Duelyst Forums

A Study of Friendship


#21

hey man. that’s tough. i’ve had something… well the effect was the same. i killed something with a person that was very, very dear to me, and i still feel like i need to compensate them for it. My advice is that even though we can’t really know what to do, just do it if it feels right. We could die at any moment, and we have to do what we have to while we have the time to. Like you said, no regrets. when you do something, think about whether or not you would be ashamed more of doing it or not doing it on your death bed. wow. that escalated quickly. Good luck my dude, and remember that all roads lead home.


#22

Hmm. Hard to say.

Initially, I had thought of it as a data structure with only a single value: the amount of time recently spent with them. But this is too simple. I think I’m still friends with people I haven’t talked to in years. Thus, at the very least it could be a linear combination:
f=(time spent together in last 5 years)*weight1+
(time spent together in last year)*weight2+
(time spent together in last month)*weight3+
(time spent together in last week)*weight4

However, this model has flaws as well. What if you spend a lot of time with a person you don’t like that much? Clearly, we need more data members (what is a data member? If a data structure is a TABLE, then a data member is a COLUMN) to account for this. But first, let’s call this linear combination of time spent “Familiarity.” But what are the properties of “familiarity?” Clearly, being very familiar with someone has a strong effect on your friendship. Stopping contact with someone can bleed away your familiarity quickly.

Or rather, now that I think of it more closely, the linear combination model is wrong. The model implies that familiarity bleeds away the most in the short term. Yet this isn’t quite the case. In my experience, the less familiar you are with someone, the faster they fade out of your life.

Since the rate of decay is dependent on the current value, the model is actually a differential equation…

Anyways, more importantly, familiarity only really matters if you like them, or have some sort of affinity (shared interests/goals/relatablility/etc). Thus, “Affinity/Liking” should be a data member as well in order to model friendship.

In fact this would make a lot of sense. How much you like a person should correlate pretty strongly with how much time you spend with them. It would make sense for a positive impression to have a positive “Affinity/Liking” value, so when the combination of yours and their Affinity/Liking is combined, we get a useful value.

This explains why you can quickly become great friends with someone in a matter of a day, while it may take weeks to become the same level of friends with someone else.

Of course, this model is based on my own subjective experiences and may not be the same for everyone. I tried to keep things as abstract as possible to encompass as many relationships as possible.

Believe it or not, you’re probably in my top 10 oldest friends now, @isbee. Give it a year or two and you’ll definitely be in the top 5! Although the concept of friends fading away feels sad, I’ve only really missed one person, ever. Not caring about much lets you do that I suppose.


#23

I appreciate your thoughtful feedback!

Your approach is fascinating. I shall ponder it at greater length when I get the chance.

Losing anyone to which I have given substantial amounts of time and heartfelt consideration leaves me the most devastated. I don’t invest my time lightly.


#24

Hmm now that you mention it, there was a guy recently. His name was Jason. He was pretty nice to me for the most part so we were pretty good friends.

I didn’t recognize the warning signs. He had some serious problems.

Some shit went down and I severed all contact. I couldn’t stop thinking of the incident for awhile, but I don’t regret what I did.


closed #25

This topic was automatically closed 5 days after the last reply. New replies are no longer allowed.