Untitled attachment

Yohan Yukiya Seseㆍ사요한・謝雪矢 (youronlyone@c.im)'s status on Sunday, 31-Mar-2024 14:17:34 JST

in reply to

I'll just write down what I can think of.

I. Gaming

A. We already have an AP-based game running. It utilises polls and let the public decide the course of action of the character.

B. I think we can further expand the use of AP in gaming. What I have in mind is, an account's statistics (following, followers, activeness [how many posts per day; which time of the day; how many characters; was there an image attached; etc.], and other metrics) determines the user's character in a game software/app. There are various formulas in place to avoid intentional spamming and gaming the system, but that's up to the gamedev to develop.

C. Another I can think of is a trading-style game based on AP accounts.

It is similar to a now-defunct game that uses a website, and puts them up in a fictional stock exchange (I can't remember the name, but it lasted almost 20+ years).

Another similar game is play.empire.kred. Their style is a user is placed in a fictional trading market. A user adds their SNS profiles/pages which will help establish their value. And if they also keep on updating, it gives them more score and increases their stock value.

We can combine both and apply it to AP accounts. Again these are fictional, no crypto, no conversion of “fictional money” to real-money (the game publisher is going to have serious trouble with SEC if that happens).

II. Emergency Response & Mapping system.

With the decentralised, distributed, federated nature of the network, different towns, and cities, can establish their own instances for emergency response & mapping systems. They either have an AI, or manual labour, monitoring the fediverse network for certain keywords, and they boost it so their own instance will see it and record it on their server.

Then their responders, or another AI system, processes this information and collate it in an easy to consume database their responders, and key people, can easily pull info from. And there will also be a map available for them, based on the information from the boosted posts.

These instances are exclusive, and the accounts boosting doesn't have to be followed by anyone other than those in the response teams. So these accounts are aware of users to watch for, users voluntarily request to be monitored for certain keywords (opt-in). OR, they tag the account in case of emergency reports.

For example, me:
“Hi @-philippinedisasterreponse We currently have 300 injuries, and counting, here at street name, city name, due to the super typhoon.”

Then their AI collates it, put it in their database. The account will probably reply to acknowledge the report. And their response teams can open their internal map and info system to see the reports and other relevant information.

It can also be useful in “trapped” situations. If the victim still have Internet access, and in such a situation hotlines are more likely flooded with calls; a post in the fediverse can at least make the rescuers aware where they are.

Embed this notice

K‮ly‬e (khird@qoto.org)'s status on Saturday, 11-Nov-2023 10:12:52 JST

K‮ly‬e

in reply to

@freemo depends. There are certain times it can be an obstacle. For example, the undo/cut/copy/paste commands are now scattered across the keyboard. Some software that uses keys based on position rather than letter value (e.g. WASD keys if you're a gamer) works automatically, some needs to be configured, and some just leaves you stuck with unusable keybinds. Importantly, early in your boot process (BIOS password, disk encryption) you'll still have to type in QWERTY because the computer doesn't know anything about your keyboard settings until after you've decrypted that information.

It will restrict your options for customising your keyboard, because many keycap sets have different profiles on a per-row basis, so keys that are in a different row between QWERTY and Dvorak will stick out. Obviously you'll touch-type most of the time, but sometimes you run into situations where you have one hand on the keyboard and one on the mouse, and you'll need to find a key with your left hand for which only your right hand has muscle memory.

Also, if you swipe-type on your phone, some of Dvorak's strengths become weaknesses. Alternating fingers turns into a lot of back-and-forth. Concentrating heavily used keys onto the homerow means you have to be very precise, and even then there are a lot of ambiguous words. For example, swiping S-O-T could be:
- soot
- snot
- snout
- stout
- shot
- shoot
- shout
and that's if you aim perfectly. If you miss by one key and type S-E-T there's a whole different set of words. The poor autocorrect can only do so much for you.

Embed this notice

Wander ΘΔ :verified_paw: (wander@packmates.org)'s status on Wednesday, 28-Jun-2023 00:32:36 JST

Wander ΘΔ :verified_paw:

in reply to

involved reached out to ask and get more information.

It's increasingly difficult for me to not side with my users if the alternative is a moderation style that's aggressive and ruthless.

But this isn't about me. When these new Reddit refugees express their concerns about something that seems to have been normalized on #mastodon or the wider #fediverse, I realize that I'm not alone.

Jabber / #XMPP did not have this problem. Neither does email, mainly because it's all private messages, but we better do something to be more responsible with our users to which we owe a sensible moderation instead of doing things on a whim with disregard of the people it affects.

#Lemmy / #Kbin needs to implement silences and Fediblock's popularity needs to be converted into Fedimute.

Fediblocking IMHO is almost only warranted if a remote instance is spammy, harassing or disruptive. Just like blocking spam servers.

Muting should be implemented in such a way that it allows the instance owner to

Public

Notices where this attachment appears