Thank you to @buercher for creating TootFinder. The service is opt-in, not opt-out, and won’t index accounts that don’t register for it. (This also means that the index will remain small unless users join—hint, hint.)
Not only do they have one of the most private search engines, it is now fully independent. Search engines not tied to either Google or Bing crawlers are good for the web in general. 👍
DDG, however, is still reliant on third parties for results, mostly Bing. With Brave's recent change, #Brave#Search is entirely independent.
On a personal level I also just prefer the results I get from Brave Search. I like how they present "discussions" in a separate area for forum/reddit threads and I like their AI summarizer. I also like how Brave has implemented DDG's !bang system, and seems to redirect quicker.
Any of you using the #Brave#search. What do you think? I just switched yesterday because #DuckDuckGo got a lot worse after Microsoft downmodded Wikipedia in #Bing (probably to make space for their shitty "AI"). So far it seems pretty good in general but very poor for local things here in Sweden.
I love how when you #search on #Mastodon it completely ignores the words you just typed and makes a list of the accounts you already follow in order of popularity.
Here's what I don't like about #firefox for #android: it doesn't render certain pages very well. Look at the image below, and you'll see what I mean. You should notice how the button to clear the search bar sticks out a bit; I actually have to press it before I can move to another result category (like Images) (thx query strings). You can also see that Firefox allows you to zoom out by a lot. Also, it's not just #Google#search that's affected; I have to deal with Firefox obstructing some of the UI in practically every #website I visit. I hope that @mozilla sees this and looks for a way to fix the bug.
Been enjoying my day or so using #Calckey so far. I can do fun things like #search for text or create a #list that lets me add accounts while not requiring me to follow them, and thus having them included on my timeline. I have been wanting that for a while now to aggregate news sources without making my timeline useless.
I still need to mess around with #antennas, but this feature seems like a straight up win over the #hashtag following on #Mastodon.
I do miss using the @IceCubesApp app but pinning the site to my home screen does work pretty well and I still get notifications. Would love to see #IceCubes support this someday, if such a thing is feasible.
Eric Elliott wrote an interesting / scary article yesterday on how to powerfully utilize #LLM's to achieve 10-20x #productivity
I've long held doubts about the ability of #AI to provide the sort of value we often place in it (e.g., web #search has been using the best of intelligent computing for years and overall still delivers only adequate results)...but I've read Elliott's articles for a while and I take seriously his confidence that this is a huge shift.
What do you all think of a #Fediverse#SearchEngine that doesn't index? You can give it a #search term and it goes out and looks through the fediverse, hopping from instance to instance. It would send you an email a few days later with the results.
Again seeing people talking about how they, as posting & reading users, feel #Search Is Necessary on Mastodon. Especially on a day like today, with spicy news!
I rarely see people discuss what search would entail from a technical standpoint, for our small-time server admin users.
How much extra processing would it require? How many people would decide that running an instance is now out of their financial reach, let alone their increased feelings of obligation to act as close to an "available at all times" admin as they can? How would it push us all toward consolidation and fewer, larger instances?
{This isn't a request for speculation or opinions. (Yes, you're correct, thank you, I insist, there's really no need to reply.)}
I'm just continually surprised at how the loudest voices—many who are in journalism, media, or tech—don't seem inclined to facet their Obviously Correct Pronouncements with such small-time nuances as "for whom, under what circumstances, with what effects?".
Nice - now #StackOverflow, which almost assuredly derives the bulk of it's traffic from Google #Search, wants to take the #code YOU gave it for free...
this enables full-text search for posts you haven't interacted with, as well as full-text search for accounts, and includes several advanced filtering operators and parser fixes.
Gesucht: Möglichkeit, mit dem Handy ISBN zu scannen und automatisch mit dem IAI-Katalog abzugleichen (möglichst schnell oder im batch), etwa wie beim Momox-Verkauf.
Gefunden: https://journal.code4lib.org/articles/5014 - aber ich brauch kein schickes App Interface von vor über 10 Jahren, ich brauche es nur für mich selbst und evt. Kolleg:innen
Ich selbst bin zum Programmieren (und Alternativen finden) zu blöde - vielleicht hat jmd eine Idee?
I am looking for a solution to scan ISBN (with my smartphone) and automatically search them in our catalogue. I need it for a very specific purpose: going through large amounts of books at fairs in a quick fashion. Only need to check availability.
I found an older Code4Lib article (see above) but I have no idea of coding and great optimism that there should be easier/newer solutions?
You want your toots to be searchable and findable, then have Tootfinder index your profile.
It works and the best thing is, it is an opt-in option and will not be indexed blindly without consent.
/ via @antiaall3s
"For the moment, we recommend that if #LLMs are used to write scholarly reviews, reviewers should disclose their use and accept full responsibility for their reports’ accuracy, tone, reasoning and originality."
PS: "For the moment" these tools can help reviewers string words together, not judge quality. We have good reasons to seek evaluative comments from human experts.
Update. I've been playing with #Elicit, one of the new #AI#search engines. Apart from answering your questions in full sentences, it cites peer-reviewed sources. When you click on one, Elicit helps you evaluate it. Quoting from a real example:
"Can I trust this paper?
• No mention found of study type
• No mention found of funding source
• No mention found of participant count
• No mention found of multiple comparisons
• No mention found of intent to treat
• No mention found of preregistration"
Having moved across from my previous instance to a new one recently, I thought I should recreate and sort my previous tips and tricks thread into a new one to pin to this profile.
It would mean the world to me if you could look to #Boost this thread around your networks and help spread these tips to as many people as possible.
The goal of this thread is to share with everyone the things that I have learnt during my time here and I will continue to add more #FediTips to it as I continue to learn more myself.
Each tip will have a video to go along with it, but if you would like to view all of them, I have made a playlist on #YouTube that has them all in it.
If you have found other tips etc that you think would be awesome for others to know, please feel free to shoot me a message (or tag me in a post) and I will do my best to create some content for it and add it here.
And with all that out of the way, let the monstrous thread commence!
A lot of people have joined Mastodon, only to then look at their home timeline and see it bare. As there is no algorithm here to fill it with content for you, it is key to learn how to find people to follow from across the Fediverse and how to follow hashtags to help you find the content you are interested in.
The search capability in Mastodon is a bit different to other platforms and so it is important to learn how the #search features work.
Here is a short video on how to best use search to find what you want and I also cover why you may sometimes not initially find what you are expecting.
Please #boost to help spread the knowledge and if you like the content I make, feel free to subscribe to the channel.