I'm souring on open source code, un-AUTHed relays, etc. I'm opening myself and my servers up to the shady scraper mafia and malicious bot army and creepy Internet stalkers (alive or mechnical), but receiving no discernable benefit in return. The people who work on private code with me can see it and nobody else has ever bothered to even look. And clients that can't AUTH are failed experiments and should simply be allowed to die. Good riddance to bad rubbish. In fact, I might AUTH my websites. You want to see it? Okay. Who are you? Have you already been here 5927401 times this hour? Asking for a fren.

If you want to know why I built #aitherboard, it's not to make a daily-driver. That's what

Imwald

Imwald

Imwald — a user-friendly Nostr client focused on relay feed browsing, publications, and relay discovery.

does quite well. It's to make a *universal app* that you can use to read, edit, cache, share, and publish things that you have trouble finding and working with, on the other apps. I'm designing it to work smoothly with _every kind_ and _every relay_ and _every markup_. Because I need that for testing other apps that I am building. And I put a board on the landing page, so that you have a place to post something, that someone else might actually see. Nobody needs to follow you, to see what you post on that board. You just have to get onto a relay they use and write something. https://aitherboard.imwald.eu/

I did this on a lark, but then I discovered that I can turn my screen lighting all the way down, and still read everything well, It feels like some sort of life hack. View quoted note →

Okay #aitherboard v0.3.0 just dropped. * Write any kind of event. * Edit/clone/replace any event. * View and reply to any event. * Delete events and npubs. * Report events and npubs. * Themes!! 🎉 * Greentext markup in the plain-text feeds and markup feeds. * Fixed responsiveness. * Advanced markup editors (handles both Markdown and Asciidoc and uses Codemirror 🤩) * Raw json always rendered with syntax highlighting. Same for repo files. And etc. Thanks to @The Beave for testing. GM

Always go full terminal theme. #aitherboard

The people who are falling in love with their chatbots are like the ones that fall in love with sex puppets, game characters, and fictional characters in movies and books. Remember those women who would cry and become depressed because a soap opera character died? LOL That you, Bro?

You bot doesn't need a client. It is the client. WTF are you building, bro?

I find it increasingly tiresome to see people claiming AI does all of the work to build websites. That only works with relatively simplistic, shallow websites, which usually end up vaporware because humans don't want to juggle 15 different widgets, just to accomplish what they want. As soon as the architecture or data structure gets a bit complex, you have to roll up your sleeves and basically whip your AI, to get the rest built. Can take days, weeks, months. It's time-consuming, expensive, and exhausting. That's why we have a veritable plague of widget vaporware. Yes, I know your AI can independently churn out 1047201740 different clients per day, like a software factory, but humans don't actually want that. Each new client presents a steep learning curve. People need AI to figure out how to use the software your AI built and they get tired of asking, > Mr. ChatGPT, sir, please look at this website and help me find the Save button. > Mr. ChatGPT, sir, why do the drop-down menus on this app not drop-down? You are using bots to build software only computers will use. That's not modern. It's B2B and IoT and we've been building that boring shit for decades, before you discovered what an IDE is. Which is fine. Huge market. Half the Internet is bots. Have fun with that. Myself, I usually prefer human company. 🤖

We at @npub1s3ht...75wz have now imported half of Project Gutenberg to our #Orly 🦉 relay (actively developed and supported by @mleku). You can see my branch on:

Gitea: Git with a cup of tea

next.orly.dev

clone of https://git.mleku.dev/mleku/next.orly.dev

The second-half import will start tonight (it's a lot!). Then we start importing Wikipedia, in the various languages offered. We'll look for more data sources, after that. Will be millions of mid-sized events, once we're done. If you want to be an #Alexandria Library mirror, you might need to pimp your hardware, but you won't have to increase your event size-limits. Our relay admin @cloud fodder is going to host our first full mirror on wss://theforest.nostr1.com and you can then arrange streams, or whatever, with him. Our subscribers are making this good work possible! 🫶🏻 All of the items are in the public domain (no copyright) or are copyleft (we include a source tag and a clear notice that the source tag has to be maintained in a clone or fork). We will be regularly scanning the original sources for changes and updating the events. Once you have the basic set, you can just sync the updates. Also feel free to maintain a set of event forks, or to archive changes. Our biggest challenge is now making the books easily discoverable, from any client that can render 30040 events, by juicing up relay searches with some high-tech engineering magic voodoo stuff. @semisol and @npub1wqfz...qsyn are both actively working on getting that going. ♥ Both of their systems are getting major updates. Expect cool stuff to get cool even harder. 🫡 That's all for now, and GM.