zlacker

[parent] [thread] 42 comments
1. keifer+(OP)[view] [source] 2026-02-03 08:19:11
Very cool. I made the top 1,000 too.

It would be interesting to see karma-per-word, as well, as a kind of succinctness density factor. Although karma points are not equivalent to quality, and you’d need to also factor in average comment length and some other things.

To use myself:

31,273 karma / 351,012 words ≈ 0.0891 karma per word

replies(10): >>atoav+p >>Kolmog+b3 >>latexr+p3 >>kqr+i4 >>genewi+t4 >>pjc50+Kb >>thomas+rj >>JKCalh+nB >>usrusr+lW >>rldjbp+Rs6
2. atoav+p[view] [source] 2026-02-03 08:23:10
>>keifer+(OP)
Or 89 milikarma/word. That is pretty good. I only got 34 milikarma/word.
replies(1): >>MrGilb+82
◧◩
3. MrGilb+82[view] [source] [discussion] 2026-02-03 08:37:03
>>atoav+p
83 milikarma/word. That's interesting, though I'm not sure what to do with it.
4. Kolmog+b3[view] [source] 2026-02-03 08:44:31
>>keifer+(OP)
It would not make much sense to compute it, if you do not subtract the karma earned through submissions.
5. latexr+p3[view] [source] 2026-02-03 08:46:01
>>keifer+(OP)
You also get karma for submissions, so that metric will be highly skewed.

The submission karma is public, so you should be able to subtract it, but that karma doesn’t seem to be the same as the one for comments (i.e. I think one point in a comment gives you one point overall, but on submissions you need two or three points to earn one in your account).

replies(1): >>keifer+C3
◧◩
6. keifer+C3[view] [source] [discussion] 2026-02-03 08:48:08
>>latexr+p3
Yeah, I started to think out what you'd need to actually get a "succinct but high-quality" score and it gets complex, fast. Karma will be bloated by popular hot takes and submissions, for starters. Then you have to determine the certain cut-offs to ensure that someone with 10 comments of 10 words each (with 100+ karma each) isn't "the most succinct."

I'm less interested in the idea as a ranking, and more as a way to evaluate my own writing, with the aim of being as succinct but high quality as possible.

replies(2): >>embedd+af >>morkal+6P
7. kqr+i4[view] [source] 2026-02-03 08:53:34
>>keifer+(OP)
Interesting. I'm at 18,176/392,187 ≈ 46 millikarma per word. I did not expect any of those numbers to be so high.

But at almost 90, I have to ask: do you have a blog I should follow?

replies(3): >>keifer+Q5 >>oefrha+iA >>JKCalh+HE
8. genewi+t4[view] [source] 2026-02-03 08:54:48
>>keifer+(OP)
1500s with 221k, so there's a real long tail
◧◩
9. keifer+Q5[view] [source] [discussion] 2026-02-03 09:04:45
>>kqr+i4
Thanks, no I have written various substacks and little blogs over the years, but I really ought to just make a personal blog and put them all there. +1 for the reminder.
replies(1): >>Imusta+S6
◧◩◪
10. Imusta+S6[view] [source] [discussion] 2026-02-03 09:13:38
>>keifer+Q5
Author here. That's actually why I had written this for myself (although compared to you, my karma's comparatively low)

I felt as if I had written quite a lot on HN and I was always referencing my past comments. I just usually write what I am thinking so I do have quite a filler words in my HN comments, but still, I wanted to see how many words have I written in the first place & I wrote this thought in another comment & thought that its pretty interesting, let's do it. Maybe others would be curious too :)

But essentially, some part of me wanted to write blog too to quickly reference it and be part of the indie-web.

>>46828331

This is something that I wrote in that comment

"I guess I can write it but I already write like this in HN. The procastination of writing specifically in a blog is something which hits me.

Is it just me or is it someone else too? Because on HN I can literally write like novels (or I may have genuinely written enough characters of a novel here, I might have to test it or something lol, got a cool idea right now to measure how many novels a person has written from just their username, time to code it)"

So you aren't alone in having the procastination around making a personal blog. This is literally why I had made this idea so if this project (or discussion) helps you in making a personal blog & helps atleast a single person (meaningfully). I would consider my project to be success :)

Have a nice day and good luck for your personal blog!

replies(1): >>keifer+m8
◧◩◪◨
11. keifer+m8[view] [source] [discussion] 2026-02-03 09:23:31
>>Imusta+S6
Aha, yeah I have dozens of text notes on my computer that are just links to a HN comment with "expand on this as a full post."

I definitely feel like I've written at least 50-100 decent blog posts as HN comments over the years.

And thanks!

12. pjc50+Kb[view] [source] 2026-02-03 09:46:59
>>keifer+(OP)
#33 here. I have written .. a lot of words. I don't know whether they're correctly excluding ">" quoted words though.

A quirky feature of HN is that you can only see detailed karma counts for your own posts. One of these days I plan to scrape all of mine so I can sort by karma and do some meta-commentary.

replies(3): >>embedd+Pe >>tasuki+gs >>ryandr+O91
◧◩
13. embedd+Pe[view] [source] [discussion] 2026-02-03 10:12:10
>>pjc50+Kb
> One of these days I plan to scrape all of mine so I can sort by karma and do some meta-commentary

Seems simple enough, while searching I came across this snippet you can paste in the console, and gives you a sorted list of most upvoted/downvoted comments: >>36107028

Sadly requires hitting news.ycombinator.com rather than the API, but only way to get the actual points as you mention.

◧◩◪
14. embedd+af[view] [source] [discussion] 2026-02-03 10:14:32
>>keifer+C3
> with the aim of being as succinct but high quality as possible

You're assuming "high karma = high quality" which, isn't always correct :) I've had wildly incorrect claims be upvoted a lot, and correct ones downvoted, seems to be more about what the subject is about and what "side you're coming from" rather than anything else sometimes. Other times it goes exactly as expected.

End effect is, I wouldn't rely on karma as a signal for quality, just "agreement at large" or something.

replies(2): >>keifer+Uf >>pc86+Fs1
◧◩◪◨
15. keifer+Uf[view] [source] [discussion] 2026-02-03 10:19:37
>>embedd+af
No, I mentioned in the grandparent comment:

> Although karma points are not equivalent to quality,

But I don't think they are totally uncorrelated to quality, either. So you'd need a way to factor karma points in without over-valuing them.

To really get specific, the only thing we're really measuring here is something like, well-written, succinct comments that are appreciated by HN users that are able to upvote. Which is not exactly super useful or insightful, but is a fun exercise.

16. thomas+rj[view] [source] 2026-02-03 10:52:29
>>keifer+(OP)
+1 and also add a feature for unique words to show how you vocab ranks
replies(3): >>s_dev+dn >>JKCalh+wB >>Yizahi+5J
◧◩
17. s_dev+dn[view] [source] [discussion] 2026-02-03 11:23:51
>>thomas+rj
> unique words to show how you vocab ranks

I think writing well with plain language would be a better indicator of worthwhile contributions than estoeric jargon that only serves to confuse or intimate. That would be a lot more difficult to measure though, the number of fancy words per post probably is a lot easier to vibe code.

replies(2): >>BLKNSL+Dp >>Imusta+M51
◧◩◪
18. BLKNSL+Dp[view] [source] [discussion] 2026-02-03 11:39:59
>>s_dev+dn
intimidate

;)

replies(1): >>s_dev+dJ
◧◩
19. tasuki+gs[view] [source] [discussion] 2026-02-03 11:55:32
>>pjc50+Kb
Karma depends entirely on when you comment: My most upvoted comments are the early ones. I check HN perhaps once a day, so my comments don't always get a lot of visibility. Perhaps it's better that way.
replies(1): >>embedd+yu
◧◩◪
20. embedd+yu[view] [source] [discussion] 2026-02-03 12:12:38
>>tasuki+gs
I mean, no, karma depends entirely on the contents of your comment :P I've have comments go from 0 to -4 to 50 to 10 over the course of one day, probably I'd say the flunctionations depends more on how controversial ("hot take") your comment is, but if it's generally good, it tends to be upvoted while more emotional/nonsense appeals tend to be downvoted. Pretty much as expected :)
replies(2): >>JKCalh+2C >>pc86+Ws1
◧◩
21. oefrha+iA[view] [source] [discussion] 2026-02-03 12:48:47
>>kqr+i4
I’m at 19186/233259, about 0.082, and I’m pretty sure most of my higher upvoted comments are on the shorter side, and my wall of texts tend to not deviate much from 1 karma, sometimes even negative. Don’t put too much stock into fake internet points.

And I really need to waste less time here, didn’t expect to be top 1500…

22. JKCalh+nB[view] [source] 2026-02-03 12:57:01
>>keifer+(OP)
Showoff. (I'm at 0.06).
◧◩
23. JKCalh+wB[view] [source] [discussion] 2026-02-03 12:58:00
>>thomas+rj
Please don't start subtracting for em-dashes though. ;-)
◧◩◪◨
24. JKCalh+2C[view] [source] [discussion] 2026-02-03 13:02:15
>>embedd+yu
Being late to a post though does you no favors. People have moved on. So it's both.

Hot takes? Aside: an example of a "third rail" post (where I seem to get the most downvotes) appears to be when I disparage UBI. I used to get hated for disparaging self-driving cars too but people beat me up less about that these days.

◧◩
25. JKCalh+HE[view] [source] [discussion] 2026-02-03 13:20:44
>>kqr+i4
Reminding me of the classic take on brevity [1]:

A prominent example of a laconism involving Philip II of Macedon was reported by the historian Plutarch. After invading southern Greece and receiving the submission of other key city-states, Philip turned his attention to Sparta and asked menacingly whether he should come as friend or foe. The reply was "Neither."

Losing patience, he sent the message:

If I invade Laconia, I shall turn you out.[4] The Spartan ephors again replied with a single word:

If.

[1] https://en.wikipedia.org/wiki/Laconic_phrase?utm_source=chat...

◧◩
26. Yizahi+5J[view] [source] [discussion] 2026-02-03 13:46:21
>>thomas+rj
It will be skewed due to all the jargon, slang and grammar errors.
replies(1): >>kqr+nK
◧◩◪◨
27. s_dev+dJ[view] [source] [discussion] 2026-02-03 13:46:50
>>BLKNSL+Dp
Yes, that is a typo!
replies(1): >>pc86+qt1
◧◩◪
28. kqr+nK[view] [source] [discussion] 2026-02-03 13:53:27
>>Yizahi+5J
But if it is equally skewed for everyone, does it really matter?
replies(1): >>pc86+3u1
◧◩◪
29. morkal+6P[view] [source] [discussion] 2026-02-03 14:18:19
>>keifer+C3
I always wondered about metrics to measure things like: does the commenter generate positive discussion, flame wars, or plain old dead threads?
30. usrusr+lW[view] [source] 2026-02-03 14:53:48
>>keifer+(OP)
Karma per word would be a terrible metric though: the short, slightly divisive clever quip tends to still net a few points positive as long as it's not all too negative, despite clearly not being great hn content. Great content isn't short, but the vote button is or of sight once you're done reading. Good long texts will certainly still get some upvotes, but rarely enough to outcompete small & clever that just goes with the flow.
replies(1): >>troad+vJ2
◧◩◪
31. Imusta+M51[view] [source] [discussion] 2026-02-03 15:37:57
>>s_dev+dn
Yes I agree with this 100%

One of the things in future I wish to personally learn is how to write concisely. My posts are large and scattered.

To me, the beauty was in the depth/content in Hackernews. I still remember the day when HN clicked to me when I was in metro. A comment clicked with me and really changed my perspective on something. It was fairly long from what I can tell (I am sorry but I am a little hazy other than I was going/returning to school and I was using hackernews)

HN comments are great the way they are. Let's keep it that way.

> That would be a lot more difficult to measure though, the number of fancy words per post probably is a lot easier to vibe code.

Agreed, I use it for prototyping but I am still learning. I hope to not vibe code as I progress and go to college for example. Currently I was constrained because I was (sad?) from my last exam not going so well & the next one being in 8 days ish.

Wish me luck :)

The only reason I vibe code is either for prototyping (for time constraints) and I just wanted to share it to the world.

I have actually written a lot about it. I hope you can read it if you have time [all comments are and will always be written by me] :)

Have a nice day! I am just happy that it can be on front page :]

◧◩
32. ryandr+O91[view] [source] [discussion] 2026-02-03 15:52:41
>>pjc50+Kb
I tried to do exactly this, but you need to hit the actual HTML news.ycombinator.com site, since karma is not visible through the API or in any of the public data dumps. And HN appears to have rather robust anti-scraping provisions. Even after trying a few different back-off strategies, my scraper was rate limited quickly, so I gave up.
replies(1): >>Imusta+gC1
◧◩◪◨
33. pc86+Fs1[view] [source] [discussion] 2026-02-03 17:07:17
>>embedd+af
I don't think it's public (it might not even be saved) but something akin to Reddit's "controversial" would be interesting as well. I've seen my comments go from double-digit positive to double-digit negative, and vice versa, in contentious threads on divisive topics. It'd be interesting to see who has the most votes regardless of +/- but the lowest total karma. Who amongst us is just pissing off everyone?
replies(1): >>Imusta+2B1
◧◩◪◨
34. pc86+Ws1[view] [source] [discussion] 2026-02-03 17:09:05
>>embedd+yu
Your comment is never going to get 75 votes on it if it only gets in front of 25 eyes, regardless of how good, bad, offensive, or insightful it is.

Especially on political threads you'll see the most milquetoast takes imaginable basically locked at the top of the page because they were commented 2 minutes after the post was made, and all the fighting happens underneath.

◧◩◪◨⬒
35. pc86+qt1[view] [source] [discussion] 2026-02-03 17:11:19
>>s_dev+dJ
You get double points for that one
◧◩◪◨
36. pc86+3u1[view] [source] [discussion] 2026-02-03 17:14:10
>>kqr+nK
It probably doesn't have a large enough effect to matter, but I would expect that it would negatively impact the people you're trying to positively impact by using this metric. If you're careful with your words, a better typist, refrain from slang, reread your posts multiple times, edit out typos or inconsistencies or rambling thoughts, this type of vocabulary ranking would "hurt" you. But if you do that you're also probably the type of person to write longer more well thought out comments. So it's probably a wash to slightly "achieving the opposite of what you want but not really enough that you'd notice it" if I had to guess.
◧◩◪◨⬒
37. Imusta+2B1[view] [source] [discussion] 2026-02-03 17:41:39
>>pc86+Fs1
https://news.ycombinator.com/user?id=proven

Here's the Query that I ran on play.clickhouse.com (Please note that both the queries that I am about to give are AI generated)

SELECT id, argMax(karma, update_time) AS current_karma FROM hackernews_changes_profiles GROUP BY id ORDER BY current_karma ASC LIMIT 5

№ id current_karma1 1 Proven -247

2 VentureHawk -89

3 adamyormark -50

4 oldpersonintx2 -45

5 sevenstar -35

So proven has the lowest karma.

While at it, Here's the query for the most upvoted (karma) people on hackernews

SELECT id, argMax(karma, update_time) AS current_karma FROM hackernews_changes_profiles GROUP BY id ORDER BY current_karma DESC LIMIT 5

id current_karma 1 tptacek 414947

2 jacquesm 235042

3 ingve 214146

4 todsacerdoti 204814

5 rbanffy 184030

I remember tptacek because he is the second person of all the people (behind dragonwriter) and he has written 14.37 Volumes worth of game of thrones.

Hope this helps. I just ran the query because I was curious to find the results :D

I can add this query to the website, but one of my worries (even with this post) is that people might try to now benchmark it which wasn't my intention & it will fail to be a good measure (someone mentioned the goodhart's law which is correct/apt here)

◧◩◪
38. Imusta+gC1[view] [source] [discussion] 2026-02-03 17:46:41
>>ryandr+O91
What about having a browser and having a tampermonkey script which can do that? I have found that works really great. Ooh this is giving me some ideas.

A lot of people have given me some really great ideas. Commenting to have them in backlog whenever I get bored and think to add addition. I would just be interested in having a dump/export of my comments personally as well (someone else wanted to/analyze all of their comments and have them in a floppy disk, maybe this can help their idea too!)

replies(1): >>ryandr+D12
◧◩◪◨
39. ryandr+D12[view] [source] [discussion] 2026-02-03 19:22:45
>>Imusta+gC1
It would be better if HN simply made karma scores available through the API. Such a strange omission--they obviously have the data.
◧◩
40. troad+vJ2[view] [source] [discussion] 2026-02-03 22:59:00
>>usrusr+lW
The irony is that karma posts are so easy. Take something most of your audience already agrees with, triple down on some reductionist caricature of it, and smother it in pithy glibness. The shorter the better. Particularly effective if you set up a false dichotomy vis-a-vis the person you're replying to. It's a reflexive style of engagement for many, and HN is not immune to it.

I aim to avoid it these days, with varying degrees of success. I don't need fictitious internet points, I want to hear other people's genuine thoughts on a subject of interest. Or sometimes just to share something I thought was neat.

But since all social media are Pavlovian conditioning for points, you rarely get any fruitful exchange. And it seems to be getting rarer and rarer, sadly.

I wonder how one would structure social media to avoid it. HN is good, but the karma system is a double edged sword. Would it increase the quality of the discussion to retain the use of points for ranking posts, but hide point counts completely? Perhaps they could be represented by words: "Positive response", "negative response", but only past -3 and +3, with no changes in wording beyond that score?

replies(1): >>bagacr+PI4
◧◩◪
41. bagacr+PI4[view] [source] [discussion] 2026-02-04 14:33:31
>>troad+vJ2
Wrt my own posts I like the karma system as feedback for how well I'm getting my point across. Helps to understand what communication style resonates with people. I'd say the biggest flaw is not that it rewards snarky popular opinions, but that it overly rewards first movers on a topic.

I do think that pithy is good. The real world also rewards people who can convey an idea succinctly. ("Healthcare for all" for example is an effective rallying cry despite lack of implementation details.)

replies(1): >>troad+XA6
42. rldjbp+Rs6[view] [source] 2026-02-04 22:45:17
>>keifer+(OP)
0.000622 here :)

somehow almost close to the top 10k despite writing one-hundredth as much as the top guys.

◧◩◪◨
43. troad+XA6[view] [source] [discussion] 2026-02-04 23:32:28
>>bagacr+PI4
If it were an effective rallying cry, it would have worked at any point in the last forty years.

Politics is not assessed in terms of how the slogans sound, but what they achieve. Universal healthcare is further away today than it was in the '90s, and Democrats are less 'rallied' than ever.

[go to top]