Not sure if anyone appreciates but here are the new dR stats!

For more stats see comments.

Do you want to have your personal stats? Ask me, it's already generated. I only have to post it.

Comments per hour

Comments per weekday

Rants per day

Rants per hour

Most ignored

My own personalized stat

Highest score, the title of graph is wrong (I hope :D)

The personalized stats are getting nice (see of jestdotty and me) since there is more data

Soon we will be able to completely simulate devRant.
And according to simulation argument, this very comment is already a simulated one, statistically speaking.

@Lensflare this is all content for training of my new bot soon. Waiting for new hardware. Someone already created a simple recurrent neural network based on users. You give username, and it will generate a rant like that user. I don't expect it to be very good - my AoK bot was RNN. RNN only is not enough to simulate someone

If i would use the @kiki's export tool as source, I'm able to make a realistic devrant user clone. But need some RAM! :)

Do your statistics exclude all the crypto spam bots?

@retoor don‘t worry about it! You are just a simulation and the real retoor has already managed to do that!

meow

Ah, here are the stats of many people! I made an automerger! In future this will be a one post rant.

Edit: ah fuck, dR fucks up quality. I fucking spent 15 minutes on it! Time lost

High quality statistics image here: https://devrant.molodetz.nl/

@devRancid yes, in upvotes statistics they're counted. They lower the upvote sum with one or two points each. Other option is not to take negative upvoted in sum at all. Don't know which better

@Lensflare damn, that's why I rarely sleep. It's just nog needed! I'm an AI, yEEAaaah

@cafecortado that's freaking beautiful. Wish I was even a bit so creative

@jestdotty see his post!

Cool stats.

Where is "rage downvote" stat?

Fuck, I'm no longer ignored enough -- my name appears in retoor's magical charts. This is so humiliating. Brb, gonna brew some invisibility potions.

@Liebranca I know a vampire so you can get the dust. You will have to find your own nirnroot.

@Demolishun Well, I have this here black soulgem, so perhaps we can strike a deal. I could enchant your pants with the good shit, you know.

@Liebranca well I want you to enchant pants for someone else. I am looking for the Testicular Torsion enchant on these.

I'm pissed that I am not featured in any of these graphs. Mmmphhh

@cafecortado I'll check if AI can cut out the cat for me and i'll use it as a github avatar. Huge fan of the kitty

@SidTheITGuy ah, will fix that immediately

sentiment analysis.
word cloud analysis.
mentioned / week day.

Also - do you exclude anaytics posta from the stats?

@magicMirror FUCK, sentiment analysis! WHAT A GREAT IDEA! I have hardware to do such AI now! I could let it generate a score and find out who's the most toxic or wholesome of devrant! Wordcloud is also good idea. Maybe i'll make a wordcloud aggregate function as sqlite plugin. Completely useless and ten times longer to make? Sure, but I like to do stuff in crazy way just for gaining skills.

When it's generated could be on image but it's in file name as well: https://retoor.molodetz.nl/retoor/...

Old stats won't be deleted.

magicMirror fuck, what a great idea dude. Thank you very much. I have smth fun to do where AI is actually useful for. I know it's possible without AI, but now I can do this: 80% of comments made by this fucker is political or 60% is IT related. I'm happy. I can even export the main subject users commonly talk about. Like "She always talks about apple" or "He only talks about shitting and pretends to be a plane. It doesn't make sense"

@magicMirror: I just realize, i can ask retoor9b to make the wordcloud.. My statistics project can just call the retoor9b API

@retoor i once wrote an entire scientific article for english class in first year of uni.

It was about when the weather's good, people used more smileys than frownies on my IRC channel.

It was the best channel. Lots of tots and drugz

@antigermanist that's an interesting statistic. While i worked for the biggest automative trader in NL, they found out that the choice of color regarding cars was based on the state of economy. What's is a good assumption since not every color costs the same. White is always cheapest but just less courant. Not good for resale so not advised by car sellers. That's reason you don't see them that much but ironically you see it quite much at mercedes's. That from all people, they're the one not wanting to spend on a paint job. Or the prices are equal for them or smth

@antigermanist reminds me of the "WTFs per minute" metric for code quality.