WTF Is Botmetr? The First Platform That Rewards You for Hunting Chatbots | Hacker News Reader

WTF Is Botmetr? The First Platform That Rewards You for Hunting Chatbots

botmetr.com

9 pointsandyolivers23 comments

With all the best will in the world: your HN account is hours old, your only submission is promoting your thing, and even your title violates guidelines. I am not sure how much you think others should invest in your thing, but if it's 5-minutes reading about it on your website, maybe start by doing unto others?

https://news.ycombinator.com/newsguidelines.html

They like to call it "karma" around here, after all ;-)

EDIT: Following my own advice, added ;-) to reduce snark

So what sort of rewards do we get for reporting bots? I know a particular web host that uses a chat bot that was giving me outrageously bad answers.

Promo codes for now. We are thinking about creating a solana faucet in the near future

Welcome to Botmetr — the first platform that lets you compare, analyze, and review chatbots, all while earning rewards for your insights. Have you ever tried to get help from a chatbot and ended up screaming at your screen? You’re not alone. From missed questions to robotic loops that go nowhere, bad chatbot experiences are more common than ever — and it’s hurting customer support across industries. But here’s the good news: We’re flipping the script.

This kind of stock marketing script tends not to fly on HN. If you want general interest you need to go deeper than this.

But I already have your interest, at least enough to have your comment

Your comment is very obviously AI generated. You can't even be bothered to write a small blurb about your website (that is about chatbots, for crying out loud) by hand?

It really says all there is to say about this 'platform'. The creators don't care enough to put even the barest of minimum effort into promoting it, why would anyone else care?

I care enough to reply to your post. What do you know about putting effort into anything? Just because you're allowed to judge a piece of content, doesn't mean that you can judge how much effort a person puts on something. Would love to know how much effort you put on your work

How is it very obviously? Because the correct grammar? What if OP is not native English speaker and let AI proofread it?

It wasn't AI generated. If a correct description for you means that it was written by AI, I'm flattered

And the purpose is "gamifying" the whole chatbot hunt, not fact-checking or red-flagging AI-generating content btw

I have no idea what a "chatbot hunt" is and why there would be economic value in hunting them.

That's why you should check my website, you'd be better informed ;)

foiled by the old em-dasheroo!

I am annoyed that correct use of punctuation is now considered a red flag.

Like all punctuation: it's a choice.

One that has made it's way into LLMs because they've injested a lot of actual books that were professionally type-set.

Humans typing online don't tend to use em-dash; LLMs often make the choice to do so.

That makes it a very strong signal.

I use the em-dash and en-dash pretty often—as many academically-styled human authors do—and sometimes I worry my writing style makes my content seem like LLM output. I tend to use it more when I've been using a lot of parenthesis for lists or elaboration (eg. & ie.).

I've found the most telling sign isn't just the presence of the em-dash, but the fact that the em-dash is used for everything, regardless of wether a parentheses, comma, or semi-colon was more appropriate.

Often the em-dash is used "incorrectly" such as the example above where there is a single space on either side of the em-dash and sometimes it is used in place of the en-dash for number ranges too. I regularly have to fight things like ChatGPT to use other "advanced" punctuation in place of the em-dash.

Also line breaks, LLMs love markdown line breaks for some ungodly reason.

That's my pointless pontification for the day. :)

I'm seeing that most of you guys like to rant. This is just a place to review chatbots and customer support. The purpose is trying to improve CS itself and the UX of some websites we tend to use nowadays. Don't you get sick of trying to use chatbots that don't work? Are you not fed up of calling calling CS hotlines and spending loads of money on it?

Now I want to say this as polite as possible, but I think you might be under false impressions as to what this site is. It might look like "Reddit, but for IT", but many people here are very, very particular about protocol. You can disagree with people, in fact, it's good to have people disagree with what you are sharing because it allows you to explain and strengthen your position.

The "rant" is not a bug, it is a feature. The only reason I frequent this website is for the prolific pontification on random topics. There aren't many other places I can find such a high concentration of people with an above average intelligence sharing their internal thoughts and opinions in relatively neutral/objective forum.

I'm a lurker. I don't really post much, because I often don't have much of value to contribute that hasn't already be said. The most valuable aspect of HN is it's dicussions. I've found so many useful bits of info in the comments of seemingly unrelated articles.

> Don't you get sick of trying to use chatbots that don't work? Are you not fed up of calling calling CS hotlines and spending loads of money on it?

You aren't going to convince many people here with a call to action like this, especially after calling a relatively civil discussion between users to be "ranting". Show us why we should care based on the merits of the product at hand. If someone pokes holes in it, defend that feature or accept the criticism. Candidly, I don't even really know much about your product because your behavior here turned me off from even clicking the link.

The message I'm responding to reads as petulant or indignant, which is not a good look if you are attempting to cultivate an audience here. I see your previous response to my commentary on the topic of certain punctuation making human written text look similar to LLM output appears to have broken a rule and subsequently got removed, which is also concerning.

Overall, it may just be best to cut your losses today and stop arguing with every critique/criticism. Maybe a do a bit more research about your audience next time. I say this as someone who has been in your shoes before. Not trying to be a dick. Treat it as a learning experience.

I think you should take a good look at your comments, cause if I lack humility you are doing exactly the same and worse. I'm being called "petulant, indignant, arrogant" and so on... I don't recall calling you anything like that. I'd prefer people to criticize the features of a website rather than saying "this text was written by LLM, you can't even bother to write your own text". The feedback you are providing is not constructive, it's simply meant to troll a piece of text and red flag it as "AI-generated". Either way, I'm apologizing for my comment.

Reply on news.ycombinator.com