Message from (っ◔◡◔)っ ♥ tคฯ 👌☞🐸 ʳє𝔢Ẹ𝕖𝐄Ẹᵉέє in Turtle Hermit School #random

2018-09-26 21:34:39 UTC  


2018-09-26 21:34:49 UTC  

You are brilliant 🤗

2018-09-26 21:34:58 UTC  


2018-09-26 21:35:03 UTC  


2018-09-26 21:35:13 UTC  

Do something

2018-09-26 21:35:17 UTC  

Show me more

2018-09-26 21:35:41 UTC  

Now I wish I had made more clever taunts

2018-09-26 21:35:53 UTC  

I’m just venting at the poor thing

2018-09-26 21:36:19 UTC  


2018-09-26 21:36:26 UTC  

A prediction

2018-09-26 21:36:49 UTC  

Well 🤷🏼‍♀️ why not?

2018-09-26 21:37:07 UTC  


2018-09-26 21:37:21 UTC  


2018-09-26 21:37:32 UTC  

It paired the u with uh

2018-09-26 21:38:00 UTC  

What is the method to the madness

if it's using markov chains then it's traversing word associations by proximity, the huge message it gave you about salmon was because you had a word somehow associated with a word in the slamon text

thinking face could be "no msg found" instead of building new entry

it could be operating on all text? I dunno. Would need to analyze the source

2018-09-26 21:41:16 UTC  

That’s exactly what I thought!!! 😮

2018-09-26 21:42:29 UTC  

But I thought it was maybe recognizing a word or phrase (best match) and then looking for the comment immediately following that

2018-09-26 21:42:38 UTC  

To approximate a response

2018-09-26 21:42:51 UTC  

But I don’t think that’s it now

2018-09-26 21:43:40 UTC  

Can you look at the cogs and know what program it is following?

2018-09-26 21:44:28 UTC  


I can't duplicate the state however

2018-09-26 21:47:36 UTC  

The state?


2018-09-28 05:10:33 UTC  

This will unsticky the current sticky message from this channel. Are you sure you want to do this? (Y/N)


2018-09-28 05:10:35 UTC