ Read More
In-reply-to » @falsifian In my opinion it was a mistake that we defined the first url field in the feed to define the URL for hashing. It should have been the last encountered one. Then, assuming append-style feeds, you could override the old URL with a new one from a certain point on:

I was not suggesting to that everyone need to setup a working webfinger endpoint, but that we take the format of nick+(sub)domain as base for generating the hashed together with the message date and content.

If we omit the protocol prefix from the way we do things now will that not solve most of the problems? In the case of gemini://gemini.ctrl-c.club/~nristen/twtxt.txt they also have a working twtxt.txt at https://ctrl-c.club/~nristen/twtxt.txt … damn I just notice the gemini. subdomain.

Okay what about defining a prefers protocol as part of the hash schema? so 1: https , 2: http 3: gemini 4: gopher ?

⤋ Read More
In-reply-to » @falsifian In my opinion it was a mistake that we defined the first url field in the feed to define the URL for hashing. It should have been the last encountered one. Then, assuming append-style feeds, you could override the old URL with a new one from a certain point on:

how little data is needed for generating the hashes? Instead of the full URL, can we makedo with just the domain (example.net) so we avoid the conflicts with gemini://, https:// and only http:// (like in my own twtxt.txt) or construct something like like a webfinger id nick@domain (also used by mastodon etc.) from the domain and nick if there, else use domain as nick as well

⤋ Read More

Google Chrome will have Gemini LLM built into the browser.

⤋ Read More

Google Chrome will have Gemini LLM built into the browser.

⤋ Read More

Gemini, le moteur ultra-Woke de Google
La semaine derniĆØre, alors qu’OpenAI faisait frĆ©mir le monde de l’intelligence artificielle avec Sora, le premier moteur permettant de crĆ©er des petites vidĆ©os rĆ©alistes Ć  partir d’une simple description textuelle, Alphabet (la maison-mĆØre de Google) lanƧait en fanfare la nouvelle version de son agent conversationnel, rĆ©ponse de l’entreprise mondiale Ć  ChatGPT d’OpenAI, elle-mĆŖme dĆ©tenue par […] ⌘ Read more

⤋ Read More

Status 2024-01-29
Friday is my day off from work, as usual. So when I’m typing this I’m
in front of the hackstation (not a battlestation, obviously) with my
third cup of coffee, writing an update again.

I’ve been doing these status updates on my Gemini log, but I’m
increasingly aware of the dropping amounts of traffic, so I’m thinking
about doing them on the blog instead, but see below for some thoughts
on Gemini.

Abstract

In which I speak about an intense week, feeling good(?), spending … ⌘ Read more

⤋ Read More

Orbbec’s Latest AI-Driven Camera with PoE priced at $479.99
The Gemini 2 XL, Orbbec’s latest addition to 3D vision technology, is engineered for a variety of robotic and AI-driven applications. This camera excels in providing accurate and dependable data in diverse lighting environments, from absolute darkness to bright outdoor conditions. Equipped with a global shutter for both RGB and IR cameras, it ensures high-quality, […] ⌘ Read more

⤋ Read More

Another part of this crisis is that I like the idea of what I was doing with gemini, but the main issue here is that hosting from my house when my internet is terminated every month for 10-15 days is a problem. Not just for my sanity, but also for reliability

⤋ Read More

H3: Instead of C3
[Updated with correct Gemlog link.]

A version of this was posted on on 2023-01-06 but I thought it might
also fit here. Go to my gemlog for somewhat more personal takes and
see what I publish first. IPv6 only!

gemini://gem.hack.org/mc/log/

As long-time readers know I have participated in the Chaos
Communication Congress (C3) in Germany every year since 2008.

Since C3 was cancelled this year I thought I’d arrange a very small
conference of my own. I would at least try to gather some friends and
acquaintances … ⌘ Read more

⤋ Read More

H3: Instead of C3
A version of this was posted on on 2023-01-06 but I thought it might
also fit here. Go to my gemlog for somewhat more personal takes and
see what I publish first. IPv6 only!

gemini://gem.hack.org/log/

As long-time readers know I have participated in the Chaos
Communication Congress (C3) in Germany every year since 2008.

Since C3 was cancelled this year I thought I’d arrange a very small
conference of my own. I would at least try to gather some friends and
acquaintances in chat and video conference and watch t … ⌘ Read more

⤋ Read More
In-reply-to » What do you feel when you listen to something you didn't believe it's true?

@eaplmx@twtxt.net This exact thing happened to me last night. I happened to be watching some random Youtube video, then this Ad came on, normally they are short 3-5s ads and I just tolerate them (sometimes) – But this particular ad was 20+ mins long! Somehow I kept listening to it too, despite my daughter telling me I could hit that ā€œSkip Adā€ button.

What was it you ask?! šŸ˜… It was one of those testimonial-style, hyped up marketing videos of some product called ā€œGemini 2ā€ (a currency trading app, allegedly), I kept watching all the way through, it was fantastic! 🤣

Then I went and read up on it! …

Short answer: TOTAL FUCKING SCAM 🤣

⤋ Read More

Tillitis Key 1
The secret project I’ve been mentioning in my gemlog is out!

This week we presented the Tillitis Key 1 at the Open Source Firmware\
Conference (OSFC). What we unveiled is a new kind of USB security
stick and a new company, a sister company to Mullvad VPN: Tillitis.

The stick is a small computer that can load and run small programs
uploded to it from a host computer. It always measures … ⌘ Read more

⤋ Read More