The bar is on the floor.

I am a stay-er, and I thought a lot about why that is, and I think most of it is just particular to me and how I operate. I have never played on a game I was not invited to, and once I am established on a game, I don’t even think about looking for a new one.

When I had more free time, I’d try out another game if I was invited, but these days I know that I only really have time for one game at a time. For better or worse, history says that means whatever game I’m playing on at that time until that game shuts down or I quit over sexual harassment (sad lol).

So there’s some interesting work done by social scientists on romantic relationships that, while they probably don’t track directly, I think they have strong parallels with Mushing and people’s relationship to a game/the people who play there, specifically John Gottman at the Relationship Research Institute.

The two things I think track best to MUs are these:

1. A ratio of positive:negative interactions during conflict
2. A ratio of successful:unsuccessful “bids for connection”

You can’t expect a game to provide a connection every single time you want one, but there is some ratio where you feel like generally speaking you’re easily able to get plugged into things when you want to, and below that, it starts to feel like it’s quite hard, even if ‘most’ times you actually are successful.

Gottmann’s numbers for couples are 5:1 and 20:1 respectively to predict a successful relationship. I think the standards are probably higher for a romantic relationship vs a game, but I put the numbers to illustrate that the gap needs to be pretty large to stave off a perceived feeling of ‘this is actually quite difficult’ from creeping in.

@KarmaBum said in Grid vs Web Scenes:

The only flaw I’ve found is that, if you then change the location using the portal, it keeps the original room desc in the scene info.

I thought this for a long time before discovering that instead of going to ‘Edit Scene’ and changing the value for Location, there is an honest-to-god ‘Change Location’ BUTTON under the ‘Play’ menu.

So, at least on the client, Ares does tell you whether the scene was started on the grid or from the web (which is always a temp room).

You can also easily check how many scenes were a temp room or not, where the grid is always not a temp room. On Shattered, the ratio of temp rooms vs not (i.e. started on the grid) was 4204 to 83.

It’s not never but that’s 50:1.

@bear_necessities said in RP Safari - Pacing Styles:

I think what is meant by starting a scene from the grid is that you can be on a MUSH client and scene/start in any room on the grid.

That is what I’m referring to, correct. I never saw it happen.

@Faraday said in RP Safari - Pacing Styles:

You can plunk yourself down on the grid and scene/start a random pickup scene --OR-- you can selectively start private scenes with only the people you want to play with.

You can RP with live/traditional pacing --OR-- you can play async across disparate schedules and timezones.

If more people are choosing the alternatives, maybe it’s just showing that the old ways were never that popular to begin with.

TEZ, THIS IS OFF TOPIC, fork me if you want.

I’m not positive this is a conscious choice people are making as much as it might be influenced by some of the game design of Ares and the web portal in particular. The Ares game I played on (just one, so grain of salt) was probably 80% ‘live’ scenes, and I don’t know that I ever saw someone start a scene from the grid a single time. It wasn’t even clear to me as a new player that you could. What was clear was that using the scene system rather than just walking into a room and posing at each other had huge benefits and you should use it.

Even knowing that you could start a scene from the grid, again, I never saw anyone do it and I never did it even though I’d just come from a game where that was the only way of RPing. Why? Because it is just so much easier to click ‘Create a Scene’ and fill out the fields, which you will need to do anyway to share it and it’s clunky to do them all from the client. You don’t have to wander the grid like a fool peering in shop windows for the right room. You can just select from the list.

Following on to this, the default privacy setting when setting up a scene in this manner is prepopulated for you as private, and therefore I wonder how many people are even thinking about whether this scene could be open or not as they’re going through the motions to get something started if they did not predetermine that they wanted it to be open before they even got to that screen. On a grid-only game, this is not something you have to think about. If you go to a public place and start RPing, it’s an open scene. If you start a scene on Ares’ webportal, unless you deliberately and intentionally set out to have an open scene, the UI assumes you meant it to be private, and people tend to follow the path of least resistance.

Conversely, if you start a scene from the grid without passing additional arguments, no matter where you are, the default is open.

@bear_necessities said in RP Safari - Pacing Styles:

If a majority of scenes on the “active scenes” list began several days ago and are still ongoing, and nearly every single scene on the “recent scene list” are also several days old (versus, say, from yesterday or even two days ago), how does that say “live” to you?

The argument is that people prefer scenes that: a) are not async, b) are taking 7-10 minutes between poses, and c) end within the day.

If we take ‘two days ago’ as 2/23, then the recent scenes for Empty Night has 2 of those and both of them show the IC and OOC date as the same, which usually means it was started and finished on the same day. Aegis Company has 4 of those and all 4 show the same IC and OOC end date.

The active scenes list being mostly ‘stale’ scenes is survivorship bias. The scenes that are not going to finish in a day stay on there; the scenes that are end up on recent scenes, and between these 2 games I’m counting 6 likely same-day real-time scenes in the last 2 days.

I’m not saying we’re drowning in games where this is happening, but I am saying “none” is not accurate.

@bear_necessities said in RP Safari - Pacing Styles:

There are very few public games on Ares, and none of them appear to have a “live scene” culture.

Empty Night and Aegis Company both have a heavy majority of ‘traditional’ pacing scenes listed in their 10+ active scenes. There are probably others.

The truth of MU*s that we’re always going around about on what behavior is expected and what behavior is not okay and who is owed what and what sort of obligation anyone has (player or staff) is that this is all free and no one owes anyone anything. The only obligations are social ones, and you only have as much right to be offended by the perceived deficit as compared to what the other party has explicitly committed to providing.

If staff has a policy of handling all requests within a two-week timeframe and yours is not, then it’s fair to be upset. If there is no policy, then whether you like the way that things are being handled or not is your personal preference and not any reflection on whether staff is good or bad. You’re certainly free to decide this doesn’t match your preference, but it’s not really fair to decry it as a miscarriage of justice.

This is why it is so important for games to have explicit policies for staff and players to follow. The only thing you have to pin down a game on whether the people there are being fair or not is how they measure up to what they have committed to. If they have not committed to anything, you have no ground to stand on, and I would not personally play on a game that did not have any clear policies on subjects I feel strongly about.

@Yam
I prefer web-based by a mile. It is much easier (for most users) to navigate theme files and fill out a form on a web page than running individual commands which all continually push data further up the buffer.

@Faraday
This white paper, published by Originality.ai, which concluded that Originality.ai is the best checker, includes 6 studies in the chart your screenshots come from. 2 of those are from 2023. The newest is from June 2024. Since the article is marked November 2025, I have to question why nothing newer is included.

If there is any takeaway from what I’ve argued in these threads, it is that specifics matter and none of this is static. That is why opposition to GenAI cannot be based on the quality of the experience, it must be based on the nature of the experience. “I don’t want to RP with GenAI because it’s not very good” is not going to age well as a position.

@Trashcan
I had to wait for my free Pangram checks to tick over, but out of curiosity I put the AI generated text above into GPTZero and Pangram.

And then I checked @Tez’s original desc.

Not a huge sample size, of course, but I thought it was interesting.

@Hobbie said in AI In Poses:

With the amount of time investment needed to get AIs these days to do absolutely anything right (see the other thread where I said AI is getting dumber), by the time you carefully cultivate an LLM to write a pose that actually makes sense within the context of the scene, before even making it fit your writing style, you might as well have just written it out yourself.

We have to let this one go. It just is not representative of the experience today. I do not use these tools, but Copilot is installed on every Windows device now and it is trivially easy to try this yourself.

I gave it a desc @Tez wrote and asked it to make a new desc in the same style for a woman.

It used too many exact phrases from the original so I simply asked it to try again.

I gave it some basic instructions about how to pose and intro’d ‘my guy’.

I realized I needed to tell it not to write like a writer (we don’t powerpose) and I was in business.

From here on all I have to do is paste a scene partner’s poses and I could keep ‘Evelyn’ rattling away perpetually with responses coming back to me in 5 seconds.

Again, this took less than 10 minutes. We need to be aware of where these tools actually are.

@Faraday
This is a question about the individual service, not the entire category. For instance, Pangram’s policy:

Pangram does not train generalized AI models like ChatGPT, and our AI detection technology is based off of a large, proprietary dataset that doesn’t include user submitted content.

Also from Pangram:

We train an initial model on a small but diverse dataset of approximately 1 million documents comprised of public and licensed human-written text. The dataset also includes AI-generated text produced by GPT-4 and other frontier language models. The result of training is a neural network capable of reliably predicting whether text was authored by human or AI.

If you refuse to use any technology that relies on machine learning, algorithms, or neural networks regardless of the specifics then obviously that is your prerogative but you are going to have a hard time using the internet at all.

@Yam
I think that some amount of mistakes in any system are acceptable. Nothing is flawless. To me the barrier that a system needs to clear is “better than any alternative”.

In AI detectors, we’ve already seen that most of the time, people unassisted get it right only 50-60% of the time. Certain detectors are performing at level where less than 1% of results are false positive. That seems better.

@Faraday said in AI Megathread:

say you have a self-driving car. Are you OK if it gets into an accident 1 out of every 100 times you drive it?

There were about 6 million auto accidents in 2022. If the self-driving car (extrapolated to the whole population) would have caused 5 million accidents, it would be better.

@Faraday said in AI Megathread:

Say you have a facial recognition program that law enforcement leans heavily on. Are you OK if it mis-identifies 1 out of every 100 suspects?

If this facial recognition program does a better job than humans, yes I am okay with it. Humans are notoriously poor eye witnesses.

Eyewitness misidentification has been a leading cause of wrongful convictions across the United States. It has played a role in 70% of the more than 375 wrongful convictions overturned by DNA evidence. In Indiana, 36% of wrongful convictions have involved mistaken eyewitness identification.

@Pavel said in AI Megathread:

but a sort of “humans relying on authorities instead of thinking” problem

There are cases when humans should rely on authorities instead of thinking. No one is advocating for completely disconnecting your brain while making any judgment, but authoritative sources can and should play a key role in decision-making.

@Faraday
There was cheating before AI and there were false accusations of cheating before AI detectors. Being falsely accused of using AI is no more serious than being accused of plagiarism.

What is the alternative?

@Gashlycrumb
That does sound frustrating, and while I’ve been in defense of the odds of AI detectors not falsely accusing people throughout this thread, it’s still worth noting that even recent studies find wide disparities between product offerings. From one of the studies already linked:

Clearly RoBERTa, the open-source offering, is not something anyone should be using. I hope there’s some sort of feedback mechanism to the administration that the particular tool they’ve selected is highly unsuited to the task.

@Faraday said in AI Megathread:

For example, from the Univ of San Diego Legal Research Center:

If we’re getting down to the level of sample size and methodology, it’s probably worth mentioning that this study looked at 88 essays and ‘recent’ in this context was May 2023, or 6 months after the release of ChatGPT. It is safe to assume the technology has progressed.

@Pavel said in AI Megathread:

are very small in scale

First study:

By using the method described above, we create 6 datasets with around 20K samples each

Second study:

The researchers built a dataset of about 2,000 human-written passages spanning six mediums: blogs, consumer reviews, news articles, novels, restaurant reviews, and résumés. They then used four popular large language models to generate AI versions of the content by using prompts designed to elicit similar text to the originals.

What would you consider an acceptable scale?

@Faraday said in AI Megathread:

You cited a study with a microscopic sample size and flawed methodology,

Fair enough, I can’t find any similar studies with a larger sample size. Most other studies find odds statistically significantly better than a coin flip, somewhere between just barely and the upper 60%s.

@Pavel said in AI Megathread:

The odds of someone familiar with AI output putting it through two different commercial AI detectors in the real world are almost laughably small

Even if we grant that only 2 events must occur, the suspicion (we’ll go with a 50/50) and a single check (with an average from the commercial offerings of 1% false positives), if you’re approaching your 50’s, you’re still more likely to die in a given year than for this to happen to you (0.5%). These tools are aware of the negative ramifications of a false positive and are biased towards not returning them.