That's pretty cute, and from looking at some of the badges I can see that a decent bit of work was put into coming up with some unique and meaningful badges. Nice work!
From what I remember from University anyone can have an internet. It just won’t be THE Internet. So feel free to set one up. If you are at the border of two countries chuck a network cable over that border.
But remember, it is your internet, so you don't have to peer equally with everyone. You can allocate bandwidth to those you prefer while limiting to those you don't. Just like THE internet.
Great fun! The gamification and cutesy badges are basically the opposite of the HN aesthetic, which gives the whole project a fun sort of tongue-in-cheek feel.
Because I don't have anything more constructive to say, I'll just point out what looks like a small error: the text for the "Early Adopter" badge always seems to say "NaN" in place of the user's name.
Thanks for the feedback - the karma thing is on purpose - you don’t lose the 4 badge because you got the 5, ironically as if gaining karma is a game of some kind!
Very neat app! One other bug report: I have a "programmer" badge for mentioning a programming language in my about page, but such isn't the case. Not sure which part of my bio it might be mistaking for a language?
edit: found it, it's reacting to the fact that my keybase ID happened to have a "VB" in the middle. You might want to change some of those language checks from |vb| to |\bvb\b| or the like.
I'm just glad HN doesn't (yet?) have strict (puritanical, think of the children/advertisers, etc) language rules. I mean we're all adults here, we can use adult language to each other.
Likewise, I don't get the recent trend of self-censoring words like suicide, police or kill. I think it's mainly a thing on tiktok? But it's leaking out from there as well.
I think I've seen it on sexual education-like videos on Instagram, too. People censored the names of genitals, probably so that automatic filters won't ban them or whatnot.
It's not that, it's more that there are words in general that TikTok does not tolerate, and there's no definitive list of those words. Some of them: gun, pissed, penis, vagina, any swear word, etc.
So since there's no definitive list and you can't make one because the content is not outright deleted (algorithm just ignores it), you end up in a situation where TikTok users self-censor just in case.
This is similar to what China does internally, but the list is definitely different.
Indeed, and it's not as if we use those words every day in informal context. Yet when it's slightly more public it's suddenly a big deal? It's a bit hypocritical to pretend to be offended by something that everyone really does.
Even the "think of the children". Are there any children that don't know these words?
I’m a bit confused by the recent adoption in saying “unalive” instead of “suicide”, I’m genuinely not sure what it accomplishes. They sound similar and mean the same thing.
At least on TikTok "unalive" is used as a replacement for "kill" in a way to circumvent the automatic language detection from banning/quarantining your content.
A rock is most definitely 'unalive', but it didn't commit suicide, nor was it murdered.
As technologists we're tuned the idea that if you throw enough code at a problem it can be solved, but currently it's very clear that the content filtering 'AI's that are out there on the big platforms are not very good at their jobs.
Especially TikTok has very aggressive content filtering, so if you want people to see your content about such topics, you need to use terms the filter doesn't yet penalize.
This is one of those topics I tend to agree with in principle.
And then every time I run into someone who feels very strongly about it ... the rate of me thinking about how it should be a rule based on their usage alone.
One of those weird things where i agree and then you meet the folks who have very strong feelings about it and I start to have second thoughts ;)
"Likewise, I don't get the recent trend of self-censoring words like suicide, police or kill. I think it's mainly a thing on tiktok? But it's leaking out from there as well."
I presume this is to that their postings won't be censored/shadow banned/deleted/muted/whatever. Kind of reminds you of China, doesn't it?
What is classified as the "F Word!"?
I have alledgedly used it twice giving me the badge, but glancing through my comments I cannot see any mention of what I perceive as the american "F word"
It wrongly thinks that I have a Bitcoin address in my about page, and thus gives me a "Bitcoiner" badge, which… I really do not want, and almost take offense at.
More seriously, I think the app is confused by the Tor Onion v3 address in my profile.
Kudos for running your site on Tor. I’ve been wanting to do that as well, but I’ve been putting it off because I don’t have any good enough content at the moment to warrant maintaining a site for. So currently I just post stuff to TikTok and Reddit etc. But when I make something good enough and I put it on a site of my own, I will make my site run over Tor also in addition to clear net.
The user RiderOfGiraffes is claimed to have used the F word. Searching through their comments I can find no such usage, but the replies to their comments are consistent with the count given. I think the count is incorrectly including occurrences in replies by other people to comments.
Added in edit:
Following the suggestion from elsewhere I've tracked down the problem. It's counting the "Fuk" in Fukushima, the "funk" in "jazz funk", and the "Fack" in "Fack to Brunt".
So that's wrong, being mislead by the results from the search engine's false matches.
Same with me, I doubt I've used the f word in comments.
But holy cow! Close to 15k submissions now.
The karma/comment ratio is obviously with total karma (including submission karma). I think if you only counted comment karma (no idea if that's possible) I wouldn't have either of the comment badges.
I also had a badge for the F-word about 30 mins ago), which I am pretty sure was wrong. However, having checked again just now it has disappeared.
I'm not sure what to make of badges being bestowed on HN. It's rather like walking into a civil service interview and finding the panel wearing Hawaiian shirts and drinking colourful cocktails. A generous gesture, regardless, and it brought a smile to my face.
Edit to add: just noticed that while I was commenting, quickthrower2 posted to say that they'd fixed the problem. Nice one!
Oh apparently you have to use fuck in two comments. And it has to be over a certain percentage of your total comments. Also you can get a badge by mentioning Rick Astley. Hilarious.
> Oh apparently you have to use fuck in two comments. And it has to be over a certain percentage of your total comments. Also you can get a badge by mentioning Rick Astley. Hilarious.
Xkcd uses alt text or titles, can't remember which, to pop up an accessibility tooltip. Clever and fun but it's quite dated and doesn't work on most (all?) mobile browsers.
Well, now I learned that there is apparently a way to change the header color and I have enough karma to do so. And I got a Gordan Ramsey badge for swearing ;)
Bug! Badges have descriptions in the title attributes on the <li> elements, but <li>s contain <img>s which also have an accessibility title and that's what gets displayed (on my Firefox 91.0.2), so I have to look at html to read the badge descriptions. I love this, btw!
It seems the names are case sensitive. While HN itself is case insensitive. Also: Is there any documentation about what different badges mean?
Edit: As it turns out you need to hover over the text to get an description, while the icon only shows "icon for x" on mouseover.
Addendum: Would it be possible to split karma into submit and comment karma? My gut says it'd not be possible as HN uses some kinda curve + only displays the combined karma on the profile
Comment karma is not visible (it used to be and it was terrible...). But submitted story karma is public so it's possible to sum it up and negate it from a user's public total karma.
(I'm also distraught about some of the numbers. I know I have a HN problem, but seeing it laid out like that...)
I second the "hover text" issue other commenters brought up - it would be great if badge details were accessible in other ways (e.g. expanding hidden text or a popup when the user clicks a badge or its main text).
EDIT:
s/Call Me Anytime/Call Me Maybe/ for extra meme value?
So got side-tracked - but pg has 156K karma, but is not on the leaderboard (/leaders). I assume that dang etc are not there either but I had not really thought about it.
If this gets popular I might have to try and find my first HN login - I want that OG badge. :-)
Which kind of is the point - I dont really want to worry about those sort of things - I am glad HN is staying off this course. But it is fun :-)
Bug report: On my badge profile [0], it looks like the Keybase proof link is correctly detected, however, I think it's also being interpreted as a bitcoin address (I do not have a BTC address in the profile despite having the badge).
162 comments
[ 2.8 ms ] story [ 239 ms ] threadHappy to add any suggestions. I don’t want to make too many more api calls to algolia though!
Because I don't have anything more constructive to say, I'll just point out what looks like a small error: the text for the "Early Adopter" badge always seems to say "NaN" in place of the user's name.
https://news.ycombinator.com/user?id=NaN
Edit:
https://news.ycombinator.com/user?id=undefined
Inquisitive - Has submitted an ask HN
Show Off - Has submitted an show HN
Out of the park! - Has submitted anything to HN with at least 100 upvotes
Won the internet! - Has submitted anything to HN with at least 500 upvotes
Marker Researcher - Has submitted a poll HN
Self Describer - Has written something in the about section of their profile
Has a life - Hasn't interacted with HN for over 180 days
Memior - About section has over 1000 characters
F Word! - Has used the F word in at least 2 comments and more than 0.1% of comments
Gordan Ramsey - Has used the F word in at least 10 comments and more than 1% of comments
Rick Roller - Has mentioned Rick Astley in a comment
Motor Mouth - Has over 10000 comments
Mute - Less than 10 comments
Flag Powers - Has enough karma to flag submissions
Colorful - Has enough karma to set their header color
Downvote Powers - Has enough karma to downvote other users
4 Figure Karma - Has reached or exceeded 1000 karma points
5 Figure Karma - Has reached or exceeded 10000 karma points
Earthly Nirvana - Has reached or exceeded 100000 karma points
Pretty Good Commenter - Has a karma/comment ratio of 5 or higher
Substantive Commenter - Has a karma/comment ratio of 10 or higher
OG - An original user of HN: someone who joined before 2008
Early Adopter - An early adopter of HN: someone who joined before 2010
Necro Linker - About page links to defunct service, e.g. hnchat.com
Call Me Anytime - Profile has a US telephone number
Bird of a feather - Has a Twitter link/reference in about page
Corporate Drone - Has a LinkedIn URL in about page
Tim Berners Lee - Has a link in about page
Git - Has a Github link in their about page
Programmer - Mentions a programming language in their about page
Top Secret - Has a Keybase link in their about page
Open Book - Has an About.me link in their about page
Beggar - Has a Patreon/Hypage/Buymeacoffe/Indiegogo/Liberapay/Kickstarter link in their about page
Bitcoiner - Has a Bitcoin Address in their about page
Wow, I guess I cannot ignore my addiction anymore. I haven’t taken a 6 months break in 8 years of HN commenting?!
Wait, what?
Achievement unlocked?
Typo or Wekman?
Karma length check seems to be wrong. 5 figure karma und 4 figure karma can be had at the same time
s/marker researcher/market researcher/
At least I think that those are valid corrections...
Fun project
edit: found it, it's reacting to the fact that my keybase ID happened to have a "VB" in the middle. You might want to change some of those language checks from |vb| to |\bvb\b| or the like.
That makes sense. If you have more than 5 figure of karma you also have more than 4 figure of karma.
Really inspired me to think about integrating something like this in one of our systems.
Likewise, I don't get the recent trend of self-censoring words like suicide, police or kill. I think it's mainly a thing on tiktok? But it's leaking out from there as well.
It's a difficult problem, I think.
edit - I guess Tiananmen Square isn't available in spell check
So since there's no definitive list and you can't make one because the content is not outright deleted (algorithm just ignores it), you end up in a situation where TikTok users self-censor just in case.
This is similar to what China does internally, but the list is definitely different.
Even the "think of the children". Are there any children that don't know these words?
Like monad or functor?
Edit: I'd be more worried about using a word like "monad" incorrectly on HN than swearing.
You know, the kinds of topics they flag as "adult situations" in TV.
A rock is most definitely 'unalive', but it didn't commit suicide, nor was it murdered.
As technologists we're tuned the idea that if you throw enough code at a problem it can be solved, but currently it's very clear that the content filtering 'AI's that are out there on the big platforms are not very good at their jobs.
And then every time I run into someone who feels very strongly about it ... the rate of me thinking about how it should be a rule based on their usage alone.
One of those weird things where i agree and then you meet the folks who have very strong feelings about it and I start to have second thoughts ;)
Granted I don't yet feel that way.
I presume this is to that their postings won't be censored/shadow banned/deleted/muted/whatever. Kind of reminds you of China, doesn't it?
processing time 2
More seriously, I think the app is confused by the Tor Onion v3 address in my profile.
The user RiderOfGiraffes is claimed to have used the F word. Searching through their comments I can find no such usage, but the replies to their comments are consistent with the count given. I think the count is incorrectly including occurrences in replies by other people to comments.
Added in edit:
Following the suggestion from elsewhere I've tracked down the problem. It's counting the "Fuk" in Fukushima, the "funk" in "jazz funk", and the "Fack" in "Fack to Brunt".
So that's wrong, being mislead by the results from the search engine's false matches.
But holy cow! Close to 15k submissions now.
The karma/comment ratio is obviously with total karma (including submission karma). I think if you only counted comment karma (no idea if that's possible) I wouldn't have either of the comment badges.
I'm not sure what to make of badges being bestowed on HN. It's rather like walking into a civil service interview and finding the panel wearing Hawaiian shirts and drinking colourful cocktails. A generous gesture, regardless, and it brought a smile to my face.
Edit to add: just noticed that while I was commenting, quickthrower2 posted to say that they'd fixed the problem. Nice one!
Congratulations on reaching your quota :)
Edit: Oh, apparently that's not what "Colorful" means at all.
Perhaps, but it would be a bit immature to do so. Or even to quote someone else doing so.
> Perhaps, but it would be a bit immature to do so. Or even to quote someone else doing so.
Yeah I shudder to think about how gamification in this way will change the discourse here.
>> Perhaps, but it would be a bit immature to do so. Or even to quote someone else doing so.
>Yeah I shudder to think about how gamification in this way will change the discourse here.
So do I. There could be some serious fucking repercussions from this.
There seems to be a bug...(Firefox on linux)
They work fine on Firefox though.
Examples that others use: https://news.ycombinator.com/topcolors
https://github.com/minimaxir/hacker-news-undocumented
I fixed the biggest cause of confusion, the F word search is now fixed, and so getting a Gordon Ramsey should be a pretty rare occurrence.
But it looks cool and works well! Nice effort!
Edit: As it turns out you need to hover over the text to get an description, while the icon only shows "icon for x" on mouseover.
Addendum: Would it be possible to split karma into submit and comment karma? My gut says it'd not be possible as HN uses some kinda curve + only displays the combined karma on the profile
(I'm also distraught about some of the numbers. I know I have a HN problem, but seeing it laid out like that...)
I second the "hover text" issue other commenters brought up - it would be great if badge details were accessible in other ways (e.g. expanding hidden text or a popup when the user clicks a badge or its main text).
EDIT:
s/Call Me Anytime/Call Me Maybe/ for extra meme value?
If this gets popular I might have to try and find my first HN login - I want that OG badge. :-)
Which kind of is the point - I dont really want to worry about those sort of things - I am glad HN is staying off this course. But it is fun :-)
[0] https://hnbadges.netlify.app/?user=tbabej