The Saltiest Users on HN • Hacker News Board

Re: The Saltiest Users on HN

#51

Post by reificator » Thu, Apr 18, 2019, 5:04 PM UTC

weberc2 wrote: post #19

msluyter wrote: post #11

I love this! I'm sort of proud to note that my salt score is -0.08, with my saltiest comment being: The iPad and iPhone are especially dangerous when it comes to accidental downvoting. Separating or enlarging the arrows would help those of us with fat fingers. However, (and this will apparently add to my salt score ;)), I'm curious how a comments like these get rated as expressing a negative sentiment?: Out of curios…

I like the idea, but I'm very skeptical of the implementation. My saltiest comments per the tool are not remotely salty, and I'm sure I have many saltier comments. Turns out NLP is hard. FWIW, my salt score is -0.09 ;)

> I like the idea, but I'm very skeptical of the implementation. My saltiest comments per the tool are not remotely salty, and I'm sure I have many saltier comments. Turns out NLP is hard.

> FWIW, my salt score is -0.09 ;)

Same salt score, same opinion on the results. My top salty comment has a score of -1.00

> > > Unfortunately it's not a bottle. It's just a plastic cup with a lid and a straw. A really big plastic cup that tapers at the bottom so you can fit it in your vehicle's cupholder.

> > In USA even the cups have muffin-tops!?

> Yep. Horrific, isn't it?

It's a joke, it's not terribly on topic, and it's a bit glib. But salty? Nah.

Fourth most "salty" comment, at a score of -0.50:

> I like having a library that I can flip to when I'm bored and want to do something productive.

Seventh most salty, at -0.45, is literally just an explanation of the second argument in Javascript's parseInt function. Ninth is just an earnest recap of a conversation about earworms.

Meanwhile the saltiest comment I saw in a quick skim is in 12th place, with a score of -0.30. I was discussing House of Cards and the then recent scandal around Kevin Spacey, and someone was putting words in my mouth and claiming I said things that I didn't actually say or believe. I was quite salty in my response(s), but here the tool marked it as 70% less salty than joking about muffin topped cups.

Re: The Saltiest Users on HN

#52

Post by imgabe » Thu, Apr 18, 2019, 5:07 PM UTC

I guess I'm not nearly as salty as I thought. Or at least, other people on here are saltier.

Re: The Saltiest Users on HN

#53

Post by DanBC » Thu, Apr 18, 2019, 5:07 PM UTC

This is cool, but my comment history contains some comments far saltier than the one that makes the list.

Re: The Saltiest Users on HN

#54

Post by CivBase » Thu, Apr 18, 2019, 5:15 PM UTC

(testing the tool with this comment, do not take this seriously - I think it's a neat experiment)

This is the worst thing I've ever seen.

I am appalled at how terrible the UI design is. My eyes are literally bleeding because it does not have enough contrast. I can't even see my "sweet" comments so I can inflate my massive ego.

I take personal offense that the site evaluated 9 of my comments as being salty. Maybe the site is just salty!

The developer should take personal responsibility for this tool by manually counting the distribution of colors in 400 bags of skittles!

Also, ordbajsare is right. Javascript is disgusting!

Here's some more words I think the algorithm dislikes: horrible, screwing, dreadful, idiot, stupid, retard

Re: The Saltiest Users on HN

#55

Post by JudgeWapner » Thu, Apr 18, 2019, 5:17 PM UTC

jeho wrote: post #29

this is hilarious

a computer converted a human's action into a number and graphed it using sophisticated numerical algorithms and analysis methods. ergo it must be true.

Re: The Saltiest Users on HN

#56

Post by logfromblammo » Thu, Apr 18, 2019, 5:17 PM UTC

weberc2 wrote: post #24

I would be interested to know what words/phrases are the best predictors of saltiness.

0 1 2 3 4 5 6 7 8 9

awful difficult terrible horrible boring

bad worse worst

stupid moron retard idiot dumb

crazy insane shocking outrageous

worthless useless

effing wtf

disgust discuss (?!)

Re: The Saltiest Users on HN

#57

Post by mindcrime » Thu, Apr 18, 2019, 5:18 PM UTC

aetherson wrote: post #27

It is my feeling that sentiment analysis has a ways to go. Here are a few of the comments I've made that the system has described as among my saltiest: Oh, sorry, "nm" means "nanometer" to me, but of course nautical miles. (Score: -0.5. Comment is entirely taking responsibility for misinterpreting someone) Well, if he was trying 1M combinations every 40 seconds, for $7 per hour, and he didn't need to use hundreds of…

My "saltiest" comment was:

Oh man, that's terrible news. R.I.P. Tim May.

So yeah, the sentiment analysis could use some work.

Re: The Saltiest Users on HN

#58

Post by krapp » Thu, Apr 18, 2019, 5:18 PM UTC

Judging by my score... this can't see flagged comments at all.

Re: The Saltiest Users on HN

#59

Post by thisisit » Thu, Apr 18, 2019, 5:21 PM UTC

My guess is pg_is_a_butt gets enough salt from his username.

But this got me curious. Who are the sweetest users?

And what kind of topics generate most salt? This will be interesting because there has been a general trend on HN where people bemoan "HN hates X" with X being cryptocurrency, Tesla, new starts on Show HN.

Re: The Saltiest Users on HN

#60

Post by Kassius509 » Thu, Apr 18, 2019, 5:36 PM UTC

aetherson wrote: post #27

It is my feeling that sentiment analysis has a ways to go. Here are a few of the comments I've made that the system has described as among my saltiest: Oh, sorry, "nm" means "nanometer" to me, but of course nautical miles. (Score: -0.5. Comment is entirely taking responsibility for misinterpreting someone) Well, if he was trying 1M combinations every 40 seconds, for $7 per hour, and he didn't need to use hundreds of…

Thanks for trying my app! And thank you for taking the time to respond!

It does have a LONG way to go. These are valid criticisms. And all areas we are working to improve on with our next ML model. (Sad vs Negative, rating numbers as "salty",

This model is pretty simple. It's using TextBlob and looking for a combination of negative sentiment (not necessarily condescending) and subjectivity. Essentially hand built heuristics derived by weighting each word in the sentence. Not a great way to make predictions.

The model is FAR from great. But great from afar. For high level (overall user saltiness) it performs better.

The unlabeled dataset of this size presents some unique challenges but in our testing of our new model (based on SOTA BERT fine-tuining & a large labeled training dataset) the results look promising. I'm really looking forward to getting it deployed.

I am encouraged by the words of @pg who said "you can and should give users an insanely great experience with an early, incomplete, buggy product, if you make up the difference with attentiveness.

Can, perhaps, but should? Yes. Over-engaging with early users is not just a permissible technique for getting growth rolling. For most successful startups it's a necessary part of the feedback loop that makes the product good. Making a better mousetrap is not an atomic operation. Even if you start the way most successful startups have, by building something you yourself need, the first thing you build is never quite right. And except in domains with big penalties for making mistakes, it's often better not to aim for perfection initially."